Update tokenizer, configuration and training reports
Browse files
reports/run/metrics.jsonl
CHANGED
|
@@ -36,3 +36,5 @@
|
|
| 36 |
{"step": 720, "tokens": 377487360, "loss": 3.769327163696289, "tokens_per_second": 944236.843989973, "lr": 0.00054, "gradient_norm": 0.5464304685592651, "benchmark": false}
|
| 37 |
{"step": 740, "tokens": 387973120, "loss": 3.776169204711914, "tokens_per_second": 987457.2237525532, "lr": 0.000555, "gradient_norm": 0.4834839999675751, "benchmark": false}
|
| 38 |
{"step": 760, "tokens": 398458880, "loss": 3.6883705139160154, "tokens_per_second": 984211.8141510511, "lr": 0.00057, "gradient_norm": 0.5399366617202759, "benchmark": false}
|
|
|
|
|
|
|
|
|
| 36 |
{"step": 720, "tokens": 377487360, "loss": 3.769327163696289, "tokens_per_second": 944236.843989973, "lr": 0.00054, "gradient_norm": 0.5464304685592651, "benchmark": false}
|
| 37 |
{"step": 740, "tokens": 387973120, "loss": 3.776169204711914, "tokens_per_second": 987457.2237525532, "lr": 0.000555, "gradient_norm": 0.4834839999675751, "benchmark": false}
|
| 38 |
{"step": 760, "tokens": 398458880, "loss": 3.6883705139160154, "tokens_per_second": 984211.8141510511, "lr": 0.00057, "gradient_norm": 0.5399366617202759, "benchmark": false}
|
| 39 |
+
{"step": 780, "tokens": 408944640, "loss": 3.716542434692383, "tokens_per_second": 982512.276017266, "lr": 0.000585, "gradient_norm": 0.46255218982696533, "benchmark": false}
|
| 40 |
+
{"step": 800, "tokens": 419430400, "loss": 3.6764266967773436, "tokens_per_second": 987572.7835053173, "lr": 0.0006000000000000001, "gradient_norm": 0.47228479385375977, "benchmark": false}
|