Update tokenizer, configuration and training reports
Browse files
reports/run/metrics.jsonl
CHANGED
|
@@ -17,3 +17,5 @@
|
|
| 17 |
{"step": 340, "tokens": 178257920, "loss": 5.24281120300293, "tokens_per_second": 1003209.9996139308, "lr": 0.000255, "gradient_norm": 1.9500349760055542, "benchmark": false}
|
| 18 |
{"step": 360, "tokens": 188743680, "loss": 5.100251770019531, "tokens_per_second": 1004851.9846189881, "lr": 0.00027, "gradient_norm": 2.0687453746795654, "benchmark": false}
|
| 19 |
{"step": 380, "tokens": 199229440, "loss": 4.99603271484375, "tokens_per_second": 1017205.9020828365, "lr": 0.000285, "gradient_norm": 1.5136432647705078, "benchmark": false}
|
|
|
|
|
|
|
|
|
| 17 |
{"step": 340, "tokens": 178257920, "loss": 5.24281120300293, "tokens_per_second": 1003209.9996139308, "lr": 0.000255, "gradient_norm": 1.9500349760055542, "benchmark": false}
|
| 18 |
{"step": 360, "tokens": 188743680, "loss": 5.100251770019531, "tokens_per_second": 1004851.9846189881, "lr": 0.00027, "gradient_norm": 2.0687453746795654, "benchmark": false}
|
| 19 |
{"step": 380, "tokens": 199229440, "loss": 4.99603271484375, "tokens_per_second": 1017205.9020828365, "lr": 0.000285, "gradient_norm": 1.5136432647705078, "benchmark": false}
|
| 20 |
+
{"step": 400, "tokens": 209715200, "loss": 4.855997848510742, "tokens_per_second": 1014320.6709609136, "lr": 0.00030000000000000003, "gradient_norm": 1.2021150588989258, "benchmark": false}
|
| 21 |
+
{"step": 420, "tokens": 220200960, "loss": 4.741484451293945, "tokens_per_second": 1007498.0805184614, "lr": 0.000315, "gradient_norm": 1.2149815559387207, "benchmark": false}
|