Auto upload 2026-08-09T07:04:11.258775 (part 3)

.gitattributes

@@ -33,3 +33,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+wandb/run-20260809_035819-cvzjg5ej/run-cvzjg5ej.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/run-20260809_050050-59pftr14/run-59pftr14.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/run-20260809_055344-aqwnomdl/run-aqwnomdl.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/run-20260809_055726-m1dnjnh6/run-m1dnjnh6.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/run-20260809_061951-oe9tdw54/run-oe9tdw54.wandb filter=lfs diff=lfs merge=lfs -text
+wandb/run-20260809_070213-uvqyddz0/run-uvqyddz0.wandb filter=lfs diff=lfs merge=lfs -text

outio/mlp-linear-9L_run/checkpoint-750/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ce4f81803861286fbc66ef29ca3a3f867174d8f7eaf0b4091acb991c9eaf7690
+size 1064

outio/mlp-linear-9L_run/checkpoint-750/tokenizer_config.json

@@ -0,0 +1,13 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "is_local": false,
+  "local_files_only": false,
+  "model_max_length": 1024,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

outio/mlp-linear-9L_run/checkpoint-750/trainer_state.json

@@ -0,0 +1,413 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0107853050219076,
+  "eval_steps": 50,
+  "global_step": 750,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.026963262554769128,
+      "grad_norm": 19.875,
+      "learning_rate": 0.0005,
+      "loss": 120.13565673828126,
+      "step": 20
+    },
+    {
+      "epoch": 0.053926525109538256,
+      "grad_norm": 24.625,
+      "learning_rate": 0.0005,
+      "loss": 102.05034790039062,
+      "step": 40
+    },
+    {
+      "epoch": 0.06740815638692282,
+      "eval_loss": 5.665860652923584,
+      "eval_runtime": 9.5524,
+      "eval_samples_per_second": 997.346,
+      "eval_steps_per_second": 12.562,
+      "step": 50
+    },
+    {
+      "epoch": 0.08088978766430738,
+      "grad_norm": 18.125,
+      "learning_rate": 0.0005,
+      "loss": 90.97202758789062,
+      "step": 60
+    },
+    {
+      "epoch": 0.10785305021907651,
+      "grad_norm": 40.75,
+      "learning_rate": 0.0005,
+      "loss": 83.10302734375,
+      "step": 80
+    },
+    {
+      "epoch": 0.13481631277384565,
+      "grad_norm": 29.625,
+      "learning_rate": 0.0005,
+      "loss": 77.51190185546875,
+      "step": 100
+    },
+    {
+      "epoch": 0.13481631277384565,
+      "eval_loss": 4.689935684204102,
+      "eval_runtime": 9.1111,
+      "eval_samples_per_second": 1045.652,
+      "eval_steps_per_second": 13.171,
+      "step": 100
+    },
+    {
+      "epoch": 0.16177957532861476,
+      "grad_norm": 19.5,
+      "learning_rate": 0.0005,
+      "loss": 73.29612426757812,
+      "step": 120
+    },
+    {
+      "epoch": 0.1887428378833839,
+      "grad_norm": 25.25,
+      "learning_rate": 0.0005,
+      "loss": 70.37640380859375,
+      "step": 140
+    },
+    {
+      "epoch": 0.20222446916076844,
+      "eval_loss": 4.263797760009766,
+      "eval_runtime": 9.4931,
+      "eval_samples_per_second": 1003.568,
+      "eval_steps_per_second": 12.641,
+      "step": 150
+    },
+    {
+      "epoch": 0.21570610043815303,
+      "grad_norm": 14.625,
+      "learning_rate": 0.0005,
+      "loss": 68.2427978515625,
+      "step": 160
+    },
+    {
+      "epoch": 0.24266936299292213,
+      "grad_norm": 13.0625,
+      "learning_rate": 0.0005,
+      "loss": 66.13766479492188,
+      "step": 180
+    },
+    {
+      "epoch": 0.2696326255476913,
+      "grad_norm": 18.5,
+      "learning_rate": 0.0005,
+      "loss": 63.81178588867188,
+      "step": 200
+    },
+    {
+      "epoch": 0.2696326255476913,
+      "eval_loss": 3.9343974590301514,
+      "eval_runtime": 9.3868,
+      "eval_samples_per_second": 1014.939,
+      "eval_steps_per_second": 12.784,
+      "step": 200
+    },
+    {
+      "epoch": 0.2965958881024604,
+      "grad_norm": 16.125,
+      "learning_rate": 0.0005,
+      "loss": 62.16768188476563,
+      "step": 220
+    },
+    {
+      "epoch": 0.3235591506572295,
+      "grad_norm": 21.25,
+      "learning_rate": 0.0005,
+      "loss": 61.1105712890625,
+      "step": 240
+    },
+    {
+      "epoch": 0.33704078193461406,
+      "eval_loss": 3.748317241668701,
+      "eval_runtime": 9.4777,
+      "eval_samples_per_second": 1005.204,
+      "eval_steps_per_second": 12.661,
+      "step": 250
+    },
+    {
+      "epoch": 0.3505224132119987,
+      "grad_norm": 42.25,
+      "learning_rate": 0.0005,
+      "loss": 60.05653686523438,
+      "step": 260
+    },
+    {
+      "epoch": 0.3774856757667678,
+      "grad_norm": 27.5,
+      "learning_rate": 0.0005,
+      "loss": 59.25806884765625,
+      "step": 280
+    },
+    {
+      "epoch": 0.4044489383215369,
+      "grad_norm": 14.375,
+      "learning_rate": 0.0005,
+      "loss": 58.334844970703124,
+      "step": 300
+    },
+    {
+      "epoch": 0.4044489383215369,
+      "eval_loss": 3.6137855052948,
+      "eval_runtime": 9.6654,
+      "eval_samples_per_second": 985.685,
+      "eval_steps_per_second": 12.415,
+      "step": 300
+    },
+    {
+      "epoch": 0.43141220087630605,
+      "grad_norm": 32.75,
+      "learning_rate": 0.0005,
+      "loss": 57.341180419921876,
+      "step": 320
+    },
+    {
+      "epoch": 0.45837546343107516,
+      "grad_norm": 12.5625,
+      "learning_rate": 0.0005,
+      "loss": 56.699591064453124,
+      "step": 340
+    },
+    {
+      "epoch": 0.4718570947084597,
+      "eval_loss": 3.492727518081665,
+      "eval_runtime": 9.4437,
+      "eval_samples_per_second": 1008.824,
+      "eval_steps_per_second": 12.707,
+      "step": 350
+    },
+    {
+      "epoch": 0.48533872598584427,
+      "grad_norm": 21.5,
+      "learning_rate": 0.0005,
+      "loss": 55.840753173828126,
+      "step": 360
+    },
+    {
+      "epoch": 0.5123019885406134,
+      "grad_norm": 14.5625,
+      "learning_rate": 0.0005,
+      "loss": 55.377203369140624,
+      "step": 380
+    },
+    {
+      "epoch": 0.5392652510953826,
+      "grad_norm": 14.6875,
+      "learning_rate": 0.0005,
+      "loss": 54.936090087890626,
+      "step": 400
+    },
+    {
+      "epoch": 0.5392652510953826,
+      "eval_loss": 3.41583251953125,
+      "eval_runtime": 9.1837,
+      "eval_samples_per_second": 1037.382,
+      "eval_steps_per_second": 13.067,
+      "step": 400
+    },
+    {
+      "epoch": 0.5662285136501517,
+      "grad_norm": 22.0,
+      "learning_rate": 0.0005,
+      "loss": 54.31259765625,
+      "step": 420
+    },
+    {
+      "epoch": 0.5931917762049208,
+      "grad_norm": 22.5,
+      "learning_rate": 0.0005,
+      "loss": 53.894476318359374,
+      "step": 440
+    },
+    {
+      "epoch": 0.6066734074823054,
+      "eval_loss": 3.3446545600891113,
+      "eval_runtime": 9.6871,
+      "eval_samples_per_second": 983.477,
+      "eval_steps_per_second": 12.388,
+      "step": 450
+    },
+    {
+      "epoch": 0.6201550387596899,
+      "grad_norm": 21.875,
+      "learning_rate": 0.0005,
+      "loss": 53.53094482421875,
+      "step": 460
+    },
+    {
+      "epoch": 0.647118301314459,
+      "grad_norm": 20.625,
+      "learning_rate": 0.0005,
+      "loss": 53.14410400390625,
+      "step": 480
+    },
+    {
+      "epoch": 0.6740815638692281,
+      "grad_norm": 19.75,
+      "learning_rate": 0.0005,
+      "loss": 52.86866455078125,
+      "step": 500
+    },
+    {
+      "epoch": 0.6740815638692281,
+      "eval_loss": 3.2967636585235596,
+      "eval_runtime": 9.5934,
+      "eval_samples_per_second": 993.08,
+      "eval_steps_per_second": 12.509,
+      "step": 500
+    },
+    {
+      "epoch": 0.7010448264239973,
+      "grad_norm": 31.0,
+      "learning_rate": 0.0005,
+      "loss": 52.4890625,
+      "step": 520
+    },
+    {
+      "epoch": 0.7280080889787665,
+      "grad_norm": 16.875,
+      "learning_rate": 0.0005,
+      "loss": 52.321435546875,
+      "step": 540
+    },
+    {
+      "epoch": 0.741489720256151,
+      "eval_loss": 3.250918388366699,
+      "eval_runtime": 9.1469,
+      "eval_samples_per_second": 1041.558,
+      "eval_steps_per_second": 13.119,
+      "step": 550
+    },
+    {
+      "epoch": 0.7549713515335356,
+      "grad_norm": 18.5,
+      "learning_rate": 0.0005,
+      "loss": 52.020050048828125,
+      "step": 560
+    },
+    {
+      "epoch": 0.7819346140883047,
+      "grad_norm": 23.875,
+      "learning_rate": 0.0005,
+      "loss": 51.72322387695313,
+      "step": 580
+    },
+    {
+      "epoch": 0.8088978766430738,
+      "grad_norm": 17.125,
+      "learning_rate": 0.0005,
+      "loss": 51.40700073242188,
+      "step": 600
+    },
+    {
+      "epoch": 0.8088978766430738,
+      "eval_loss": 3.21195650100708,
+      "eval_runtime": 9.4995,
+      "eval_samples_per_second": 1002.9,
+      "eval_steps_per_second": 12.632,
+      "step": 600
+    },
+    {
+      "epoch": 0.8358611391978429,
+      "grad_norm": 18.125,
+      "learning_rate": 0.0005,
+      "loss": 51.1299072265625,
+      "step": 620
+    },
+    {
+      "epoch": 0.8628244017526121,
+      "grad_norm": 18.875,
+      "learning_rate": 0.0005,
+      "loss": 50.86155700683594,
+      "step": 640
+    },
+    {
+      "epoch": 0.8763060330299967,
+      "eval_loss": 3.1778173446655273,
+      "eval_runtime": 9.5123,
+      "eval_samples_per_second": 1001.54,
+      "eval_steps_per_second": 12.615,
+      "step": 650
+    },
+    {
+      "epoch": 0.8897876643073812,
+      "grad_norm": 30.125,
+      "learning_rate": 0.0005,
+      "loss": 50.78846435546875,
+      "step": 660
+    },
+    {
+      "epoch": 0.9167509268621503,
+      "grad_norm": 23.0,
+      "learning_rate": 0.0005,
+      "loss": 50.50668640136719,
+      "step": 680
+    },
+    {
+      "epoch": 0.9437141894169194,
+      "grad_norm": 18.25,
+      "learning_rate": 0.0005,
+      "loss": 50.196780395507815,
+      "step": 700
+    },
+    {
+      "epoch": 0.9437141894169194,
+      "eval_loss": 3.135035991668701,
+      "eval_runtime": 9.3872,
+      "eval_samples_per_second": 1014.894,
+      "eval_steps_per_second": 12.783,
+      "step": 700
+    },
+    {
+      "epoch": 0.9706774519716885,
+      "grad_norm": 21.75,
+      "learning_rate": 0.0005,
+      "loss": 49.94974365234375,
+      "step": 720
+    },
+    {
+      "epoch": 0.9976407145264578,
+      "grad_norm": 25.125,
+      "learning_rate": 0.0005,
+      "loss": 49.74850769042969,
+      "step": 740
+    },
+    {
+      "epoch": 1.0107853050219076,
+      "eval_loss": 3.108103036880493,
+      "eval_runtime": 9.4956,
+      "eval_samples_per_second": 1003.31,
+      "eval_steps_per_second": 12.637,
+      "step": 750
+    }
+  ],
+  "logging_steps": 20,
+  "max_steps": 750,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 4354347480907776.0,
+  "train_batch_size": 80,
+  "trial_name": null,
+  "trial_params": null
+}

outio/mlp-linear-9L_run/checkpoint-750/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa9985a7c7f521fa2a6f7d16c7d254b6d5e243707cd393fa6ea6dfe9b9926147
+size 4920

outio/mlp-linear-9L_run/config.json

@@ -0,0 +1,35 @@
+{
+  "activation": "linear",
+  "architectures": [
+    "TinyLlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "dtype": "bfloat16",
+  "eos_token_id": 2,
+  "head_dim": 32,
+  "hidden_act": "silu",
+  "hidden_size": 128,
+  "initializer_range": 0.02,
+  "intermediate_size": 256,
+  "max_position_embeddings": 512,
+  "mlp_bias": false,
+  "mlp_type": "mlp",
+  "model_type": "tiny_llama",
+  "num_attention_heads": 4,
+  "num_hidden_layers": 9,
+  "num_key_value_heads": 4,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "rope_theta": 10000.0,
+    "rope_type": "default"
+  },
+  "tie_word_embeddings": true,
+  "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
+  "transformers_version": "5.15.0.dev0",
+  "use_cache": false,
+  "vocab_size": 4096
+}

outio/mlp-linear-9L_run/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b5e551dcb45402eb46f36dadcb4bb5f569e4a98cac7b8ef431d7311497cad45
+size 4010544

outio/mlp-linear-9L_run/tokenizer_config.json

@@ -0,0 +1,13 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "is_local": false,
+  "local_files_only": false,
+  "model_max_length": 1024,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

outio/mlp-linear-9L_run/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fa9985a7c7f521fa2a6f7d16c7d254b6d5e243707cd393fa6ea6dfe9b9926147
+size 4920

outio/sweep_summary.json

@@ -0,0 +1,8 @@
+[
+  {
+    "variant": "mlp-linear-9L",
+    "eval_loss": 3.108103036880493,
+    "out": "outio/mlp-linear-9L_run",
+    "run_name": "LM-mlp-linear-9L-2.0M-20260809-055343"
+  }
+]

sweep.py

@@ -0,0 +1,161 @@
+#!/usr/bin/env python3
+"""Sweep explicit GLU / MLP variants with identical data and hyperparameters."""
+import argparse
+import copy
+import json
+import re
+import time
+import os
+from pathlib import Path
+
+import yaml
+import wandb
+import torch
+from transformers import AutoTokenizer, set_seed
+from exp import TinyLlamaConfig, TinyLlamaForCausalLM, build_dataset, create_trainer
+
+
+def format_param_count(total_params: int) -> str:
+    """Return human‑readable string with M or B suffix, 1 decimal."""
+    if total_params >= 1e9:
+        return f"{total_params / 1e9:.1f}B"
+    else:
+        return f"{total_params / 1e6:.1f}M"
+
+
+def parse_variant(variant: str):
+    """
+    Parse variant string into (prefix, activation, layers).
+    Formats:
+        glu-silu        -> ('glu', 'silu', None)
+        mlp-s10-10L     -> ('mlp', 's10', 10)
+        glu-relu-8L     -> ('glu', 'relu', 8)
+    """
+    # Pattern: optional layers at end with 'L'
+    match = re.fullmatch(r'(glu|mlp)-([a-zA-Z0-9]+)(?:-(\d+)L)?', variant)
+    if not match:
+        raise ValueError(
+            f"Invalid variant format: '{variant}'. "
+            "Expected: <glu|mlp>-<activation>[-<layers>L] e.g. glu-silu-10L"
+        )
+    prefix, act, layers_str = match.groups()
+    layers = int(layers_str) if layers_str is not None else None
+    return prefix, act, layers
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--config", required=True, help="Base YAML config")
+    parser.add_argument(
+        "--variants",
+        nargs="+",
+        required=True,
+        help="List of variants: e.g. glu-silu-10L mlp-relu-8L"
+    )
+    parser.add_argument("--push", action="store_true")
+    args = parser.parse_args()
+
+    with open(args.config) as f:
+        base = yaml.safe_load(f)
+
+    seed = base.get("training", {}).get("seed", 42)
+    set_seed(seed)
+
+    tok_name = base["model"].get("tokenizer_name", "meta-llama/Llama-2-7b-hf")
+    tokenizer = AutoTokenizer.from_pretrained(tok_name)
+    if tokenizer.pad_token is None:
+        tokenizer.pad_token = tokenizer.eos_token
+
+    msl = base["model"].get("max_position_embeddings", 512)
+    train_ds = build_dataset(
+        tokenizer,
+        max_seq_len=msl,
+        split="train",
+        max_samples=None   # limit to 50k samples
+    )
+    eval_ds = build_dataset(
+        tokenizer,
+        max_seq_len=msl,
+        split="validation",
+        max_samples=None   # no limit for eval
+    )
+
+    results = []
+
+    for variant in args.variants:
+        prefix, act, layers = parse_variant(variant)
+
+        # Validation: mlp-situglu and mlp-waleed are banned
+        if prefix == "mlp" and act in ("situglu", "waleed"):
+            raise ValueError(
+                f"Activation '{act}' requires a gated architecture (GLU). "
+                f"Please use 'glu-{act}' instead."
+            )
+
+        # Build config overrides
+        cfg = copy.deepcopy(base)
+        cfg["model"]["mlp_type"] = prefix
+        cfg["model"]["activation"] = act
+        if layers is not None:
+            cfg["model"]["num_hidden_layers"] = layers
+
+        # Create a descriptive suffix for folders / run names
+        layer_suffix = f"-{layers}L" if layers is not None else ""
+        variant_label = f"{prefix}-{act}{layer_suffix}"
+
+        # Unique output directory
+        out_dir = Path(cfg["training"]["output_dir"]).parent / f"{variant_label}_run"
+        cfg["training"]["output_dir"] = str(out_dir)
+
+        # Re-seed for reproducibility across variants
+        set_seed(seed)
+
+        print(f"\n{'='*60}\n>>> Variant: {variant_label} | Out: {out_dir}\n{'='*60}")
+
+        # Instantiate model – no explicit _attn_implementation (use default)
+        config = TinyLlamaConfig(**cfg["model"])
+        model = TinyLlamaForCausalLM(config)
+        # Cast to bfloat16 if desired (we keep the same as before)
+        model = model.to(torch.bfloat16)
+
+        total_params = sum(p.numel() for p in model.parameters())
+        param_str = format_param_count(total_params)
+        timestamp = time.strftime("%Y%m%d-%H%M%S")
+        run_name = f"LM-{variant_label}-{param_str}-{timestamp}"
+        cfg["training"]["run_name"] = run_name
+
+        # Also update hub_model_id to include variant and layers
+        hub_id_base = cfg["training"].get("hub_model_id", "tiny-llama-lab")
+        cfg["training"]["hub_model_id"] = f"{hub_id_base}-{variant_label}"
+
+        # Ensure a fresh WandB run – remove any global WANDB_RUN_ID
+        os.environ.pop("WANDB_RUN_ID", None)
+
+        trainer = create_trainer(model, tokenizer, cfg, train_ds, eval_ds)
+
+        try:
+            trainer.train()
+            metrics = trainer.evaluate()
+            results.append({
+                "variant": variant_label,
+                "eval_loss": metrics.get("eval_loss"),
+                "out": str(out_dir),
+                "run_name": run_name,
+            })
+            trainer.save_model(str(out_dir))
+            if args.push or cfg["training"].get("push_to_hub", False):
+                trainer.push_to_hub()
+        finally:
+            # Explicitly finish WandB run to avoid re‑using the same run
+            wandb.finish()
+
+    # Save summary
+    summary = Path(base["training"]["output_dir"]).parent / "sweep_summary.json"
+    summary.write_text(json.dumps(results, indent=2))
+    print("\nSweep complete:")
+    for r in results:
+        print(f"  {r['variant']:20s}  eval_loss={r['eval_loss']:.4f}")
+
+
+if __name__ == "__main__":
+    main()

train.py

@@ -0,0 +1,70 @@
+#!/usr/bin/env python3
+"""Train one TinyLlama variant from a YAML config."""
+import argparse
+import yaml
+import torch
+
+from transformers import AutoTokenizer, set_seed
+from exp import TinyLlamaConfig, TinyLlamaForCausalLM, build_dataset, create_trainer
+
+
+def main():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--config", required=True, help="Path to YAML config")
+    parser.add_argument("--push", action="store_true", help="Push final model to HF Hub")
+    args = parser.parse_args()
+
+    with open(args.config) as f:
+        cfg = yaml.safe_load(f)
+
+    # Explicit seed before any randomness
+    seed = cfg.get("training", {}).get("seed", 42)
+    set_seed(seed)
+
+    model_cfg = cfg["model"]
+    train_cfg = cfg.get("training", {})
+
+    # Tokenizer
+    tok_name = model_cfg.pop("tokenizer_name", "meta-llama/Llama-2-7b-hf")
+    tokenizer = AutoTokenizer.from_pretrained(tok_name)
+    if tokenizer.pad_token is None:
+        tokenizer.pad_token = tokenizer.eos_token
+
+    # Model – the config must contain mlp_type and activation
+    tiny_config = TinyLlamaConfig(**model_cfg)
+    # No explicit attention implementation – let transformers pick the default
+    model = TinyLlamaForCausalLM(tiny_config)
+    model = model.to(torch.bfloat16)
+
+    n_params = sum(p.numel() for p in model.parameters()) / 1e6
+    print(f"Model: {n_params:.2f}M params | MLP type: {tiny_config.mlp_type} | Activation: {tiny_config.activation}")
+
+    # Data
+    msl = model_cfg.get("max_position_embeddings", 512)
+    train_ds = build_dataset(
+        tokenizer,
+        max_seq_len=msl,
+        split="train",
+        max_samples=None   # limit to 50k samples
+    )
+    eval_ds = build_dataset(
+        tokenizer,
+        max_seq_len=msl,
+        split="validation",
+        max_samples=None   # no limit for eval
+    )
+
+    # Train
+    trainer = create_trainer(model, tokenizer, cfg, train_ds, eval_ds)
+    trainer.train()
+
+    # Save & push
+    out = train_cfg.get("output_dir", "./out")
+    trainer.save_model(out)
+    if args.push or train_cfg.get("push_to_hub", False):
+        trainer.push_to_hub()
+    print(f"Done. Artifacts in {out}")
+
+
+if __name__ == "__main__":
+    main()

wandb/debug-internal.log

@@ -0,0 +1,40 @@
+{"time":"2026-08-09T07:02:13.954953014Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-09T07:02:13.955105829Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-09T07:02:14.215123169Z","level":"INFO","msg":"stream: created new stream","id":"uvqyddz0"}
+{"time":"2026-08-09T07:02:14.215184619Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-09T07:02:14.215266968Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-09T07:02:14.215275401Z","level":"INFO","msg":"writer: started","stream_id":"uvqyddz0"}
+{"time":"2026-08-09T07:02:14.215296287Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-09T07:02:15.114109279Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-09T07:02:15.211087322Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T07:02:30.114358029Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":0,"events_lines":2,"console_offset":1,"console_lines":4,"uploaded_len":2}
+{"time":"2026-08-09T07:02:30.230090173Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T07:02:45.11459263Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":2,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T07:02:45.228934504Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T07:02:56.598899602Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":737}
+{"time":"2026-08-09T07:02:56.598941378Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":1}
+{"time":"2026-08-09T07:02:56.606194892Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":3095}
+{"time":"2026-08-09T07:02:56.606379301Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":14}
+{"time":"2026-08-09T07:02:56.610759144Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":4507}
+{"time":"2026-08-09T07:02:56.610912588Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":13}
+{"time":"2026-08-09T07:02:56.613175719Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":5301}
+{"time":"2026-08-09T07:02:56.618404048Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":1932}
+{"time":"2026-08-09T07:02:56.629287315Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":10255}
+{"time":"2026-08-09T07:02:56.629322845Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":1}
+{"time":"2026-08-09T07:02:56.63322527Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":11868}
+{"time":"2026-08-09T07:02:56.633338533Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":10}
+{"time":"2026-08-09T07:02:56.636529541Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":13176}
+{"time":"2026-08-09T07:02:56.63797994Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":477}
+{"time":"2026-08-09T07:02:56.643113777Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":14979}
+{"time":"2026-08-09T07:02:56.643188638Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":12}
+{"time":"2026-08-09T07:03:00.155597928Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":4,"events_lines":2,"console_offset":4,"console_lines":2}
+{"time":"2026-08-09T07:03:01.144833138Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T07:03:15.114830048Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":6,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T07:03:15.243621253Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T07:03:30.114703175Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":8,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T07:03:30.219274044Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T07:03:40.29903607Z","level":"ERROR","msg":"runupserter: failed to upload changes","error":"POST https://api.wandb.ai/graphql giving up after 1 attempt(s): context canceled"}
+{"time":"2026-08-09T07:03:40.30029433Z","level":"ERROR","msg":"runfiles: CreateRunFiles returned error: POST https://api.wandb.ai/graphql giving up after 1 attempt(s): context canceled"}
+{"time":"2026-08-09T07:03:40.300493766Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-09T07:03:40.322418888Z","level":"INFO","msg":"filestream: sending request","total_files":3,"history_offset":1,"history_lines":1,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T07:03:40.3225757Z","level":"ERROR+4","msg":"filestream: fatal error: filestream: error making HTTP request: POST https://api.wandb.ai/files/deepnevro-deepnevro/huggingface/uvqyddz0/file_stream giving up after 1 attempt(s): context canceled. got response: <nil>"}

wandb/debug.log

@@ -0,0 +1,27 @@
+2026-08-09 07:02:13,953 INFO    MainThread:474098 [wandb_setup.py:_flush():81] Current SDK version is 0.28.1
+2026-08-09 07:02:13,953 INFO    MainThread:474098 [wandb_setup.py:_flush():81] Configure stats pid to 474098
+2026-08-09 07:02:13,953 INFO    MainThread:474098 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-08-09 07:02:13,953 INFO    MainThread:474098 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/Activation/wandb/run-20260809_070213-uvqyddz0/logs/debug.log
+2026-08-09 07:02:13,953 INFO    MainThread:474098 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/Activation/wandb/run-20260809_070213-uvqyddz0/logs/debug-internal.log
+2026-08-09 07:02:13,953 INFO    MainThread:474098 [wandb_init.py:init():772] calling init triggers
+2026-08-09 07:02:13,953 INFO    MainThread:474098 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+config: {'_wandb': {}}
+2026-08-09 07:02:13,953 INFO    MainThread:474098 [wandb_init.py:init():820] starting backend
+2026-08-09 07:02:13,953 INFO    MainThread:474098 [wandb_init.py:init():826] Connected to an existing wandb-core service via WANDB_SERVICE
+2026-08-09 07:02:13,953 INFO    MainThread:474098 [wandb_init.py:init():835] sending inform_init request
+2026-08-09 07:02:14,215 INFO    MainThread:474098 [wandb_init.py:init():840] backend started and connected
+2026-08-09 07:02:14,218 INFO    MainThread:474098 [wandb_init.py:init():910] updated telemetry
+2026-08-09 07:02:14,225 INFO    MainThread:474098 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-09 07:02:14,452 INFO    MainThread:474098 [wandb_init.py:init():978] starting run threads in backend
+2026-08-09 07:02:14,525 INFO    MainThread:474098 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-09 07:02:14,525 INFO    MainThread:474098 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-09 07:02:14,525 INFO    MainThread:474098 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-09 07:02:14,525 INFO    MainThread:474098 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-09 07:02:14,528 INFO    MainThread:474098 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-09 07:02:14,529 INFO    MainThread:474098 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.15.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 94, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'glu', 'activation': 'tanh', 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/glu-tanh-94L_run', 'per_device_train_batch_size': 128, 'num_train_epochs': 1, 'max_steps': 1500, 'learning_rate': 0.001, 'lr_scheduler_type': 'constant', 'lr_scheduler_kwargs': None, 'warmup_steps': 0, 'optim': 'adamw_torch_fused', 'optim_args': None, 'weight_decay': 0.01, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 4, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-glu-tanh-94L-15.9M-20260809-070212', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 50, 'eval_delay': 0, 'per_device_eval_batch_size': 128, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': True, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/A-glu-tanh-94L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
+2026-08-09 07:02:14,532 INFO    MainThread:474098 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 15949440 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x14c4299b9450>>
+2026-08-09 07:02:14,532 INFO    MainThread:474098 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 15949440 None
+2026-08-09 07:03:39,932 INFO    MainThread:474098 [wandb_run.py:_finish():2383] finishing run deepnevro-deepnevro/huggingface/uvqyddz0
+2026-08-09 07:03:39,933 INFO    MainThread:474098 [wandb_run.py:_atexit_cleanup():2588] got exitcode: 0
+2026-08-09 07:03:39,933 INFO    MainThread:474098 [wandb_run.py:_restore():2570] restore
+2026-08-09 07:03:39,933 INFO    MainThread:474098 [wandb_run.py:_restore():2576] restore done

wandb/run-20260809_035819-cvzjg5ej/files/config.yaml

@@ -0,0 +1,433 @@
+_name_or_path:
+    value: ""
+_wandb:
+    value:
+        cli_version: 0.28.1
+        e:
+            oz75ofyjp8ovu1ftttzp3wyx4l4oquya:
+                args:
+                    - --config
+                    - configs/baseline.yaml
+                    - --variants
+                    - mlp-s10-94L
+                    - --push
+                codePath: sweep.py
+                codePathLocal: sweep.py
+                cpu_count: 112
+                cpu_count_logical: 224
+                cudaVersion: "12.4"
+                disk:
+                    /:
+                        total: "1560765693952"
+                        used: "708205846528"
+                email: deepnevro@gmail.com
+                executable: /mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python
+                git:
+                    commit: 34b8d2e8f9a0c5751333310e69fa0c1056381deb
+                    remote: https://github.com/deepnevro/Activation.git
+                gpu: NVIDIA H100 80GB HBM3
+                gpu_count: 8
+                gpu_nvidia:
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-39c684a5-fde6-83d7-1663-0859795881ae
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-132944c4-b689-2b5f-89a4-d730401677ab
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bc6c3e3c-9b90-09ca-c034-774961847c54
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea
+                host: deeplens-k3s-node1
+                memory:
+                    total: "2164089937920"
+                os: Linux-5.15.0-126-generic-x86_64-with-glibc2.35
+                program: /mnt/data/zainulabideen/zain-exp/notebooks/Activation/sweep.py
+                python: CPython 3.11.15
+                root: /mnt/data/zainulabideen/zain-exp/notebooks/Activation
+                startedAt: "2026-08-09T03:58:19.280139Z"
+                writerId: oz75ofyjp8ovu1ftttzp3wyx4l4oquya
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.11.15
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 19
+                - 41
+                - 62
+                - 66
+            "4": 3.11.15
+            "5": 0.28.1
+            "6": 5.15.0.dev0
+            "9":
+                "1": transformers_trainer
+            "12": 0.28.1
+            "13": linux-x86_64
+accelerator_config:
+    value:
+        dispatch_batches: null
+        even_batches: true
+        gradient_accumulation_kwargs: null
+        non_blocking: false
+        split_batches: false
+        use_seedable_sampler: true
+activation:
+    value: s10
+adam_beta1:
+    value: 0.9
+adam_beta2:
+    value: 0.999
+adam_epsilon:
+    value: 1e-08
+architectures:
+    value: null
+attention_bias:
+    value: false
+attention_dropout:
+    value: 0
+auto_find_batch_size:
+    value: false
+average_tokens_across_devices:
+    value: true
+batch_eval_metrics:
+    value: false
+bf16:
+    value: true
+bf16_full_eval:
+    value: false
+bos_token_id:
+    value: 1
+chunk_size_feed_forward:
+    value: 0
+data_seed:
+    value: 42
+dataloader_drop_last:
+    value: false
+dataloader_in_order:
+    value: true
+dataloader_multiprocessing_context:
+    value: null
+dataloader_num_workers:
+    value: 0
+dataloader_persistent_workers:
+    value: false
+dataloader_pin_memory:
+    value: true
+dataloader_prefetch_factor:
+    value: null
+ddp_backend:
+    value: null
+ddp_broadcast_buffers:
+    value: null
+ddp_bucket_cap_mb:
+    value: null
+ddp_find_unused_parameters:
+    value: null
+ddp_static_graph:
+    value: null
+ddp_timeout:
+    value: 1800
+debug:
+    value: []
+deepspeed:
+    value: null
+disable_tqdm:
+    value: false
+do_eval:
+    value: true
+do_predict:
+    value: false
+do_train:
+    value: false
+dtype:
+    value: null
+enable_jit_checkpoint:
+    value: false
+eos_token_id:
+    value: 2
+eval_accumulation_steps:
+    value: null
+eval_delay:
+    value: 0
+eval_do_concat_batches:
+    value: true
+eval_on_start:
+    value: false
+eval_steps:
+    value: 50
+eval_strategy:
+    value: steps
+eval_use_gather_object:
+    value: false
+fp16:
+    value: false
+fp16_full_eval:
+    value: false
+fsdp:
+    value: null
+fsdp_config:
+    value: null
+full_determinism:
+    value: false
+gradient_accumulation_steps:
+    value: 4
+gradient_checkpointing:
+    value: false
+gradient_checkpointing_kwargs:
+    value: null
+greater_is_better:
+    value: null
+head_dim:
+    value: 32
+hidden_act:
+    value: silu
+hidden_size:
+    value: 128
+hub_always_push:
+    value: false
+hub_model_id:
+    value: w-ahmad/A-mlp-s10-94L
+hub_private_repo:
+    value: null
+hub_revision:
+    value: null
+hub_strategy:
+    value: every_save
+hub_token:
+    value: <HUB_TOKEN>
+id2label:
+    value:
+        "0": LABEL_0
+        "1": LABEL_1
+ignore_data_skip:
+    value: false
+include_for_metrics:
+    value: []
+include_num_input_tokens_seen:
+    value: "no"
+initializer_range:
+    value: 0.02
+intermediate_size:
+    value: 256
+is_encoder_decoder:
+    value: false
+label_names:
+    value: null
+label_smoothing_factor:
+    value: 0
+label2id:
+    value:
+        LABEL_0: 0
+        LABEL_1: 1
+learning_rate:
+    value: 0.001
+length_column_name:
+    value: length
+liger_kernel_config:
+    value: null
+load_best_model_at_end:
+    value: false
+local_rank:
+    value: -1
+log_level:
+    value: passive
+log_level_replica:
+    value: warning
+log_on_each_node:
+    value: true
+logging_first_step:
+    value: false
+logging_nan_inf_filter:
+    value: true
+logging_steps:
+    value: 20
+logging_strategy:
+    value: steps
+lr_scheduler_kwargs:
+    value: null
+lr_scheduler_type:
+    value: constant
+max_grad_norm:
+    value: 1
+max_position_embeddings:
+    value: 512
+max_steps:
+    value: 1500
+metric_for_best_model:
+    value: null
+mlp_bias:
+    value: false
+mlp_type:
+    value: mlp
+model/num_parameters:
+    value: 15949440
+model_type:
+    value: tiny_llama
+neftune_noise_alpha:
+    value: null
+num_attention_heads:
+    value: 4
+num_hidden_layers:
+    value: 94
+num_key_value_heads:
+    value: 4
+num_train_epochs:
+    value: 1
+optim:
+    value: adamw_torch_fused
+optim_args:
+    value: null
+optim_target_modules:
+    value: null
+output_attentions:
+    value: false
+output_dir:
+    value: out/mlp-s10-94L_run
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: 0
+parallelism_config:
+    value: null
+per_device_eval_batch_size:
+    value: 128
+per_device_train_batch_size:
+    value: 128
+prediction_loss_only:
+    value: false
+pretraining_tp:
+    value: 1
+problem_type:
+    value: null
+project:
+    value: huggingface
+push_to_hub:
+    value: true
+remove_unused_columns:
+    value: false
+report_to:
+    value:
+        - wandb
+restore_callback_states_from_checkpoint:
+    value: false
+resume_from_checkpoint:
+    value: null
+return_dict:
+    value: true
+rms_norm_eps:
+    value: 1e-06
+rope_parameters:
+    value:
+        rope_theta: 10000
+        rope_type: default
+run_name:
+    value: LM-mlp-s10-94L-15.9M-20260809-035818
+save_on_each_node:
+    value: false
+save_only_model:
+    value: false
+save_steps:
+    value: 100
+save_strategy:
+    value: steps
+save_total_limit:
+    value: null
+seed:
+    value: 42
+skip_memory_metrics:
+    value: true
+tf32:
+    value: null
+tie_word_embeddings:
+    value: true
+tokenizer_name:
+    value: w-ahmad/tiny-stories-tokenizer
+torch_compile:
+    value: false
+torch_compile_backend:
+    value: null
+torch_compile_mode:
+    value: null
+torch_empty_cache_steps:
+    value: null
+trackio_bucket_id:
+    value: null
+trackio_space_id:
+    value: null
+trackio_static_space_id:
+    value: null
+train_sampling_strategy:
+    value: random
+transformers_version:
+    value: 5.15.0.dev0
+use_cache:
+    value: false
+use_cpu:
+    value: false
+use_liger_kernel:
+    value: false
+vocab_size:
+    value: 4096
+warmup_steps:
+    value: 0
+weight_decay:
+    value: 0.01

