5 lines · 118 BytesRaw
| 1 | step 300 loss 2.41 grad_norm 1.2 lr 3.0e-4 |
| 2 | step 380 loss 2.39 grad_norm 48.7 |
| 3 | step 400 loss nan grad_norm inf |
| 4 | DIVERGED |
| 1 | step 300 loss 2.41 grad_norm 1.2 lr 3.0e-4 |
| 2 | step 380 loss 2.39 grad_norm 48.7 |
| 3 | step 400 loss nan grad_norm inf |
| 4 | DIVERGED |