Update tokenizer, configuration and training reports
Browse files
reports/run/metrics.jsonl
CHANGED
|
@@ -7,3 +7,5 @@
|
|
| 7 |
{"step": 140, "tokens": 73400320, "loss": 7.291810607910156, "tokens_per_second": 1015522.4437225041, "lr": 0.00010500000000000002, "gradient_norm": 0.8104191422462463, "benchmark": false}
|
| 8 |
{"step": 160, "tokens": 83886080, "loss": 7.000241088867187, "tokens_per_second": 1024867.7143696964, "lr": 0.00012, "gradient_norm": 0.8440332412719727, "benchmark": false}
|
| 9 |
{"step": 180, "tokens": 94371840, "loss": 6.713601684570312, "tokens_per_second": 1024219.9743449177, "lr": 0.000135, "gradient_norm": 1.245676040649414, "benchmark": false}
|
|
|
|
|
|
|
|
|
| 7 |
{"step": 140, "tokens": 73400320, "loss": 7.291810607910156, "tokens_per_second": 1015522.4437225041, "lr": 0.00010500000000000002, "gradient_norm": 0.8104191422462463, "benchmark": false}
|
| 8 |
{"step": 160, "tokens": 83886080, "loss": 7.000241088867187, "tokens_per_second": 1024867.7143696964, "lr": 0.00012, "gradient_norm": 0.8440332412719727, "benchmark": false}
|
| 9 |
{"step": 180, "tokens": 94371840, "loss": 6.713601684570312, "tokens_per_second": 1024219.9743449177, "lr": 0.000135, "gradient_norm": 1.245676040649414, "benchmark": false}
|
| 10 |
+
{"step": 200, "tokens": 104857600, "loss": 6.462795257568359, "tokens_per_second": 1021215.5608316573, "lr": 0.00015000000000000001, "gradient_norm": 2.098832607269287, "benchmark": false}
|
| 11 |
+
{"step": 220, "tokens": 115343360, "loss": 6.25281867980957, "tokens_per_second": 1011047.7764669958, "lr": 0.000165, "gradient_norm": 2.103592872619629, "benchmark": false}
|