| 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 | |
| 0,11.160570,-1.000000,0.000007,53.779457,568.490356,28820.189849,3.251040,416.581604,3.715232,101.714302,9.922016,10.500352,21026.937500 | |