Update tokenizer, configuration and training reports
Browse files
reports/run/metrics.jsonl
CHANGED
|
@@ -29,3 +29,5 @@
|
|
| 29 |
{"step": 580, "tokens": 304087040, "loss": 4.041319274902344, "tokens_per_second": 991743.1745662729, "lr": 0.000435, "gradient_norm": 0.7584083676338196, "benchmark": false}
|
| 30 |
{"step": 600, "tokens": 314572800, "loss": 4.01019515991211, "tokens_per_second": 948723.8898603223, "lr": 0.00045, "gradient_norm": 0.695301353931427, "benchmark": false}
|
| 31 |
{"step": 620, "tokens": 325058560, "loss": 3.9603034973144533, "tokens_per_second": 987068.105185979, "lr": 0.000465, "gradient_norm": 0.595501720905304, "benchmark": false}
|
|
|
|
|
|
|
|
|
| 29 |
{"step": 580, "tokens": 304087040, "loss": 4.041319274902344, "tokens_per_second": 991743.1745662729, "lr": 0.000435, "gradient_norm": 0.7584083676338196, "benchmark": false}
|
| 30 |
{"step": 600, "tokens": 314572800, "loss": 4.01019515991211, "tokens_per_second": 948723.8898603223, "lr": 0.00045, "gradient_norm": 0.695301353931427, "benchmark": false}
|
| 31 |
{"step": 620, "tokens": 325058560, "loss": 3.9603034973144533, "tokens_per_second": 987068.105185979, "lr": 0.000465, "gradient_norm": 0.595501720905304, "benchmark": false}
|
| 32 |
+
{"step": 640, "tokens": 335544320, "loss": 3.9002197265625, "tokens_per_second": 937157.4389623627, "lr": 0.00048, "gradient_norm": 0.6494060754776001, "benchmark": false}
|
| 33 |
+
{"step": 660, "tokens": 346030080, "loss": 3.87745361328125, "tokens_per_second": 987337.2396404969, "lr": 0.000495, "gradient_norm": 0.5577827095985413, "benchmark": false}
|