Update tokenizer, configuration and training reports
Browse files
reports/run/metrics.jsonl
CHANGED
|
@@ -19,3 +19,5 @@
|
|
| 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}
|
|
|
|
|
|
|
|
|
| 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}
|
| 22 |
+
{"step": 440, "tokens": 230686720, "loss": 4.623250198364258, "tokens_per_second": 1011900.1103533852, "lr": 0.00033, "gradient_norm": 1.3369032144546509, "benchmark": false}
|
| 23 |
+
{"step": 460, "tokens": 241172480, "loss": 4.510102844238281, "tokens_per_second": 988569.6225406213, "lr": 0.00034500000000000004, "gradient_norm": 1.1452369689941406, "benchmark": false}
|