Update tokenizer, configuration and training reports
Browse files
reports/run/metrics.jsonl
CHANGED
|
@@ -52,3 +52,5 @@
|
|
| 52 |
{"step": 1040, "tokens": 545259520, "loss": 3.485460662841797, "tokens_per_second": 996745.0972588192, "lr": 0.0007800000000000001, "gradient_norm": 0.3457220494747162, "benchmark": false}
|
| 53 |
{"step": 1060, "tokens": 555745280, "loss": 3.48607177734375, "tokens_per_second": 996269.3046775112, "lr": 0.000795, "gradient_norm": 0.37579262256622314, "benchmark": false}
|
| 54 |
{"step": 1080, "tokens": 566231040, "loss": 3.5009490966796877, "tokens_per_second": 993456.0662475126, "lr": 0.0008100000000000001, "gradient_norm": 0.3382250666618347, "benchmark": false}
|
|
|
|
|
|
|
|
|
| 52 |
{"step": 1040, "tokens": 545259520, "loss": 3.485460662841797, "tokens_per_second": 996745.0972588192, "lr": 0.0007800000000000001, "gradient_norm": 0.3457220494747162, "benchmark": false}
|
| 53 |
{"step": 1060, "tokens": 555745280, "loss": 3.48607177734375, "tokens_per_second": 996269.3046775112, "lr": 0.000795, "gradient_norm": 0.37579262256622314, "benchmark": false}
|
| 54 |
{"step": 1080, "tokens": 566231040, "loss": 3.5009490966796877, "tokens_per_second": 993456.0662475126, "lr": 0.0008100000000000001, "gradient_norm": 0.3382250666618347, "benchmark": false}
|
| 55 |
+
{"step": 1100, "tokens": 576716800, "loss": 3.4624401092529298, "tokens_per_second": 990775.8492234668, "lr": 0.0008250000000000001, "gradient_norm": 0.3198245167732239, "benchmark": false}
|
| 56 |
+
{"step": 1120, "tokens": 587202560, "loss": 3.4398429870605467, "tokens_per_second": 850654.6348279766, "lr": 0.0008400000000000001, "gradient_norm": 0.3153975307941437, "benchmark": false}
|