wandb/run-20260809_035819-cvzjg5ej/files/output.log

@@ -0,0 +1,221 @@
+[transformers] `use_return_dict` is deprecated! Use `return_dict` instead!
+[INFO] Causal mask (float with -inf) applied to all attention layers.
+/mnt/data/zainulabideen/zain-exp/notebooks/Activation/exp.py:487: UserWarning: std(): degrees of freedom is <= 0. Correction should be strictly less than the reduction factor (input numel divided by output numel). (Triggered internally at /pytorch/aten/src/ATen/native/ReduceOps.cpp:1831.)
+  "std": tensor.std().item(),
+  7%|██▌                                    | 100/1500 [03:58<47:48,  2.05s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '28.57', 'grad_norm': '3.75', 'learning_rate': '0.001', 'epoch': '0.01078', 'train/total_time_seconds': '35.3', 'train/time_per_step_avg': '1.765', 'train/epoch_time_elapsed': '43.74', 'train/estimated_remaining_minutes': '43.53', 'train/global/act/norm': '8.829e+04', 'train/global/act/mean': '0.01651', 'train/global/act/std': '0.4295', 'train/global/act/max_abs': '8.337', 'train/global/act/frac_near_dtype_limit': '0', 'train/global/act/frac_near_user_limit': '0', 'train/global/grad/norm': '5.516', 'train/global/grad/mean': '-3.54e-06', 'train/global/grad/std': '0.0006905', 'train/global/grad/max_abs': '0.1084', 'train/global/grad/frac_near_dtype_limit': '0', 'train/global/grad/frac_near_user_limit': '0', 'train/global/param/norm': '174.8', 'train/global/param/mean': '0.001534', 'train/global/param/std': '0.04376', 'train/global/param/max_abs': '1', 'train/global/param/frac_near_dtype_limit': '0', 'train/global/param/frac_near_user_limit': '0', 'train/layer_model_layers_27/act/norm': '8942', 'train/layer_model_layers_27/act/mean': '0.018', 'train/layer_model_layers_27/act/std': '0.4279', 'train/layer_model_layers_27/act/max_abs': '5.094', 'train/layer_model_layers_27/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_27/act/frac_near_user_limit': '0', 'train/layer_model_layers_27/grad/norm': '0.3325', 'train/layer_model_layers_27/grad/mean': '-1.208e-05', 'train/layer_model_layers_27/grad/std': '0.0004102', 'train/layer_model_layers_27/grad/max_abs': '0.008545', 'train/layer_model_layers_27/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_27/grad/frac_near_user_limit': '0', 'train/layer_model_layers_87/act/norm': '9222', 'train/layer_model_layers_87/act/mean': '0.01927', 'train/layer_model_layers_87/act/std': '0.4411', 'train/layer_model_layers_87/act/max_abs': '4.281', 'train/layer_model_layers_87/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_87/act/frac_near_user_limit': '0', 'train/layer_model_layers_87/grad/norm': '0.2037', 'train/layer_model_layers_87/grad/mean': '-1.154e-06', 'train/layer_model_layers_87/grad/std': '0.0002515', 'train/layer_model_layers_87/grad/max_abs': '0.005646', 'train/layer_model_layers_87/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_87/grad/frac_near_user_limit': '0', 'train/layer_model_layers_53/act/norm': '9063', 'train/layer_model_layers_53/act/mean': '0.02554', 'train/layer_model_layers_53/act/std': '0.4333', 'train/layer_model_layers_53/act/max_abs': '4.562', 'train/layer_model_layers_53/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_53/act/frac_near_user_limit': '0', 'train/layer_model_layers_53/grad/norm': '0.2643', 'train/layer_model_layers_53/grad/mean': '-7.417e-07', 'train/layer_model_layers_53/grad/std': '0.0003263', 'train/layer_model_layers_53/grad/max_abs': '0.007202', 'train/layer_model_layers_53/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_53/grad/frac_near_user_limit': '0', 'train/layer__model_layers_31/param/norm': '17.93', 'train/layer__model_layers_31/param/mean': '0.001673', 'train/layer__model_layers_31/param/std': '0.04423', 'train/layer__model_layers_31/param/max_abs': '1', 'train/layer__model_layers_31/param/frac_near_dtype_limit': '0', 'train/layer__model_layers_31/param/frac_near_user_limit': '0', 'train/layer_model_layers_24/act/norm': '8922', 'train/layer_model_layers_24/act/mean': '0.01278', 'train/layer_model_layers_24/act/std': '0.4268', 'train/layer_model_layers_24/act/max_abs': '5', 'train/layer_model_layers_24/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_24/act/frac_near_user_limit': '0', 'train/layer_model_layers_24/grad/norm': '0.3614', 'train/layer_model_layers_24/grad/mean': '-1.145e-05', 'train/layer_model_layers_24/grad/std': '0.000446', 'train/layer_model_layers_24/grad/max_abs': '0.007385', 'train/layer_model_layers_24/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_24/grad/frac_near_user_limit': '0', 'train/layer_model_layers_39/act/norm': '8998', 'train/layer_model_layers_39/act/mean': '0.02845', 'train/layer_model_layers_39/act/std': '0.43
+{'loss': '24.28', 'grad_norm': '0.2637', 'learning_rate': '0.001', 'epoch': '0.02157', 'train/total_time_seconds': '67.59', 'train/time_per_step_avg': '1.69', 'train/epoch_time_elapsed': '84.33', 'train/estimated_remaining_minutes': '41.11'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '5.909', 'eval_runtime': '16.38', 'eval_samples_per_second': '581.5', 'eval_steps_per_second': '4.578', 'epoch': '0.02696', 'train/total_time_seconds': '83.86', 'train/time_per_step_avg': '1.677', 'train/epoch_time_elapsed': '120.9', 'train/estimated_remaining_minutes': '40.53'}
+{'loss': '23.61', 'grad_norm': '2.031', 'learning_rate': '0.001', 'epoch': '0.03235', 'train/total_time_seconds': '99.98', 'train/time_per_step_avg': '1.666', 'train/epoch_time_elapsed': '140.8', 'train/estimated_remaining_minutes': '39.99'}
+{'loss': '22.8', 'grad_norm': '3.938', 'learning_rate': '0.001', 'epoch': '0.04314', 'train/total_time_seconds': '132.2', 'train/time_per_step_avg': '1.653', 'train/epoch_time_elapsed': '180.9', 'train/estimated_remaining_minutes': '39.12'}
+{'loss': '22.14', 'grad_norm': '4.5', 'learning_rate': '0.001', 'epoch': '0.05392', 'train/total_time_seconds': '164.8', 'train/time_per_step_avg': '1.648', 'train/epoch_time_elapsed': '221.1', 'train/estimated_remaining_minutes': '38.45'}
+{'eval_loss': '5.443', 'eval_runtime': '16.35', 'eval_samples_per_second': '582.7', 'eval_steps_per_second': '4.587', 'epoch': '0.05392', 'train/total_time_seconds': '164.8', 'train/time_per_step_avg': '1.648', 'train/epoch_time_elapsed': '237.4', 'train/estimated_remaining_minutes': '38.45'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 17.57it/s]
+ 13%|█████▏                                 | 200/1500 [07:51<43:21,  2.00s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '21.52', 'grad_norm': '9.062', 'learning_rate': '0.001', 'epoch': '0.06471', 'train/total_time_seconds': '197', 'train/time_per_step_avg': '1.617', 'train/epoch_time_elapsed': '277.5', 'train/estimated_remaining_minutes': '37.76'}
+{'loss': '20.92', 'grad_norm': '8.062', 'learning_rate': '0.001', 'epoch': '0.07549', 'train/total_time_seconds': '229.2', 'train/time_per_step_avg': '1.616', 'train/epoch_time_elapsed': '317.3', 'train/estimated_remaining_minutes': '37.11'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '5.031', 'eval_runtime': '16.31', 'eval_samples_per_second': '584.1', 'eval_steps_per_second': '4.598', 'epoch': '0.08088', 'train/total_time_seconds': '245.3', 'train/time_per_step_avg': '1.615', 'train/epoch_time_elapsed': '353.5', 'train/estimated_remaining_minutes': '36.8'}
+{'loss': '20.13', 'grad_norm': '3.828', 'learning_rate': '0.001', 'epoch': '0.08628', 'train/total_time_seconds': '261.5', 'train/time_per_step_avg': '1.615', 'train/epoch_time_elapsed': '373.7', 'train/estimated_remaining_minutes': '36.5'}
+{'loss': '19.28', 'grad_norm': '5.875', 'learning_rate': '0.001', 'epoch': '0.09706', 'train/total_time_seconds': '293.8', 'train/time_per_step_avg': '1.616', 'train/epoch_time_elapsed': '413.8', 'train/estimated_remaining_minutes': '35.91'}
+{'loss': '18.32', 'grad_norm': '3.797', 'learning_rate': '0.001', 'epoch': '0.1078', 'train/total_time_seconds': '326.1', 'train/time_per_step_avg': '1.613', 'train/epoch_time_elapsed': '454', 'train/estimated_remaining_minutes': '35.33'}
+{'eval_loss': '4.459', 'eval_runtime': '16.53', 'eval_samples_per_second': '576.4', 'eval_steps_per_second': '4.538', 'epoch': '0.1078', 'train/total_time_seconds': '326.1', 'train/time_per_step_avg': '1.613', 'train/epoch_time_elapsed': '470.6', 'train/estimated_remaining_minutes': '35.33'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 18.81it/s]
+ 20%|███████▊                               | 300/1500 [11:45<40:08,  2.01s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '17.48', 'grad_norm': '3.125', 'learning_rate': '0.001', 'epoch': '0.1186', 'train/total_time_seconds': '358.7', 'train/time_per_step_avg': '1.617', 'train/epoch_time_elapsed': '511.2', 'train/estimated_remaining_minutes': '34.78'}
+{'loss': '16.86', 'grad_norm': '3.781', 'learning_rate': '0.001', 'epoch': '0.1294', 'train/total_time_seconds': '391', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '551.3', 'train/estimated_remaining_minutes': '34.22'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '4.038', 'eval_runtime': '16.74', 'eval_samples_per_second': '569', 'eval_steps_per_second': '4.479', 'epoch': '0.1348', 'train/total_time_seconds': '407.2', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '588.1', 'train/estimated_remaining_minutes': '33.93'}
+{'loss': '16.16', 'grad_norm': '4.812', 'learning_rate': '0.001', 'epoch': '0.1402', 'train/total_time_seconds': '423.4', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '608.2', 'train/estimated_remaining_minutes': '33.65'}
+{'loss': '15.61', 'grad_norm': '2.453', 'learning_rate': '0.001', 'epoch': '0.151', 'train/total_time_seconds': '455.8', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '648.3', 'train/estimated_remaining_minutes': '33.1'}
+{'loss': '15.26', 'grad_norm': '5.031', 'learning_rate': '0.001', 'epoch': '0.1618', 'train/total_time_seconds': '488.1', 'train/time_per_step_avg': '1.62', 'train/epoch_time_elapsed': '688.5', 'train/estimated_remaining_minutes': '32.54'}
+{'eval_loss': '3.779', 'eval_runtime': '16.48', 'eval_samples_per_second': '578', 'eval_steps_per_second': '4.55', 'epoch': '0.1618', 'train/total_time_seconds': '488.1', 'train/time_per_step_avg': '1.62', 'train/epoch_time_elapsed': '705', 'train/estimated_remaining_minutes': '32.54'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 15.70it/s]
+ 27%|██████████▍                            | 400/1500 [15:39<36:40,  2.00s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '14.86', 'grad_norm': '2.531', 'learning_rate': '0.001', 'epoch': '0.1726', 'train/total_time_seconds': '520.4', 'train/time_per_step_avg': '1.617', 'train/epoch_time_elapsed': '745.3', 'train/estimated_remaining_minutes': '31.98'}
+{'loss': '14.42', 'grad_norm': '1.883', 'learning_rate': '0.001', 'epoch': '0.1833', 'train/total_time_seconds': '553', 'train/time_per_step_avg': '1.62', 'train/epoch_time_elapsed': '785.7', 'train/estimated_remaining_minutes': '31.44'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '3.509', 'eval_runtime': '16.51', 'eval_samples_per_second': '577.1', 'eval_steps_per_second': '4.543', 'epoch': '0.1887', 'train/total_time_seconds': '569.2', 'train/time_per_step_avg': '1.62', 'train/epoch_time_elapsed': '822.2', 'train/estimated_remaining_minutes': '31.17'}
+{'loss': '14.03', 'grad_norm': '2.219', 'learning_rate': '0.001', 'epoch': '0.1941', 'train/total_time_seconds': '585.3', 'train/time_per_step_avg': '1.62', 'train/epoch_time_elapsed': '842.4', 'train/estimated_remaining_minutes': '30.89'}
+{'loss': '13.66', 'grad_norm': '1.617', 'learning_rate': '0.001', 'epoch': '0.2049', 'train/total_time_seconds': '617.7', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '882.4', 'train/estimated_remaining_minutes': '30.34'}
+{'loss': '13.35', 'grad_norm': '1.086', 'learning_rate': '0.001', 'epoch': '0.2157', 'train/total_time_seconds': '650', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '922.6', 'train/estimated_remaining_minutes': '29.79'}
+{'eval_loss': '3.323', 'eval_runtime': '16.5', 'eval_samples_per_second': '577.3', 'eval_steps_per_second': '4.545', 'epoch': '0.2157', 'train/total_time_seconds': '650', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '939.1', 'train/estimated_remaining_minutes': '29.79'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 19.13it/s]
+ 33%|█████████████                          | 500/1500 [19:33<33:17,  2.00s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '13.05', 'grad_norm': '2.656', 'learning_rate': '0.001', 'epoch': '0.2265', 'train/total_time_seconds': '682.3', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '979.4', 'train/estimated_remaining_minutes': '29.24'}
+{'loss': '12.78', 'grad_norm': '2.047', 'learning_rate': '0.001', 'epoch': '0.2373', 'train/total_time_seconds': '714.9', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '1020', 'train/estimated_remaining_minutes': '28.7'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '3.121', 'eval_runtime': '16.44', 'eval_samples_per_second': '579.5', 'eval_steps_per_second': '4.562', 'epoch': '0.2427', 'train/total_time_seconds': '731.1', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '1056', 'train/estimated_remaining_minutes': '28.43'}
+{'loss': '12.46', 'grad_norm': '2.344', 'learning_rate': '0.001', 'epoch': '0.248', 'train/total_time_seconds': '747.2', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '1076', 'train/estimated_remaining_minutes': '28.16'}
+{'loss': '12.18', 'grad_norm': '1.734', 'learning_rate': '0.001', 'epoch': '0.2588', 'train/total_time_seconds': '779.6', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '1116', 'train/estimated_remaining_minutes': '27.61'}
+{'loss': '11.92', 'grad_norm': '1.656', 'learning_rate': '0.001', 'epoch': '0.2696', 'train/total_time_seconds': '811.8', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '1156', 'train/estimated_remaining_minutes': '27.06'}
+{'eval_loss': '2.951', 'eval_runtime': '16.42', 'eval_samples_per_second': '580.3', 'eval_steps_per_second': '4.568', 'epoch': '0.2696', 'train/total_time_seconds': '811.8', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '1173', 'train/estimated_remaining_minutes': '27.06'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 15.22it/s]
+/mnt/data/zainulabideen/zain-exp/notebooks/Activation/exp.py:487: UserWarning: std(): degrees of freedom is <= 0. Correction should be strictly less than the reduction factor (input numel divided by output numel). (Triggered internally at /pytorch/aten/src/ATen/native/ReduceOps.cpp:1831.)
+  "std": tensor.std().item(),
+ 40%|███████████████▌                       | 600/1500 [23:29<30:04,  2.00s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '11.68', 'grad_norm': '1.523', 'learning_rate': '0.001', 'epoch': '0.2804', 'train/total_time_seconds': '846.4', 'train/time_per_step_avg': '1.641', 'train/epoch_time_elapsed': '1216', 'train/estimated_remaining_minutes': '26.58', 'train/global/act/norm': '2.451e+05', 'train/global/act/mean': '-0.04372', 'train/global/act/std': '1.193', 'train/global/act/max_abs': '47.5', 'train/global/act/frac_near_dtype_limit': '0', 'train/global/act/frac_near_user_limit': '0', 'train/global/grad/norm': '0.8004', 'train/global/grad/mean': '-7.534e-07', 'train/global/grad/std': '0.0001002', 'train/global/grad/max_abs': '0.01471', 'train/global/grad/frac_near_dtype_limit': '0', 'train/global/grad/frac_near_user_limit': '0', 'train/global/param/norm': '202.1', 'train/global/param/mean': '0.001283', 'train/global/param/std': '0.0506', 'train/global/param/max_abs': '1', 'train/global/param/frac_near_dtype_limit': '0', 'train/global/param/frac_near_user_limit': '0', 'train/layer_model_layers_27/act/norm': '2.303e+04', 'train/layer_model_layers_27/act/mean': '0.01982', 'train/layer_model_layers_27/act/std': '1.102', 'train/layer_model_layers_27/act/max_abs': '44.25', 'train/layer_model_layers_27/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_27/act/frac_near_user_limit': '0', 'train/layer_model_layers_27/grad/norm': '0.02829', 'train/layer_model_layers_27/grad/mean': '-3.765e-07', 'train/layer_model_layers_27/grad/std': '3.491e-05', 'train/layer_model_layers_27/grad/max_abs': '0.001175', 'train/layer_model_layers_27/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_27/grad/frac_near_user_limit': '0', 'train/layer_model_layers_87/act/norm': '2.417e+04', 'train/layer_model_layers_87/act/mean': '0.003093', 'train/layer_model_layers_87/act/std': '1.157', 'train/layer_model_layers_87/act/max_abs': '33', 'train/layer_model_layers_87/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_87/act/frac_near_user_limit': '0', 'train/layer_model_layers_87/grad/norm': '0.08349', 'train/layer_model_layers_87/grad/mean': '-6.282e-07', 'train/layer_model_layers_87/grad/std': '0.0001032', 'train/layer_model_layers_87/grad/max_abs': '0.001923', 'train/layer_model_layers_87/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_87/grad/frac_near_user_limit': '0', 'train/layer_model_layers_53/act/norm': '2.161e+04', 'train/layer_model_layers_53/act/mean': '0.0156', 'train/layer_model_layers_53/act/std': '1.036', 'train/layer_model_layers_53/act/max_abs': '43', 'train/layer_model_layers_53/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_53/act/frac_near_user_limit': '0', 'train/layer_model_layers_53/grad/norm': '0.04708', 'train/layer_model_layers_53/grad/mean': '-3.519e-07', 'train/layer_model_layers_53/grad/std': '5.811e-05', 'train/layer_model_layers_53/grad/max_abs': '0.001633', 'train/layer_model_layers_53/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_53/grad/frac_near_user_limit': '0', 'train/layer__model_layers_31/param/norm': '18.88', 'train/layer__model_layers_31/param/mean': '0.001684', 'train/layer__model_layers_31/param/std': '0.04658', 'train/layer__model_layers_31/param/max_abs': '1', 'train/layer__model_layers_31/param/frac_near_dtype_limit': '0', 'train/layer__model_layers_31/param/frac_near_user_limit': '0', 'train/layer_model_layers_24/act/norm': '2.373e+04', 'train/layer_model_layers_24/act/mean': '0.003067', 'train/layer_model_layers_24/act/std': '1.136', 'train/layer_model_layers_24/act/max_abs': '44.5', 'train/layer_model_layers_24/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_24/act/frac_near_user_limit': '0', 'train/layer_model_layers_24/grad/norm': '0.03601', 'train/layer_model_layers_24/grad/mean': '-3.842e-07', 'train/layer_model_layers_24/grad/std': '4.446e-05', 'train/layer_model_layers_24/grad/max_abs': '0.001152', 'train/layer_model_layers_24/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_24/grad/frac_near_user_limit': '0', 'train/layer_model_layers_39/act/norm': '2.249e+04', 'train/layer_model_layers_39/act/mean': '0.01775', 'train/layer_mode
+{'loss': '11.46', 'grad_norm': '2.297', 'learning_rate': '0.001', 'epoch': '0.2912', 'train/total_time_seconds': '878.8', 'train/time_per_step_avg': '1.639', 'train/epoch_time_elapsed': '1256', 'train/estimated_remaining_minutes': '26.04'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.814', 'eval_runtime': '16.35', 'eval_samples_per_second': '582.7', 'eval_steps_per_second': '4.587', 'epoch': '0.2966', 'train/total_time_seconds': '894.9', 'train/time_per_step_avg': '1.639', 'train/epoch_time_elapsed': '1292', 'train/estimated_remaining_minutes': '25.76'}
+{'loss': '11.25', 'grad_norm': '1.844', 'learning_rate': '0.001', 'epoch': '0.302', 'train/total_time_seconds': '911.1', 'train/time_per_step_avg': '1.639', 'train/epoch_time_elapsed': '1312', 'train/estimated_remaining_minutes': '25.49'}
+{'loss': '11.02', 'grad_norm': '1.961', 'learning_rate': '0.001', 'epoch': '0.3128', 'train/total_time_seconds': '943.7', 'train/time_per_step_avg': '1.641', 'train/epoch_time_elapsed': '1352', 'train/estimated_remaining_minutes': '24.95'}
+{'loss': '10.83', 'grad_norm': '2.109', 'learning_rate': '0.001', 'epoch': '0.3235', 'train/total_time_seconds': '976', 'train/time_per_step_avg': '1.642', 'train/epoch_time_elapsed': '1393', 'train/estimated_remaining_minutes': '24.4'}
+{'eval_loss': '2.694', 'eval_runtime': '16.51', 'eval_samples_per_second': '577', 'eval_steps_per_second': '4.542', 'epoch': '0.3235', 'train/total_time_seconds': '976', 'train/time_per_step_avg': '1.642', 'train/epoch_time_elapsed': '1409', 'train/estimated_remaining_minutes': '24.4'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 17.02it/s]
+ 47%|██████████████████▏                    | 700/1500 [27:23<27:10,  2.04s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '10.67', 'grad_norm': '1.258', 'learning_rate': '0.001', 'epoch': '0.3343', 'train/total_time_seconds': '1008', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '1449', 'train/estimated_remaining_minutes': '23.85'}
+{'loss': '10.49', 'grad_norm': '1.938', 'learning_rate': '0.001', 'epoch': '0.3451', 'train/total_time_seconds': '1041', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '1489', 'train/estimated_remaining_minutes': '23.3'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.583', 'eval_runtime': '16.38', 'eval_samples_per_second': '581.5', 'eval_steps_per_second': '4.577', 'epoch': '0.3505', 'train/total_time_seconds': '1057', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '1526', 'train/estimated_remaining_minutes': '23.03'}
+{'loss': '10.31', 'grad_norm': '1.477', 'learning_rate': '0.001', 'epoch': '0.3559', 'train/total_time_seconds': '1073', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '1546', 'train/estimated_remaining_minutes': '22.76'}
+{'loss': '10.16', 'grad_norm': '1.609', 'learning_rate': '0.001', 'epoch': '0.3667', 'train/total_time_seconds': '1105', 'train/time_per_step_avg': '1.616', 'train/epoch_time_elapsed': '1586', 'train/estimated_remaining_minutes': '22.21'}
+{'loss': '10', 'grad_norm': '2.078', 'learning_rate': '0.001', 'epoch': '0.3775', 'train/total_time_seconds': '1138', 'train/time_per_step_avg': '1.616', 'train/epoch_time_elapsed': '1626', 'train/estimated_remaining_minutes': '21.67'}
+{'eval_loss': '2.485', 'eval_runtime': '16.64', 'eval_samples_per_second': '572.4', 'eval_steps_per_second': '4.506', 'epoch': '0.3775', 'train/total_time_seconds': '1138', 'train/time_per_step_avg': '1.616', 'train/epoch_time_elapsed': '1643', 'train/estimated_remaining_minutes': '21.67'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 14.85it/s]
+ 53%|████████████████████▊                  | 800/1500 [31:18<23:21,  2.00s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '9.83', 'grad_norm': '1.82', 'learning_rate': '0.001', 'epoch': '0.3882', 'train/total_time_seconds': '1170', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '1684', 'train/estimated_remaining_minutes': '21.13'}
+{'loss': '9.726', 'grad_norm': '1.672', 'learning_rate': '0.001', 'epoch': '0.399', 'train/total_time_seconds': '1203', 'train/time_per_step_avg': '1.62', 'train/epoch_time_elapsed': '1724', 'train/estimated_remaining_minutes': '20.58'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.396', 'eval_runtime': '16.48', 'eval_samples_per_second': '578', 'eval_steps_per_second': '4.55', 'epoch': '0.4044', 'train/total_time_seconds': '1219', 'train/time_per_step_avg': '1.62', 'train/epoch_time_elapsed': '1760', 'train/estimated_remaining_minutes': '20.31'}
+{'loss': '9.586', 'grad_norm': '1.5', 'learning_rate': '0.001', 'epoch': '0.4098', 'train/total_time_seconds': '1235', 'train/time_per_step_avg': '1.62', 'train/epoch_time_elapsed': '1781', 'train/estimated_remaining_minutes': '20.04'}
+{'loss': '9.475', 'grad_norm': '1.492', 'learning_rate': '0.001', 'epoch': '0.4206', 'train/total_time_seconds': '1267', 'train/time_per_step_avg': '1.621', 'train/epoch_time_elapsed': '1821', 'train/estimated_remaining_minutes': '19.5'}
+{'loss': '9.304', 'grad_norm': '1.789', 'learning_rate': '0.001', 'epoch': '0.4314', 'train/total_time_seconds': '1300', 'train/time_per_step_avg': '1.62', 'train/epoch_time_elapsed': '1861', 'train/estimated_remaining_minutes': '18.95'}
+{'eval_loss': '2.319', 'eval_runtime': '16.36', 'eval_samples_per_second': '582.3', 'eval_steps_per_second': '4.584', 'epoch': '0.4314', 'train/total_time_seconds': '1300', 'train/time_per_step_avg': '1.62', 'train/epoch_time_elapsed': '1877', 'train/estimated_remaining_minutes': '18.95'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 18.96it/s]
+ 60%|███████████████████████▍               | 900/1500 [35:11<19:54,  1.99s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '9.215', 'grad_norm': '1.969', 'learning_rate': '0.001', 'epoch': '0.4422', 'train/total_time_seconds': '1332', 'train/time_per_step_avg': '1.617', 'train/epoch_time_elapsed': '1918', 'train/estimated_remaining_minutes': '18.41'}
+{'loss': '9.11', 'grad_norm': '1.266', 'learning_rate': '0.001', 'epoch': '0.453', 'train/total_time_seconds': '1364', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '1958', 'train/estimated_remaining_minutes': '17.87'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.256', 'eval_runtime': '16.63', 'eval_samples_per_second': '572.7', 'eval_steps_per_second': '4.509', 'epoch': '0.4583', 'train/total_time_seconds': '1381', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '1995', 'train/estimated_remaining_minutes': '17.6'}
+{'loss': '8.99', 'grad_norm': '1.531', 'learning_rate': '0.001', 'epoch': '0.4637', 'train/total_time_seconds': '1397', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '2015', 'train/estimated_remaining_minutes': '17.32'}
+{'loss': '8.903', 'grad_norm': '1.469', 'learning_rate': '0.001', 'epoch': '0.4745', 'train/total_time_seconds': '1429', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '2055', 'train/estimated_remaining_minutes': '16.78'}
+{'loss': '8.798', 'grad_norm': '1.562', 'learning_rate': '0.001', 'epoch': '0.4853', 'train/total_time_seconds': '1461', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '2095', 'train/estimated_remaining_minutes': '16.24'}
+{'eval_loss': '2.199', 'eval_runtime': '16.44', 'eval_samples_per_second': '579.6', 'eval_steps_per_second': '4.563', 'epoch': '0.4853', 'train/total_time_seconds': '1461', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '2111', 'train/estimated_remaining_minutes': '16.24'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 15.39it/s]
+ 67%|█████████████████████████▎            | 1000/1500 [39:05<16:54,  2.03s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '8.712', 'grad_norm': '1.352', 'learning_rate': '0.001', 'epoch': '0.4961', 'train/total_time_seconds': '1494', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '2152', 'train/estimated_remaining_minutes': '15.69'}
+{'loss': '8.657', 'grad_norm': '1.375', 'learning_rate': '0.001', 'epoch': '0.5069', 'train/total_time_seconds': '1526', 'train/time_per_step_avg': '1.616', 'train/epoch_time_elapsed': '2192', 'train/estimated_remaining_minutes': '15.15'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.152', 'eval_runtime': '16.52', 'eval_samples_per_second': '576.9', 'eval_steps_per_second': '4.541', 'epoch': '0.5123', 'train/total_time_seconds': '1542', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '2228', 'train/estimated_remaining_minutes': '14.88'}
+{'loss': '8.597', 'grad_norm': '1.547', 'learning_rate': '0.001', 'epoch': '0.5177', 'train/total_time_seconds': '1559', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '2248', 'train/estimated_remaining_minutes': '14.61'}
+{'loss': '8.528', 'grad_norm': '1.383', 'learning_rate': '0.001', 'epoch': '0.5284', 'train/total_time_seconds': '1591', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '2289', 'train/estimated_remaining_minutes': '14.07'}
+{'loss': '8.455', 'grad_norm': '1.391', 'learning_rate': '0.001', 'epoch': '0.5392', 'train/total_time_seconds': '1623', 'train/time_per_step_avg': '1.617', 'train/epoch_time_elapsed': '2329', 'train/estimated_remaining_minutes': '13.53'}
+{'eval_loss': '2.105', 'eval_runtime': '16.47', 'eval_samples_per_second': '578.5', 'eval_steps_per_second': '4.554', 'epoch': '0.5392', 'train/total_time_seconds': '1623', 'train/time_per_step_avg': '1.617', 'train/epoch_time_elapsed': '2345', 'train/estimated_remaining_minutes': '13.53'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 17.20it/s]
+/mnt/data/zainulabideen/zain-exp/notebooks/Activation/exp.py:487: UserWarning: std(): degrees of freedom is <= 0. Correction should be strictly less than the reduction factor (input numel divided by output numel). (Triggered internally at /pytorch/aten/src/ATen/native/ReduceOps.cpp:1831.)
+  "std": tensor.std().item(),
+ 73%|███████████████████████████▊          | 1100/1500 [43:02<13:16,  1.99s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '8.371', 'grad_norm': '1.609', 'learning_rate': '0.001', 'epoch': '0.55', 'train/total_time_seconds': '1658', 'train/time_per_step_avg': '1.639', 'train/epoch_time_elapsed': '2388', 'train/estimated_remaining_minutes': '13', 'train/global/act/norm': '1.817e+05', 'train/global/act/mean': '-0.06364', 'train/global/act/std': '0.882', 'train/global/act/max_abs': '30', 'train/global/act/frac_near_dtype_limit': '0', 'train/global/act/frac_near_user_limit': '0', 'train/global/grad/norm': '0.7324', 'train/global/grad/mean': '-1.664e-07', 'train/global/grad/std': '9.171e-05', 'train/global/grad/max_abs': '0.01361', 'train/global/grad/frac_near_dtype_limit': '0', 'train/global/grad/frac_near_user_limit': '0', 'train/global/param/norm': '225.7', 'train/global/param/mean': '0.001164', 'train/global/param/std': '0.05651', 'train/global/param/max_abs': '1', 'train/global/param/frac_near_dtype_limit': '0', 'train/global/param/frac_near_user_limit': '0', 'train/layer_model_layers_27/act/norm': '1.39e+04', 'train/layer_model_layers_27/act/mean': '0.002256', 'train/layer_model_layers_27/act/std': '0.6657', 'train/layer_model_layers_27/act/max_abs': '17.75', 'train/layer_model_layers_27/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_27/act/frac_near_user_limit': '0', 'train/layer_model_layers_27/grad/norm': '0.0356', 'train/layer_model_layers_27/grad/mean': '-5.762e-07', 'train/layer_model_layers_27/grad/std': '4.393e-05', 'train/layer_model_layers_27/grad/max_abs': '0.00174', 'train/layer_model_layers_27/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_27/grad/frac_near_user_limit': '0', 'train/layer_model_layers_87/act/norm': '1.856e+04', 'train/layer_model_layers_87/act/mean': '-0.00713', 'train/layer_model_layers_87/act/std': '0.8882', 'train/layer_model_layers_87/act/max_abs': '12.12', 'train/layer_model_layers_87/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_87/act/frac_near_user_limit': '0', 'train/layer_model_layers_87/grad/norm': '0.0918', 'train/layer_model_layers_87/grad/mean': '9.888e-07', 'train/layer_model_layers_87/grad/std': '0.0001133', 'train/layer_model_layers_87/grad/max_abs': '0.001129', 'train/layer_model_layers_87/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_87/grad/frac_near_user_limit': '0', 'train/layer_model_layers_53/act/norm': '1.384e+04', 'train/layer_model_layers_53/act/mean': '-0.0009267', 'train/layer_model_layers_53/act/std': '0.6624', 'train/layer_model_layers_53/act/max_abs': '15.31', 'train/layer_model_layers_53/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_53/act/frac_near_user_limit': '0', 'train/layer_model_layers_53/grad/norm': '0.04177', 'train/layer_model_layers_53/grad/mean': '-1.384e-06', 'train/layer_model_layers_53/grad/std': '5.152e-05', 'train/layer_model_layers_53/grad/max_abs': '0.000927', 'train/layer_model_layers_53/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_53/grad/frac_near_user_limit': '0', 'train/layer__model_layers_31/param/norm': '20.27', 'train/layer__model_layers_31/param/mean': '0.001609', 'train/layer__model_layers_31/param/std': '0.05002', 'train/layer__model_layers_31/param/max_abs': '1', 'train/layer__model_layers_31/param/frac_near_dtype_limit': '0', 'train/layer__model_layers_31/param/frac_near_user_limit': '0', 'train/layer_model_layers_24/act/norm': '1.446e+04', 'train/layer_model_layers_24/act/mean': '-0.01199', 'train/layer_model_layers_24/act/std': '0.6924', 'train/layer_model_layers_24/act/max_abs': '17.75', 'train/layer_model_layers_24/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_24/act/frac_near_user_limit': '0', 'train/layer_model_layers_24/grad/norm': '0.0361', 'train/layer_model_layers_24/grad/mean': '-6.869e-07', 'train/layer_model_layers_24/grad/std': '4.46e-05', 'train/layer_model_layers_24/grad/max_abs': '0.001503', 'train/layer_model_layers_24/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_24/grad/frac_near_user_limit': '0', 'train/layer_model_layers_39/act/norm': '1.377e+04', 'train/layer_model_layers_39/act/mean': '0.001056', 'train/layer_m
+{'loss': '8.291', 'grad_norm': '1.344', 'learning_rate': '0.001', 'epoch': '0.5608', 'train/total_time_seconds': '1690', 'train/time_per_step_avg': '1.638', 'train/epoch_time_elapsed': '2428', 'train/estimated_remaining_minutes': '12.46'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.071', 'eval_runtime': '16.32', 'eval_samples_per_second': '583.7', 'eval_steps_per_second': '4.595', 'epoch': '0.5662', 'train/total_time_seconds': '1706', 'train/time_per_step_avg': '1.636', 'train/epoch_time_elapsed': '2464', 'train/estimated_remaining_minutes': '12.19'}
+{'loss': '8.234', 'grad_norm': '1.461', 'learning_rate': '0.001', 'epoch': '0.5716', 'train/total_time_seconds': '1722', 'train/time_per_step_avg': '1.636', 'train/epoch_time_elapsed': '2484', 'train/estimated_remaining_minutes': '11.91'}
+{'loss': '8.194', 'grad_norm': '1.414', 'learning_rate': '0.001', 'epoch': '0.5824', 'train/total_time_seconds': '1755', 'train/time_per_step_avg': '1.64', 'train/epoch_time_elapsed': '2525', 'train/estimated_remaining_minutes': '11.37'}
+{'loss': '8.149', 'grad_norm': '1.367', 'learning_rate': '0.001', 'epoch': '0.5932', 'train/total_time_seconds': '1787', 'train/time_per_step_avg': '1.639', 'train/epoch_time_elapsed': '2565', 'train/estimated_remaining_minutes': '10.83'}
+{'eval_loss': '2.038', 'eval_runtime': '16.58', 'eval_samples_per_second': '574.5', 'eval_steps_per_second': '4.523', 'epoch': '0.5932', 'train/total_time_seconds': '1787', 'train/time_per_step_avg': '1.639', 'train/epoch_time_elapsed': '2581', 'train/estimated_remaining_minutes': '10.83'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 14.79it/s]
+ 80%|██████████████████████████████▍       | 1200/1500 [47:30<12:36,  2.52s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '8.088', 'grad_norm': '1.18', 'learning_rate': '0.001', 'epoch': '0.6039', 'train/total_time_seconds': '1819', 'train/time_per_step_avg': '1.618', 'train/epoch_time_elapsed': '2622', 'train/estimated_remaining_minutes': '10.29'}
+{'loss': '8.043', 'grad_norm': '1.367', 'learning_rate': '0.001', 'epoch': '0.6147', 'train/total_time_seconds': '1852', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '2662', 'train/estimated_remaining_minutes': '9.746'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.006', 'eval_runtime': '19.6', 'eval_samples_per_second': '486.2', 'eval_steps_per_second': '3.827', 'epoch': '0.6201', 'train/total_time_seconds': '1868', 'train/time_per_step_avg': '1.619', 'train/epoch_time_elapsed': '2702', 'train/estimated_remaining_minutes': '9.475'}
+{'loss': '8.001', 'grad_norm': '1.289', 'learning_rate': '0.001', 'epoch': '0.6255', 'train/total_time_seconds': '1890', 'train/time_per_step_avg': '1.675', 'train/epoch_time_elapsed': '2727', 'train/estimated_remaining_minutes': '9.231'}
+{'loss': '7.95', 'grad_norm': '1.164', 'learning_rate': '0.001', 'epoch': '0.6363', 'train/total_time_seconds': '1933', 'train/time_per_step_avg': '1.778', 'train/epoch_time_elapsed': '2778', 'train/estimated_remaining_minutes': '8.735'}
+{'loss': '7.893', 'grad_norm': '1.211', 'learning_rate': '0.001', 'epoch': '0.6471', 'train/total_time_seconds': '1976', 'train/time_per_step_avg': '1.891', 'train/epoch_time_elapsed': '2830', 'train/estimated_remaining_minutes': '8.234'}
+{'eval_loss': '1.98', 'eval_runtime': '20.34', 'eval_samples_per_second': '468.4', 'eval_steps_per_second': '3.687', 'epoch': '0.6471', 'train/total_time_seconds': '1976', 'train/time_per_step_avg': '1.891', 'train/epoch_time_elapsed': '2850', 'train/estimated_remaining_minutes': '8.234'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 16.31it/s]
+ 87%|████████████████████████████████▉     | 1300/1500 [52:27<08:45,  2.63s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '7.88', 'grad_norm': '1.344', 'learning_rate': '0.001', 'epoch': '0.6579', 'train/total_time_seconds': '2020', 'train/time_per_step_avg': '2.002', 'train/epoch_time_elapsed': '2902', 'train/estimated_remaining_minutes': '7.725'}
+{'loss': '7.828', 'grad_norm': '1.273', 'learning_rate': '0.001', 'epoch': '0.6686', 'train/total_time_seconds': '2063', 'train/time_per_step_avg': '2.111', 'train/epoch_time_elapsed': '2954', 'train/estimated_remaining_minutes': '7.209'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '1.956', 'eval_runtime': '20.26', 'eval_samples_per_second': '470.3', 'eval_steps_per_second': '3.702', 'epoch': '0.674', 'train/total_time_seconds': '2084', 'train/time_per_step_avg': '2.163', 'train/epoch_time_elapsed': '2999', 'train/estimated_remaining_minutes': '6.947'}
+{'loss': '7.784', 'grad_norm': '1.188', 'learning_rate': '0.001', 'epoch': '0.6794', 'train/total_time_seconds': '2105', 'train/time_per_step_avg': '2.154', 'train/epoch_time_elapsed': '3024', 'train/estimated_remaining_minutes': '6.683'}
+{'loss': '7.767', 'grad_norm': '1.203', 'learning_rate': '0.001', 'epoch': '0.6902', 'train/total_time_seconds': '2148', 'train/time_per_step_avg': '2.157', 'train/epoch_time_elapsed': '3075', 'train/estimated_remaining_minutes': '6.154'}
+{'loss': '7.714', 'grad_norm': '1.219', 'learning_rate': '0.001', 'epoch': '0.701', 'train/total_time_seconds': '2192', 'train/time_per_step_avg': '2.16', 'train/epoch_time_elapsed': '3127', 'train/estimated_remaining_minutes': '5.621'}
+{'eval_loss': '1.935', 'eval_runtime': '20.2', 'eval_samples_per_second': '471.7', 'eval_steps_per_second': '3.713', 'epoch': '0.701', 'train/total_time_seconds': '2192', 'train/time_per_step_avg': '2.16', 'train/epoch_time_elapsed': '3147', 'train/estimated_remaining_minutes': '5.621'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 13.90it/s]
+ 93%|███████████████████████████████████▍  | 1400/1500 [57:24<04:21,  2.62s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '7.675', 'grad_norm': '1.32', 'learning_rate': '0.001', 'epoch': '0.7118', 'train/total_time_seconds': '2235', 'train/time_per_step_avg': '2.155', 'train/epoch_time_elapsed': '3199', 'train/estimated_remaining_minutes': '5.08'}
+{'loss': '7.656', 'grad_norm': '1.211', 'learning_rate': '0.001', 'epoch': '0.7226', 'train/total_time_seconds': '2278', 'train/time_per_step_avg': '2.153', 'train/epoch_time_elapsed': '3250', 'train/estimated_remaining_minutes': '4.534'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '1.912', 'eval_runtime': '19.71', 'eval_samples_per_second': '483.3', 'eval_steps_per_second': '3.804', 'epoch': '0.728', 'train/total_time_seconds': '2301', 'train/time_per_step_avg': '2.164', 'train/epoch_time_elapsed': '3296', 'train/estimated_remaining_minutes': '4.26'}
+{'loss': '7.623', 'grad_norm': '1.188', 'learning_rate': '0.001', 'epoch': '0.7334', 'train/total_time_seconds': '2323', 'train/time_per_step_avg': '2.177', 'train/epoch_time_elapsed': '3322', 'train/estimated_remaining_minutes': '3.985'}
+{'loss': '7.605', 'grad_norm': '1.109', 'learning_rate': '0.001', 'epoch': '0.7441', 'train/total_time_seconds': '2366', 'train/time_per_step_avg': '2.179', 'train/epoch_time_elapsed': '3373', 'train/estimated_remaining_minutes': '3.429'}
+{'loss': '7.559', 'grad_norm': '1.109', 'learning_rate': '0.001', 'epoch': '0.7549', 'train/total_time_seconds': '2410', 'train/time_per_step_avg': '2.175', 'train/epoch_time_elapsed': '3425', 'train/estimated_remaining_minutes': '2.869'}
+{'eval_loss': '1.894', 'eval_runtime': '19.41', 'eval_samples_per_second': '490.9', 'eval_steps_per_second': '3.864', 'epoch': '0.7549', 'train/total_time_seconds': '2410', 'train/time_per_step_avg': '2.175', 'train/epoch_time_elapsed': '3444', 'train/estimated_remaining_minutes': '2.869'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 19.84it/s]
+100%|████████████████████████████████████| 1500/1500 [1:01:59<00:00,  1.99s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '7.519', 'grad_norm': '1.344', 'learning_rate': '0.001', 'epoch': '0.7657', 'train/total_time_seconds': '2453', 'train/time_per_step_avg': '2.181', 'train/epoch_time_elapsed': '3496', 'train/estimated_remaining_minutes': '2.303'}
+{'loss': '7.509', 'grad_norm': '1.367', 'learning_rate': '0.001', 'epoch': '0.7765', 'train/total_time_seconds': '2496', 'train/time_per_step_avg': '2.183', 'train/epoch_time_elapsed': '3547', 'train/estimated_remaining_minutes': '1.734'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '1.876', 'eval_runtime': '19.25', 'eval_samples_per_second': '494.8', 'eval_steps_per_second': '3.895', 'epoch': '0.7819', 'train/total_time_seconds': '2519', 'train/time_per_step_avg': '2.182', 'train/epoch_time_elapsed': '3593', 'train/estimated_remaining_minutes': '1.448'}
+{'loss': '7.472', 'grad_norm': '1.078', 'learning_rate': '0.001', 'epoch': '0.7873', 'train/total_time_seconds': '2541', 'train/time_per_step_avg': '2.184', 'train/epoch_time_elapsed': '3619', 'train/estimated_remaining_minutes': '1.16'}
+{'loss': '7.451', 'grad_norm': '1.148', 'learning_rate': '0.001', 'epoch': '0.7981', 'train/total_time_seconds': '2576', 'train/time_per_step_avg': '2.101', 'train/epoch_time_elapsed': '3662', 'train/estimated_remaining_minutes': '0.5802'}
+{'loss': '7.419', 'grad_norm': '1.117', 'learning_rate': '0.001', 'epoch': '0.8088', 'train/total_time_seconds': '2609', 'train/time_per_step_avg': '1.989', 'train/epoch_time_elapsed': '3702', 'train/estimated_remaining_minutes': '0'}
+{'eval_loss': '1.862', 'eval_runtime': '16.4', 'eval_samples_per_second': '581.1', 'eval_steps_per_second': '4.574', 'epoch': '0.8088', 'train/total_time_seconds': '2609', 'train/time_per_step_avg': '1.989', 'train/epoch_time_elapsed': '3718', 'train/estimated_remaining_minutes': '0'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 14.85it/s]
+100%|████████████████████████████████████| 1500/1500 [1:01:59<00:00,  2.48s/it]
+{'train_runtime': '3719', 'train_samples_per_second': '206.5', 'train_steps_per_second': '0.403', 'train_loss': '11.67', 'epoch': '0.8088', 'train/total_time_seconds': '2609', 'train/time_per_step_avg': '1.989', 'train/epoch_time_elapsed': '3719', 'train/estimated_remaining_minutes': '0'}
+100%|██████████████████████████████████████████| 75/75 [00:16<00:00,  4.68it/s]
+[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 15.86it/s]
+[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 18.91it/s]
+Found 7 files to upload
+  Preparing   ████████████████████  7 / 7 ✓
+  Uploading   ████████████████████  2 / 2 files ✓
+  Committing  ████████░░░░░░░░░░░░  3 / 7
+  Preparing   ████████████████████  7 / 7 ✓
+  Uploading   ████████████████████  2 / 2 files ✓
+  Committing  ████████░░░░░░░░░░░░  3 / 7
+  Preparing   ████████████████████  7 / 7 ✓
+  Uploading   ████████████████████  2 / 2 files ✓
+  Committing  ████████████████████  7 / 7 ✓
+[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 18.37it/s]
+Found 7 files to upload
+  Preparing   ████████████████████  7 / 7 ✓
+  Uploading   ████████████████████  2 / 2 files ✓
+  Committing  ████████████████████  7 / 7 ✓
+No files have been modified since last commit. Skipping to prevent empty commit.

