587 B
587 B
| 1 | step | epoch | loss | learning_rate | grad_norm | mean_token_accuracy |
|---|---|---|---|---|---|---|
| 2 | 10 | 0.16 | 1.8803634643554688 | 9.678584095202468e-05 | 1.182433009147644 | 0.582773034274578 |
| 3 | 20 | 0.32 | 1.1098207473754882 | 8.202965893490878e-05 | 0.5895639657974243 | 0.7326480627059937 |
| 4 | 30 | 0.48 | 0.9109317779541015 | 5.896403794053679e-05 | 0.4854218065738678 | 0.772037148475647 |
| 5 | 40 | 0.64 | 0.8902366638183594 | 3.357288090445827e-05 | 0.4858759939670563 | 0.775420868396759 |
| 6 | 50 | 0.8 | 0.8645750999450683 | 1.2443403456474017e-05 | 0.4035021960735321 | 0.7814616322517395 |
| 7 | 60 | 0.96 | 0.9060385704040528 | 1.0572157452321097e-06 | 0.37886306643486023 | 0.7762055635452271 |