Auto upload zain 2026-08-18T15:44:15.901384 (part 4)

zain/Activation/wandb/run-20260818_154123-mv5l58cq/files/output.log

@@ -121,5 +121,32 @@ Writing model shards: 100%|██████████| 1/1 [00:00<00:00, 162
   - 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, 161.94it/s]
- 45%|████▍     | 1120/2500 [02:19<03:10,  7.25it/s], ?it/s]
+ 48%|████▊     | 1200/2500 [02:31<01:07, 19.29i[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': '3.577', 'grad_norm': '1.203', 'learning_rate': '0.0003', 'epoch': '0.03775', 'train/total_time_seconds': '24.25', 'train/time_per_step_avg': '0.02068', 'train/epoch_time_elapsed': '139.9', 'train/estimated_remaining_minutes': '0.4979'}
+{'loss': '3.551', 'grad_norm': '1.305', 'learning_rate': '0.0003', 'epoch': '0.03842', 'train/total_time_seconds': '24.65', 'train/time_per_step_avg': '0.02051', 'train/epoch_time_elapsed': '140.8', 'train/estimated_remaining_minutes': '0.4901'}
+{'loss': '3.549', 'grad_norm': '1.312', 'learning_rate': '0.0003', 'epoch': '0.0391', 'train/total_time_seconds': '25.11', 'train/time_per_step_avg': '0.02099', 'train/epoch_time_elapsed': '141.8', 'train/estimated_remaining_minutes': '0.4835'}
+{'loss': '3.565', 'grad_norm': '1.273', 'learning_rate': '0.0003', 'epoch': '0.03977', 'train/total_time_seconds': '25.63', 'train/time_per_step_avg': '0.02198', 'train/epoch_time_elapsed': '142.9', 'train/estimated_remaining_minutes': '0.4778'}
+{'loss': '3.533', 'grad_norm': '1.336', 'learning_rate': '0.0003', 'epoch': '0.04044', 'train/total_time_seconds': '26.12', 'train/time_per_step_avg': '0.02269', 'train/epoch_time_elapsed': '143.9', 'train/estimated_remaining_minutes': '0.4716'}
+  - 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': '3.536', 'eval_runtime': '7.835', 'eval_samples_per_second': '1216', 'eval_steps_per_second': '1.276', 'epoch': '0.04044', 'train/total_time_seconds': '26.12', 'train/time_per_step_avg': '0.02269', 'train/epoch_time_elapsed': '151.7', 'train/estimated_remaining_minutes': '0.4716'}
+  - 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, 162.48it/s]
+ 52%|█████▏    | 1300/2500 [02:44<00:53, 22.32i[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': '3.504', 'grad_norm': '1.312', 'learning_rate': '0.0003', 'epoch': '0.04112', 'train/total_time_seconds': '26.53', 'train/time_per_step_avg': '0.02284', 'train/epoch_time_elapsed': '152.6', 'train/estimated_remaining_minutes': '0.4639'}
+{'loss': '3.522', 'grad_norm': '1.359', 'learning_rate': '0.0003', 'epoch': '0.04179', 'train/total_time_seconds': '26.94', 'train/time_per_step_avg': '0.0229', 'train/epoch_time_elapsed': '153.5', 'train/estimated_remaining_minutes': '0.4562'}
+{'loss': '3.532', 'grad_norm': '1.383', 'learning_rate': '0.0003', 'epoch': '0.04247', 'train/total_time_seconds': '27.35', 'train/time_per_step_avg': '0.02235', 'train/epoch_time_elapsed': '154.4', 'train/estimated_remaining_minutes': '0.4486'}
+{'loss': '3.509', 'grad_norm': '1.25', 'learning_rate': '0.0003', 'epoch': '0.04314', 'train/total_time_seconds': '27.76', 'train/time_per_step_avg': '0.02132', 'train/epoch_time_elapsed': '155.3', 'train/estimated_remaining_minutes': '0.441'}
+{'loss': '3.5', 'grad_norm': '1.273', 'learning_rate': '0.0003', 'epoch': '0.04382', 'train/total_time_seconds': '28.17', 'train/time_per_step_avg': '0.0205', 'train/epoch_time_elapsed': '156.2', 'train/estimated_remaining_minutes': '0.4334'}
+  - 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': '3.494', 'eval_runtime': '7.953', 'eval_samples_per_second': '1198', 'eval_steps_per_second': '1.257', 'epoch': '0.04382', 'train/total_time_seconds': '28.17', 'train/time_per_step_avg': '0.0205', 'train/epoch_time_elapsed': '164.2', 'train/estimated_remaining_minutes': '0.4334'}
+  - 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, 163.97it/s]
+ 56%|█████▌    | 1400/2500 [02:48<00:49, 22.33it/s], ?it/s]
+{'loss': '3.467', 'grad_norm': '1.305', 'learning_rate': '0.0003', 'epoch': '0.04449', 'train/total_time_seconds': '28.58', 'train/time_per_step_avg': '0.02053', 'train/epoch_time_elapsed': '165.1', 'train/estimated_remaining_minutes': '0.4259'}
+{'loss': '3.492', 'grad_norm': '1.344', 'learning_rate': '0.0003', 'epoch': '0.04516', 'train/total_time_seconds': '29', 'train/time_per_step_avg': '0.02059', 'train/epoch_time_elapsed': '166', 'train/estimated_remaining_minutes': '0.4184'}
+{'loss': '3.471', 'grad_norm': '1.133', 'learning_rate': '0.0003', 'epoch': '0.04584', 'train/total_time_seconds': '29.41', 'train/time_per_step_avg': '0.02059', 'train/epoch_time_elapsed': '166.9', 'train/estimated_remaining_minutes': '0.4108'}
+{'loss': '3.467', 'grad_norm': '1.188', 'learning_rate': '0.0003', 'epoch': '0.04651', 'train/total_time_seconds': '29.82', 'train/time_per_step_avg': '0.02062', 'train/epoch_time_elapsed': '167.8', 'train/estimated_remaining_minutes': '0.4034'}
+{'loss': '3.445', 'grad_norm': '1.445', 'learning_rate': '0.0003', 'epoch': '0.04719', 'train/total_time_seconds': '30.23', 'train/time_per_step_avg': '0.02061', 'train/epoch_time_elapsed': '168.7', 'train/estimated_remaining_minutes': '0.3959'}
+ 60%|██████    | 6/10 [00:04<00:03,  1.23it/s]