wandb/run-20260809_035819-cvzjg5ej/files/requirements.txt

@@ -0,0 +1,149 @@
+asttokens==3.0.1
+comm==0.2.3
+debugpy==1.8.21
+decorator==5.3.1
+executing==2.2.1
+nest-asyncio==1.6.0
+parso==0.8.7
+platformdirs==4.11.0
+psutil==7.2.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+Pygments==2.20.0
+pyzmq==27.1.0
+setuptools==83.0.0
+six==1.17.0
+tornado==6.5.7
+traitlets==5.15.0
+fsspec==2026.4.0
+wcwidth==0.8.2
+ipython_pygments_lexers==1.1.1
+jedi==0.20.0
+jupyter_core==5.9.1
+matplotlib-inline==0.2.2
+pexpect==4.9.0
+prompt_toolkit==3.0.53
+python-dateutil==2.9.0.post0
+stack_data==0.6.3
+wheel==0.47.0
+jupyter_client==8.9.1
+pip==26.1.2
+ipython==9.15.0
+ipykernel==7.2.0
+threadpoolctl==3.6.0
+pyparsing==3.3.2
+typing_extensions==4.15.0
+Jinja2==3.1.6
+narwhals==2.24.0
+kiwisolver==1.5.0
+joblib==1.5.3
+fonttools==4.63.0
+cycler==0.12.1
+scipy==1.17.1
+pandas==3.0.5
+contourpy==1.3.3
+scikit-learn==1.9.0
+matplotlib==3.11.1
+urllib3==2.7.0
+tqdm==4.70.0
+idna==3.18
+charset-normalizer==3.4.9
+certifi==2026.7.22
+requests==2.34.2
+seaborn==0.13.2
+uv==0.12.0
+shellingham==1.5.4
+mpmath==1.3.0
+attrs==26.1.0
+hf-xet==1.5.2
+nvidia-nccl-cu12==2.21.5
+MarkupSafe==3.0.3
+regex==2026.7.19
+importlib_metadata==9.0.0
+httpcore==1.0.9
+annotated-doc==0.0.5
+multidict==6.7.1
+aiohttp==3.14.3
+aiosignal==1.4.0
+xxhash==3.8.1
+aiohappyeyeballs==2.7.1
+mdurl==0.1.2
+cuda-toolkit==13.0.3.0
+networkx==3.6.1
+PyYAML==6.0.3
+nvidia-cufile==1.15.1.6
+typer==0.27.0
+torchaudio==2.6.0+cu124
+rich==15.0.0
+nvidia-cufft-cu12==11.2.1.3
+h11==0.16.0
+dill==0.4.1
+cuda-pathfinder==1.6.0
+filelock==3.29.0
+nvidia-nvtx-cu12==12.4.127
+httpx==0.28.1
+anyio==4.14.2
+numpy==2.4.4
+yarl==1.24.5
+click==8.4.2
+triton==3.2.0
+frozenlist==1.8.0
+zipp==4.1.0
+propcache==0.5.2
+tokenizers==0.22.2
+markdown-it-py==4.2.0
+nvidia-cuda-runtime==13.0.96
+cuda-bindings==13.3.1
+nvidia-cuda-cupti==13.0.85
+torch==2.6.0+cu124
+multiprocess==0.70.19
+pillow==12.2.0
+transformers==5.15.0.dev0
+wandb==0.28.1
+nvidia-curand==10.4.0.35
+sympy==1.13.1
+nvidia-cusparse==12.6.3.3
+nvidia-cuda-nvrtc==13.0.88
+typing-inspection==0.4.2
+nvidia-cusolver==12.0.4.66
+nvidia-cufft==12.0.0.61
+nvidia-cudnn-cu13==9.20.0.48
+nvidia-cublas==13.1.1.3
+pyarrow==25.0.0
+evaluate==0.4.6
+diffusers==0.39.0
+pydantic==2.13.4
+annotated-types==0.8.0
+protobuf==7.35.1
+sentry-sdk==2.66.1
+einops==0.8.2
+packaging==26.2
+nvidia-nvjitlink-cu12==12.4.127
+nvidia-curand-cu12==10.3.5.147
+nvidia-cusparselt-cu12==0.6.2
+nvidia-cusparse-cu12==12.3.1.170
+nvidia-cuda-runtime-cu12==12.4.127
+torchvision==0.21.0+cu124
+nvidia-cuda-nvrtc-cu12==12.4.127
+nvidia-cuda-cupti-cu12==12.4.127
+nvidia-cusolver-cu12==11.6.1.9
+nvidia-cublas-cu12==12.4.5.8
+nvidia-cudnn-cu12==9.1.0.70
+huggingface_hub==1.26.0
+datasets==5.0.1
+safetensors==0.8.0
+accelerate==1.14.0
+pydantic_core==2.46.4
+ninja==1.13.0
+autocommand==2.2.2
+backports.tarfile==1.2.0
+importlib_metadata==8.7.1
+jaraco.text==4.0.0
+jaraco.context==6.1.0
+jaraco.functools==4.4.0
+more-itertools==10.8.0
+packaging==26.0
+platformdirs==4.4.0
+tomli==2.4.0
+wheel==0.46.3
+zipp==3.23.0

wandb/run-20260809_035819-cvzjg5ej/files/wandb-metadata.json

@@ -0,0 +1,96 @@
+{
+  "os":  "Linux-5.15.0-126-generic-x86_64-with-glibc2.35",
+  "python":  "CPython 3.11.15",
+  "startedAt":  "2026-08-09T03:58:19.280139Z",
+  "args":  [
+    "--config",
+    "configs/baseline.yaml",
+    "--variants",
+    "mlp-s10-94L",
+    "--push"
+  ],
+  "program":  "/mnt/data/zainulabideen/zain-exp/notebooks/Activation/sweep.py",
+  "codePath":  "sweep.py",
+  "codePathLocal":  "sweep.py",
+  "git":  {
+    "remote":  "https://github.com/deepnevro/Activation.git",
+    "commit":  "34b8d2e8f9a0c5751333310e69fa0c1056381deb"
+  },
+  "email":  "deepnevro@gmail.com",
+  "root":  "/mnt/data/zainulabideen/zain-exp/notebooks/Activation",
+  "host":  "deeplens-k3s-node1",
+  "executable":  "/mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python",
+  "cpu_count":  112,
+  "cpu_count_logical":  224,
+  "gpu":  "NVIDIA H100 80GB HBM3",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1560765693952",
+      "used":  "708205846528"
+    }
+  },
+  "memory":  {
+    "total":  "2164089937920"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-39c684a5-fde6-83d7-1663-0859795881ae"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-132944c4-b689-2b5f-89a4-d730401677ab"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bc6c3e3c-9b90-09ca-c034-774961847c54"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea"
+    }
+  ],
+  "cudaVersion":  "12.4",
+  "writerId":  "oz75ofyjp8ovu1ftttzp3wyx4l4oquya"
+}

wandb/run-20260809_035819-cvzjg5ej/logs/debug-core.log

@@ -0,0 +1,58 @@
+{"time":"2026-08-09T03:56:53.721154509Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpyqomfhgp/port-2869678.txt","pid":2869678,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-08-09T03:56:53.722258921Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":2869678}
+{"time":"2026-08-09T03:56:53.722234577Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-2869678-2909401-2521299324/socket","Net":"unix"}}
+{"time":"2026-08-09T03:56:53.900147851Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-08-09T03:58:19.204299995Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"2(@)"}
+{"time":"2026-08-09T03:58:19.282614427Z","level":"INFO","msg":"handleInformInit: received","streamId":"cvzjg5ej","id":"2(@)"}
+{"time":"2026-08-09T03:58:19.544211049Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"cvzjg5ej","id":"2(@)"}
+{"time":"2026-08-09T03:58:24.897517362Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"xv8tfjtqu29x"}
+{"time":"2026-08-09T05:00:38.67636271Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"xv8tfjtqu29x"}
+{"time":"2026-08-09T05:00:40.520576799Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"xv8tfjtqu29x"}
+{"time":"2026-08-09T05:00:40.552184373Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"cvzjg5ej","id":"2(@)"}
+{"time":"2026-08-09T05:00:40.553143275Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"cvzjg5ej","id":"2(@)"}
+{"time":"2026-08-09T05:00:42.608126464Z","level":"INFO","msg":"processOutgoingData: finished","id":"2(@)"}
+{"time":"2026-08-09T05:00:42.608106419Z","level":"INFO","msg":"connection: closing","id":"2(@)"}
+{"time":"2026-08-09T05:00:42.608216462Z","level":"INFO","msg":"connection: closed successfully","id":"2(@)"}
+{"time":"2026-08-09T05:00:42.608222262Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"2(@)"}
+{"time":"2026-08-09T05:00:50.241711721Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"3(@)"}
+{"time":"2026-08-09T05:00:50.316234404Z","level":"INFO","msg":"handleInformInit: received","streamId":"59pftr14","id":"3(@)"}
+{"time":"2026-08-09T05:00:50.57530289Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"59pftr14","id":"3(@)"}
+{"time":"2026-08-09T05:00:55.905488223Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"4jxnuihlia82"}
+{"time":"2026-08-09T05:57:13.765576248Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"4jxnuihlia82"}
+{"time":"2026-08-09T05:57:15.666017984Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"4jxnuihlia82"}
+{"time":"2026-08-09T05:57:15.999725293Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"59pftr14","id":"3(@)"}
+{"time":"2026-08-09T05:57:16.001240777Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"59pftr14","id":"3(@)"}
+{"time":"2026-08-09T05:57:18.068849513Z","level":"INFO","msg":"connection: closing","id":"3(@)"}
+{"time":"2026-08-09T05:57:18.068936758Z","level":"INFO","msg":"connection: closed successfully","id":"3(@)"}
+{"time":"2026-08-09T05:57:18.068853914Z","level":"INFO","msg":"processOutgoingData: finished","id":"3(@)"}
+{"time":"2026-08-09T05:57:18.068948948Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"3(@)"}
+{"time":"2026-08-09T05:57:26.287713024Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"4(@)"}
+{"time":"2026-08-09T05:57:26.365087395Z","level":"INFO","msg":"handleInformInit: received","streamId":"m1dnjnh6","id":"4(@)"}
+{"time":"2026-08-09T05:57:26.623707494Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"m1dnjnh6","id":"4(@)"}
+{"time":"2026-08-09T05:57:31.962348796Z","level":"INFO","msg":"connection: cancelling request","id":"4(@)","requestId":"rz2ldpq16nht"}
+{"time":"2026-08-09T07:02:02.153459338Z","level":"INFO","msg":"connection: cancelling request","id":"4(@)","requestId":"rz2ldpq16nht"}
+{"time":"2026-08-09T07:02:04.144735717Z","level":"INFO","msg":"connection: cancelling request","id":"4(@)","requestId":"rz2ldpq16nht"}
+{"time":"2026-08-09T07:02:04.180963791Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"m1dnjnh6","id":"4(@)"}
+{"time":"2026-08-09T07:02:04.181861619Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"m1dnjnh6","id":"4(@)"}
+{"time":"2026-08-09T07:02:06.233097548Z","level":"INFO","msg":"connection: closing","id":"4(@)"}
+{"time":"2026-08-09T07:02:06.233183816Z","level":"INFO","msg":"connection: closed successfully","id":"4(@)"}
+{"time":"2026-08-09T07:02:06.233105065Z","level":"INFO","msg":"processOutgoingData: finished","id":"4(@)"}
+{"time":"2026-08-09T07:02:06.233192773Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"4(@)"}
+{"time":"2026-08-09T07:02:13.885910885Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"5(@)"}
+{"time":"2026-08-09T07:02:13.954769181Z","level":"INFO","msg":"handleInformInit: received","streamId":"uvqyddz0","id":"5(@)"}
+{"time":"2026-08-09T07:02:14.215272022Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"uvqyddz0","id":"5(@)"}
+{"time":"2026-08-09T07:02:19.530617395Z","level":"INFO","msg":"connection: cancelling request","id":"5(@)","requestId":"fk71ydt1h8f7"}
+{"time":"2026-08-09T07:03:39.9336748Z","level":"INFO","msg":"connection: cancelling request","id":"5(@)","requestId":"fk71ydt1h8f7"}
+{"time":"2026-08-09T07:03:40.29901782Z","level":"INFO","msg":"connection: closing","id":"5(@)"}
+{"time":"2026-08-09T07:03:40.299111085Z","level":"INFO","msg":"connection: closed successfully","id":"5(@)"}
+{"time":"2026-08-09T07:03:40.299007456Z","level":"INFO","msg":"processOutgoingData: finished","id":"5(@)"}
+{"time":"2026-08-09T07:03:40.299122073Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"5(@)"}
+{"time":"2026-08-09T07:03:42.349833633Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-08-09T07:03:42.349942779Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-08-09T07:03:42.34985531Z","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-08-09T07:03:42.349954994Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-08-09T07:03:42.353940558Z","level":"INFO","msg":"server: parent process exited, terminating service process"}
+{"time":"2026-08-09T07:03:42.353988499Z","level":"INFO","msg":"server: is shutting down"}
+{"time":"2026-08-09T07:03:42.354128892Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-2869678-2909401-2521299324/socket","Net":"unix"}}
+{"time":"2026-08-09T07:03:42.354189401Z","level":"INFO","msg":"server: forced shutdown"}
+{"time":"2026-08-09T07:03:42.354198506Z","level":"ERROR","msg":"main: Serve() returned error","error":"forced shutdown"}

