| step,train_loss,val_loss,lr,grad_norm,dt_ms,tok_per_sec,timer_data,timer_fwd,timer_loss,timer_bwd,timer_clip,timer_optim,mem_gpu_mb |
| 0,2.749795,-1.000000,0.000006,0.369903,23021.306641,45548.066249,13.072288,13853.379288,21.374208,9039.333008,0.003712,62.146366,108268.937500 |
| 1,2.769012,-1.000000,0.000012,0.369288,18315.851562,57249.645010,8.078784,9006.841492,13.126144,9223.936829,0.003200,48.987518,109836.937500 |
| 2,2.739646,-1.000000,0.000018,0.368004,17922.433594,58506.340365,8.335008,9489.957993,12.847552,8347.856201,0.003552,49.063423,109836.937500 |
| 3,2.713669,-1.000000,0.000024,0.336734,12403.550781,84538.372801,7.035776,3995.475113,12.814368,8324.248291,0.003936,49.441536,109836.937500 |
|
|
