Update tokenizer, configuration and training reports
Browse files
reports/run/metrics.jsonl
CHANGED
|
@@ -27,3 +27,5 @@
|
|
| 27 |
{"step": 540, "tokens": 283115520, "loss": 4.178365707397461, "tokens_per_second": 987055.5145676823, "lr": 0.00040500000000000003, "gradient_norm": 0.9119740724563599, "benchmark": false}
|
| 28 |
{"step": 560, "tokens": 293601280, "loss": 4.115958404541016, "tokens_per_second": 986620.6540439959, "lr": 0.00042000000000000007, "gradient_norm": 0.8756295442581177, "benchmark": false}
|
| 29 |
{"step": 580, "tokens": 304087040, "loss": 4.041319274902344, "tokens_per_second": 991743.1745662729, "lr": 0.000435, "gradient_norm": 0.7584083676338196, "benchmark": false}
|
|
|
|
|
|
|
|
|
| 27 |
{"step": 540, "tokens": 283115520, "loss": 4.178365707397461, "tokens_per_second": 987055.5145676823, "lr": 0.00040500000000000003, "gradient_norm": 0.9119740724563599, "benchmark": false}
|
| 28 |
{"step": 560, "tokens": 293601280, "loss": 4.115958404541016, "tokens_per_second": 986620.6540439959, "lr": 0.00042000000000000007, "gradient_norm": 0.8756295442581177, "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}
|