wandb/run-20260809_035819-cvzjg5ej/logs/debug-internal.log

@@ -0,0 +1,535 @@
+{"time":"2026-08-09T03:58:19.282783063Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-09T03:58:19.283092454Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-09T03:58:19.544029023Z","level":"INFO","msg":"stream: created new stream","id":"cvzjg5ej"}
+{"time":"2026-08-09T03:58:19.544113293Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-09T03:58:19.544204994Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-09T03:58:19.544224319Z","level":"INFO","msg":"writer: started","stream_id":"cvzjg5ej"}
+{"time":"2026-08-09T03:58:19.544236358Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-09T03:58:20.49399574Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-09T03:58:20.585402614Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T03:58:35.494346021Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":0,"events_lines":2,"console_offset":1,"console_lines":4,"uploaded_len":2}
+{"time":"2026-08-09T03:58:35.60993089Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T03:58:50.494317925Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":2,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T03:58:50.598887827Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T03:59:03.877770359Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":430}
+{"time":"2026-08-09T03:59:03.878186402Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":15}
+{"time":"2026-08-09T03:59:03.883932073Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":2508}
+{"time":"2026-08-09T03:59:03.883962742Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":1}
+{"time":"2026-08-09T03:59:03.894578099Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":5231}
+{"time":"2026-08-09T03:59:03.89461253Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":1}
+{"time":"2026-08-09T03:59:03.896155708Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":5757}
+{"time":"2026-08-09T03:59:03.911398768Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":1574}
+{"time":"2026-08-09T03:59:03.912624837Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":7504}
+{"time":"2026-08-09T03:59:03.912762308Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":23}
+{"time":"2026-08-09T03:59:03.930332533Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":11798}
+{"time":"2026-08-09T03:59:03.933304856Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":220}
+{"time":"2026-08-09T03:59:03.934700228Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":12563}
+{"time":"2026-08-09T03:59:03.93682743Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":829}
+{"time":"2026-08-09T03:59:03.944832244Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":14752}
+{"time":"2026-08-09T03:59:03.949817464Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":444}
+{"time":"2026-08-09T03:59:03.951711505Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":15916}
+{"time":"2026-08-09T03:59:03.954435107Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":963}
+{"time":"2026-08-09T03:59:03.954578098Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":16937}
+{"time":"2026-08-09T03:59:03.955021925Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":119}
+{"time":"2026-08-09T03:59:05.531386979Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":4,"events_lines":2,"console_offset":4,"console_lines":2}
+{"time":"2026-08-09T03:59:06.77622938Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T03:59:20.494664032Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":6,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T03:59:20.585865699Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T03:59:35.494257506Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":8,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T03:59:35.590824829Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T03:59:50.509459052Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":10,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T03:59:51.514395751Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:00:05.494701492Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":12,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T04:00:05.595719639Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:00:20.494249891Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":14,"events_lines":2,"console_offset":6,"console_lines":2}
+{"time":"2026-08-09T04:00:20.621562887Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:00:35.517415448Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":16,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T04:00:36.625730885Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:00:50.515257082Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":1,"events_offset":18,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T04:00:51.502793371Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:01:05.494754271Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":20,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T04:01:05.610106165Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:01:20.494400825Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":22,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T04:01:20.601141197Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:01:35.512242154Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":4,"history_lines":1,"events_offset":24,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T04:01:36.40914958Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:01:50.494503273Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":26,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T04:01:50.59999154Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:02:05.515046749Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":28,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T04:02:06.39796112Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:02:20.509714476Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":30,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T04:02:21.325931349Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:02:35.494334742Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":32,"events_lines":2,"console_offset":7,"console_lines":10}
+{"time":"2026-08-09T04:02:35.59436135Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:02:50.494250642Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":34,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:02:50.613518689Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:03:05.512934467Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":36,"events_lines":2,"console_offset":16,"console_lines":2}
+{"time":"2026-08-09T04:03:06.548287522Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:03:20.494529587Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":38,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:03:20.582860961Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:03:35.494527188Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":40,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:03:35.607972831Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:03:50.509533438Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":42,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:03:51.431848812Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:04:05.494917144Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":44,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:04:05.596754607Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:04:20.507762047Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":46,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:04:21.447073466Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:04:35.509308038Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":1,"events_offset":48,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:04:36.471928253Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:04:50.494530256Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":50,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:04:50.592073322Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:05:05.494375076Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":52,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:05:05.604310138Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:05:20.512379195Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":1,"events_offset":54,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:05:21.491774019Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:05:35.494839434Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":56,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:05:35.592901032Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:05:50.49492532Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":58,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:05:50.59888656Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:06:05.514147554Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":12,"history_lines":1,"events_offset":60,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:06:06.480804234Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:06:20.514491453Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":1,"events_offset":62,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T04:06:21.452884273Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:06:35.494286866Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":64,"events_lines":2,"console_offset":18,"console_lines":11}
+{"time":"2026-08-09T04:06:35.609203544Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:06:50.494175318Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":66,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:06:50.614345309Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:07:05.515682555Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":1,"events_offset":68,"events_lines":2,"console_offset":28,"console_lines":2}
+{"time":"2026-08-09T04:07:06.427671059Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:07:20.494222615Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":70,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:07:20.573884809Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:07:35.507952716Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":72,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:07:36.475332313Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:07:50.494214091Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":74,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:07:50.753409176Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:08:05.494655816Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":76,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:08:05.586476128Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:08:20.511325739Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":1,"events_offset":78,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:08:21.410353753Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:08:35.509606509Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":17,"history_lines":1,"events_offset":80,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:08:36.420905539Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:08:50.494669295Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":82,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:08:50.604423316Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:09:05.494987144Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":84,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:09:05.607036127Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:09:20.510454935Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":18,"history_lines":1,"events_offset":86,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:09:21.483617947Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:09:35.494837249Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":88,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:09:35.610548227Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:09:50.507950635Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":19,"history_lines":1,"events_offset":90,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:09:51.511448364Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:10:05.520637653Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":20,"history_lines":1,"events_offset":92,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T04:10:06.438117864Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:10:20.49470281Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":94,"events_lines":2,"console_offset":30,"console_lines":11}
+{"time":"2026-08-09T04:10:20.613679557Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:10:35.494489332Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":96,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:10:35.60676065Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:10:50.50862038Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":21,"history_lines":1,"events_offset":98,"events_lines":2,"console_offset":40,"console_lines":2}
+{"time":"2026-08-09T04:10:51.485985122Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:11:05.494135887Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":100,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:11:05.612280741Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:11:20.494686423Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":102,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:11:20.598556666Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:11:35.515806086Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":22,"history_lines":1,"events_offset":104,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:11:36.484944909Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:11:50.494589134Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":106,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:11:50.596223304Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:12:05.513002951Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":23,"history_lines":1,"events_offset":108,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:12:06.484107421Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:12:20.494320474Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":110,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:12:20.602551441Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:12:35.508125938Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":24,"history_lines":1,"events_offset":112,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:12:36.475135091Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:12:50.494789433Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":114,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:12:50.602319198Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:13:05.512828195Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":25,"history_lines":1,"events_offset":116,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:13:06.512719752Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:13:20.494722423Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":118,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:13:20.603843019Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:13:35.494491559Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":120,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:13:35.616937106Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:13:50.509869659Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":26,"history_lines":1,"events_offset":122,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:13:51.486433037Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:14:05.512616517Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":27,"history_lines":1,"events_offset":124,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T04:14:06.69457877Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:14:20.494738585Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":126,"events_lines":2,"console_offset":42,"console_lines":11}
+{"time":"2026-08-09T04:14:20.598086436Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:14:35.494861868Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":128,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:14:35.606033152Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:14:50.50832981Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":28,"history_lines":1,"events_offset":130,"events_lines":2,"console_offset":52,"console_lines":2}
+{"time":"2026-08-09T04:14:51.451962725Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:15:05.494325431Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":132,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:15:05.599151827Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:15:20.512939886Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":29,"history_lines":1,"events_offset":134,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:15:21.39376644Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:15:35.494565716Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":136,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:15:35.597861377Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:15:50.494985286Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":138,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:15:50.61366858Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:16:05.509505556Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":30,"history_lines":1,"events_offset":140,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:16:06.464859955Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:16:20.512387578Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":31,"history_lines":1,"events_offset":142,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:16:21.456484085Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:16:35.494311848Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":144,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:16:35.614320532Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:16:50.494123741Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":146,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:16:50.597183977Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:17:05.507650032Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":32,"history_lines":1,"events_offset":148,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:17:06.496156926Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:17:20.494465778Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":150,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:17:20.608687507Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:17:35.494337018Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":152,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:17:35.598855029Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:17:50.511814815Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":33,"history_lines":1,"events_offset":154,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:17:51.379590142Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:18:05.507926178Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":34,"history_lines":1,"events_offset":156,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T04:18:06.425217866Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:18:20.494329994Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":158,"events_lines":2,"console_offset":54,"console_lines":13}
+{"time":"2026-08-09T04:18:20.597550528Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:18:35.494357619Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":160,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:18:35.590558781Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:18:50.526101792Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":1,"events_offset":162,"events_lines":2,"console_offset":66,"console_lines":2}
+{"time":"2026-08-09T04:18:51.756248392Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:19:05.494443616Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":164,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:19:05.600417147Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:19:20.509453223Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":36,"history_lines":1,"events_offset":166,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:19:21.44743209Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:19:35.494186649Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":168,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:19:35.598016278Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:19:50.494255354Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":170,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:19:50.607645062Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:20:05.512253299Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":37,"history_lines":1,"events_offset":172,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:20:06.424682607Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:20:20.514248726Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":38,"history_lines":1,"events_offset":174,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:20:21.478103921Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:20:35.494263985Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":176,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:20:35.622961052Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:20:50.494962002Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":178,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:20:50.611850033Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:21:05.517835067Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":39,"history_lines":1,"events_offset":180,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:21:06.404943823Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:21:20.494738486Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":182,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:21:20.602524281Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:21:35.508789164Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":40,"history_lines":1,"events_offset":184,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:21:36.443085607Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:21:50.511158411Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":41,"history_lines":1,"events_offset":186,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T04:21:51.460178906Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:22:05.494482305Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":188,"events_lines":2,"console_offset":68,"console_lines":11}
+{"time":"2026-08-09T04:22:05.586362541Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:22:20.494728352Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":190,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:22:20.599759017Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:22:35.513754334Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":1,"events_offset":192,"events_lines":2,"console_offset":78,"console_lines":2}
+{"time":"2026-08-09T04:22:36.516635185Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:22:50.494586997Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":194,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:22:50.571528834Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:23:05.494079314Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":196,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:23:05.598391284Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:23:20.514134103Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":43,"history_lines":1,"events_offset":198,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:23:21.457824889Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:23:35.494479121Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":200,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:23:35.592895767Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:23:50.518950272Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":1,"events_offset":202,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:23:51.404637571Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:24:05.494720966Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":204,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:24:05.600554603Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:24:20.51001913Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":45,"history_lines":1,"events_offset":206,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:24:21.455271551Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:24:35.49440308Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":208,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:24:35.587198082Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:24:50.512521872Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":1,"events_offset":210,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:24:51.470179091Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:25:05.494790615Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":212,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:25:05.614937066Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:25:20.494768189Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":214,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:25:20.603889882Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:25:35.510484888Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":47,"history_lines":1,"events_offset":216,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:25:36.442214296Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:25:50.512621688Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":48,"history_lines":1,"events_offset":218,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T04:25:51.447687898Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:26:05.494610601Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":220,"events_lines":2,"console_offset":80,"console_lines":11}
+{"time":"2026-08-09T04:26:05.591575531Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:26:20.49440059Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":222,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:26:20.578020301Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:26:35.508448912Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":1,"events_offset":224,"events_lines":2,"console_offset":90,"console_lines":2}
+{"time":"2026-08-09T04:26:36.403987032Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:26:50.494361014Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":226,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:26:50.602435073Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:27:05.507911257Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":50,"history_lines":1,"events_offset":228,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:27:06.445082363Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:27:20.494195074Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":230,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:27:20.614605046Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:27:35.494248784Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":232,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:27:35.607893177Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:27:50.514285947Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":51,"history_lines":1,"events_offset":234,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:27:51.395463717Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:28:05.514892856Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":52,"history_lines":1,"events_offset":236,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:28:06.423194446Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:28:20.49418678Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":238,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:28:20.597735182Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:28:35.494119279Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":240,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:28:35.592493251Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:28:50.5094627Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":53,"history_lines":1,"events_offset":242,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:28:51.45483569Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:29:05.494391144Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":244,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:29:05.591119154Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:29:20.494298466Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":246,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:29:20.596120363Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:29:35.513763328Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":54,"history_lines":1,"events_offset":248,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:29:36.465513234Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:29:50.513046942Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":55,"history_lines":1,"events_offset":250,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T04:29:51.472667426Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:30:05.494740686Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":252,"events_lines":2,"console_offset":92,"console_lines":11}
+{"time":"2026-08-09T04:30:05.586971324Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:30:20.515798418Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":56,"history_lines":1,"events_offset":254,"events_lines":2,"console_offset":102,"console_lines":2}
+{"time":"2026-08-09T04:30:21.494651995Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:30:35.494231232Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":256,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:30:35.58729982Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:30:50.494976906Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":258,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:30:50.601902681Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:31:05.510769779Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":57,"history_lines":1,"events_offset":260,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:31:06.780111879Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:31:20.49453644Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":262,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:31:20.599562246Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:31:35.510004044Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":58,"history_lines":1,"events_offset":264,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:31:36.458139374Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:31:50.494509411Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":266,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:31:50.575654427Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:32:05.512223467Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":59,"history_lines":1,"events_offset":268,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:32:06.469562043Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:32:20.494205295Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":270,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:32:20.584293928Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:32:35.509793195Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":60,"history_lines":1,"events_offset":272,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:32:36.535545937Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:32:50.494041072Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":274,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:32:50.594918103Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:33:05.49403774Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":276,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:33:05.607036499Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:33:20.511846379Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":61,"history_lines":1,"events_offset":278,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:33:21.48470859Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:33:35.515062737Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":62,"history_lines":1,"events_offset":280,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T04:33:36.506124978Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:33:50.4946323Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":282,"events_lines":2,"console_offset":104,"console_lines":11}
+{"time":"2026-08-09T04:33:50.612161047Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:34:05.494640716Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":284,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:34:05.587261158Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:34:20.507829114Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":63,"history_lines":1,"events_offset":286,"events_lines":2,"console_offset":114,"console_lines":2}
+{"time":"2026-08-09T04:34:21.464551249Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:34:35.494345284Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":288,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:34:35.606869158Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:34:50.494397686Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":290,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:34:50.599232901Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:35:05.508409123Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":64,"history_lines":1,"events_offset":292,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:35:06.433911202Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:35:20.494170669Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":294,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:35:20.613977234Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:35:35.509564094Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":65,"history_lines":1,"events_offset":296,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:35:36.475546977Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:35:50.508108638Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":66,"history_lines":1,"events_offset":298,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:35:51.45083907Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:36:05.494407838Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":300,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:36:05.588011312Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:36:20.494349239Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":302,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:36:20.614846145Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:36:35.510501251Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":67,"history_lines":1,"events_offset":304,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:36:36.571048121Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:36:50.494472539Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":306,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:36:50.609659468Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:37:05.494605434Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":308,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:37:05.590686834Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:37:20.510275452Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":68,"history_lines":1,"events_offset":310,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:37:21.512921238Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:37:35.50955353Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":69,"history_lines":1,"events_offset":312,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T04:37:36.500587981Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:37:50.494388687Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":314,"events_lines":2,"console_offset":116,"console_lines":13}
+{"time":"2026-08-09T04:37:50.598720388Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:38:05.494193752Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":316,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:38:05.638784236Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:38:20.530757676Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":70,"history_lines":1,"events_offset":318,"events_lines":2,"console_offset":128,"console_lines":2}
+{"time":"2026-08-09T04:38:21.632845005Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:38:35.494438726Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":320,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:38:35.594398657Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:38:50.514389068Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":71,"history_lines":1,"events_offset":322,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:38:51.399815543Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:39:05.494294314Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":324,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:39:05.605503032Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:39:20.494680808Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":326,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:39:20.611296788Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:39:35.510121829Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":72,"history_lines":1,"events_offset":328,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:39:36.465679131Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:39:50.512824888Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":73,"history_lines":1,"events_offset":330,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:39:51.478305625Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:40:05.494062759Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":332,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:40:05.580589634Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:40:20.494165168Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":334,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:40:20.607065019Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:40:35.512178504Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":74,"history_lines":1,"events_offset":336,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:40:36.602134021Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:40:50.494375721Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":338,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:40:50.583296822Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:41:05.518526338Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":75,"history_lines":1,"events_offset":340,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:41:06.435876732Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:41:20.494762288Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":342,"events_lines":2,"console_offset":130,"console_lines":6}
+{"time":"2026-08-09T04:41:20.603430974Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:41:35.514811938Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":76,"history_lines":1,"events_offset":344,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T04:41:36.473044784Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:41:50.494076612Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":346,"events_lines":2,"console_offset":131,"console_lines":1}
+{"time":"2026-08-09T04:41:50.601570222Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:42:05.511952413Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":77,"history_lines":1,"events_offset":348,"events_lines":2,"console_offset":136,"console_lines":6}
+{"time":"2026-08-09T04:42:06.491073426Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:42:20.494758077Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":350,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:42:20.606830814Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:42:35.494254813Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":352,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:42:35.595459447Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:42:50.509597712Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":78,"history_lines":1,"events_offset":354,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:42:51.385361692Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:43:05.494296853Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":356,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:43:05.584770185Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:43:20.494194739Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":358,"events_lines":2,"console_offset":142,"console_lines":2}
+{"time":"2026-08-09T04:43:20.605028176Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:43:35.511553269Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":79,"history_lines":1,"events_offset":360,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:43:36.422803214Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:43:50.51209143Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":80,"history_lines":1,"events_offset":362,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:43:51.371685359Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:44:05.494414492Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":364,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:44:05.591512538Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:44:20.494476569Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":366,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:44:20.61786034Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:44:35.494661644Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":368,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:44:35.600856267Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:44:50.512310947Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":81,"history_lines":1,"events_offset":370,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:44:51.463291334Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:45:05.494886009Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":372,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:45:05.584389307Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:45:20.494304778Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":374,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:45:20.594809211Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:45:35.510247414Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":82,"history_lines":1,"events_offset":376,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:45:36.477447729Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:45:50.515143514Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":83,"history_lines":1,"events_offset":378,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T04:45:51.542141686Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:46:05.494993318Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":380,"events_lines":2,"console_offset":143,"console_lines":10}
+{"time":"2026-08-09T04:46:05.60243871Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:46:20.494284253Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":382,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:46:20.608258694Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:46:35.494646706Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":384,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:46:35.598907205Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:46:50.514394328Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":84,"history_lines":1,"events_offset":386,"events_lines":2,"console_offset":152,"console_lines":2}
+{"time":"2026-08-09T04:46:51.494084865Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:47:05.494697595Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":388,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:47:05.586587514Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:47:20.494438098Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":390,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:47:20.597334702Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:47:35.521023173Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":85,"history_lines":1,"events_offset":392,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:47:36.445223277Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:47:50.494788572Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":394,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:47:50.590619221Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:48:05.49423104Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":396,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:48:05.600318316Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:48:20.509807197Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":86,"history_lines":1,"events_offset":398,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:48:21.489740194Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:48:35.494208141Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":400,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:48:35.591006311Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:48:50.515494589Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":87,"history_lines":1,"events_offset":402,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:48:51.427014966Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:49:05.494430456Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":404,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:49:05.619853108Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:49:20.494243574Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":406,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:49:20.601941856Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:49:35.513932029Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":88,"history_lines":1,"events_offset":408,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:49:36.457240216Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:49:50.494940016Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":410,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:49:50.581649644Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:50:05.494720247Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":412,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:50:05.608214779Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:50:20.494063881Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":414,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:50:20.603378552Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:50:35.519647341Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":89,"history_lines":1,"events_offset":416,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:50:36.503626552Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:50:50.51423598Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":90,"history_lines":1,"events_offset":418,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T04:50:51.459574025Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:51:05.494376593Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":420,"events_lines":2,"console_offset":154,"console_lines":11}
+{"time":"2026-08-09T04:51:05.591793843Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:51:20.494529619Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":422,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:51:20.639974097Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:51:35.494240177Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":424,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:51:35.59915761Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:51:50.508256175Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":91,"history_lines":1,"events_offset":426,"events_lines":2,"console_offset":164,"console_lines":2}
+{"time":"2026-08-09T04:51:51.432054336Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:52:05.494231656Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":428,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:52:05.590635379Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:52:20.494183843Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":430,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:52:20.609362605Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:52:35.512240353Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":92,"history_lines":1,"events_offset":432,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:52:36.547000959Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:52:50.494409211Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":434,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:52:50.611077694Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:53:05.494665223Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":436,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:53:05.604913862Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:53:20.513692859Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":93,"history_lines":1,"events_offset":438,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:53:21.370170293Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:53:35.494265173Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":440,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:53:35.588156095Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:53:50.515309054Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":94,"history_lines":1,"events_offset":442,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:53:51.508264221Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:54:05.494267129Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":444,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:54:05.598594866Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:54:20.494465833Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":446,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:54:20.599852885Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:54:35.512705924Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":95,"history_lines":1,"events_offset":448,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:54:36.45915457Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:54:50.494329668Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":450,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:54:50.606394944Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:55:05.494173382Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":452,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:55:05.606685612Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:55:20.494519313Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":454,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:55:20.600694415Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:55:35.509767756Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":96,"history_lines":1,"events_offset":456,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:55:36.518053017Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:55:50.511111663Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":97,"history_lines":1,"events_offset":458,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T04:55:51.460606406Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:56:05.494354453Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":460,"events_lines":2,"console_offset":166,"console_lines":11}
+{"time":"2026-08-09T04:56:05.588540788Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:56:20.49440425Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":462,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:56:20.595050994Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:56:35.494373973Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":464,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:56:35.609327704Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:56:50.51145138Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":98,"history_lines":1,"events_offset":466,"events_lines":2,"console_offset":176,"console_lines":2}
+{"time":"2026-08-09T04:56:51.609244462Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:57:05.494213007Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":468,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:57:05.598515761Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:57:20.494450349Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":470,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:57:20.596086948Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:57:35.512624981Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":99,"history_lines":1,"events_offset":472,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:57:36.509474508Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:57:50.494590228Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":474,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:57:50.582329386Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:58:05.494166836Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":476,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:58:05.584219865Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:58:20.512387587Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":100,"history_lines":1,"events_offset":478,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:58:21.442742736Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:58:35.494461513Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":480,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:58:35.704810548Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:58:50.511948147Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":101,"history_lines":1,"events_offset":482,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:58:51.416645299Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:59:05.494670676Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":484,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:59:05.594712461Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:59:20.494633402Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":486,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:59:20.592347073Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:59:35.510861324Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":102,"history_lines":1,"events_offset":488,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:59:36.435711548Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T04:59:50.494866252Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":490,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T04:59:50.60422558Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:00:05.510416009Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":103,"history_lines":1,"events_offset":492,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:00:06.405297236Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:00:20.512054289Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":104,"history_lines":2,"events_offset":494,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:00:21.415448411Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:00:35.494696267Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":496,"events_lines":2,"console_offset":178,"console_lines":13}
+{"time":"2026-08-09T05:00:35.595572837Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:00:39.446368581Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-09T05:00:39.464038823Z","level":"INFO","msg":"filestream: sending request","total_files":3,"history_offset":106,"history_lines":1,"console_offset":190,"console_lines":31,"uploaded_len":3,"complete":true,"exit_code":0}
+{"time":"2026-08-09T05:00:40.496960895Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:00:40.498370506Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2026-08-09T05:00:40.552225292Z","level":"INFO","msg":"stream: finishing up"}
+{"time":"2026-08-09T05:00:40.552256718Z","level":"INFO","msg":"handler: closed"}
+{"time":"2026-08-09T05:00:40.552367204Z","level":"INFO","msg":"sender: closed"}
+{"time":"2026-08-09T05:00:40.552371118Z","level":"INFO","msg":"stream: all finished"}

wandb/run-20260809_035819-cvzjg5ej/logs/debug.log

@@ -0,0 +1,28 @@
+2026-08-09 03:58:19,281 INFO    MainThread:2921752 [wandb_setup.py:_flush():81] Current SDK version is 0.28.1
+2026-08-09 03:58:19,281 INFO    MainThread:2921752 [wandb_setup.py:_flush():81] Configure stats pid to 2921752
+2026-08-09 03:58:19,281 INFO    MainThread:2921752 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-08-09 03:58:19,281 INFO    MainThread:2921752 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/Activation/wandb/run-20260809_035819-cvzjg5ej/logs/debug.log
+2026-08-09 03:58:19,281 INFO    MainThread:2921752 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/Activation/wandb/run-20260809_035819-cvzjg5ej/logs/debug-internal.log
+2026-08-09 03:58:19,281 INFO    MainThread:2921752 [wandb_init.py:init():772] calling init triggers
+2026-08-09 03:58:19,281 INFO    MainThread:2921752 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+config: {'_wandb': {}}
+2026-08-09 03:58:19,281 INFO    MainThread:2921752 [wandb_init.py:init():820] starting backend
+2026-08-09 03:58:19,281 INFO    MainThread:2921752 [wandb_init.py:init():826] Connected to an existing wandb-core service via WANDB_SERVICE
+2026-08-09 03:58:19,281 INFO    MainThread:2921752 [wandb_init.py:init():835] sending inform_init request
+2026-08-09 03:58:19,544 INFO    MainThread:2921752 [wandb_init.py:init():840] backend started and connected
+2026-08-09 03:58:19,547 INFO    MainThread:2921752 [wandb_init.py:init():910] updated telemetry
+2026-08-09 03:58:19,554 INFO    MainThread:2921752 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-09 03:58:19,820 INFO    MainThread:2921752 [wandb_init.py:init():978] starting run threads in backend
+2026-08-09 03:58:19,892 INFO    MainThread:2921752 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-09 03:58:19,892 INFO    MainThread:2921752 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-09 03:58:19,892 INFO    MainThread:2921752 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-09 03:58:19,892 INFO    MainThread:2921752 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-09 03:58:19,895 INFO    MainThread:2921752 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-09 03:58:19,896 INFO    MainThread:2921752 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.15.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 94, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'mlp', 'activation': 's10', 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/mlp-s10-94L_run', 'per_device_train_batch_size': 128, 'num_train_epochs': 1, 'max_steps': 1500, 'learning_rate': 0.001, 'lr_scheduler_type': 'constant', 'lr_scheduler_kwargs': None, 'warmup_steps': 0, 'optim': 'adamw_torch_fused', 'optim_args': None, 'weight_decay': 0.01, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 4, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-mlp-s10-94L-15.9M-20260809-035818', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 50, 'eval_delay': 0, 'per_device_eval_batch_size': 128, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': True, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/A-mlp-s10-94L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
+2026-08-09 03:58:19,899 INFO    MainThread:2921752 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 15949440 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x14605416ecd0>>
+2026-08-09 03:58:19,899 INFO    MainThread:2921752 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 15949440 None
+2026-08-09 05:00:38,675 INFO    MainThread:2921752 [wandb_run.py:_finish():2383] finishing run deepnevro-deepnevro/huggingface/cvzjg5ej
+2026-08-09 05:00:38,675 INFO    MainThread:2921752 [wandb_run.py:_atexit_cleanup():2588] got exitcode: 0
+2026-08-09 05:00:38,676 INFO    MainThread:2921752 [wandb_run.py:_restore():2570] restore
+2026-08-09 05:00:38,676 INFO    MainThread:2921752 [wandb_run.py:_restore():2576] restore done
+2026-08-09 05:00:40,551 INFO    MainThread:2921752 [wandb_run.py:_footer_sync_info():3993] logging synced files

wandb/run-20260809_035819-cvzjg5ej/run-cvzjg5ej.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42e28ca89c157db3fd0e1b8d55b132aed26283c09abd2a439124929bd964170e
+size 17095417

wandb/run-20260809_050050-59pftr14/files/config.yaml

@@ -0,0 +1,433 @@
+_name_or_path:
+    value: ""
+_wandb:
+    value:
+        cli_version: 0.28.1
+        e:
+            ou0ddnt36zj23smd7g76a30q6z1cg54s:
+                args:
+                    - --config
+                    - configs/baseline.yaml
+                    - --variants
+                    - glu-linear-94L
+                    - --push
+                codePath: sweep.py
+                codePathLocal: sweep.py
+                cpu_count: 112
+                cpu_count_logical: 224
+                cudaVersion: "12.4"
+                disk:
+                    /:
+                        total: "1560765693952"
+                        used: "708235583488"
+                email: deepnevro@gmail.com
+                executable: /mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python
+                git:
+                    commit: 34b8d2e8f9a0c5751333310e69fa0c1056381deb
+                    remote: https://github.com/deepnevro/Activation.git
+                gpu: NVIDIA H100 80GB HBM3
+                gpu_count: 8
+                gpu_nvidia:
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-39c684a5-fde6-83d7-1663-0859795881ae
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-132944c4-b689-2b5f-89a4-d730401677ab
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bc6c3e3c-9b90-09ca-c034-774961847c54
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea
+                host: deeplens-k3s-node1
+                memory:
+                    total: "2164089937920"
+                os: Linux-5.15.0-126-generic-x86_64-with-glibc2.35
+                program: /mnt/data/zainulabideen/zain-exp/notebooks/Activation/sweep.py
+                python: CPython 3.11.15
+                root: /mnt/data/zainulabideen/zain-exp/notebooks/Activation
+                startedAt: "2026-08-09T05:00:50.313572Z"
+                writerId: ou0ddnt36zj23smd7g76a30q6z1cg54s
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.11.15
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 19
+                - 41
+                - 62
+                - 66
+            "4": 3.11.15
+            "5": 0.28.1
+            "6": 5.15.0.dev0
+            "9":
+                "1": transformers_trainer
+            "12": 0.28.1
+            "13": linux-x86_64
+accelerator_config:
+    value:
+        dispatch_batches: null
+        even_batches: true
+        gradient_accumulation_kwargs: null
+        non_blocking: false
+        split_batches: false
+        use_seedable_sampler: true
+activation:
+    value: linear
+adam_beta1:
+    value: 0.9
+adam_beta2:
+    value: 0.999
+adam_epsilon:
+    value: 1e-08
+architectures:
+    value: null
+attention_bias:
+    value: false
+attention_dropout:
+    value: 0
+auto_find_batch_size:
+    value: false
+average_tokens_across_devices:
+    value: true
+batch_eval_metrics:
+    value: false
+bf16:
+    value: true
+bf16_full_eval:
+    value: false
+bos_token_id:
+    value: 1
+chunk_size_feed_forward:
+    value: 0
+data_seed:
+    value: 42
+dataloader_drop_last:
+    value: false
+dataloader_in_order:
+    value: true
+dataloader_multiprocessing_context:
+    value: null
+dataloader_num_workers:
+    value: 0
+dataloader_persistent_workers:
+    value: false
+dataloader_pin_memory:
+    value: true
+dataloader_prefetch_factor:
+    value: null
+ddp_backend:
+    value: null
+ddp_broadcast_buffers:
+    value: null
+ddp_bucket_cap_mb:
+    value: null
+ddp_find_unused_parameters:
+    value: null
+ddp_static_graph:
+    value: null
+ddp_timeout:
+    value: 1800
+debug:
+    value: []
+deepspeed:
+    value: null
+disable_tqdm:
+    value: false
+do_eval:
+    value: true
+do_predict:
+    value: false
+do_train:
+    value: false
+dtype:
+    value: null
+enable_jit_checkpoint:
+    value: false
+eos_token_id:
+    value: 2
+eval_accumulation_steps:
+    value: null
+eval_delay:
+    value: 0
+eval_do_concat_batches:
+    value: true
+eval_on_start:
+    value: false
+eval_steps:
+    value: 50
+eval_strategy:
+    value: steps
+eval_use_gather_object:
+    value: false
+fp16:
+    value: false
+fp16_full_eval:
+    value: false
+fsdp:
+    value: null
+fsdp_config:
+    value: null
+full_determinism:
+    value: false
+gradient_accumulation_steps:
+    value: 4
+gradient_checkpointing:
+    value: false
+gradient_checkpointing_kwargs:
+    value: null
+greater_is_better:
+    value: null
+head_dim:
+    value: 32
+hidden_act:
+    value: silu
+hidden_size:
+    value: 128
+hub_always_push:
+    value: false
+hub_model_id:
+    value: w-ahmad/A-glu-linear-94L
+hub_private_repo:
+    value: null
+hub_revision:
+    value: null
+hub_strategy:
+    value: every_save
+hub_token:
+    value: <HUB_TOKEN>
+id2label:
+    value:
+        "0": LABEL_0
+        "1": LABEL_1
+ignore_data_skip:
+    value: false
+include_for_metrics:
+    value: []
+include_num_input_tokens_seen:
+    value: "no"
+initializer_range:
+    value: 0.02
+intermediate_size:
+    value: 256
+is_encoder_decoder:
+    value: false
+label_names:
+    value: null
+label_smoothing_factor:
+    value: 0
+label2id:
+    value:
+        LABEL_0: 0
+        LABEL_1: 1
+learning_rate:
+    value: 0.001
+length_column_name:
+    value: length
+liger_kernel_config:
+    value: null
+load_best_model_at_end:
+    value: false
+local_rank:
+    value: -1
+log_level:
+    value: passive
+log_level_replica:
+    value: warning
+log_on_each_node:
+    value: true
+logging_first_step:
+    value: false
+logging_nan_inf_filter:
+    value: true
+logging_steps:
+    value: 20
+logging_strategy:
+    value: steps
+lr_scheduler_kwargs:
+    value: null
+lr_scheduler_type:
+    value: constant
+max_grad_norm:
+    value: 1
+max_position_embeddings:
+    value: 512
+max_steps:
+    value: 1500
+metric_for_best_model:
+    value: null
+mlp_bias:
+    value: false
+mlp_type:
+    value: glu
+model/num_parameters:
+    value: 15949440
+model_type:
+    value: tiny_llama
+neftune_noise_alpha:
+    value: null
+num_attention_heads:
+    value: 4
+num_hidden_layers:
+    value: 94
+num_key_value_heads:
+    value: 4
+num_train_epochs:
+    value: 1
+optim:
+    value: adamw_torch_fused
+optim_args:
+    value: null
+optim_target_modules:
+    value: null
+output_attentions:
+    value: false
+output_dir:
+    value: out/glu-linear-94L_run
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: 0
+parallelism_config:
+    value: null
+per_device_eval_batch_size:
+    value: 128
+per_device_train_batch_size:
+    value: 128
+prediction_loss_only:
+    value: false
+pretraining_tp:
+    value: 1
+problem_type:
+    value: null
+project:
+    value: huggingface
+push_to_hub:
+    value: true
+remove_unused_columns:
+    value: false
+report_to:
+    value:
+        - wandb
+restore_callback_states_from_checkpoint:
+    value: false
+resume_from_checkpoint:
+    value: null
+return_dict:
+    value: true
+rms_norm_eps:
+    value: 1e-06
+rope_parameters:
+    value:
+        rope_theta: 10000
+        rope_type: default
+run_name:
+    value: LM-glu-linear-94L-15.9M-20260809-050049
+save_on_each_node:
+    value: false
+save_only_model:
+    value: false
+save_steps:
+    value: 100
+save_strategy:
+    value: steps
+save_total_limit:
+    value: null
+seed:
+    value: 42
+skip_memory_metrics:
+    value: true
+tf32:
+    value: null
+tie_word_embeddings:
+    value: true
+tokenizer_name:
+    value: w-ahmad/tiny-stories-tokenizer
+torch_compile:
+    value: false
+torch_compile_backend:
+    value: null
+torch_compile_mode:
+    value: null
+torch_empty_cache_steps:
+    value: null
+trackio_bucket_id:
+    value: null
+trackio_space_id:
+    value: null
+trackio_static_space_id:
+    value: null
+train_sampling_strategy:
+    value: random
+transformers_version:
+    value: 5.15.0.dev0
+use_cache:
+    value: false
+use_cpu:
+    value: false
+use_liger_kernel:
+    value: false
+vocab_size:
+    value: 4096
+warmup_steps:
+    value: 0
+weight_decay:
+    value: 0.01

