step=10598 loss=4.8752
Browse files
checkpoints/ckpt_20260304_134602/model.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7db8d7cd5bf95d01590909ad0603bff514a16cdfce232c83e119aad7132cf93d
|
| 3 |
+
size 186874811
|
checkpoints/ckpt_20260304_134602/training_state.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"global_step": 10597,
|
| 3 |
+
"best_loss": 4.875235967338085,
|
| 4 |
+
"last_save_ts": "2026-03-04T13:26:31.765843",
|
| 5 |
+
"tokens_seen": 173621248,
|
| 6 |
+
"completed_datasets": [
|
| 7 |
+
"data/train/000_wikitext2.parquet",
|
| 8 |
+
"data/train/001_python_code.parquet"
|
| 9 |
+
],
|
| 10 |
+
"current_dataset": "data/train/002_python_code.parquet",
|
| 11 |
+
"current_dataset_tokens": 55902208
|
| 12 |
+
}
|