zain/Activation/wandb/run-20260818_154123-mv5l58cq/logs/debug-internal.log

@@ -25,3 +25,7 @@
 {"time":"2026-08-18T15:43:24.158079844Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
 {"time":"2026-08-18T15:43:38.83860739Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":59,"history_lines":6,"events_offset":15,"events_lines":2,"console_offset":94,"console_lines":25}
 {"time":"2026-08-18T15:43:39.143362026Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-18T15:43:53.838665388Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":65,"history_lines":6,"events_offset":17,"events_lines":2,"console_offset":112,"console_lines":1}
+{"time":"2026-08-18T15:43:54.247445569Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}
+{"time":"2026-08-18T15:44:08.838763618Z","level":"INFO","msg":"filestream: sending request","total_files":4,"history_offset":71,"history_lines":7,"events_offset":19,"events_lines":2,"console_offset":118,"console_lines":28}
+{"time":"2026-08-18T15:44:09.153705104Z","level":"INFO","msg":"filestream: request sent","status":"200 OK"}

zain/Activation/wandb/run-20260818_154123-mv5l58cq/run-mv5l58cq.wandb

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a03b0a9cd461461bf6ecfc826aa7b12aec4b60ca0cf5eff644a38e4fc4da79b
-size 196608
+oid sha256:71543266672b8902d8f53bf98d8573083b255ec5d61a4c7ad3fd42f8697363ec
+size 262144