wandb/run-20260809_050050-59pftr14/files/output.log

@@ -0,0 +1,221 @@
+[transformers] `use_return_dict` is deprecated! Use `return_dict` instead!
+[INFO] Causal mask (float with -inf) applied to all attention layers.
+/mnt/data/zainulabideen/zain-exp/notebooks/Activation/exp.py:487: UserWarning: std(): degrees of freedom is <= 0. Correction should be strictly less than the reduction factor (input numel divided by output numel). (Triggered internally at /pytorch/aten/src/ATen/native/ReduceOps.cpp:1831.)
+  "std": tensor.std().item(),
+  7%|██▌                                    | 100/1500 [03:44<43:52,  1.88s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '28.45', 'grad_norm': '3.625', 'learning_rate': '0.001', 'epoch': '0.01078', 'train/total_time_seconds': '33.14', 'train/time_per_step_avg': '1.657', 'train/epoch_time_elapsed': '41.52', 'train/estimated_remaining_minutes': '40.88', 'train/global/act/norm': '8.914e+04', 'train/global/act/mean': '-0.002708', 'train/global/act/std': '0.4187', 'train/global/act/max_abs': '8.323', 'train/global/act/frac_near_dtype_limit': '0', 'train/global/act/frac_near_user_limit': '0', 'train/global/grad/norm': '3.6', 'train/global/grad/mean': '-7.779e-08', 'train/global/grad/std': '0.0004509', 'train/global/grad/max_abs': '0.1279', 'train/global/grad/frac_near_dtype_limit': '0', 'train/global/grad/frac_near_user_limit': '0', 'train/global/param/norm': '174.8', 'train/global/param/mean': '0.001521', 'train/global/param/std': '0.04376', 'train/global/param/max_abs': '1', 'train/global/param/frac_near_dtype_limit': '0', 'train/global/param/frac_near_user_limit': '0', 'train/layer__model_layers_49/param/norm': '17.93', 'train/layer__model_layers_49/param/mean': '0.001503', 'train/layer__model_layers_49/param/std': '0.04425', 'train/layer__model_layers_49/param/max_abs': '1', 'train/layer__model_layers_49/param/frac_near_dtype_limit': '0', 'train/layer__model_layers_49/param/frac_near_user_limit': '0', 'train/layer__model_layers_7/param/norm': '17.94', 'train/layer__model_layers_7/param/mean': '0.00151', 'train/layer__model_layers_7/param/std': '0.04425', 'train/layer__model_layers_7/param/max_abs': '1', 'train/layer__model_layers_7/param/frac_near_dtype_limit': '0', 'train/layer__model_layers_7/param/frac_near_user_limit': '0', 'train/layer__model_layers_86/param/norm': '17.93', 'train/layer__model_layers_86/param/mean': '0.00157', 'train/layer__model_layers_86/param/std': '0.04425', 'train/layer__model_layers_86/param/max_abs': '1', 'train/layer__model_layers_86/param/frac_near_dtype_limit': '0', 'train/layer__model_layers_86/param/frac_near_user_limit': '0', 'train/layer_model_layers_4/act/norm': '8928', 'train/layer_model_layers_4/act/mean': '-0.007839', 'train/layer_model_layers_4/act/std': '0.4122', 'train/layer_model_layers_4/act/max_abs': '5.469', 'train/layer_model_layers_4/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_4/act/frac_near_user_limit': '0', 'train/layer_model_layers_4/grad/norm': '0.7242', 'train/layer_model_layers_4/grad/mean': '-1.734e-06', 'train/layer_model_layers_4/grad/std': '0.0008939', 'train/layer_model_layers_4/grad/max_abs': '0.01489', 'train/layer_model_layers_4/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_4/grad/frac_near_user_limit': '0', 'train/layer_model_layers_40/act/norm': '9081', 'train/layer_model_layers_40/act/mean': '0.000482', 'train/layer_model_layers_40/act/std': '0.4191', 'train/layer_model_layers_40/act/max_abs': '4.875', 'train/layer_model_layers_40/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_40/act/frac_near_user_limit': '0', 'train/layer_model_layers_40/grad/norm': '0.1891', 'train/layer_model_layers_40/grad/mean': '-4.847e-07', 'train/layer_model_layers_40/grad/std': '0.0002335', 'train/layer_model_layers_40/grad/max_abs': '0.006805', 'train/layer_model_layers_40/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_40/grad/frac_near_user_limit': '0', 'train/layer_model_layers_34/act/norm': '9060', 'train/layer_model_layers_34/act/mean': '-0.002399', 'train/layer_model_layers_34/act/std': '0.4181', 'train/layer_model_layers_34/act/max_abs': '5.188', 'train/layer_model_layers_34/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_34/act/frac_near_user_limit': '0', 'train/layer_model_layers_34/grad/norm': '0.229', 'train/layer_model_layers_34/grad/mean': '1.348e-07', 'train/layer_model_layers_34/grad/std': '0.0002827', 'train/layer_model_layers_34/grad/max_abs': '0.007446', 'train/layer_model_layers_34/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_34/grad/frac_near_user_limit': '0', 'train/layer_model_layers_27/act/norm': '9016', 'train/layer_model_layers_27/act/mean': '-0.006148', 'train/layer_model_layers_27/act/
+{'loss': '24.25', 'grad_norm': '0.2471', 'learning_rate': '0.001', 'epoch': '0.02157', 'train/total_time_seconds': '62.96', 'train/time_per_step_avg': '1.574', 'train/epoch_time_elapsed': '79.29', 'train/estimated_remaining_minutes': '38.3'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '5.901', 'eval_runtime': '15.77', 'eval_samples_per_second': '604.3', 'eval_steps_per_second': '4.757', 'epoch': '0.02696', 'train/total_time_seconds': '77.87', 'train/time_per_step_avg': '1.557', 'train/epoch_time_elapsed': '113.9', 'train/estimated_remaining_minutes': '37.63'}
+{'loss': '23.61', 'grad_norm': '1.281', 'learning_rate': '0.001', 'epoch': '0.03235', 'train/total_time_seconds': '92.78', 'train/time_per_step_avg': '1.546', 'train/epoch_time_elapsed': '132.7', 'train/estimated_remaining_minutes': '37.11'}
+{'loss': '22.99', 'grad_norm': '1.477', 'learning_rate': '0.001', 'epoch': '0.04314', 'train/total_time_seconds': '122.6', 'train/time_per_step_avg': '1.532', 'train/epoch_time_elapsed': '170.2', 'train/estimated_remaining_minutes': '36.26'}
+{'loss': '22.73', 'grad_norm': '1.641', 'learning_rate': '0.001', 'epoch': '0.05392', 'train/total_time_seconds': '152.4', 'train/time_per_step_avg': '1.524', 'train/epoch_time_elapsed': '207.7', 'train/estimated_remaining_minutes': '35.56'}
+{'eval_loss': '5.594', 'eval_runtime': '15.83', 'eval_samples_per_second': '601.7', 'eval_steps_per_second': '4.737', 'epoch': '0.05392', 'train/total_time_seconds': '152.4', 'train/time_per_step_avg': '1.524', 'train/epoch_time_elapsed': '223.6', 'train/estimated_remaining_minutes': '35.56'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 14.58it/s]
+ 13%|█████▏                                 | 200/1500 [07:24<40:31,  1.87s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '21.94', 'grad_norm': '2.391', 'learning_rate': '0.001', 'epoch': '0.06471', 'train/total_time_seconds': '182.3', 'train/time_per_step_avg': '1.491', 'train/epoch_time_elapsed': '261.6', 'train/estimated_remaining_minutes': '34.94'}
+{'loss': '21.14', 'grad_norm': '1.93', 'learning_rate': '0.001', 'epoch': '0.07549', 'train/total_time_seconds': '212.4', 'train/time_per_step_avg': '1.494', 'train/epoch_time_elapsed': '299.4', 'train/estimated_remaining_minutes': '34.38'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '5.011', 'eval_runtime': '15.69', 'eval_samples_per_second': '607.3', 'eval_steps_per_second': '4.781', 'epoch': '0.08088', 'train/total_time_seconds': '227.3', 'train/time_per_step_avg': '1.494', 'train/epoch_time_elapsed': '333.9', 'train/estimated_remaining_minutes': '34.09'}
+{'loss': '20.02', 'grad_norm': '2.234', 'learning_rate': '0.001', 'epoch': '0.08628', 'train/total_time_seconds': '242.2', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '352.9', 'train/estimated_remaining_minutes': '33.81'}
+{'loss': '18.95', 'grad_norm': '5.344', 'learning_rate': '0.001', 'epoch': '0.09706', 'train/total_time_seconds': '272.1', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '390.4', 'train/estimated_remaining_minutes': '33.26'}
+{'loss': '18.17', 'grad_norm': '2.844', 'learning_rate': '0.001', 'epoch': '0.1078', 'train/total_time_seconds': '302', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '427.9', 'train/estimated_remaining_minutes': '32.71'}
+{'eval_loss': '4.437', 'eval_runtime': '15.72', 'eval_samples_per_second': '606.2', 'eval_steps_per_second': '4.772', 'epoch': '0.1078', 'train/total_time_seconds': '302', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '443.7', 'train/estimated_remaining_minutes': '32.71'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 15.25it/s]
+ 20%|███████▊                               | 300/1500 [11:03<37:25,  1.87s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '17.4', 'grad_norm': '1.891', 'learning_rate': '0.001', 'epoch': '0.1186', 'train/total_time_seconds': '331.8', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '481.4', 'train/estimated_remaining_minutes': '32.18'}
+{'loss': '16.68', 'grad_norm': '3.359', 'learning_rate': '0.001', 'epoch': '0.1294', 'train/total_time_seconds': '361.7', 'train/time_per_step_avg': '1.493', 'train/epoch_time_elapsed': '519', 'train/estimated_remaining_minutes': '31.65'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '3.988', 'eval_runtime': '15.76', 'eval_samples_per_second': '604.5', 'eval_steps_per_second': '4.758', 'epoch': '0.1348', 'train/total_time_seconds': '376.8', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '553.9', 'train/estimated_remaining_minutes': '31.4'}
+{'loss': '15.99', 'grad_norm': '4.188', 'learning_rate': '0.001', 'epoch': '0.1402', 'train/total_time_seconds': '391.8', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '572.7', 'train/estimated_remaining_minutes': '31.14'}
+{'loss': '15.49', 'grad_norm': '2.156', 'learning_rate': '0.001', 'epoch': '0.151', 'train/total_time_seconds': '421.7', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '610.2', 'train/estimated_remaining_minutes': '30.62'}
+{'loss': '14.99', 'grad_norm': '1.516', 'learning_rate': '0.001', 'epoch': '0.1618', 'train/total_time_seconds': '451.5', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '647.6', 'train/estimated_remaining_minutes': '30.1'}
+{'eval_loss': '3.689', 'eval_runtime': '15.74', 'eval_samples_per_second': '605.3', 'eval_steps_per_second': '4.765', 'epoch': '0.1618', 'train/total_time_seconds': '451.5', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '663.3', 'train/estimated_remaining_minutes': '30.1'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 14.87it/s]
+ 27%|██████████▍                            | 400/1500 [14:43<34:11,  1.86s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '14.5', 'grad_norm': '1.875', 'learning_rate': '0.001', 'epoch': '0.1726', 'train/total_time_seconds': '481.3', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '701.1', 'train/estimated_remaining_minutes': '29.58'}
+{'loss': '14.09', 'grad_norm': '2.656', 'learning_rate': '0.001', 'epoch': '0.1833', 'train/total_time_seconds': '511.2', 'train/time_per_step_avg': '1.494', 'train/epoch_time_elapsed': '738.7', 'train/estimated_remaining_minutes': '29.07'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '3.426', 'eval_runtime': '15.72', 'eval_samples_per_second': '606', 'eval_steps_per_second': '4.77', 'epoch': '0.1887', 'train/total_time_seconds': '526.1', 'train/time_per_step_avg': '1.492', 'train/epoch_time_elapsed': '773.1', 'train/estimated_remaining_minutes': '28.81'}
+{'loss': '13.7', 'grad_norm': '2.859', 'learning_rate': '0.001', 'epoch': '0.1941', 'train/total_time_seconds': '541', 'train/time_per_step_avg': '1.492', 'train/epoch_time_elapsed': '791.8', 'train/estimated_remaining_minutes': '28.55'}
+{'loss': '13.35', 'grad_norm': '1.562', 'learning_rate': '0.001', 'epoch': '0.2049', 'train/total_time_seconds': '571.1', 'train/time_per_step_avg': '1.494', 'train/epoch_time_elapsed': '829.6', 'train/estimated_remaining_minutes': '28.05'}
+{'loss': '12.98', 'grad_norm': '1.531', 'learning_rate': '0.001', 'epoch': '0.2157', 'train/total_time_seconds': '601', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '866.9', 'train/estimated_remaining_minutes': '27.54'}
+{'eval_loss': '3.222', 'eval_runtime': '15.78', 'eval_samples_per_second': '603.6', 'eval_steps_per_second': '4.752', 'epoch': '0.2157', 'train/total_time_seconds': '601', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '882.7', 'train/estimated_remaining_minutes': '27.54'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 17.38it/s]
+ 33%|█████████████                          | 500/1500 [18:23<31:25,  1.89s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '12.66', 'grad_norm': '2.312', 'learning_rate': '0.001', 'epoch': '0.2265', 'train/total_time_seconds': '630.8', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '920.4', 'train/estimated_remaining_minutes': '27.03'}
+{'loss': '12.41', 'grad_norm': '1.805', 'learning_rate': '0.001', 'epoch': '0.2373', 'train/total_time_seconds': '660.7', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '958.1', 'train/estimated_remaining_minutes': '26.53'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '3.017', 'eval_runtime': '15.63', 'eval_samples_per_second': '609.6', 'eval_steps_per_second': '4.799', 'epoch': '0.2427', 'train/total_time_seconds': '675.7', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '992.8', 'train/estimated_remaining_minutes': '26.28'}
+{'loss': '12.05', 'grad_norm': '1.867', 'learning_rate': '0.001', 'epoch': '0.248', 'train/total_time_seconds': '690.6', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '1011', 'train/estimated_remaining_minutes': '26.02'}
+{'loss': '11.73', 'grad_norm': '1.406', 'learning_rate': '0.001', 'epoch': '0.2588', 'train/total_time_seconds': '720.5', 'train/time_per_step_avg': '1.494', 'train/epoch_time_elapsed': '1049', 'train/estimated_remaining_minutes': '25.52'}
+{'loss': '11.41', 'grad_norm': '1.531', 'learning_rate': '0.001', 'epoch': '0.2696', 'train/total_time_seconds': '750.4', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '1087', 'train/estimated_remaining_minutes': '25.01'}
+{'eval_loss': '2.821', 'eval_runtime': '16', 'eval_samples_per_second': '595.4', 'eval_steps_per_second': '4.687', 'epoch': '0.2696', 'train/total_time_seconds': '750.4', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '1103', 'train/estimated_remaining_minutes': '25.01'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00,  3.41it/s]
+/mnt/data/zainulabideen/zain-exp/notebooks/Activation/exp.py:487: UserWarning: std(): degrees of freedom is <= 0. Correction should be strictly less than the reduction factor (input numel divided by output numel). (Triggered internally at /pytorch/aten/src/ATen/native/ReduceOps.cpp:1831.)
+  "std": tensor.std().item(),
+ 40%|███████████████▌                       | 600/1500 [22:06<28:08,  1.88s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '11.16', 'grad_norm': '1.398', 'learning_rate': '0.001', 'epoch': '0.2804', 'train/total_time_seconds': '782.9', 'train/time_per_step_avg': '1.521', 'train/epoch_time_elapsed': '1144', 'train/estimated_remaining_minutes': '24.59', 'train/global/act/norm': '2.108e+05', 'train/global/act/mean': '-0.04442', 'train/global/act/std': '0.9892', 'train/global/act/max_abs': '30.5', 'train/global/act/frac_near_dtype_limit': '0', 'train/global/act/frac_near_user_limit': '0', 'train/global/grad/norm': '0.8868', 'train/global/grad/mean': '-3.285e-08', 'train/global/grad/std': '0.000111', 'train/global/grad/max_abs': '0.02441', 'train/global/grad/frac_near_dtype_limit': '0', 'train/global/grad/frac_near_user_limit': '0', 'train/global/param/norm': '204.6', 'train/global/param/mean': '0.001524', 'train/global/param/std': '0.05121', 'train/global/param/max_abs': '1', 'train/global/param/frac_near_dtype_limit': '0', 'train/global/param/frac_near_user_limit': '0', 'train/layer__model_layers_49/param/norm': '19.65', 'train/layer__model_layers_49/param/mean': '0.001461', 'train/layer__model_layers_49/param/std': '0.0485', 'train/layer__model_layers_49/param/max_abs': '1', 'train/layer__model_layers_49/param/frac_near_dtype_limit': '0', 'train/layer__model_layers_49/param/frac_near_user_limit': '0', 'train/layer__model_layers_7/param/norm': '18.96', 'train/layer__model_layers_7/param/mean': '0.001547', 'train/layer__model_layers_7/param/std': '0.04675', 'train/layer__model_layers_7/param/max_abs': '1', 'train/layer__model_layers_7/param/frac_near_dtype_limit': '0', 'train/layer__model_layers_7/param/frac_near_user_limit': '0', 'train/layer__model_layers_86/param/norm': '22.84', 'train/layer__model_layers_86/param/mean': '0.001594', 'train/layer__model_layers_86/param/std': '0.05639', 'train/layer__model_layers_86/param/max_abs': '1', 'train/layer__model_layers_86/param/frac_near_dtype_limit': '0', 'train/layer__model_layers_86/param/frac_near_user_limit': '0', 'train/layer_model_layers_4/act/norm': '2.117e+04', 'train/layer_model_layers_4/act/mean': '-0.001465', 'train/layer_model_layers_4/act/std': '0.9766', 'train/layer_model_layers_4/act/max_abs': '17.88', 'train/layer_model_layers_4/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_4/act/frac_near_user_limit': '0', 'train/layer_model_layers_4/grad/norm': '0.06447', 'train/layer_model_layers_4/grad/mean': '8.797e-08', 'train/layer_model_layers_4/grad/std': '7.95e-05', 'train/layer_model_layers_4/grad/max_abs': '0.001686', 'train/layer_model_layers_4/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_4/grad/frac_near_user_limit': '0', 'train/layer_model_layers_40/act/norm': '1.85e+04', 'train/layer_model_layers_40/act/mean': '0.001928', 'train/layer_model_layers_40/act/std': '0.8529', 'train/layer_model_layers_40/act/max_abs': '20.25', 'train/layer_model_layers_40/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_40/act/frac_near_user_limit': '0', 'train/layer_model_layers_40/grad/norm': '0.0276', 'train/layer_model_layers_40/grad/mean': '3.439e-08', 'train/layer_model_layers_40/grad/std': '3.406e-05', 'train/layer_model_layers_40/grad/max_abs': '0.0007095', 'train/layer_model_layers_40/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_40/grad/frac_near_user_limit': '0', 'train/layer_model_layers_34/act/norm': '1.832e+04', 'train/layer_model_layers_34/act/mean': '0.002268', 'train/layer_model_layers_34/act/std': '0.8442', 'train/layer_model_layers_34/act/max_abs': '20.25', 'train/layer_model_layers_34/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_34/act/frac_near_user_limit': '0', 'train/layer_model_layers_34/grad/norm': '0.02417', 'train/layer_model_layers_34/grad/mean': '1.063e-07', 'train/layer_model_layers_34/grad/std': '2.985e-05', 'train/layer_model_layers_34/grad/max_abs': '0.0006027', 'train/layer_model_layers_34/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_34/grad/frac_near_user_limit': '0', 'train/layer_model_layers_27/act/norm': '1.909e+04', 'train/layer_model_layers_27/act/mean': '-0.001571', 'train/layer
+{'loss': '10.94', 'grad_norm': '1.609', 'learning_rate': '0.001', 'epoch': '0.2912', 'train/total_time_seconds': '812.8', 'train/time_per_step_avg': '1.521', 'train/epoch_time_elapsed': '1181', 'train/estimated_remaining_minutes': '24.08'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.668', 'eval_runtime': '15.7', 'eval_samples_per_second': '606.9', 'eval_steps_per_second': '4.777', 'epoch': '0.2966', 'train/total_time_seconds': '827.8', 'train/time_per_step_avg': '1.521', 'train/epoch_time_elapsed': '1216', 'train/estimated_remaining_minutes': '23.83'}
+{'loss': '10.69', 'grad_norm': '1.742', 'learning_rate': '0.001', 'epoch': '0.302', 'train/total_time_seconds': '842.7', 'train/time_per_step_avg': '1.521', 'train/epoch_time_elapsed': '1235', 'train/estimated_remaining_minutes': '23.58'}
+{'loss': '10.45', 'grad_norm': '1.227', 'learning_rate': '0.001', 'epoch': '0.3128', 'train/total_time_seconds': '872.5', 'train/time_per_step_avg': '1.52', 'train/epoch_time_elapsed': '1273', 'train/estimated_remaining_minutes': '23.07'}
+{'loss': '10.23', 'grad_norm': '1.359', 'learning_rate': '0.001', 'epoch': '0.3235', 'train/total_time_seconds': '902.4', 'train/time_per_step_avg': '1.519', 'train/epoch_time_elapsed': '1310', 'train/estimated_remaining_minutes': '22.56'}
+{'eval_loss': '2.535', 'eval_runtime': '15.93', 'eval_samples_per_second': '597.9', 'eval_steps_per_second': '4.707', 'epoch': '0.3235', 'train/total_time_seconds': '902.4', 'train/time_per_step_avg': '1.519', 'train/epoch_time_elapsed': '1326', 'train/estimated_remaining_minutes': '22.56'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 16.86it/s]
+ 47%|██████████████████▏                    | 700/1500 [25:47<24:58,  1.87s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '10.04', 'grad_norm': '1.656', 'learning_rate': '0.001', 'epoch': '0.3343', 'train/total_time_seconds': '932.5', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '1364', 'train/estimated_remaining_minutes': '22.06'}
+{'loss': '9.852', 'grad_norm': '1.203', 'learning_rate': '0.001', 'epoch': '0.3451', 'train/total_time_seconds': '962.4', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '1402', 'train/estimated_remaining_minutes': '21.55'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.426', 'eval_runtime': '15.89', 'eval_samples_per_second': '599.5', 'eval_steps_per_second': '4.719', 'epoch': '0.3505', 'train/total_time_seconds': '977.4', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '1437', 'train/estimated_remaining_minutes': '21.3'}
+{'loss': '9.675', 'grad_norm': '1.219', 'learning_rate': '0.001', 'epoch': '0.3559', 'train/total_time_seconds': '992.3', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '1455', 'train/estimated_remaining_minutes': '21.05'}
+{'loss': '9.502', 'grad_norm': '1.18', 'learning_rate': '0.001', 'epoch': '0.3667', 'train/total_time_seconds': '1022', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '1493', 'train/estimated_remaining_minutes': '20.54'}
+{'loss': '9.35', 'grad_norm': '1.031', 'learning_rate': '0.001', 'epoch': '0.3775', 'train/total_time_seconds': '1052', 'train/time_per_step_avg': '1.497', 'train/epoch_time_elapsed': '1531', 'train/estimated_remaining_minutes': '20.04'}
+{'eval_loss': '2.323', 'eval_runtime': '15.89', 'eval_samples_per_second': '599.7', 'eval_steps_per_second': '4.721', 'epoch': '0.3775', 'train/total_time_seconds': '1052', 'train/time_per_step_avg': '1.497', 'train/epoch_time_elapsed': '1546', 'train/estimated_remaining_minutes': '20.04'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 15.13it/s]
+ 53%|████████████████████▊                  | 800/1500 [29:27<22:14,  1.91s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '9.192', 'grad_norm': '1.344', 'learning_rate': '0.001', 'epoch': '0.3882', 'train/total_time_seconds': '1082', 'train/time_per_step_avg': '1.494', 'train/epoch_time_elapsed': '1584', 'train/estimated_remaining_minutes': '19.53'}
+{'loss': '9.084', 'grad_norm': '1.227', 'learning_rate': '0.001', 'epoch': '0.399', 'train/total_time_seconds': '1112', 'train/time_per_step_avg': '1.493', 'train/epoch_time_elapsed': '1622', 'train/estimated_remaining_minutes': '19.03'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.241', 'eval_runtime': '16', 'eval_samples_per_second': '595.5', 'eval_steps_per_second': '4.688', 'epoch': '0.4044', 'train/total_time_seconds': '1127', 'train/time_per_step_avg': '1.493', 'train/epoch_time_elapsed': '1657', 'train/estimated_remaining_minutes': '18.78'}
+{'loss': '8.964', 'grad_norm': '1.156', 'learning_rate': '0.001', 'epoch': '0.4098', 'train/total_time_seconds': '1142', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '1676', 'train/estimated_remaining_minutes': '18.53'}
+{'loss': '8.877', 'grad_norm': '1.297', 'learning_rate': '0.001', 'epoch': '0.4206', 'train/total_time_seconds': '1172', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '1714', 'train/estimated_remaining_minutes': '18.03'}
+{'loss': '8.725', 'grad_norm': '1.203', 'learning_rate': '0.001', 'epoch': '0.4314', 'train/total_time_seconds': '1202', 'train/time_per_step_avg': '1.497', 'train/epoch_time_elapsed': '1751', 'train/estimated_remaining_minutes': '17.53'}
+{'eval_loss': '2.179', 'eval_runtime': '15.97', 'eval_samples_per_second': '596.5', 'eval_steps_per_second': '4.696', 'epoch': '0.4314', 'train/total_time_seconds': '1202', 'train/time_per_step_avg': '1.497', 'train/epoch_time_elapsed': '1767', 'train/estimated_remaining_minutes': '17.53'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 17.56it/s]
+ 60%|███████████████████████▍               | 900/1500 [33:08<19:04,  1.91s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '8.653', 'grad_norm': '1.078', 'learning_rate': '0.001', 'epoch': '0.4422', 'train/total_time_seconds': '1232', 'train/time_per_step_avg': '1.498', 'train/epoch_time_elapsed': '1805', 'train/estimated_remaining_minutes': '17.02'}
+{'loss': '8.572', 'grad_norm': '1.219', 'learning_rate': '0.001', 'epoch': '0.453', 'train/total_time_seconds': '1262', 'train/time_per_step_avg': '1.499', 'train/epoch_time_elapsed': '1843', 'train/estimated_remaining_minutes': '16.52'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.128', 'eval_runtime': '15.83', 'eval_samples_per_second': '601.9', 'eval_steps_per_second': '4.738', 'epoch': '0.4583', 'train/total_time_seconds': '1277', 'train/time_per_step_avg': '1.499', 'train/epoch_time_elapsed': '1878', 'train/estimated_remaining_minutes': '16.27'}
+{'loss': '8.466', 'grad_norm': '1.094', 'learning_rate': '0.001', 'epoch': '0.4637', 'train/total_time_seconds': '1291', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '1896', 'train/estimated_remaining_minutes': '16.02'}
+{'loss': '8.38', 'grad_norm': '1.008', 'learning_rate': '0.001', 'epoch': '0.4745', 'train/total_time_seconds': '1321', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '1934', 'train/estimated_remaining_minutes': '15.52'}
+{'loss': '8.304', 'grad_norm': '0.9336', 'learning_rate': '0.001', 'epoch': '0.4853', 'train/total_time_seconds': '1352', 'train/time_per_step_avg': '1.498', 'train/epoch_time_elapsed': '1972', 'train/estimated_remaining_minutes': '15.02'}
+{'eval_loss': '2.072', 'eval_runtime': '15.82', 'eval_samples_per_second': '602.1', 'eval_steps_per_second': '4.74', 'epoch': '0.4853', 'train/total_time_seconds': '1352', 'train/time_per_step_avg': '1.498', 'train/epoch_time_elapsed': '1988', 'train/estimated_remaining_minutes': '15.02'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 13.94it/s]
+ 67%|█████████████████████████▎            | 1000/1500 [36:49<15:56,  1.91s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '8.227', 'grad_norm': '1.008', 'learning_rate': '0.001', 'epoch': '0.4961', 'train/total_time_seconds': '1381', 'train/time_per_step_avg': '1.497', 'train/epoch_time_elapsed': '2026', 'train/estimated_remaining_minutes': '14.51'}
+{'loss': '8.188', 'grad_norm': '0.8984', 'learning_rate': '0.001', 'epoch': '0.5069', 'train/total_time_seconds': '1411', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '2064', 'train/estimated_remaining_minutes': '14.01'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '2.034', 'eval_runtime': '15.98', 'eval_samples_per_second': '596.1', 'eval_steps_per_second': '4.693', 'epoch': '0.5123', 'train/total_time_seconds': '1426', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '2098', 'train/estimated_remaining_minutes': '13.76'}
+{'loss': '8.129', 'grad_norm': '0.9219', 'learning_rate': '0.001', 'epoch': '0.5177', 'train/total_time_seconds': '1441', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '2117', 'train/estimated_remaining_minutes': '13.51'}
+{'loss': '8.076', 'grad_norm': '0.9453', 'learning_rate': '0.001', 'epoch': '0.5284', 'train/total_time_seconds': '1471', 'train/time_per_step_avg': '1.497', 'train/epoch_time_elapsed': '2155', 'train/estimated_remaining_minutes': '13.01'}
+{'loss': '8.016', 'grad_norm': '1.039', 'learning_rate': '0.001', 'epoch': '0.5392', 'train/total_time_seconds': '1501', 'train/time_per_step_avg': '1.494', 'train/epoch_time_elapsed': '2193', 'train/estimated_remaining_minutes': '12.51'}
+{'eval_loss': '2.001', 'eval_runtime': '15.8', 'eval_samples_per_second': '603.1', 'eval_steps_per_second': '4.748', 'epoch': '0.5392', 'train/total_time_seconds': '1501', 'train/time_per_step_avg': '1.494', 'train/epoch_time_elapsed': '2209', 'train/estimated_remaining_minutes': '12.51'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 18.81it/s]
+/mnt/data/zainulabideen/zain-exp/notebooks/Activation/exp.py:487: UserWarning: std(): degrees of freedom is <= 0. Correction should be strictly less than the reduction factor (input numel divided by output numel). (Triggered internally at /pytorch/aten/src/ATen/native/ReduceOps.cpp:1831.)
+  "std": tensor.std().item(),
+ 73%|███████████████████████████▊          | 1100/1500 [40:32<12:34,  1.89s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '7.937', 'grad_norm': '0.9922', 'learning_rate': '0.001', 'epoch': '0.55', 'train/total_time_seconds': '1533', 'train/time_per_step_avg': '1.519', 'train/epoch_time_elapsed': '2249', 'train/estimated_remaining_minutes': '12.03', 'train/global/act/norm': '1.788e+05', 'train/global/act/mean': '-0.05405', 'train/global/act/std': '0.8384', 'train/global/act/max_abs': '26.12', 'train/global/act/frac_near_dtype_limit': '0', 'train/global/act/frac_near_user_limit': '0', 'train/global/grad/norm': '0.6355', 'train/global/grad/mean': '-1.679e-08', 'train/global/grad/std': '7.957e-05', 'train/global/grad/max_abs': '0.01282', 'train/global/grad/frac_near_dtype_limit': '0', 'train/global/grad/frac_near_user_limit': '0', 'train/global/param/norm': '231.5', 'train/global/param/mean': '0.001526', 'train/global/param/std': '0.05795', 'train/global/param/max_abs': '1', 'train/global/param/frac_near_dtype_limit': '0', 'train/global/param/frac_near_user_limit': '0', 'train/layer__model_layers_49/param/norm': '21.78', 'train/layer__model_layers_49/param/mean': '0.001486', 'train/layer__model_layers_49/param/std': '0.05377', 'train/layer__model_layers_49/param/max_abs': '1', 'train/layer__model_layers_49/param/frac_near_dtype_limit': '0', 'train/layer__model_layers_49/param/frac_near_user_limit': '0', 'train/layer__model_layers_7/param/norm': '19.6', 'train/layer__model_layers_7/param/mean': '0.001541', 'train/layer__model_layers_7/param/std': '0.04837', 'train/layer__model_layers_7/param/max_abs': '1', 'train/layer__model_layers_7/param/frac_near_dtype_limit': '0', 'train/layer__model_layers_7/param/frac_near_user_limit': '0', 'train/layer__model_layers_86/param/norm': '25.44', 'train/layer__model_layers_86/param/mean': '0.001714', 'train/layer__model_layers_86/param/std': '0.06281', 'train/layer__model_layers_86/param/max_abs': '1', 'train/layer__model_layers_86/param/frac_near_dtype_limit': '0', 'train/layer__model_layers_86/param/frac_near_user_limit': '0', 'train/layer_model_layers_4/act/norm': '1.538e+04', 'train/layer_model_layers_4/act/mean': '-0.002358', 'train/layer_model_layers_4/act/std': '0.7093', 'train/layer_model_layers_4/act/max_abs': '8.812', 'train/layer_model_layers_4/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_4/act/frac_near_user_limit': '0', 'train/layer_model_layers_4/grad/norm': '0.06301', 'train/layer_model_layers_4/grad/mean': '9.569e-08', 'train/layer_model_layers_4/grad/std': '7.779e-05', 'train/layer_model_layers_4/grad/max_abs': '0.001793', 'train/layer_model_layers_4/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_4/grad/frac_near_user_limit': '0', 'train/layer_model_layers_40/act/norm': '1.37e+04', 'train/layer_model_layers_40/act/mean': '-0.002146', 'train/layer_model_layers_40/act/std': '0.6324', 'train/layer_model_layers_40/act/max_abs': '9.375', 'train/layer_model_layers_40/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_40/act/frac_near_user_limit': '0', 'train/layer_model_layers_40/grad/norm': '0.03881', 'train/layer_model_layers_40/grad/mean': '-1.765e-08', 'train/layer_model_layers_40/grad/std': '4.79e-05', 'train/layer_model_layers_40/grad/max_abs': '0.001312', 'train/layer_model_layers_40/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_40/grad/frac_near_user_limit': '0', 'train/layer_model_layers_34/act/norm': '1.354e+04', 'train/layer_model_layers_34/act/mean': '0.002037', 'train/layer_model_layers_34/act/std': '0.6245', 'train/layer_model_layers_34/act/max_abs': '9.312', 'train/layer_model_layers_34/act/frac_near_dtype_limit': '0', 'train/layer_model_layers_34/act/frac_near_user_limit': '0', 'train/layer_model_layers_34/grad/norm': '0.03821', 'train/layer_model_layers_34/grad/mean': '1.768e-08', 'train/layer_model_layers_34/grad/std': '4.715e-05', 'train/layer_model_layers_34/grad/max_abs': '0.001289', 'train/layer_model_layers_34/grad/frac_near_dtype_limit': '0', 'train/layer_model_layers_34/grad/frac_near_user_limit': '0', 'train/layer_model_layers_27/act/norm': '1.372e+04', 'train/layer_model_layers_27/act/mean': '-0.004381', 'train/laye
+{'loss': '7.884', 'grad_norm': '0.9062', 'learning_rate': '0.001', 'epoch': '0.5608', 'train/total_time_seconds': '1563', 'train/time_per_step_avg': '1.519', 'train/epoch_time_elapsed': '2287', 'train/estimated_remaining_minutes': '11.52'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '1.969', 'eval_runtime': '15.81', 'eval_samples_per_second': '602.8', 'eval_steps_per_second': '4.745', 'epoch': '0.5662', 'train/total_time_seconds': '1578', 'train/time_per_step_avg': '1.519', 'train/epoch_time_elapsed': '2322', 'train/estimated_remaining_minutes': '11.27'}
+{'loss': '7.831', 'grad_norm': '0.9609', 'learning_rate': '0.001', 'epoch': '0.5716', 'train/total_time_seconds': '1593', 'train/time_per_step_avg': '1.519', 'train/epoch_time_elapsed': '2340', 'train/estimated_remaining_minutes': '11.02'}
+{'loss': '7.804', 'grad_norm': '0.9102', 'learning_rate': '0.001', 'epoch': '0.5824', 'train/total_time_seconds': '1623', 'train/time_per_step_avg': '1.519', 'train/epoch_time_elapsed': '2378', 'train/estimated_remaining_minutes': '10.52'}
+{'loss': '7.771', 'grad_norm': '0.8984', 'learning_rate': '0.001', 'epoch': '0.5932', 'train/total_time_seconds': '1653', 'train/time_per_step_avg': '1.519', 'train/epoch_time_elapsed': '2416', 'train/estimated_remaining_minutes': '10.02'}
+{'eval_loss': '1.945', 'eval_runtime': '15.89', 'eval_samples_per_second': '599.7', 'eval_steps_per_second': '4.721', 'epoch': '0.5932', 'train/total_time_seconds': '1653', 'train/time_per_step_avg': '1.519', 'train/epoch_time_elapsed': '2432', 'train/estimated_remaining_minutes': '10.02'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 13.85it/s]
+ 80%|██████████████████████████████▍       | 1200/1500 [44:13<09:25,  1.89s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '7.725', 'grad_norm': '1.133', 'learning_rate': '0.001', 'epoch': '0.6039', 'train/total_time_seconds': '1683', 'train/time_per_step_avg': '1.494', 'train/epoch_time_elapsed': '2470', 'train/estimated_remaining_minutes': '9.516'}
+{'loss': '7.69', 'grad_norm': '0.8555', 'learning_rate': '0.001', 'epoch': '0.6147', 'train/total_time_seconds': '1713', 'train/time_per_step_avg': '1.497', 'train/epoch_time_elapsed': '2508', 'train/estimated_remaining_minutes': '9.015'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '1.92', 'eval_runtime': '16.04', 'eval_samples_per_second': '593.8', 'eval_steps_per_second': '4.675', 'epoch': '0.6201', 'train/total_time_seconds': '1728', 'train/time_per_step_avg': '1.497', 'train/epoch_time_elapsed': '2543', 'train/estimated_remaining_minutes': '8.764'}
+{'loss': '7.65', 'grad_norm': '0.8359', 'learning_rate': '0.001', 'epoch': '0.6255', 'train/total_time_seconds': '1743', 'train/time_per_step_avg': '1.497', 'train/epoch_time_elapsed': '2562', 'train/estimated_remaining_minutes': '8.513'}
+{'loss': '7.613', 'grad_norm': '0.9141', 'learning_rate': '0.001', 'epoch': '0.6363', 'train/total_time_seconds': '1773', 'train/time_per_step_avg': '1.496', 'train/epoch_time_elapsed': '2599', 'train/estimated_remaining_minutes': '8.011'}
+{'loss': '7.558', 'grad_norm': '0.8867', 'learning_rate': '0.001', 'epoch': '0.6471', 'train/total_time_seconds': '1802', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '2637', 'train/estimated_remaining_minutes': '7.51'}
+{'eval_loss': '1.897', 'eval_runtime': '15.85', 'eval_samples_per_second': '601.2', 'eval_steps_per_second': '4.733', 'epoch': '0.6471', 'train/total_time_seconds': '1802', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '2653', 'train/estimated_remaining_minutes': '7.51'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 16.66it/s]
+ 87%|████████████████████████████████▉     | 1300/1500 [47:53<06:17,  1.89s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '7.555', 'grad_norm': '0.9883', 'learning_rate': '0.001', 'epoch': '0.6579', 'train/total_time_seconds': '1832', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '2691', 'train/estimated_remaining_minutes': '7.009'}
+{'loss': '7.521', 'grad_norm': '0.8711', 'learning_rate': '0.001', 'epoch': '0.6686', 'train/total_time_seconds': '1862', 'train/time_per_step_avg': '1.493', 'train/epoch_time_elapsed': '2728', 'train/estimated_remaining_minutes': '6.507'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '1.878', 'eval_runtime': '15.95', 'eval_samples_per_second': '597.2', 'eval_steps_per_second': '4.701', 'epoch': '0.674', 'train/total_time_seconds': '1877', 'train/time_per_step_avg': '1.493', 'train/epoch_time_elapsed': '2763', 'train/estimated_remaining_minutes': '6.257'}
+{'loss': '7.483', 'grad_norm': '0.9219', 'learning_rate': '0.001', 'epoch': '0.6794', 'train/total_time_seconds': '1892', 'train/time_per_step_avg': '1.492', 'train/epoch_time_elapsed': '2782', 'train/estimated_remaining_minutes': '6.006'}
+{'loss': '7.478', 'grad_norm': '0.9141', 'learning_rate': '0.001', 'epoch': '0.6902', 'train/total_time_seconds': '1922', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '2819', 'train/estimated_remaining_minutes': '5.506'}
+{'loss': '7.426', 'grad_norm': '0.8438', 'learning_rate': '0.001', 'epoch': '0.701', 'train/total_time_seconds': '1952', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '2857', 'train/estimated_remaining_minutes': '5.005'}
+{'eval_loss': '1.861', 'eval_runtime': '15.84', 'eval_samples_per_second': '601.6', 'eval_steps_per_second': '4.736', 'epoch': '0.701', 'train/total_time_seconds': '1952', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '2873', 'train/estimated_remaining_minutes': '5.005'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 13.14it/s]
+ 93%|███████████████████████████████████▍  | 1400/1500 [51:42<03:07,  1.87s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '7.397', 'grad_norm': '0.875', 'learning_rate': '0.001', 'epoch': '0.7118', 'train/total_time_seconds': '1982', 'train/time_per_step_avg': '1.495', 'train/epoch_time_elapsed': '2910', 'train/estimated_remaining_minutes': '4.504'}
+{'loss': '7.381', 'grad_norm': '0.8477', 'learning_rate': '0.001', 'epoch': '0.7226', 'train/total_time_seconds': '2013', 'train/time_per_step_avg': '1.504', 'train/epoch_time_elapsed': '2949', 'train/estimated_remaining_minutes': '4.005'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '1.846', 'eval_runtime': '16.77', 'eval_samples_per_second': '568', 'eval_steps_per_second': '4.471', 'epoch': '0.728', 'train/total_time_seconds': '2032', 'train/time_per_step_avg': '1.553', 'train/epoch_time_elapsed': '2989', 'train/estimated_remaining_minutes': '3.763'}
+{'loss': '7.354', 'grad_norm': '0.9492', 'learning_rate': '0.001', 'epoch': '0.7334', 'train/total_time_seconds': '2049', 'train/time_per_step_avg': '1.574', 'train/epoch_time_elapsed': '3010', 'train/estimated_remaining_minutes': '3.516'}
+{'loss': '7.347', 'grad_norm': '0.7656', 'learning_rate': '0.001', 'epoch': '0.7441', 'train/total_time_seconds': '2079', 'train/time_per_step_avg': '1.572', 'train/epoch_time_elapsed': '3048', 'train/estimated_remaining_minutes': '3.013'}
+{'loss': '7.306', 'grad_norm': '0.8125', 'learning_rate': '0.001', 'epoch': '0.7549', 'train/total_time_seconds': '2109', 'train/time_per_step_avg': '1.574', 'train/epoch_time_elapsed': '3086', 'train/estimated_remaining_minutes': '2.511'}
+{'eval_loss': '1.832', 'eval_runtime': '16.11', 'eval_samples_per_second': '591.5', 'eval_steps_per_second': '4.657', 'epoch': '0.7549', 'train/total_time_seconds': '2109', 'train/time_per_step_avg': '1.574', 'train/epoch_time_elapsed': '3102', 'train/estimated_remaining_minutes': '2.511'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 18.65it/s]
+100%|██████████████████████████████████████| 1500/1500 [56:00<00:00,  2.45s/it][transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+{'loss': '7.275', 'grad_norm': '0.8125', 'learning_rate': '0.001', 'epoch': '0.7657', 'train/total_time_seconds': '2139', 'train/time_per_step_avg': '1.575', 'train/epoch_time_elapsed': '3140', 'train/estimated_remaining_minutes': '2.009'}
+{'loss': '7.275', 'grad_norm': '0.75', 'learning_rate': '0.001', 'epoch': '0.7765', 'train/total_time_seconds': '2169', 'train/time_per_step_avg': '1.568', 'train/epoch_time_elapsed': '3178', 'train/estimated_remaining_minutes': '1.507'}
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+{'eval_loss': '1.819', 'eval_runtime': '19.27', 'eval_samples_per_second': '494.3', 'eval_steps_per_second': '3.891', 'epoch': '0.7819', 'train/total_time_seconds': '2189', 'train/time_per_step_avg': '1.569', 'train/epoch_time_elapsed': '3221', 'train/estimated_remaining_minutes': '1.258'}
+{'loss': '7.239', 'grad_norm': '0.832', 'learning_rate': '0.001', 'epoch': '0.7873', 'train/total_time_seconds': '2210', 'train/time_per_step_avg': '1.604', 'train/epoch_time_elapsed': '3245', 'train/estimated_remaining_minutes': '1.009'}
+{'loss': '7.227', 'grad_norm': '0.8672', 'learning_rate': '0.001', 'epoch': '0.7981', 'train/total_time_seconds': '2249', 'train/time_per_step_avg': '1.698', 'train/epoch_time_elapsed': '3293', 'train/estimated_remaining_minutes': '0.5065'}
+{'loss': '7.199', 'grad_norm': '0.8008', 'learning_rate': '0.001', 'epoch': '0.8088', 'train/total_time_seconds': '2290', 'train/time_per_step_avg': '1.804', 'train/epoch_time_elapsed': '3341', 'train/estimated_remaining_minutes': '0'}
+{'eval_loss': '1.807', 'eval_runtime': '18.87', 'eval_samples_per_second': '504.9', 'eval_steps_per_second': '3.975', 'epoch': '0.8088', 'train/total_time_seconds': '2290', 'train/time_per_step_avg': '1.804', 'train/epoch_time_elapsed': '3360', 'train/estimated_remaining_minutes': '0'}
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 11.91it/s]
+100%|██████████████████████████████████████| 1500/1500 [56:01<00:00,  2.24s/it]
+{'train_runtime': '3361', 'train_samples_per_second': '228.5', 'train_steps_per_second': '0.446', 'train_loss': '11.32', 'epoch': '0.8088', 'train/total_time_seconds': '2290', 'train/time_per_step_avg': '1.804', 'train/epoch_time_elapsed': '3360', 'train/estimated_remaining_minutes': '0'}
+100%|██████████████████████████████████████████| 75/75 [00:18<00:00,  4.06it/s]
+[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 15.65it/s]
+[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 17.96it/s]
+Found 7 files to upload
+  Preparing   ████████████████████  7 / 7 ✓
+  Uploading   ████████████████████  2 / 2 files ✓
+  Committing  ████████░░░░░░░░░░░░  3 / 7
+  Preparing   ████████████████████  7 / 7 ✓
+  Uploading   ████████████████████  2 / 2 files ✓
+  Committing  ████████░░░░░░░░░░░░  3 / 7
+  Preparing   ████████████████████  7 / 7 ✓
+  Uploading   ████████████████████  2 / 2 files ✓
+  Committing  ████████████████████  7 / 7 ✓
+[transformers] TinyLlamaForCausalLM has generative capabilities, as `prepare_inputs_for_generation` is explicitly defined. However, it doesn't directly inherit from `GenerationMixin`. From 👉v4.50👈 onwards, `PreTrainedModel` will NOT inherit from `GenerationMixin`, and this model will lose the ability to call `generate` and other related functions.
+  - If you're using `trust_remote_code=True`, you can get rid of this warning by loading the model with an auto class. See https://huggingface.co/docs/transformers/en/model_doc/auto#auto-classes
+  - If you are the owner of the model architecture code, please modify your model class such that it inherits from `GenerationMixin` (after `PreTrainedModel`, otherwise you'll get an exception).
+  - If you are not the owner of the model architecture class, please contact the model code owner to update it.
+Writing model shards: 100%|██████████████████████| 1/1 [00:00<00:00, 16.32it/s]
+Found 7 files to upload
+  Preparing   ████████████████████  7 / 7 ✓
+  Uploading   ████████████████████  2 / 2 files ✓
+  Committing  ████████████████████  7 / 7 ✓
+No files have been modified since last commit. Skipping to prevent empty commit.

