Update tokenizer, configuration and training reports
Browse files
reports/run/metrics.jsonl
CHANGED
|
@@ -81,3 +81,5 @@
|
|
| 81 |
{"step": 1620, "tokens": 849346560, "loss": 3.268468475341797, "tokens_per_second": 846468.1827140138, "lr": 0.0012150000000000002, "gradient_norm": 0.25949716567993164, "benchmark": false}
|
| 82 |
{"step": 1640, "tokens": 859832320, "loss": 3.2780303955078125, "tokens_per_second": 988943.4693897894, "lr": 0.00123, "gradient_norm": 0.2221362441778183, "benchmark": false}
|
| 83 |
{"step": 1660, "tokens": 870318080, "loss": 3.2745872497558595, "tokens_per_second": 982150.9602999857, "lr": 0.001245, "gradient_norm": 0.2505227327346802, "benchmark": false}
|
|
|
|
|
|
|
|
|
| 81 |
{"step": 1620, "tokens": 849346560, "loss": 3.268468475341797, "tokens_per_second": 846468.1827140138, "lr": 0.0012150000000000002, "gradient_norm": 0.25949716567993164, "benchmark": false}
|
| 82 |
{"step": 1640, "tokens": 859832320, "loss": 3.2780303955078125, "tokens_per_second": 988943.4693897894, "lr": 0.00123, "gradient_norm": 0.2221362441778183, "benchmark": false}
|
| 83 |
{"step": 1660, "tokens": 870318080, "loss": 3.2745872497558595, "tokens_per_second": 982150.9602999857, "lr": 0.001245, "gradient_norm": 0.2505227327346802, "benchmark": false}
|
| 84 |
+
{"step": 1680, "tokens": 880803840, "loss": 3.2768791198730467, "tokens_per_second": 986788.4008772722, "lr": 0.00126, "gradient_norm": 0.23133485019207, "benchmark": false}
|
| 85 |
+
{"step": 1700, "tokens": 891289600, "loss": 3.2841224670410156, "tokens_per_second": 935487.5310804343, "lr": 0.001275, "gradient_norm": 0.26458561420440674, "benchmark": false}
|