Update tokenizer, configuration and training reports
Browse files
reports/run/metrics.jsonl
CHANGED
|
@@ -42,3 +42,5 @@
|
|
| 42 |
{"step": 840, "tokens": 440401920, "loss": 3.6467517852783202, "tokens_per_second": 990496.904193152, "lr": 0.00063, "gradient_norm": 0.4072583019733429, "benchmark": false}
|
| 43 |
{"step": 860, "tokens": 450887680, "loss": 3.6281856536865233, "tokens_per_second": 983591.4808985759, "lr": 0.000645, "gradient_norm": 0.3906756639480591, "benchmark": false}
|
| 44 |
{"step": 880, "tokens": 461373440, "loss": 3.5856857299804688, "tokens_per_second": 984787.8953277065, "lr": 0.00066, "gradient_norm": 0.3721332550048828, "benchmark": false}
|
|
|
|
|
|
|
|
|
| 42 |
{"step": 840, "tokens": 440401920, "loss": 3.6467517852783202, "tokens_per_second": 990496.904193152, "lr": 0.00063, "gradient_norm": 0.4072583019733429, "benchmark": false}
|
| 43 |
{"step": 860, "tokens": 450887680, "loss": 3.6281856536865233, "tokens_per_second": 983591.4808985759, "lr": 0.000645, "gradient_norm": 0.3906756639480591, "benchmark": false}
|
| 44 |
{"step": 880, "tokens": 461373440, "loss": 3.5856857299804688, "tokens_per_second": 984787.8953277065, "lr": 0.00066, "gradient_norm": 0.3721332550048828, "benchmark": false}
|
| 45 |
+
{"step": 900, "tokens": 471859200, "loss": 3.576865005493164, "tokens_per_second": 986523.1621472359, "lr": 0.000675, "gradient_norm": 0.43096911907196045, "benchmark": false}
|
| 46 |
+
{"step": 920, "tokens": 482344960, "loss": 3.551728057861328, "tokens_per_second": 985259.8203679301, "lr": 0.0006900000000000001, "gradient_norm": 0.3959476351737976, "benchmark": false}
|