wandb/run-20260809_050050-59pftr14/files/requirements.txt

@@ -0,0 +1,149 @@
+asttokens==3.0.1
+comm==0.2.3
+debugpy==1.8.21
+decorator==5.3.1
+executing==2.2.1
+nest-asyncio==1.6.0
+parso==0.8.7
+platformdirs==4.11.0
+psutil==7.2.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+Pygments==2.20.0
+pyzmq==27.1.0
+setuptools==83.0.0
+six==1.17.0
+tornado==6.5.7
+traitlets==5.15.0
+fsspec==2026.4.0
+wcwidth==0.8.2
+ipython_pygments_lexers==1.1.1
+jedi==0.20.0
+jupyter_core==5.9.1
+matplotlib-inline==0.2.2
+pexpect==4.9.0
+prompt_toolkit==3.0.53
+python-dateutil==2.9.0.post0
+stack_data==0.6.3
+wheel==0.47.0
+jupyter_client==8.9.1
+pip==26.1.2
+ipython==9.15.0
+ipykernel==7.2.0
+threadpoolctl==3.6.0
+pyparsing==3.3.2
+typing_extensions==4.15.0
+Jinja2==3.1.6
+narwhals==2.24.0
+kiwisolver==1.5.0
+joblib==1.5.3
+fonttools==4.63.0
+cycler==0.12.1
+scipy==1.17.1
+pandas==3.0.5
+contourpy==1.3.3
+scikit-learn==1.9.0
+matplotlib==3.11.1
+urllib3==2.7.0
+tqdm==4.70.0
+idna==3.18
+charset-normalizer==3.4.9
+certifi==2026.7.22
+requests==2.34.2
+seaborn==0.13.2
+uv==0.12.0
+shellingham==1.5.4
+mpmath==1.3.0
+attrs==26.1.0
+hf-xet==1.5.2
+nvidia-nccl-cu12==2.21.5
+MarkupSafe==3.0.3
+regex==2026.7.19
+importlib_metadata==9.0.0
+httpcore==1.0.9
+annotated-doc==0.0.5
+multidict==6.7.1
+aiohttp==3.14.3
+aiosignal==1.4.0
+xxhash==3.8.1
+aiohappyeyeballs==2.7.1
+mdurl==0.1.2
+cuda-toolkit==13.0.3.0
+networkx==3.6.1
+PyYAML==6.0.3
+nvidia-cufile==1.15.1.6
+typer==0.27.0
+torchaudio==2.6.0+cu124
+rich==15.0.0
+nvidia-cufft-cu12==11.2.1.3
+h11==0.16.0
+dill==0.4.1
+cuda-pathfinder==1.6.0
+filelock==3.29.0
+nvidia-nvtx-cu12==12.4.127
+httpx==0.28.1
+anyio==4.14.2
+numpy==2.4.4
+yarl==1.24.5
+click==8.4.2
+triton==3.2.0
+frozenlist==1.8.0
+zipp==4.1.0
+propcache==0.5.2
+tokenizers==0.22.2
+markdown-it-py==4.2.0
+nvidia-cuda-runtime==13.0.96
+cuda-bindings==13.3.1
+nvidia-cuda-cupti==13.0.85
+torch==2.6.0+cu124
+multiprocess==0.70.19
+pillow==12.2.0
+transformers==5.15.0.dev0
+wandb==0.28.1
+nvidia-curand==10.4.0.35
+sympy==1.13.1
+nvidia-cusparse==12.6.3.3
+nvidia-cuda-nvrtc==13.0.88
+typing-inspection==0.4.2
+nvidia-cusolver==12.0.4.66
+nvidia-cufft==12.0.0.61
+nvidia-cudnn-cu13==9.20.0.48
+nvidia-cublas==13.1.1.3
+pyarrow==25.0.0
+evaluate==0.4.6
+diffusers==0.39.0
+pydantic==2.13.4
+annotated-types==0.8.0
+protobuf==7.35.1
+sentry-sdk==2.66.1
+einops==0.8.2
+packaging==26.2
+nvidia-nvjitlink-cu12==12.4.127
+nvidia-curand-cu12==10.3.5.147
+nvidia-cusparselt-cu12==0.6.2
+nvidia-cusparse-cu12==12.3.1.170
+nvidia-cuda-runtime-cu12==12.4.127
+torchvision==0.21.0+cu124
+nvidia-cuda-nvrtc-cu12==12.4.127
+nvidia-cuda-cupti-cu12==12.4.127
+nvidia-cusolver-cu12==11.6.1.9
+nvidia-cublas-cu12==12.4.5.8
+nvidia-cudnn-cu12==9.1.0.70
+huggingface_hub==1.26.0
+datasets==5.0.1
+safetensors==0.8.0
+accelerate==1.14.0
+pydantic_core==2.46.4
+ninja==1.13.0
+autocommand==2.2.2
+backports.tarfile==1.2.0
+importlib_metadata==8.7.1
+jaraco.text==4.0.0
+jaraco.context==6.1.0
+jaraco.functools==4.4.0
+more-itertools==10.8.0
+packaging==26.0
+platformdirs==4.4.0
+tomli==2.4.0
+wheel==0.46.3
+zipp==3.23.0

wandb/run-20260809_050050-59pftr14/files/wandb-metadata.json

@@ -0,0 +1,96 @@
+{
+  "os":  "Linux-5.15.0-126-generic-x86_64-with-glibc2.35",
+  "python":  "CPython 3.11.15",
+  "startedAt":  "2026-08-09T05:00:50.313572Z",
+  "args":  [
+    "--config",
+    "configs/baseline.yaml",
+    "--variants",
+    "glu-linear-94L",
+    "--push"
+  ],
+  "program":  "/mnt/data/zainulabideen/zain-exp/notebooks/Activation/sweep.py",
+  "codePath":  "sweep.py",
+  "codePathLocal":  "sweep.py",
+  "git":  {
+    "remote":  "https://github.com/deepnevro/Activation.git",
+    "commit":  "34b8d2e8f9a0c5751333310e69fa0c1056381deb"
+  },
+  "email":  "deepnevro@gmail.com",
+  "root":  "/mnt/data/zainulabideen/zain-exp/notebooks/Activation",
+  "host":  "deeplens-k3s-node1",
+  "executable":  "/mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python",
+  "cpu_count":  112,
+  "cpu_count_logical":  224,
+  "gpu":  "NVIDIA H100 80GB HBM3",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1560765693952",
+      "used":  "708235583488"
+    }
+  },
+  "memory":  {
+    "total":  "2164089937920"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-39c684a5-fde6-83d7-1663-0859795881ae"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-132944c4-b689-2b5f-89a4-d730401677ab"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bc6c3e3c-9b90-09ca-c034-774961847c54"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea"
+    }
+  ],
+  "cudaVersion":  "12.4",
+  "writerId":  "ou0ddnt36zj23smd7g76a30q6z1cg54s"
+}

wandb/run-20260809_050050-59pftr14/logs/debug-core.log

@@ -0,0 +1,58 @@
+{"time":"2026-08-09T03:56:53.721154509Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpyqomfhgp/port-2869678.txt","pid":2869678,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-08-09T03:56:53.722258921Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":2869678}
+{"time":"2026-08-09T03:56:53.722234577Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-2869678-2909401-2521299324/socket","Net":"unix"}}
+{"time":"2026-08-09T03:56:53.900147851Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-08-09T03:58:19.204299995Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"2(@)"}
+{"time":"2026-08-09T03:58:19.282614427Z","level":"INFO","msg":"handleInformInit: received","streamId":"cvzjg5ej","id":"2(@)"}
+{"time":"2026-08-09T03:58:19.544211049Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"cvzjg5ej","id":"2(@)"}
+{"time":"2026-08-09T03:58:24.897517362Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"xv8tfjtqu29x"}
+{"time":"2026-08-09T05:00:38.67636271Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"xv8tfjtqu29x"}
+{"time":"2026-08-09T05:00:40.520576799Z","level":"INFO","msg":"connection: cancelling request","id":"2(@)","requestId":"xv8tfjtqu29x"}
+{"time":"2026-08-09T05:00:40.552184373Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"cvzjg5ej","id":"2(@)"}
+{"time":"2026-08-09T05:00:40.553143275Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"cvzjg5ej","id":"2(@)"}
+{"time":"2026-08-09T05:00:42.608126464Z","level":"INFO","msg":"processOutgoingData: finished","id":"2(@)"}
+{"time":"2026-08-09T05:00:42.608106419Z","level":"INFO","msg":"connection: closing","id":"2(@)"}
+{"time":"2026-08-09T05:00:42.608216462Z","level":"INFO","msg":"connection: closed successfully","id":"2(@)"}
+{"time":"2026-08-09T05:00:42.608222262Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"2(@)"}
+{"time":"2026-08-09T05:00:50.241711721Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"3(@)"}
+{"time":"2026-08-09T05:00:50.316234404Z","level":"INFO","msg":"handleInformInit: received","streamId":"59pftr14","id":"3(@)"}
+{"time":"2026-08-09T05:00:50.57530289Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"59pftr14","id":"3(@)"}
+{"time":"2026-08-09T05:00:55.905488223Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"4jxnuihlia82"}
+{"time":"2026-08-09T05:57:13.765576248Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"4jxnuihlia82"}
+{"time":"2026-08-09T05:57:15.666017984Z","level":"INFO","msg":"connection: cancelling request","id":"3(@)","requestId":"4jxnuihlia82"}
+{"time":"2026-08-09T05:57:15.999725293Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"59pftr14","id":"3(@)"}
+{"time":"2026-08-09T05:57:16.001240777Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"59pftr14","id":"3(@)"}
+{"time":"2026-08-09T05:57:18.068849513Z","level":"INFO","msg":"connection: closing","id":"3(@)"}
+{"time":"2026-08-09T05:57:18.068936758Z","level":"INFO","msg":"connection: closed successfully","id":"3(@)"}
+{"time":"2026-08-09T05:57:18.068853914Z","level":"INFO","msg":"processOutgoingData: finished","id":"3(@)"}
+{"time":"2026-08-09T05:57:18.068948948Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"3(@)"}
+{"time":"2026-08-09T05:57:26.287713024Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"4(@)"}
+{"time":"2026-08-09T05:57:26.365087395Z","level":"INFO","msg":"handleInformInit: received","streamId":"m1dnjnh6","id":"4(@)"}
+{"time":"2026-08-09T05:57:26.623707494Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"m1dnjnh6","id":"4(@)"}
+{"time":"2026-08-09T05:57:31.962348796Z","level":"INFO","msg":"connection: cancelling request","id":"4(@)","requestId":"rz2ldpq16nht"}
+{"time":"2026-08-09T07:02:02.153459338Z","level":"INFO","msg":"connection: cancelling request","id":"4(@)","requestId":"rz2ldpq16nht"}
+{"time":"2026-08-09T07:02:04.144735717Z","level":"INFO","msg":"connection: cancelling request","id":"4(@)","requestId":"rz2ldpq16nht"}
+{"time":"2026-08-09T07:02:04.180963791Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"m1dnjnh6","id":"4(@)"}
+{"time":"2026-08-09T07:02:04.181861619Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"m1dnjnh6","id":"4(@)"}
+{"time":"2026-08-09T07:02:06.233097548Z","level":"INFO","msg":"connection: closing","id":"4(@)"}
+{"time":"2026-08-09T07:02:06.233183816Z","level":"INFO","msg":"connection: closed successfully","id":"4(@)"}
+{"time":"2026-08-09T07:02:06.233105065Z","level":"INFO","msg":"processOutgoingData: finished","id":"4(@)"}
+{"time":"2026-08-09T07:02:06.233192773Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"4(@)"}
+{"time":"2026-08-09T07:02:13.885910885Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"5(@)"}
+{"time":"2026-08-09T07:02:13.954769181Z","level":"INFO","msg":"handleInformInit: received","streamId":"uvqyddz0","id":"5(@)"}
+{"time":"2026-08-09T07:02:14.215272022Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"uvqyddz0","id":"5(@)"}
+{"time":"2026-08-09T07:02:19.530617395Z","level":"INFO","msg":"connection: cancelling request","id":"5(@)","requestId":"fk71ydt1h8f7"}
+{"time":"2026-08-09T07:03:39.9336748Z","level":"INFO","msg":"connection: cancelling request","id":"5(@)","requestId":"fk71ydt1h8f7"}
+{"time":"2026-08-09T07:03:40.29901782Z","level":"INFO","msg":"connection: closing","id":"5(@)"}
+{"time":"2026-08-09T07:03:40.299111085Z","level":"INFO","msg":"connection: closed successfully","id":"5(@)"}
+{"time":"2026-08-09T07:03:40.299007456Z","level":"INFO","msg":"processOutgoingData: finished","id":"5(@)"}
+{"time":"2026-08-09T07:03:40.299122073Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"5(@)"}
+{"time":"2026-08-09T07:03:42.349833633Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-08-09T07:03:42.349942779Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-08-09T07:03:42.34985531Z","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-08-09T07:03:42.349954994Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-08-09T07:03:42.353940558Z","level":"INFO","msg":"server: parent process exited, terminating service process"}
+{"time":"2026-08-09T07:03:42.353988499Z","level":"INFO","msg":"server: is shutting down"}
+{"time":"2026-08-09T07:03:42.354128892Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-2869678-2909401-2521299324/socket","Net":"unix"}}
+{"time":"2026-08-09T07:03:42.354189401Z","level":"INFO","msg":"server: forced shutdown"}
+{"time":"2026-08-09T07:03:42.354198506Z","level":"ERROR","msg":"main: Serve() returned error","error":"forced shutdown"}

