| 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 |
| 0,11.091398,-1.000000,0.000600,19.278999,6757.555176,155170.912072,12.984256,2399.408794,16.897408,4221.621109,0.003872,64.055298,61116.937500 |
| 1,9.663713,-1.000000,0.000600,31.900805,6334.794922,165526.431863,10.534208,2036.898731,13.676480,4201.504593,0.003488,50.892769,61244.937500 |
| 2,9.032335,-1.000000,0.000330,107.822159,6410.564941,163569.983236,10.558080,2038.739952,13.808160,4275.160324,0.003488,50.965855,61244.937500 |
|
|
