Update tokenizer, configuration and training reports
Browse files
reports/run/metrics.jsonl
CHANGED
|
@@ -31,3 +31,5 @@
|
|
| 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}
|
|
|
|
|
|
|
|
|
| 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}
|
| 34 |
+
{"step": 680, "tokens": 356515840, "loss": 3.830519104003906, "tokens_per_second": 986269.7550769978, "lr": 0.00051, "gradient_norm": 0.5967057347297668, "benchmark": false}
|
| 35 |
+
{"step": 700, "tokens": 367001600, "loss": 3.826114273071289, "tokens_per_second": 992132.7181165003, "lr": 0.000525, "gradient_norm": 0.5667921900749207, "benchmark": false}
|