wandb/run-20260809_050050-59pftr14/logs/debug-internal.log

@@ -0,0 +1,485 @@
+{"time":"2026-08-09T05:00:50.316382325Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-09T05:00:50.316525497Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-09T05:00:50.575130432Z","level":"INFO","msg":"stream: created new stream","id":"59pftr14"}
+{"time":"2026-08-09T05:00:50.575204351Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-09T05:00:50.575297084Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-09T05:00:50.575306959Z","level":"INFO","msg":"writer: started","stream_id":"59pftr14"}
+{"time":"2026-08-09T05:00:50.575368938Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-09T05:00:51.480743748Z","level":"INFO","msg":"filestream: sending request","total_files":1,"console_offset":0,"console_lines":1}
+{"time":"2026-08-09T05:00:51.561240291Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:01:06.481345427Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":0,"events_lines":2,"console_offset":0,"console_lines":5,"uploaded_len":2}
+{"time":"2026-08-09T05:01:06.588785368Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:01:21.481259493Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":2,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:01:21.590035378Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:01:32.681908293Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":384}
+{"time":"2026-08-09T05:01:32.68198076Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":1}
+{"time":"2026-08-09T05:01:32.691539462Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":2462}
+{"time":"2026-08-09T05:01:32.692719035Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":335}
+{"time":"2026-08-09T05:01:32.698314928Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":4822}
+{"time":"2026-08-09T05:01:32.698424691Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":14}
+{"time":"2026-08-09T05:01:32.702256488Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":6358}
+{"time":"2026-08-09T05:01:32.702387381Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":15}
+{"time":"2026-08-09T05:01:32.706713558Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":7253}
+{"time":"2026-08-09T05:01:32.708196871Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":93}
+{"time":"2026-08-09T05:01:32.712656212Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":8892}
+{"time":"2026-08-09T05:01:32.725765023Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":5668}
+{"time":"2026-08-09T05:01:32.727393292Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":14924}
+{"time":"2026-08-09T05:01:32.727423186Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":1}
+{"time":"2026-08-09T05:01:32.727732623Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":15045}
+{"time":"2026-08-09T05:01:32.727835605Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":9}
+{"time":"2026-08-09T05:01:32.730442968Z","level":"INFO","msg":"flowcontrol: backed up, offloading to disk","recordNumber":16070}
+{"time":"2026-08-09T05:01:32.737046002Z","level":"INFO","msg":"flowcontrol: unblocked","totalOffloaded":2678}
+{"time":"2026-08-09T05:01:36.519407838Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":0,"history_lines":1,"events_offset":4,"events_lines":2,"console_offset":4,"console_lines":2}
+{"time":"2026-08-09T05:01:37.706004293Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:01:51.481365298Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":6,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:01:51.596320975Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:02:06.480925245Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":8,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:02:06.586636942Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:02:21.498041897Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":1,"history_lines":1,"events_offset":10,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:02:22.532929656Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:02:36.481659249Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":12,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:02:36.621746818Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:02:51.507187931Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":2,"history_lines":1,"events_offset":14,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:02:52.538965872Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:03:06.497711137Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":3,"history_lines":1,"events_offset":16,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:03:07.406445161Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:03:21.481660818Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":18,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:03:21.644645315Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:03:36.481585895Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":20,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:03:36.591305248Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:03:51.499410115Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":4,"history_lines":1,"events_offset":22,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:03:52.420339049Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:04:06.481223309Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":24,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:04:06.586916017Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:04:21.499976608Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":5,"history_lines":1,"events_offset":26,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:04:22.440880355Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:04:36.503115163Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":6,"history_lines":1,"events_offset":28,"events_lines":2,"console_offset":4,"console_lines":1}
+{"time":"2026-08-09T05:04:37.434101869Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:04:51.480868859Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":30,"events_lines":2,"console_offset":6,"console_lines":11}
+{"time":"2026-08-09T05:04:51.58771127Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:05:06.48149862Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":32,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:05:06.617849821Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:05:21.498001668Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":7,"history_lines":1,"events_offset":34,"events_lines":2,"console_offset":16,"console_lines":2}
+{"time":"2026-08-09T05:05:22.532059405Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:05:36.481272157Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":36,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:05:36.611686864Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:05:51.50349838Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":8,"history_lines":1,"events_offset":38,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:05:52.42872915Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:06:06.481067013Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":40,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:06:06.624376428Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:06:21.481451995Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":42,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:06:21.580400913Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:06:36.498750788Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":9,"history_lines":1,"events_offset":44,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:06:37.479364541Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:06:51.499950008Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":10,"history_lines":1,"events_offset":46,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:06:52.424905167Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:07:06.480837832Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":48,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:07:06.5879285Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:07:21.497769405Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":11,"history_lines":1,"events_offset":50,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:07:22.420651885Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:07:36.481679224Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":52,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:07:36.586207773Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:07:51.48168829Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":54,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:07:51.608316231Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:08:06.497385296Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":12,"history_lines":1,"events_offset":56,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:08:07.414492478Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:08:21.498696248Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":13,"history_lines":1,"events_offset":58,"events_lines":2,"console_offset":16,"console_lines":1}
+{"time":"2026-08-09T05:08:22.472074441Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:08:36.481080877Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":60,"events_lines":2,"console_offset":18,"console_lines":11}
+{"time":"2026-08-09T05:08:36.593043174Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:08:51.481213802Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":62,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:08:51.601597348Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:09:06.498299771Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":14,"history_lines":1,"events_offset":64,"events_lines":2,"console_offset":28,"console_lines":2}
+{"time":"2026-08-09T05:09:07.594657221Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:09:21.480997622Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":66,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:09:21.587300995Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:09:36.500970166Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":15,"history_lines":1,"events_offset":68,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:09:37.412755991Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:09:51.481568492Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":70,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:09:51.587311446Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:10:06.503035144Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":16,"history_lines":1,"events_offset":72,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:10:07.401707034Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:10:21.481629645Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":74,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:10:21.652792794Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:10:36.504128018Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":17,"history_lines":1,"events_offset":76,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:10:37.451086196Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:10:51.481374598Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":78,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:10:51.616703256Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:11:06.498458682Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":18,"history_lines":1,"events_offset":80,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:11:07.69317861Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:11:21.481191292Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":82,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:11:21.58886328Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:11:36.481561709Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":84,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:11:36.583857842Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:11:51.497386796Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":19,"history_lines":1,"events_offset":86,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:11:52.440204013Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:12:06.497327915Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":20,"history_lines":1,"events_offset":88,"events_lines":2,"console_offset":28,"console_lines":1}
+{"time":"2026-08-09T05:12:07.396293595Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:12:21.481158326Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":90,"events_lines":2,"console_offset":30,"console_lines":11}
+{"time":"2026-08-09T05:12:21.592522674Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:12:36.503425179Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":21,"history_lines":1,"events_offset":92,"events_lines":2,"console_offset":40,"console_lines":2}
+{"time":"2026-08-09T05:12:37.416814559Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:12:51.481041129Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":94,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T05:12:51.610448799Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:13:06.481201194Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":96,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T05:13:06.625500113Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:13:21.498075795Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":22,"history_lines":1,"events_offset":98,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T05:13:22.560395397Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:13:36.481411011Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":100,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T05:13:36.605888576Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:13:51.500833721Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":23,"history_lines":1,"events_offset":102,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T05:13:52.407787379Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:14:06.497449559Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":24,"history_lines":1,"events_offset":104,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T05:14:07.390239326Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:14:21.48119166Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":106,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T05:14:21.638691331Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:14:36.481262857Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":108,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T05:14:36.634845955Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:14:51.501864135Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":25,"history_lines":1,"events_offset":110,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T05:14:52.366198559Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:15:06.481264357Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":112,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T05:15:06.577661402Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:15:21.498953568Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":26,"history_lines":1,"events_offset":114,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T05:15:22.437573941Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:15:36.501716856Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":27,"history_lines":1,"events_offset":116,"events_lines":2,"console_offset":40,"console_lines":1}
+{"time":"2026-08-09T05:15:37.369161631Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:15:51.481510049Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":118,"events_lines":2,"console_offset":42,"console_lines":11}
+{"time":"2026-08-09T05:15:51.63079258Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:16:06.481670382Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":120,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:16:06.612349198Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:16:21.497081864Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":28,"history_lines":1,"events_offset":122,"events_lines":2,"console_offset":52,"console_lines":2}
+{"time":"2026-08-09T05:16:22.429585439Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:16:36.481386631Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":124,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:16:36.716895363Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:16:51.498344862Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":29,"history_lines":1,"events_offset":126,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:16:52.573229517Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:17:06.481679973Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":128,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:17:06.6380613Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:17:21.481192289Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":130,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:17:21.615467345Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:17:36.498032747Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":30,"history_lines":1,"events_offset":132,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:17:37.37259259Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:17:51.501849113Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":31,"history_lines":1,"events_offset":134,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:17:52.461736682Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:18:06.481331592Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":136,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:18:06.664869101Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:18:21.50046419Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":32,"history_lines":1,"events_offset":138,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:18:22.389205003Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:18:36.481352846Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":140,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:18:36.69619412Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:18:51.48085544Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":142,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:18:51.599415662Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:19:06.501115812Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":33,"history_lines":1,"events_offset":144,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:19:07.486241738Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:19:21.499147294Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":34,"history_lines":1,"events_offset":146,"events_lines":2,"console_offset":52,"console_lines":1}
+{"time":"2026-08-09T05:19:22.438925205Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:19:36.481577527Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":148,"events_lines":2,"console_offset":54,"console_lines":13}
+{"time":"2026-08-09T05:19:36.608439891Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:19:51.481485963Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":150,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T05:19:51.659884173Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:20:06.525040764Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":35,"history_lines":1,"events_offset":152,"events_lines":2,"console_offset":66,"console_lines":2}
+{"time":"2026-08-09T05:20:07.675864743Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:20:21.481552724Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":154,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T05:20:21.581914998Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:20:36.500684642Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":36,"history_lines":1,"events_offset":156,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T05:20:37.523239345Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:20:51.481557042Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":158,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T05:20:51.618146016Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:21:06.481365631Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":160,"events_lines":2,"console_offset":68,"console_lines":2}
+{"time":"2026-08-09T05:21:06.603391244Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:21:21.505549859Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":37,"history_lines":1,"events_offset":162,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T05:21:22.433779828Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:21:36.502373763Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":38,"history_lines":1,"events_offset":164,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T05:21:37.46846612Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:21:51.481185661Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":166,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T05:21:51.678879589Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:22:06.504549408Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":39,"history_lines":1,"events_offset":168,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T05:22:07.431785778Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:22:21.480925067Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":170,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T05:22:21.617766324Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:22:36.481037106Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":172,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T05:22:36.631472656Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:22:51.498329537Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":40,"history_lines":1,"events_offset":174,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T05:22:52.376455957Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:23:06.502499439Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":41,"history_lines":1,"events_offset":176,"events_lines":2,"console_offset":66,"console_lines":1}
+{"time":"2026-08-09T05:23:07.406128001Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:23:21.481598535Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":178,"events_lines":2,"console_offset":69,"console_lines":10}
+{"time":"2026-08-09T05:23:21.621966825Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:23:36.498090022Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":42,"history_lines":1,"events_offset":180,"events_lines":2,"console_offset":78,"console_lines":2}
+{"time":"2026-08-09T05:23:37.601595059Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:23:51.481759024Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":182,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T05:23:51.586781813Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:24:06.481276767Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":184,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T05:24:06.598097548Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:24:21.507460244Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":43,"history_lines":1,"events_offset":186,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T05:24:22.440295279Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:24:36.481283213Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":188,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T05:24:36.610184037Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:24:51.502397715Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":44,"history_lines":1,"events_offset":190,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T05:24:52.438748096Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:25:06.501113438Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":45,"history_lines":1,"events_offset":192,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T05:25:07.42127244Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:25:21.481510164Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":194,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T05:25:21.669561341Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:25:36.481578249Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":196,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T05:25:36.594439235Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:25:51.498405234Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":46,"history_lines":1,"events_offset":198,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T05:25:52.378590133Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:26:06.480986539Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":200,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T05:26:06.638024923Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:26:21.50308839Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":47,"history_lines":1,"events_offset":202,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T05:26:22.404563682Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:26:36.48164935Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":204,"events_lines":2,"console_offset":80,"console_lines":6}
+{"time":"2026-08-09T05:26:36.608214624Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:26:51.498904739Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":48,"history_lines":1,"events_offset":206,"events_lines":2,"console_offset":78,"console_lines":1}
+{"time":"2026-08-09T05:26:52.54906005Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:27:06.481195872Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":208,"events_lines":2,"console_offset":81,"console_lines":1}
+{"time":"2026-08-09T05:27:06.694166718Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:27:21.504577552Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":49,"history_lines":1,"events_offset":210,"events_lines":2,"console_offset":86,"console_lines":6}
+{"time":"2026-08-09T05:27:22.424460773Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:27:36.481110203Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":212,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T05:27:36.604006133Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:27:51.481765731Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":214,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T05:27:51.611864716Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:28:06.498158704Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":50,"history_lines":1,"events_offset":216,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T05:28:07.409376378Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:28:21.481511802Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":218,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T05:28:21.562499861Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:28:36.50710607Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":51,"history_lines":1,"events_offset":220,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T05:28:37.487154585Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:28:51.497901326Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":52,"history_lines":1,"events_offset":222,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T05:28:52.446794195Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:29:06.481006896Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":224,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T05:29:06.604939782Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:29:21.48132648Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":226,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T05:29:21.675663702Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:29:36.502055175Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":53,"history_lines":1,"events_offset":228,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T05:29:37.492772989Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:29:51.48098667Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":230,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T05:29:51.566571956Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:30:06.498304058Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":54,"history_lines":1,"events_offset":232,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T05:30:07.476602804Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:30:21.497009392Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":55,"history_lines":1,"events_offset":234,"events_lines":2,"console_offset":90,"console_lines":1}
+{"time":"2026-08-09T05:30:22.426875321Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:30:36.481198656Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":236,"events_lines":2,"console_offset":92,"console_lines":11}
+{"time":"2026-08-09T05:30:36.569501322Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:30:51.481362955Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":238,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:30:51.58612004Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:31:06.498109113Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":56,"history_lines":1,"events_offset":240,"events_lines":2,"console_offset":102,"console_lines":2}
+{"time":"2026-08-09T05:31:07.465651065Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:31:21.481204233Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":242,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:31:21.607697687Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:31:36.501843353Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":57,"history_lines":1,"events_offset":244,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:31:37.484272619Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:31:51.481294572Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":246,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:31:51.569699481Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:32:06.481267101Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":248,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:32:06.564697645Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:32:21.501309751Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":58,"history_lines":1,"events_offset":250,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:32:22.400233409Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:32:36.502665345Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":59,"history_lines":1,"events_offset":252,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:32:37.539187444Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:32:51.481000342Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":254,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:32:51.589307243Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:33:06.497296324Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":60,"history_lines":1,"events_offset":256,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:33:07.410411774Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:33:21.480896144Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":258,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:33:21.557619437Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:33:36.481359105Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":260,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:33:36.586477785Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:33:51.501293775Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":61,"history_lines":1,"events_offset":262,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:33:52.446357727Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:34:06.501566811Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":62,"history_lines":1,"events_offset":264,"events_lines":2,"console_offset":102,"console_lines":1}
+{"time":"2026-08-09T05:34:07.431632319Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:34:21.481356503Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":266,"events_lines":2,"console_offset":104,"console_lines":11}
+{"time":"2026-08-09T05:34:21.58737243Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:34:36.481254313Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":268,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:34:36.596785368Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:34:51.498620247Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":63,"history_lines":1,"events_offset":270,"events_lines":2,"console_offset":114,"console_lines":2}
+{"time":"2026-08-09T05:34:52.447346801Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:35:06.481384808Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":272,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:35:06.584314432Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:35:21.499491427Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":64,"history_lines":1,"events_offset":274,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:35:22.406403243Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:35:36.481510694Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":276,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:35:36.570794108Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:35:51.502728659Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":65,"history_lines":1,"events_offset":278,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:35:52.433176071Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:36:06.481082158Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":280,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:36:06.56595802Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:36:21.502575907Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":66,"history_lines":1,"events_offset":282,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:36:22.396774635Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:36:36.481194169Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":284,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:36:36.564324046Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:36:51.507585877Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":67,"history_lines":1,"events_offset":286,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:36:52.450052493Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:37:06.481177189Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":288,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:37:06.562873382Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:37:21.481773468Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":290,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:37:21.578202075Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:37:36.496352178Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":68,"history_lines":1,"events_offset":292,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:37:37.378367849Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:37:51.498182421Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":69,"history_lines":1,"events_offset":294,"events_lines":2,"console_offset":114,"console_lines":1}
+{"time":"2026-08-09T05:37:52.501913241Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:38:06.481276719Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":296,"events_lines":2,"console_offset":116,"console_lines":13}
+{"time":"2026-08-09T05:38:06.575845243Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:38:21.517547479Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":70,"history_lines":1,"events_offset":298,"events_lines":2,"console_offset":128,"console_lines":2}
+{"time":"2026-08-09T05:38:22.68375147Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:38:36.48169975Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":300,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:38:36.58233458Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:38:51.481639785Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":302,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:38:51.57204535Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:39:06.502117161Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":71,"history_lines":1,"events_offset":304,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:39:07.408069696Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:39:21.481127559Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":306,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:39:21.584984533Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:39:36.500833886Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":72,"history_lines":1,"events_offset":308,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:39:37.3542055Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:39:51.499790298Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":73,"history_lines":1,"events_offset":310,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:39:52.423275386Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:40:06.481402259Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":312,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:40:06.574340065Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:40:21.481696586Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":314,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:40:21.585374046Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:40:36.500319073Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":74,"history_lines":1,"events_offset":316,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:40:37.416603366Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:40:51.481735415Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":318,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:40:51.568608859Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:41:06.481327101Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":320,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:41:06.577508114Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:41:21.504914431Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":75,"history_lines":1,"events_offset":322,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:41:22.589623821Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:41:36.500920471Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":76,"history_lines":1,"events_offset":324,"events_lines":2,"console_offset":128,"console_lines":1}
+{"time":"2026-08-09T05:41:37.445500069Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:41:51.481699831Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":326,"events_lines":2,"console_offset":130,"console_lines":11}
+{"time":"2026-08-09T05:41:51.589802594Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:42:06.499496996Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":77,"history_lines":1,"events_offset":328,"events_lines":2,"console_offset":140,"console_lines":2}
+{"time":"2026-08-09T05:42:07.490323728Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:42:21.481007487Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":330,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T05:42:21.573910407Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:42:36.480944829Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":332,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T05:42:36.58648518Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:42:51.502651759Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":78,"history_lines":1,"events_offset":334,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T05:42:52.407092114Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:43:06.48131406Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":336,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T05:43:06.572964826Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:43:21.502659005Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":79,"history_lines":1,"events_offset":338,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T05:43:22.460952902Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:43:36.498438868Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":80,"history_lines":1,"events_offset":340,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T05:43:37.435539856Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:43:51.481499575Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":342,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T05:43:51.581657982Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:44:06.480992155Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":344,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T05:44:06.575942046Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:44:21.499285771Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":81,"history_lines":1,"events_offset":346,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T05:44:22.48502359Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:44:36.481415886Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":348,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T05:44:36.590027025Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:44:51.498199637Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":82,"history_lines":1,"events_offset":350,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T05:44:52.522574549Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:45:06.507362609Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":83,"history_lines":1,"events_offset":352,"events_lines":2,"console_offset":140,"console_lines":1}
+{"time":"2026-08-09T05:45:07.437618138Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:45:21.481223976Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":354,"events_lines":2,"console_offset":142,"console_lines":11}
+{"time":"2026-08-09T05:45:21.571863157Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:45:36.481250148Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":356,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:45:36.588306221Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:45:51.498140841Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":84,"history_lines":1,"events_offset":358,"events_lines":2,"console_offset":152,"console_lines":2}
+{"time":"2026-08-09T05:45:52.40392286Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:46:06.481151661Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":360,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:46:06.574924044Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:46:21.50224276Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":85,"history_lines":1,"events_offset":362,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:46:22.468575161Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:46:36.48151792Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":364,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:46:36.557472174Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:46:51.481659212Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":366,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:46:51.583561336Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:47:06.503251857Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":86,"history_lines":1,"events_offset":368,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:47:07.524513415Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:47:21.498223553Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":87,"history_lines":1,"events_offset":370,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:47:22.436952446Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:47:36.481235917Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":372,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:47:36.58953031Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:47:51.509767914Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":88,"history_lines":1,"events_offset":374,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:47:52.472129123Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:48:06.480977725Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":376,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:48:06.572015048Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:48:21.481635627Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":378,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:48:21.572405344Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:48:36.522493716Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":89,"history_lines":1,"events_offset":380,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:48:37.446677623Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:48:51.501843279Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":90,"history_lines":1,"events_offset":382,"events_lines":2,"console_offset":152,"console_lines":1}
+{"time":"2026-08-09T05:48:52.491223677Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:49:06.481489796Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":384,"events_lines":2,"console_offset":154,"console_lines":11}
+{"time":"2026-08-09T05:49:06.586343176Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:49:21.499852391Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":91,"history_lines":1,"events_offset":386,"events_lines":2,"console_offset":164,"console_lines":2}
+{"time":"2026-08-09T05:49:22.479462348Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:49:36.481384045Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":388,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:49:36.5774241Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:49:51.481275624Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":390,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:49:51.585688059Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:50:06.501668608Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":92,"history_lines":1,"events_offset":392,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:50:07.367218244Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:50:21.481792859Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":394,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:50:21.570177867Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:50:36.481176295Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":396,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:50:36.58018685Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:50:51.496247888Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":93,"history_lines":1,"events_offset":398,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:50:52.481878699Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:51:06.497846529Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":94,"history_lines":1,"events_offset":400,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:51:07.407970871Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:51:21.481238939Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":402,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:51:21.59538021Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:51:36.481689771Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":404,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:51:36.583888684Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:51:51.499980191Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":95,"history_lines":1,"events_offset":406,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:51:52.427907364Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:52:06.481473609Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":408,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:52:06.597914659Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:52:21.499728804Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":96,"history_lines":1,"events_offset":410,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:52:22.366340564Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:52:36.498067451Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":97,"history_lines":1,"events_offset":412,"events_lines":2,"console_offset":164,"console_lines":1}
+{"time":"2026-08-09T05:52:37.449598672Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:52:51.481617633Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":414,"events_lines":2,"console_offset":166,"console_lines":11}
+{"time":"2026-08-09T05:52:51.581562046Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:53:06.481233846Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":416,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:53:06.57963203Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:53:21.499870228Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":98,"history_lines":1,"events_offset":418,"events_lines":2,"console_offset":176,"console_lines":2}
+{"time":"2026-08-09T05:53:22.41340383Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:53:36.481163388Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":420,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:53:36.57497954Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:53:51.497905926Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":99,"history_lines":1,"events_offset":422,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:53:52.368218191Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:54:06.481748072Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":424,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:54:06.578247243Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:54:21.481590391Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":426,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:54:21.595103309Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:54:36.504538021Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":100,"history_lines":1,"events_offset":428,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:54:37.426563022Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:54:51.480970743Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":430,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:54:51.573296853Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:55:06.505576648Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":101,"history_lines":1,"events_offset":432,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:55:07.439584344Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:55:21.481626327Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":434,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:55:21.573957985Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:55:36.481205668Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":436,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:55:36.566940604Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:55:51.502848806Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":102,"history_lines":1,"events_offset":438,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:55:52.39689707Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:56:06.481314164Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":440,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:56:06.569483549Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:56:21.481079544Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":442,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:56:21.588552813Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:56:36.497668927Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":103,"history_lines":1,"events_offset":444,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:56:37.466258007Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:56:51.502614572Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":104,"history_lines":2,"events_offset":446,"events_lines":2,"console_offset":176,"console_lines":1}
+{"time":"2026-08-09T05:56:52.423617477Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:57:06.481096852Z","level":"INFO","msg":"filestream: sending request","total_files":2,"events_offset":448,"events_lines":2,"console_offset":178,"console_lines":13}
+{"time":"2026-08-09T05:57:06.567967592Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:57:14.637598114Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-09T05:57:14.659564926Z","level":"INFO","msg":"filestream: sending request","total_files":3,"history_offset":106,"history_lines":1,"console_offset":190,"console_lines":31,"uploaded_len":3,"complete":true,"exit_code":0}
+{"time":"2026-08-09T05:57:15.639628923Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:57:15.641071554Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2026-08-09T05:57:15.999773399Z","level":"INFO","msg":"stream: finishing up"}
+{"time":"2026-08-09T05:57:15.999820775Z","level":"INFO","msg":"handler: closed"}
+{"time":"2026-08-09T05:57:15.999975766Z","level":"INFO","msg":"sender: closed"}
+{"time":"2026-08-09T05:57:15.999980748Z","level":"INFO","msg":"stream: all finished"}

wandb/run-20260809_050050-59pftr14/logs/debug.log

@@ -0,0 +1,28 @@
+2026-08-09 05:00:50,314 INFO    MainThread:3518447 [wandb_setup.py:_flush():81] Current SDK version is 0.28.1
+2026-08-09 05:00:50,314 INFO    MainThread:3518447 [wandb_setup.py:_flush():81] Configure stats pid to 3518447
+2026-08-09 05:00:50,314 INFO    MainThread:3518447 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-08-09 05:00:50,314 INFO    MainThread:3518447 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/Activation/wandb/run-20260809_050050-59pftr14/logs/debug.log
+2026-08-09 05:00:50,314 INFO    MainThread:3518447 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/Activation/wandb/run-20260809_050050-59pftr14/logs/debug-internal.log
+2026-08-09 05:00:50,315 INFO    MainThread:3518447 [wandb_init.py:init():772] calling init triggers
+2026-08-09 05:00:50,315 INFO    MainThread:3518447 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+config: {'_wandb': {}}
+2026-08-09 05:00:50,315 INFO    MainThread:3518447 [wandb_init.py:init():820] starting backend
+2026-08-09 05:00:50,315 INFO    MainThread:3518447 [wandb_init.py:init():826] Connected to an existing wandb-core service via WANDB_SERVICE
+2026-08-09 05:00:50,315 INFO    MainThread:3518447 [wandb_init.py:init():835] sending inform_init request
+2026-08-09 05:00:50,575 INFO    MainThread:3518447 [wandb_init.py:init():840] backend started and connected
+2026-08-09 05:00:50,579 INFO    MainThread:3518447 [wandb_init.py:init():910] updated telemetry
+2026-08-09 05:00:50,585 INFO    MainThread:3518447 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-09 05:00:50,829 INFO    MainThread:3518447 [wandb_init.py:init():978] starting run threads in backend
+2026-08-09 05:00:50,902 INFO    MainThread:3518447 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-09 05:00:50,902 INFO    MainThread:3518447 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-09 05:00:50,902 INFO    MainThread:3518447 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-09 05:00:50,903 INFO    MainThread:3518447 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-09 05:00:50,905 INFO    MainThread:3518447 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-09 05:00:50,906 INFO    MainThread:3518447 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.15.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 94, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'glu', 'activation': 'linear', 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'out/glu-linear-94L_run', 'per_device_train_batch_size': 128, 'num_train_epochs': 1, 'max_steps': 1500, 'learning_rate': 0.001, 'lr_scheduler_type': 'constant', 'lr_scheduler_kwargs': None, 'warmup_steps': 0, 'optim': 'adamw_torch_fused', 'optim_args': None, 'weight_decay': 0.01, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 4, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-glu-linear-94L-15.9M-20260809-050049', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 50, 'eval_delay': 0, 'per_device_eval_batch_size': 128, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': True, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/A-glu-linear-94L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
+2026-08-09 05:00:50,910 INFO    MainThread:3518447 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 15949440 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x15208aba5590>>
+2026-08-09 05:00:50,910 INFO    MainThread:3518447 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 15949440 None
+2026-08-09 05:57:13,764 INFO    MainThread:3518447 [wandb_run.py:_finish():2383] finishing run deepnevro-deepnevro/huggingface/59pftr14
+2026-08-09 05:57:13,765 INFO    MainThread:3518447 [wandb_run.py:_atexit_cleanup():2588] got exitcode: 0
+2026-08-09 05:57:13,765 INFO    MainThread:3518447 [wandb_run.py:_restore():2570] restore
+2026-08-09 05:57:13,765 INFO    MainThread:3518447 [wandb_run.py:_restore():2576] restore done
+2026-08-09 05:57:15,998 INFO    MainThread:3518447 [wandb_run.py:_footer_sync_info():3993] logging synced files

wandb/run-20260809_050050-59pftr14/run-59pftr14.wandb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e15846733ea8053359a05fee6d71eff0f773e49c1dc69509b91f5d7618e26991
+size 18403101

wandb/run-20260809_055134-ppvwto7l/files/config.yaml

@@ -0,0 +1,431 @@
+_name_or_path:
+    value: ""
+_wandb:
+    value:
+        cli_version: 0.28.1
+        e:
+            9oiz0iyxjogu9rr8lhkve9eqryk4v0ll:
+                args:
+                    - --config
+                    - /mnt/data/zainulabideen/zain-exp/notebooks/Activation/configs/baseline1.yaml
+                    - --variants
+                    - mlp-linear-9L
+                    - --push
+                codePath: sweep.py
+                codePathLocal: sweep.py
+                cpu_count: 112
+                cpu_count_logical: 224
+                cudaVersion: "12.4"
+                disk:
+                    /:
+                        total: "1560765693952"
+                        used: "708272566272"
+                email: deepnevro@gmail.com
+                executable: /mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python
+                git:
+                    commit: 34b8d2e8f9a0c5751333310e69fa0c1056381deb
+                    remote: https://github.com/deepnevro/Activation.git
+                gpu: NVIDIA H100 80GB HBM3
+                gpu_count: 8
+                gpu_nvidia:
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-39c684a5-fde6-83d7-1663-0859795881ae
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-132944c4-b689-2b5f-89a4-d730401677ab
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bc6c3e3c-9b90-09ca-c034-774961847c54
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea
+                host: deeplens-k3s-node1
+                memory:
+                    total: "2164089937920"
+                os: Linux-5.15.0-126-generic-x86_64-with-glibc2.35
+                program: /mnt/data/zainulabideen/zain-exp/notebooks/Activation/sweep.py
+                python: CPython 3.11.15
+                root: /mnt/data/zainulabideen/zain-exp/notebooks/Activation
+                startedAt: "2026-08-09T05:51:34.110899Z"
+                writerId: 9oiz0iyxjogu9rr8lhkve9eqryk4v0ll
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.11.15
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 19
+                - 66
+            "4": 3.11.15
+            "5": 0.28.1
+            "6": 5.15.0.dev0
+            "9":
+                "1": transformers_trainer
+            "12": 0.28.1
+            "13": linux-x86_64
+accelerator_config:
+    value:
+        dispatch_batches: null
+        even_batches: true
+        gradient_accumulation_kwargs: null
+        non_blocking: false
+        split_batches: false
+        use_seedable_sampler: true
+activation:
+    value: linear
+adam_beta1:
+    value: 0.9
+adam_beta2:
+    value: 0.999
+adam_epsilon:
+    value: 1e-08
+architectures:
+    value: null
+attention_bias:
+    value: false
+attention_dropout:
+    value: 0
+auto_find_batch_size:
+    value: false
+average_tokens_across_devices:
+    value: true
+batch_eval_metrics:
+    value: false
+bf16:
+    value: true
+bf16_full_eval:
+    value: false
+bos_token_id:
+    value: 1
+chunk_size_feed_forward:
+    value: 0
+data_seed:
+    value: 42
+dataloader_drop_last:
+    value: false
+dataloader_in_order:
+    value: true
+dataloader_multiprocessing_context:
+    value: null
+dataloader_num_workers:
+    value: 0
+dataloader_persistent_workers:
+    value: false
+dataloader_pin_memory:
+    value: true
+dataloader_prefetch_factor:
+    value: null
+ddp_backend:
+    value: null
+ddp_broadcast_buffers:
+    value: null
+ddp_bucket_cap_mb:
+    value: null
+ddp_find_unused_parameters:
+    value: null
+ddp_static_graph:
+    value: null
+ddp_timeout:
+    value: 1800
+debug:
+    value: []
+deepspeed:
+    value: null
+disable_tqdm:
+    value: false
+do_eval:
+    value: true
+do_predict:
+    value: false
+do_train:
+    value: false
+dtype:
+    value: null
+enable_jit_checkpoint:
+    value: false
+eos_token_id:
+    value: 2
+eval_accumulation_steps:
+    value: null
+eval_delay:
+    value: 0
+eval_do_concat_batches:
+    value: true
+eval_on_start:
+    value: false
+eval_steps:
+    value: 50
+eval_strategy:
+    value: steps
+eval_use_gather_object:
+    value: false
+fp16:
+    value: false
+fp16_full_eval:
+    value: false
+fsdp:
+    value: null
+fsdp_config:
+    value: null
+full_determinism:
+    value: false
+gradient_accumulation_steps:
+    value: 16
+gradient_checkpointing:
+    value: false
+gradient_checkpointing_kwargs:
+    value: null
+greater_is_better:
+    value: null
+head_dim:
+    value: 32
+hidden_act:
+    value: silu
+hidden_size:
+    value: 128
+hub_always_push:
+    value: false
+hub_model_id:
+    value: w-ahmad/finale-mlp-linear-9L
+hub_private_repo:
+    value: null
+hub_revision:
+    value: null
+hub_strategy:
+    value: every_save
+hub_token:
+    value: <HUB_TOKEN>
+id2label:
+    value:
+        "0": LABEL_0
+        "1": LABEL_1
+ignore_data_skip:
+    value: false
+include_for_metrics:
+    value: []
+include_num_input_tokens_seen:
+    value: "no"
+initializer_range:
+    value: 0.02
+intermediate_size:
+    value: 256
+is_encoder_decoder:
+    value: false
+label_names:
+    value: null
+label_smoothing_factor:
+    value: 0
+label2id:
+    value:
+        LABEL_0: 0
+        LABEL_1: 1
+learning_rate:
+    value: 0.0005
+length_column_name:
+    value: length
+liger_kernel_config:
+    value: null
+load_best_model_at_end:
+    value: false
+local_rank:
+    value: -1
+log_level:
+    value: passive
+log_level_replica:
+    value: warning
+log_on_each_node:
+    value: true
+logging_first_step:
+    value: false
+logging_nan_inf_filter:
+    value: true
+logging_steps:
+    value: 20
+logging_strategy:
+    value: steps
+lr_scheduler_kwargs:
+    value: null
+lr_scheduler_type:
+    value: constant
+max_grad_norm:
+    value: 1
+max_position_embeddings:
+    value: 512
+max_steps:
+    value: 750
+metric_for_best_model:
+    value: null
+mlp_bias:
+    value: false
+mlp_type:
+    value: mlp
+model/num_parameters:
+    value: 2001280
+model_type:
+    value: tiny_llama
+neftune_noise_alpha:
+    value: null
+num_attention_heads:
+    value: 4
+num_hidden_layers:
+    value: 9
+num_key_value_heads:
+    value: 4
+num_train_epochs:
+    value: 1
+optim:
+    value: adamw_torch_fused
+optim_args:
+    value: null
+optim_target_modules:
+    value: null
+output_attentions:
+    value: false
+output_dir:
+    value: outio/mlp-linear-9L_run
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: 0
+parallelism_config:
+    value: null
+per_device_eval_batch_size:
+    value: 80
+per_device_train_batch_size:
+    value: 80
+prediction_loss_only:
+    value: false
+pretraining_tp:
+    value: 1
+problem_type:
+    value: null
+project:
+    value: huggingface
+push_to_hub:
+    value: true
+remove_unused_columns:
+    value: false
+report_to:
+    value:
+        - wandb
+restore_callback_states_from_checkpoint:
+    value: false
+resume_from_checkpoint:
+    value: null
+return_dict:
+    value: true
+rms_norm_eps:
+    value: 1e-06
+rope_parameters:
+    value:
+        rope_theta: 10000
+        rope_type: default
+run_name:
+    value: LM-mlp-linear-9L-2.0M-20260809-055132
+save_on_each_node:
+    value: false
+save_only_model:
+    value: false
+save_steps:
+    value: 100
+save_strategy:
+    value: steps
+save_total_limit:
+    value: null
+seed:
+    value: 42
+skip_memory_metrics:
+    value: true
+tf32:
+    value: null
+tie_word_embeddings:
+    value: true
+tokenizer_name:
+    value: w-ahmad/tiny-stories-tokenizer
+torch_compile:
+    value: false
+torch_compile_backend:
+    value: null
+torch_compile_mode:
+    value: null
+torch_empty_cache_steps:
+    value: null
+trackio_bucket_id:
+    value: null
+trackio_space_id:
+    value: null
+trackio_static_space_id:
+    value: null
+train_sampling_strategy:
+    value: random
+transformers_version:
+    value: 5.15.0.dev0
+use_cache:
+    value: false
+use_cpu:
+    value: false
+use_liger_kernel:
+    value: false
+vocab_size:
+    value: 4096
+warmup_steps:
+    value: 0
+weight_decay:
+    value: 0.01

wandb/run-20260809_055134-ppvwto7l/files/requirements.txt

@@ -0,0 +1,149 @@
+asttokens==3.0.1
+comm==0.2.3
+debugpy==1.8.21
+decorator==5.3.1
+executing==2.2.1
+nest-asyncio==1.6.0
+parso==0.8.7
+platformdirs==4.11.0
+psutil==7.2.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+Pygments==2.20.0
+pyzmq==27.1.0
+setuptools==83.0.0
+six==1.17.0
+tornado==6.5.7
+traitlets==5.15.0
+fsspec==2026.4.0
+wcwidth==0.8.2
+ipython_pygments_lexers==1.1.1
+jedi==0.20.0
+jupyter_core==5.9.1
+matplotlib-inline==0.2.2
+pexpect==4.9.0
+prompt_toolkit==3.0.53
+python-dateutil==2.9.0.post0
+stack_data==0.6.3
+wheel==0.47.0
+jupyter_client==8.9.1
+pip==26.1.2
+ipython==9.15.0
+ipykernel==7.2.0
+threadpoolctl==3.6.0
+pyparsing==3.3.2
+typing_extensions==4.15.0
+Jinja2==3.1.6
+narwhals==2.24.0
+kiwisolver==1.5.0
+joblib==1.5.3
+fonttools==4.63.0
+cycler==0.12.1
+scipy==1.17.1
+pandas==3.0.5
+contourpy==1.3.3
+scikit-learn==1.9.0
+matplotlib==3.11.1
+urllib3==2.7.0
+tqdm==4.70.0
+idna==3.18
+charset-normalizer==3.4.9
+certifi==2026.7.22
+requests==2.34.2
+seaborn==0.13.2
+uv==0.12.0
+shellingham==1.5.4
+mpmath==1.3.0
+attrs==26.1.0
+hf-xet==1.5.2
+nvidia-nccl-cu12==2.21.5
+MarkupSafe==3.0.3
+regex==2026.7.19
+importlib_metadata==9.0.0
+httpcore==1.0.9
+annotated-doc==0.0.5
+multidict==6.7.1
+aiohttp==3.14.3
+aiosignal==1.4.0
+xxhash==3.8.1
+aiohappyeyeballs==2.7.1
+mdurl==0.1.2
+cuda-toolkit==13.0.3.0
+networkx==3.6.1
+PyYAML==6.0.3
+nvidia-cufile==1.15.1.6
+typer==0.27.0
+torchaudio==2.6.0+cu124
+rich==15.0.0
+nvidia-cufft-cu12==11.2.1.3
+h11==0.16.0
+dill==0.4.1
+cuda-pathfinder==1.6.0
+filelock==3.29.0
+nvidia-nvtx-cu12==12.4.127
+httpx==0.28.1
+anyio==4.14.2
+numpy==2.4.4
+yarl==1.24.5
+click==8.4.2
+triton==3.2.0
+frozenlist==1.8.0
+zipp==4.1.0
+propcache==0.5.2
+tokenizers==0.22.2
+markdown-it-py==4.2.0
+nvidia-cuda-runtime==13.0.96
+cuda-bindings==13.3.1
+nvidia-cuda-cupti==13.0.85
+torch==2.6.0+cu124
+multiprocess==0.70.19
+pillow==12.2.0
+transformers==5.15.0.dev0
+wandb==0.28.1
+nvidia-curand==10.4.0.35
+sympy==1.13.1
+nvidia-cusparse==12.6.3.3
+nvidia-cuda-nvrtc==13.0.88
+typing-inspection==0.4.2
+nvidia-cusolver==12.0.4.66
+nvidia-cufft==12.0.0.61
+nvidia-cudnn-cu13==9.20.0.48
+nvidia-cublas==13.1.1.3
+pyarrow==25.0.0
+evaluate==0.4.6
+diffusers==0.39.0
+pydantic==2.13.4
+annotated-types==0.8.0
+protobuf==7.35.1
+sentry-sdk==2.66.1
+einops==0.8.2
+packaging==26.2
+nvidia-nvjitlink-cu12==12.4.127
+nvidia-curand-cu12==10.3.5.147
+nvidia-cusparselt-cu12==0.6.2
+nvidia-cusparse-cu12==12.3.1.170
+nvidia-cuda-runtime-cu12==12.4.127
+torchvision==0.21.0+cu124
+nvidia-cuda-nvrtc-cu12==12.4.127
+nvidia-cuda-cupti-cu12==12.4.127
+nvidia-cusolver-cu12==11.6.1.9
+nvidia-cublas-cu12==12.4.5.8
+nvidia-cudnn-cu12==9.1.0.70
+huggingface_hub==1.26.0
+datasets==5.0.1
+safetensors==0.8.0
+accelerate==1.14.0
+pydantic_core==2.46.4
+ninja==1.13.0
+autocommand==2.2.2
+backports.tarfile==1.2.0
+importlib_metadata==8.7.1
+jaraco.text==4.0.0
+jaraco.context==6.1.0
+jaraco.functools==4.4.0
+more-itertools==10.8.0
+packaging==26.0
+platformdirs==4.4.0
+tomli==2.4.0
+wheel==0.46.3
+zipp==3.23.0

wandb/run-20260809_055134-ppvwto7l/files/wandb-metadata.json

@@ -0,0 +1,96 @@
+{
+  "os":  "Linux-5.15.0-126-generic-x86_64-with-glibc2.35",
+  "python":  "CPython 3.11.15",
+  "startedAt":  "2026-08-09T05:51:34.110899Z",
+  "args":  [
+    "--config",
+    "/mnt/data/zainulabideen/zain-exp/notebooks/Activation/configs/baseline1.yaml",
+    "--variants",
+    "mlp-linear-9L",
+    "--push"
+  ],
+  "program":  "/mnt/data/zainulabideen/zain-exp/notebooks/Activation/sweep.py",
+  "codePath":  "sweep.py",
+  "codePathLocal":  "sweep.py",
+  "git":  {
+    "remote":  "https://github.com/deepnevro/Activation.git",
+    "commit":  "34b8d2e8f9a0c5751333310e69fa0c1056381deb"
+  },
+  "email":  "deepnevro@gmail.com",
+  "root":  "/mnt/data/zainulabideen/zain-exp/notebooks/Activation",
+  "host":  "deeplens-k3s-node1",
+  "executable":  "/mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python",
+  "cpu_count":  112,
+  "cpu_count_logical":  224,
+  "gpu":  "NVIDIA H100 80GB HBM3",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1560765693952",
+      "used":  "708272566272"
+    }
+  },
+  "memory":  {
+    "total":  "2164089937920"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-39c684a5-fde6-83d7-1663-0859795881ae"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-132944c4-b689-2b5f-89a4-d730401677ab"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bc6c3e3c-9b90-09ca-c034-774961847c54"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea"
+    }
+  ],
+  "cudaVersion":  "12.4",
+  "writerId":  "9oiz0iyxjogu9rr8lhkve9eqryk4v0ll"
+}

