Auto-sync beta checkpoint from H100 pretraining runner
Browse files
continuous/beta/SHA256SUMS
CHANGED
|
@@ -1 +1 @@
|
|
| 1 |
-
|
|
|
|
| 1 |
+
77631d7770d7780790f03b4a76940914e8df2ee9ff2ba2e1987166fc5faf6eef model.safetensors
|
continuous/beta/config.json
CHANGED
|
@@ -5,7 +5,7 @@
|
|
| 5 |
"n_heads": 16,
|
| 6 |
"n_kv_heads": 4,
|
| 7 |
"d_ffn": 2816,
|
| 8 |
-
"max_seq_len":
|
| 9 |
"rope_theta": 1000000.0,
|
| 10 |
"rms_norm_eps": 1e-05,
|
| 11 |
"tie_word_embeddings": true,
|
|
|
|
| 5 |
"n_heads": 16,
|
| 6 |
"n_kv_heads": 4,
|
| 7 |
"d_ffn": 2816,
|
| 8 |
+
"max_seq_len": 512,
|
| 9 |
"rope_theta": 1000000.0,
|
| 10 |
"rms_norm_eps": 1e-05,
|
| 11 |
"tie_word_embeddings": true,
|
continuous/beta/manifest.json
CHANGED
|
@@ -1,26 +1,8 @@
|
|
| 1 |
{
|
| 2 |
"tier": "beta",
|
| 3 |
-
"
|
| 4 |
-
"
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"
|
| 8 |
-
"final_loss": 0.09809950739145279,
|
| 9 |
-
"tokens": 2847,
|
| 10 |
-
"checkpoint": "/Volumes/One Touch/Helios/checkpoints/beta_continuous/latest",
|
| 11 |
-
"category": "all",
|
| 12 |
-
"resumed_from_step": 1685,
|
| 13 |
-
"total_steps": 1717,
|
| 14 |
-
"baseline_initial_loss": 10.615724563598633,
|
| 15 |
-
"previous_loss": 0.10182087868452072,
|
| 16 |
-
"loss_improvement_pct": 99.08,
|
| 17 |
-
"previous_ppl": null,
|
| 18 |
-
"heldout_ppl": null,
|
| 19 |
-
"sha256_digest": "ed459609599e566ed9ed5b79b93e5259470b1bef4e718cb597728248ec8f90d9",
|
| 20 |
-
"health_status": "HEALTHY",
|
| 21 |
-
"datasets_synced": 177,
|
| 22 |
-
"updated_at": "2026-11-10 04:50:15 UTC",
|
| 23 |
-
"trained_topics_summary": "History 37% \u00b7 Coding 37% \u00b7 Math 12% \u00b7 Agents 10% \u00b7 Geography 2% \u00b7 Tagalog 1% \u00b7 Science 1% \u00b7 Law 1% \u00b7 General Web 1%",
|
| 24 |
-
"cursor_range": "Rec #21,900\u2192#24,900 (0% Dup \u00b7 Ep 1)",
|
| 25 |
-
"hardware_profile": "Thermal-Safe Shift (7793MB free \u00b7 bsz=1, seq=128, yield=16ms)"
|
| 26 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"tier": "beta",
|
| 3 |
+
"step": 1967,
|
| 4 |
+
"loss": 4.4208,
|
| 5 |
+
"sha256": "77631d7770d7780790f03b4a76940914e8df2ee9ff2ba2e1987166fc5faf6eef",
|
| 6 |
+
"timestamp": "2026-11-10T06:55:40.480905+00:00",
|
| 7 |
+
"hardware": "COLAB_GPU"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 8 |
}
|
continuous/beta/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1344499216
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:77631d7770d7780790f03b4a76940914e8df2ee9ff2ba2e1987166fc5faf6eef
|
| 3 |
size 1344499216
|