wandb/run-20260809_055134-ppvwto7l/files/wandb-summary.json

@@ -0,0 +1 @@
+{"_runtime":9,"_wandb":{"runtime":9}}

wandb/run-20260809_055134-ppvwto7l/logs/debug-core.log

@@ -0,0 +1,20 @@
+{"time":"2026-08-09T05:51:33.863685591Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmpsns932q3/port-4000224.txt","pid":4000224,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-08-09T05:51:33.864202947Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":4000224}
+{"time":"2026-08-09T05:51:33.86420271Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-4000224-4001478-2427168253/socket","Net":"unix"}}
+{"time":"2026-08-09T05:51:34.042568412Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-08-09T05:51:34.118653333Z","level":"INFO","msg":"handleInformInit: received","streamId":"ppvwto7l","id":"1(@)"}
+{"time":"2026-08-09T05:51:34.4012495Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"ppvwto7l","id":"1(@)"}
+{"time":"2026-08-09T05:51:39.812793282Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"yucvjm7hzbh6"}
+{"time":"2026-08-09T05:51:44.150194715Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"yucvjm7hzbh6"}
+{"time":"2026-08-09T05:51:44.547236455Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"yucvjm7hzbh6"}
+{"time":"2026-08-09T05:51:44.5482416Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"ppvwto7l","id":"1(@)"}
+{"time":"2026-08-09T05:51:44.549062518Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"ppvwto7l","id":"1(@)"}
+{"time":"2026-08-09T05:51:44.550900825Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2026-08-09T05:51:44.550911085Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2026-08-09T05:51:44.550915633Z","level":"INFO","msg":"server: is shutting down"}
+{"time":"2026-08-09T05:51:44.550916381Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-08-09T05:51:44.550929029Z","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-08-09T05:51:44.550961866Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-08-09T05:51:44.550965473Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-08-09T05:51:44.551073051Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-4000224-4001478-2427168253/socket","Net":"unix"}}
+{"time":"2026-08-09T05:51:44.551115443Z","level":"INFO","msg":"server: all connections closed"}

wandb/run-20260809_055134-ppvwto7l/logs/debug-internal.log

@@ -0,0 +1,17 @@
+{"time":"2026-08-09T05:51:34.118928528Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-09T05:51:34.119531712Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-09T05:51:34.401063124Z","level":"INFO","msg":"stream: created new stream","id":"ppvwto7l"}
+{"time":"2026-08-09T05:51:34.401149755Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-09T05:51:34.401243417Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-09T05:51:34.401264715Z","level":"INFO","msg":"writer: started","stream_id":"ppvwto7l"}
+{"time":"2026-08-09T05:51:34.401273087Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-09T05:51:39.847970826Z","level":"INFO","msg":"filestream: sending request","total_files":0,"uploaded_len":1}
+{"time":"2026-08-09T05:51:39.944211852Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:51:44.459083269Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-09T05:51:44.459268582Z","level":"INFO","msg":"filestream: sending request","total_files":1,"uploaded_len":3,"complete":true,"exit_code":0}
+{"time":"2026-08-09T05:51:44.544669385Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:51:44.546111568Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2026-08-09T05:51:44.548260935Z","level":"INFO","msg":"stream: finishing up"}
+{"time":"2026-08-09T05:51:44.548274074Z","level":"INFO","msg":"handler: closed"}
+{"time":"2026-08-09T05:51:44.548332472Z","level":"INFO","msg":"sender: closed"}
+{"time":"2026-08-09T05:51:44.548335957Z","level":"INFO","msg":"stream: all finished"}

wandb/run-20260809_055134-ppvwto7l/logs/debug.log

@@ -0,0 +1,27 @@
+2026-08-09 05:51:34,116 INFO    MainThread:4000224 [wandb_setup.py:_flush():81] Current SDK version is 0.28.1
+2026-08-09 05:51:34,116 INFO    MainThread:4000224 [wandb_setup.py:_flush():81] Configure stats pid to 4000224
+2026-08-09 05:51:34,116 INFO    MainThread:4000224 [wandb_setup.py:_flush():81] Loading settings from environment variables
+2026-08-09 05:51:34,116 INFO    MainThread:4000224 [wandb_init.py:setup_run_log_directory():729] Logging user logs to /mnt/data/zainulabideen/zain-exp/notebooks/Activation/wandb/run-20260809_055134-ppvwto7l/logs/debug.log
+2026-08-09 05:51:34,116 INFO    MainThread:4000224 [wandb_init.py:setup_run_log_directory():730] Logging internal logs to /mnt/data/zainulabideen/zain-exp/notebooks/Activation/wandb/run-20260809_055134-ppvwto7l/logs/debug-internal.log
+2026-08-09 05:51:34,117 INFO    MainThread:4000224 [wandb_init.py:init():772] calling init triggers
+2026-08-09 05:51:34,117 INFO    MainThread:4000224 [wandb_init.py:init():777] wandb.init called with sweep_config: {}
+config: {'_wandb': {}}
+2026-08-09 05:51:34,117 INFO    MainThread:4000224 [wandb_init.py:init():820] starting backend
+2026-08-09 05:51:34,117 INFO    MainThread:4000224 [wandb_init.py:init():835] sending inform_init request
+2026-08-09 05:51:34,401 INFO    MainThread:4000224 [wandb_init.py:init():840] backend started and connected
+2026-08-09 05:51:34,403 INFO    MainThread:4000224 [wandb_init.py:init():910] updated telemetry
+2026-08-09 05:51:34,410 INFO    MainThread:4000224 [wandb_init.py:init():933] communicating run to backend with 90.0 second timeout
+2026-08-09 05:51:34,638 INFO    MainThread:4000224 [wandb_init.py:init():978] starting run threads in backend
+2026-08-09 05:51:34,712 INFO    MainThread:4000224 [wandb_run.py:_console_start():2621] atexit reg
+2026-08-09 05:51:34,712 INFO    MainThread:4000224 [wandb_run.py:_redirect():2471] redirect: wrap_raw
+2026-08-09 05:51:34,712 INFO    MainThread:4000224 [wandb_run.py:_redirect():2540] Wrapping output streams.
+2026-08-09 05:51:34,712 INFO    MainThread:4000224 [wandb_run.py:_redirect():2563] Redirects installed.
+2026-08-09 05:51:34,715 INFO    MainThread:4000224 [wandb_init.py:init():1016] run started, returning control to user process
+2026-08-09 05:51:34,716 INFO    MainThread:4000224 [wandb_run.py:_config_callback():1346] config_cb None None {'transformers_version': '5.15.0.dev0', 'architectures': None, 'output_hidden_states': False, 'return_dict': True, 'dtype': None, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'problem_type': None, 'vocab_size': 4096, 'hidden_size': 128, 'intermediate_size': 256, 'num_hidden_layers': 9, 'num_attention_heads': 4, 'num_key_value_heads': 4, 'hidden_act': 'silu', 'max_position_embeddings': 512, 'initializer_range': 0.02, 'rms_norm_eps': 1e-06, 'use_cache': False, 'pad_token_id': 0, 'bos_token_id': 1, 'eos_token_id': 2, 'pretraining_tp': 1, 'tie_word_embeddings': True, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'attention_bias': False, 'attention_dropout': 0.0, 'mlp_bias': False, 'head_dim': 32, '_name_or_path': '', 'tokenizer_name': 'w-ahmad/tiny-stories-tokenizer', 'mlp_type': 'mlp', 'activation': 'linear', 'model_type': 'tiny_llama', 'output_attentions': False, 'output_dir': 'outio/mlp-linear-9L_run', 'per_device_train_batch_size': 80, 'num_train_epochs': 1, 'max_steps': 750, 'learning_rate': 0.0005, 'lr_scheduler_type': 'constant', 'lr_scheduler_kwargs': None, 'warmup_steps': 0, 'optim': 'adamw_torch_fused', 'optim_args': None, 'weight_decay': 0.01, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'optim_target_modules': None, 'gradient_accumulation_steps': 16, 'average_tokens_across_devices': True, 'max_grad_norm': 1.0, 'label_smoothing_factor': 0.0, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'use_liger_kernel': False, 'liger_kernel_config': None, 'neftune_noise_alpha': None, 'torch_empty_cache_steps': None, 'auto_find_batch_size': False, 'logging_strategy': 'steps', 'logging_steps': 20, 'logging_first_step': False, 'log_on_each_node': True, 'logging_nan_inf_filter': True, 'include_num_input_tokens_seen': 'no', 'log_level': 'passive', 'log_level_replica': 'warning', 'disable_tqdm': False, 'report_to': ['wandb'], 'run_name': 'LM-mlp-linear-9L-2.0M-20260809-055132', 'project': 'huggingface', 'trackio_space_id': None, 'trackio_bucket_id': None, 'trackio_static_space_id': None, 'eval_strategy': 'steps', 'eval_steps': 50, 'eval_delay': 0, 'per_device_eval_batch_size': 80, 'prediction_loss_only': False, 'eval_on_start': False, 'eval_do_concat_batches': True, 'eval_use_gather_object': False, 'eval_accumulation_steps': None, 'include_for_metrics': [], 'batch_eval_metrics': False, 'save_only_model': False, 'save_strategy': 'steps', 'save_steps': 100, 'save_on_each_node': False, 'save_total_limit': None, 'enable_jit_checkpoint': False, 'push_to_hub': True, 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_model_id': 'w-ahmad/finale-mlp-linear-9L', 'hub_strategy': 'every_save', 'hub_always_push': False, 'hub_revision': None, 'load_best_model_at_end': False, 'metric_for_best_model': None, 'greater_is_better': None, 'ignore_data_skip': False, 'restore_callback_states_from_checkpoint': False, 'full_determinism': False, 'seed': 42, 'data_seed': 42, 'use_cpu': False, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'dataloader_drop_last': False, 'dataloader_num_workers': 0, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'dataloader_prefetch_factor': None, 'dataloader_multiprocessing_context': None, 'dataloader_in_order': True, 'remove_unused_columns': False, 'label_names': None, 'train_sampling_strategy': 'random', 'length_column_name': 'length', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'ddp_static_graph': None, 'ddp_backend': None, 'ddp_timeout': 1800, 'fsdp': None, 'fsdp_config': None, 'deepspeed': None, 'debug': [], 'skip_memory_metrics': True, 'do_train': False, 'do_eval': True, 'do_predict': False, 'resume_from_checkpoint': None, 'local_rank': -1}
+2026-08-09 05:51:34,718 INFO    MainThread:4000224 [wandb_config.py:__setitem__():155] [no run ID] config set model/num_parameters = 2001280 - <bound method Run._config_callback of <wandb.sdk.wandb_run.Run object at 0x149ea5fc4bd0>>
+2026-08-09 05:51:34,718 INFO    MainThread:4000224 [wandb_run.py:_config_callback():1346] config_cb model/num_parameters 2001280 None
+2026-08-09 05:51:44,149 INFO    MainThread:4000224 [wandb_run.py:_finish():2383] finishing run deepnevro-deepnevro/huggingface/ppvwto7l
+2026-08-09 05:51:44,149 INFO    MainThread:4000224 [wandb_run.py:_atexit_cleanup():2588] got exitcode: 0
+2026-08-09 05:51:44,149 INFO    MainThread:4000224 [wandb_run.py:_restore():2570] restore
+2026-08-09 05:51:44,149 INFO    MainThread:4000224 [wandb_run.py:_restore():2576] restore done
+2026-08-09 05:51:44,547 INFO    MainThread:4000224 [wandb_run.py:_footer_sync_info():3993] logging synced files

wandb/run-20260809_055154-xr7l4yvo/files/config.yaml

@@ -0,0 +1,431 @@
+_name_or_path:
+    value: ""
+_wandb:
+    value:
+        cli_version: 0.28.1
+        e:
+            78a21vuu11ek41ydatzvn60d6p6iz5jh:
+                args:
+                    - --config
+                    - /mnt/data/zainulabideen/zain-exp/notebooks/Activation/configs/baseline1.yaml
+                    - --variants
+                    - mlp-tanh-9L
+                    - --push
+                codePath: sweep.py
+                codePathLocal: sweep.py
+                cpu_count: 112
+                cpu_count_logical: 224
+                cudaVersion: "12.4"
+                disk:
+                    /:
+                        total: "1560765693952"
+                        used: "708272943104"
+                email: deepnevro@gmail.com
+                executable: /mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python
+                git:
+                    commit: 34b8d2e8f9a0c5751333310e69fa0c1056381deb
+                    remote: https://github.com/deepnevro/Activation.git
+                gpu: NVIDIA H100 80GB HBM3
+                gpu_count: 8
+                gpu_nvidia:
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-39c684a5-fde6-83d7-1663-0859795881ae
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-132944c4-b689-2b5f-89a4-d730401677ab
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-bc6c3e3c-9b90-09ca-c034-774961847c54
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9
+                    - architecture: Hopper
+                      cudaCores: 16896
+                      memoryTotal: "85520809984"
+                      name: NVIDIA H100 80GB HBM3
+                      uuid: GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea
+                host: deeplens-k3s-node1
+                memory:
+                    total: "2164089937920"
+                os: Linux-5.15.0-126-generic-x86_64-with-glibc2.35
+                program: /mnt/data/zainulabideen/zain-exp/notebooks/Activation/sweep.py
+                python: CPython 3.11.15
+                root: /mnt/data/zainulabideen/zain-exp/notebooks/Activation
+                startedAt: "2026-08-09T05:51:54.554743Z"
+                writerId: 78a21vuu11ek41ydatzvn60d6p6iz5jh
+        m:
+            - "1": train/global_step
+              "6":
+                - 3
+              "7": []
+            - "2": '*'
+              "5": 1
+              "6":
+                - 1
+              "7": []
+        python_version: 3.11.15
+        t:
+            "1":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "2":
+                - 1
+                - 5
+                - 11
+                - 41
+                - 49
+                - 51
+                - 53
+                - 71
+            "3":
+                - 2
+                - 7
+                - 13
+                - 19
+                - 66
+            "4": 3.11.15
+            "5": 0.28.1
+            "6": 5.15.0.dev0
+            "9":
+                "1": transformers_trainer
+            "12": 0.28.1
+            "13": linux-x86_64
+accelerator_config:
+    value:
+        dispatch_batches: null
+        even_batches: true
+        gradient_accumulation_kwargs: null
+        non_blocking: false
+        split_batches: false
+        use_seedable_sampler: true
+activation:
+    value: tanh
+adam_beta1:
+    value: 0.9
+adam_beta2:
+    value: 0.999
+adam_epsilon:
+    value: 1e-08
+architectures:
+    value: null
+attention_bias:
+    value: false
+attention_dropout:
+    value: 0
+auto_find_batch_size:
+    value: false
+average_tokens_across_devices:
+    value: true
+batch_eval_metrics:
+    value: false
+bf16:
+    value: true
+bf16_full_eval:
+    value: false
+bos_token_id:
+    value: 1
+chunk_size_feed_forward:
+    value: 0
+data_seed:
+    value: 42
+dataloader_drop_last:
+    value: false
+dataloader_in_order:
+    value: true
+dataloader_multiprocessing_context:
+    value: null
+dataloader_num_workers:
+    value: 0
+dataloader_persistent_workers:
+    value: false
+dataloader_pin_memory:
+    value: true
+dataloader_prefetch_factor:
+    value: null
+ddp_backend:
+    value: null
+ddp_broadcast_buffers:
+    value: null
+ddp_bucket_cap_mb:
+    value: null
+ddp_find_unused_parameters:
+    value: null
+ddp_static_graph:
+    value: null
+ddp_timeout:
+    value: 1800
+debug:
+    value: []
+deepspeed:
+    value: null
+disable_tqdm:
+    value: false
+do_eval:
+    value: true
+do_predict:
+    value: false
+do_train:
+    value: false
+dtype:
+    value: null
+enable_jit_checkpoint:
+    value: false
+eos_token_id:
+    value: 2
+eval_accumulation_steps:
+    value: null
+eval_delay:
+    value: 0
+eval_do_concat_batches:
+    value: true
+eval_on_start:
+    value: false
+eval_steps:
+    value: 50
+eval_strategy:
+    value: steps
+eval_use_gather_object:
+    value: false
+fp16:
+    value: false
+fp16_full_eval:
+    value: false
+fsdp:
+    value: null
+fsdp_config:
+    value: null
+full_determinism:
+    value: false
+gradient_accumulation_steps:
+    value: 16
+gradient_checkpointing:
+    value: false
+gradient_checkpointing_kwargs:
+    value: null
+greater_is_better:
+    value: null
+head_dim:
+    value: 32
+hidden_act:
+    value: silu
+hidden_size:
+    value: 128
+hub_always_push:
+    value: false
+hub_model_id:
+    value: w-ahmad/finale-mlp-tanh-9L
+hub_private_repo:
+    value: null
+hub_revision:
+    value: null
+hub_strategy:
+    value: every_save
+hub_token:
+    value: <HUB_TOKEN>
+id2label:
+    value:
+        "0": LABEL_0
+        "1": LABEL_1
+ignore_data_skip:
+    value: false
+include_for_metrics:
+    value: []
+include_num_input_tokens_seen:
+    value: "no"
+initializer_range:
+    value: 0.02
+intermediate_size:
+    value: 256
+is_encoder_decoder:
+    value: false
+label_names:
+    value: null
+label_smoothing_factor:
+    value: 0
+label2id:
+    value:
+        LABEL_0: 0
+        LABEL_1: 1
+learning_rate:
+    value: 0.0005
+length_column_name:
+    value: length
+liger_kernel_config:
+    value: null
+load_best_model_at_end:
+    value: false
+local_rank:
+    value: -1
+log_level:
+    value: passive
+log_level_replica:
+    value: warning
+log_on_each_node:
+    value: true
+logging_first_step:
+    value: false
+logging_nan_inf_filter:
+    value: true
+logging_steps:
+    value: 20
+logging_strategy:
+    value: steps
+lr_scheduler_kwargs:
+    value: null
+lr_scheduler_type:
+    value: constant
+max_grad_norm:
+    value: 1
+max_position_embeddings:
+    value: 512
+max_steps:
+    value: 750
+metric_for_best_model:
+    value: null
+mlp_bias:
+    value: false
+mlp_type:
+    value: mlp
+model/num_parameters:
+    value: 2001280
+model_type:
+    value: tiny_llama
+neftune_noise_alpha:
+    value: null
+num_attention_heads:
+    value: 4
+num_hidden_layers:
+    value: 9
+num_key_value_heads:
+    value: 4
+num_train_epochs:
+    value: 1
+optim:
+    value: adamw_torch_fused
+optim_args:
+    value: null
+optim_target_modules:
+    value: null
+output_attentions:
+    value: false
+output_dir:
+    value: outio/mlp-tanh-9L_run
+output_hidden_states:
+    value: false
+pad_token_id:
+    value: 0
+parallelism_config:
+    value: null
+per_device_eval_batch_size:
+    value: 80
+per_device_train_batch_size:
+    value: 80
+prediction_loss_only:
+    value: false
+pretraining_tp:
+    value: 1
+problem_type:
+    value: null
+project:
+    value: huggingface
+push_to_hub:
+    value: true
+remove_unused_columns:
+    value: false
+report_to:
+    value:
+        - wandb
+restore_callback_states_from_checkpoint:
+    value: false
+resume_from_checkpoint:
+    value: null
+return_dict:
+    value: true
+rms_norm_eps:
+    value: 1e-06
+rope_parameters:
+    value:
+        rope_theta: 10000
+        rope_type: default
+run_name:
+    value: LM-mlp-tanh-9L-2.0M-20260809-055153
+save_on_each_node:
+    value: false
+save_only_model:
+    value: false
+save_steps:
+    value: 100
+save_strategy:
+    value: steps
+save_total_limit:
+    value: null
+seed:
+    value: 42
+skip_memory_metrics:
+    value: true
+tf32:
+    value: null
+tie_word_embeddings:
+    value: true
+tokenizer_name:
+    value: w-ahmad/tiny-stories-tokenizer
+torch_compile:
+    value: false
+torch_compile_backend:
+    value: null
+torch_compile_mode:
+    value: null
+torch_empty_cache_steps:
+    value: null
+trackio_bucket_id:
+    value: null
+trackio_space_id:
+    value: null
+trackio_static_space_id:
+    value: null
+train_sampling_strategy:
+    value: random
+transformers_version:
+    value: 5.15.0.dev0
+use_cache:
+    value: false
+use_cpu:
+    value: false
+use_liger_kernel:
+    value: false
+vocab_size:
+    value: 4096
+warmup_steps:
+    value: 0
+weight_decay:
+    value: 0.01

wandb/run-20260809_055154-xr7l4yvo/files/requirements.txt

@@ -0,0 +1,149 @@
+asttokens==3.0.1
+comm==0.2.3
+debugpy==1.8.21
+decorator==5.3.1
+executing==2.2.1
+nest-asyncio==1.6.0
+parso==0.8.7
+platformdirs==4.11.0
+psutil==7.2.2
+ptyprocess==0.7.0
+pure_eval==0.2.3
+Pygments==2.20.0
+pyzmq==27.1.0
+setuptools==83.0.0
+six==1.17.0
+tornado==6.5.7
+traitlets==5.15.0
+fsspec==2026.4.0
+wcwidth==0.8.2
+ipython_pygments_lexers==1.1.1
+jedi==0.20.0
+jupyter_core==5.9.1
+matplotlib-inline==0.2.2
+pexpect==4.9.0
+prompt_toolkit==3.0.53
+python-dateutil==2.9.0.post0
+stack_data==0.6.3
+wheel==0.47.0
+jupyter_client==8.9.1
+pip==26.1.2
+ipython==9.15.0
+ipykernel==7.2.0
+threadpoolctl==3.6.0
+pyparsing==3.3.2
+typing_extensions==4.15.0
+Jinja2==3.1.6
+narwhals==2.24.0
+kiwisolver==1.5.0
+joblib==1.5.3
+fonttools==4.63.0
+cycler==0.12.1
+scipy==1.17.1
+pandas==3.0.5
+contourpy==1.3.3
+scikit-learn==1.9.0
+matplotlib==3.11.1
+urllib3==2.7.0
+tqdm==4.70.0
+idna==3.18
+charset-normalizer==3.4.9
+certifi==2026.7.22
+requests==2.34.2
+seaborn==0.13.2
+uv==0.12.0
+shellingham==1.5.4
+mpmath==1.3.0
+attrs==26.1.0
+hf-xet==1.5.2
+nvidia-nccl-cu12==2.21.5
+MarkupSafe==3.0.3
+regex==2026.7.19
+importlib_metadata==9.0.0
+httpcore==1.0.9
+annotated-doc==0.0.5
+multidict==6.7.1
+aiohttp==3.14.3
+aiosignal==1.4.0
+xxhash==3.8.1
+aiohappyeyeballs==2.7.1
+mdurl==0.1.2
+cuda-toolkit==13.0.3.0
+networkx==3.6.1
+PyYAML==6.0.3
+nvidia-cufile==1.15.1.6
+typer==0.27.0
+torchaudio==2.6.0+cu124
+rich==15.0.0
+nvidia-cufft-cu12==11.2.1.3
+h11==0.16.0
+dill==0.4.1
+cuda-pathfinder==1.6.0
+filelock==3.29.0
+nvidia-nvtx-cu12==12.4.127
+httpx==0.28.1
+anyio==4.14.2
+numpy==2.4.4
+yarl==1.24.5
+click==8.4.2
+triton==3.2.0
+frozenlist==1.8.0
+zipp==4.1.0
+propcache==0.5.2
+tokenizers==0.22.2
+markdown-it-py==4.2.0
+nvidia-cuda-runtime==13.0.96
+cuda-bindings==13.3.1
+nvidia-cuda-cupti==13.0.85
+torch==2.6.0+cu124
+multiprocess==0.70.19
+pillow==12.2.0
+transformers==5.15.0.dev0
+wandb==0.28.1
+nvidia-curand==10.4.0.35
+sympy==1.13.1
+nvidia-cusparse==12.6.3.3
+nvidia-cuda-nvrtc==13.0.88
+typing-inspection==0.4.2
+nvidia-cusolver==12.0.4.66
+nvidia-cufft==12.0.0.61
+nvidia-cudnn-cu13==9.20.0.48
+nvidia-cublas==13.1.1.3
+pyarrow==25.0.0
+evaluate==0.4.6
+diffusers==0.39.0
+pydantic==2.13.4
+annotated-types==0.8.0
+protobuf==7.35.1
+sentry-sdk==2.66.1
+einops==0.8.2
+packaging==26.2
+nvidia-nvjitlink-cu12==12.4.127
+nvidia-curand-cu12==10.3.5.147
+nvidia-cusparselt-cu12==0.6.2
+nvidia-cusparse-cu12==12.3.1.170
+nvidia-cuda-runtime-cu12==12.4.127
+torchvision==0.21.0+cu124
+nvidia-cuda-nvrtc-cu12==12.4.127
+nvidia-cuda-cupti-cu12==12.4.127
+nvidia-cusolver-cu12==11.6.1.9
+nvidia-cublas-cu12==12.4.5.8
+nvidia-cudnn-cu12==9.1.0.70
+huggingface_hub==1.26.0
+datasets==5.0.1
+safetensors==0.8.0
+accelerate==1.14.0
+pydantic_core==2.46.4
+ninja==1.13.0
+autocommand==2.2.2
+backports.tarfile==1.2.0
+importlib_metadata==8.7.1
+jaraco.text==4.0.0
+jaraco.context==6.1.0
+jaraco.functools==4.4.0
+more-itertools==10.8.0
+packaging==26.0
+platformdirs==4.4.0
+tomli==2.4.0
+wheel==0.46.3
+zipp==3.23.0

wandb/run-20260809_055154-xr7l4yvo/files/wandb-metadata.json

@@ -0,0 +1,96 @@
+{
+  "os":  "Linux-5.15.0-126-generic-x86_64-with-glibc2.35",
+  "python":  "CPython 3.11.15",
+  "startedAt":  "2026-08-09T05:51:54.554743Z",
+  "args":  [
+    "--config",
+    "/mnt/data/zainulabideen/zain-exp/notebooks/Activation/configs/baseline1.yaml",
+    "--variants",
+    "mlp-tanh-9L",
+    "--push"
+  ],
+  "program":  "/mnt/data/zainulabideen/zain-exp/notebooks/Activation/sweep.py",
+  "codePath":  "sweep.py",
+  "codePathLocal":  "sweep.py",
+  "git":  {
+    "remote":  "https://github.com/deepnevro/Activation.git",
+    "commit":  "34b8d2e8f9a0c5751333310e69fa0c1056381deb"
+  },
+  "email":  "deepnevro@gmail.com",
+  "root":  "/mnt/data/zainulabideen/zain-exp/notebooks/Activation",
+  "host":  "deeplens-k3s-node1",
+  "executable":  "/mnt/data/zainulabideen/zain-exp/notebooks/my_env/bin/python",
+  "cpu_count":  112,
+  "cpu_count_logical":  224,
+  "gpu":  "NVIDIA H100 80GB HBM3",
+  "gpu_count":  8,
+  "disk":  {
+    "/":  {
+      "total":  "1560765693952",
+      "used":  "708272943104"
+    }
+  },
+  "memory":  {
+    "total":  "2164089937920"
+  },
+  "gpu_nvidia":  [
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-39c684a5-fde6-83d7-1663-0859795881ae"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-68012e5a-38b6-b643-0ca6-62fb66720bf3"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-132944c4-b689-2b5f-89a4-d730401677ab"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-2df386cc-6d26-d0e2-7a2d-a057b0d95864"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bfa16575-1d94-1aa2-4537-2c93433f42ef"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-bc6c3e3c-9b90-09ca-c034-774961847c54"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-00a441e1-7c95-e7d6-4c35-43d6b291aea9"
+    },
+    {
+      "name":  "NVIDIA H100 80GB HBM3",
+      "memoryTotal":  "85520809984",
+      "cudaCores":  16896,
+      "architecture":  "Hopper",
+      "uuid":  "GPU-1c4d29a2-4647-6fce-d8fc-0c5ecfbbd6ea"
+    }
+  ],
+  "cudaVersion":  "12.4",
+  "writerId":  "78a21vuu11ek41ydatzvn60d6p6iz5jh"
+}

wandb/run-20260809_055154-xr7l4yvo/files/wandb-summary.json

@@ -0,0 +1 @@
+{"_wandb":{"runtime":9},"_runtime":9}

wandb/run-20260809_055154-xr7l4yvo/logs/debug-core.log

@@ -0,0 +1,20 @@
+{"time":"2026-08-09T05:51:54.304723821Z","level":"INFO","msg":"main: starting server","port-filename":"/tmp/tmphnmz3lmh/port-4003770.txt","pid":4003770,"detached":false,"idle-timeout":600000000000,"log-level":0,"disable-analytics":false,"shutdown-on-parent-exit":false,"enable-dcgm-profiling":false}
+{"time":"2026-08-09T05:51:54.305175155Z","level":"INFO","msg":"server: will exit if parent process dies","ppid":4003770}
+{"time":"2026-08-09T05:51:54.305174806Z","level":"INFO","msg":"server: accepting connections","addr":{"Name":"/tmp/wandb-4003770-4005207-1456478744/socket","Net":"unix"}}
+{"time":"2026-08-09T05:51:54.482042592Z","level":"INFO","msg":"connection: ManageConnectionData: new connection created","id":"1(@)"}
+{"time":"2026-08-09T05:51:54.557130119Z","level":"INFO","msg":"handleInformInit: received","streamId":"xr7l4yvo","id":"1(@)"}
+{"time":"2026-08-09T05:51:54.818993787Z","level":"INFO","msg":"handleInformInit: stream started","streamId":"xr7l4yvo","id":"1(@)"}
+{"time":"2026-08-09T05:52:00.223396575Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"8vvtss3n0twh"}
+{"time":"2026-08-09T05:52:04.735899112Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"8vvtss3n0twh"}
+{"time":"2026-08-09T05:52:05.165131163Z","level":"INFO","msg":"connection: cancelling request","id":"1(@)","requestId":"8vvtss3n0twh"}
+{"time":"2026-08-09T05:52:05.166076718Z","level":"INFO","msg":"handleInformFinish: finish message received","streamId":"xr7l4yvo","id":"1(@)"}
+{"time":"2026-08-09T05:52:05.166573614Z","level":"INFO","msg":"handleInformFinish: stream closed","streamId":"xr7l4yvo","id":"1(@)"}
+{"time":"2026-08-09T05:52:05.168275247Z","level":"INFO","msg":"handleInformTeardown: server teardown initiated","id":"1(@)"}
+{"time":"2026-08-09T05:52:05.168291473Z","level":"INFO","msg":"handleInformTeardown: server shutdown complete","id":"1(@)"}
+{"time":"2026-08-09T05:52:05.168299429Z","level":"INFO","msg":"server: is shutting down"}
+{"time":"2026-08-09T05:52:05.168310652Z","level":"INFO","msg":"processOutgoingData: finished","id":"1(@)"}
+{"time":"2026-08-09T05:52:05.168297313Z","level":"INFO","msg":"connection: closing","id":"1(@)"}
+{"time":"2026-08-09T05:52:05.168365845Z","level":"INFO","msg":"connection: closed successfully","id":"1(@)"}
+{"time":"2026-08-09T05:52:05.16837009Z","level":"INFO","msg":"connection: ManageConnectionData: connection closed","id":"1(@)"}
+{"time":"2026-08-09T05:52:05.168457232Z","level":"INFO","msg":"server: listener closed","addr":{"Name":"/tmp/wandb-4003770-4005207-1456478744/socket","Net":"unix"}}
+{"time":"2026-08-09T05:52:05.168490421Z","level":"INFO","msg":"server: all connections closed"}

wandb/run-20260809_055154-xr7l4yvo/logs/debug-internal.log

@@ -0,0 +1,17 @@
+{"time":"2026-08-09T05:51:54.557323835Z","level":"INFO","msg":"wandb-core"}
+{"time":"2026-08-09T05:51:54.557613003Z","level":"INFO","msg":"stream: starting","core version":"0.28.1"}
+{"time":"2026-08-09T05:51:54.818810378Z","level":"INFO","msg":"stream: created new stream","id":"xr7l4yvo"}
+{"time":"2026-08-09T05:51:54.818911698Z","level":"INFO","msg":"handler: started"}
+{"time":"2026-08-09T05:51:54.818986607Z","level":"INFO","msg":"stream: started"}
+{"time":"2026-08-09T05:51:54.81900645Z","level":"INFO","msg":"writer: started","stream_id":"xr7l4yvo"}
+{"time":"2026-08-09T05:51:54.819025589Z","level":"INFO","msg":"sender: started"}
+{"time":"2026-08-09T05:52:00.274471268Z","level":"INFO","msg":"filestream: sending request","total_files":0,"uploaded_len":1}
+{"time":"2026-08-09T05:52:00.383855621Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:52:05.064200274Z","level":"INFO","msg":"fileTransfer: Close: file transfer manager closed"}
+{"time":"2026-08-09T05:52:05.064412769Z","level":"INFO","msg":"filestream: sending request","total_files":1,"uploaded_len":3,"complete":true,"exit_code":0}
+{"time":"2026-08-09T05:52:05.162443319Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-09T05:52:05.163842042Z","level":"INFO","msg":"handler: operation stats","stats":{}}
+{"time":"2026-08-09T05:52:05.166110839Z","level":"INFO","msg":"stream: finishing up"}
+{"time":"2026-08-09T05:52:05.16614337Z","level":"INFO","msg":"handler: closed"}
+{"time":"2026-08-09T05:52:05.166217191Z","level":"INFO","msg":"sender: closed"}
+{"time":"2026-08-09T05:52:05.166221159Z","level":"INFO","msg":"stream: all finished"}