| 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,-1.000000,0.000006,0.369922,19128.255859,54818.170967,18.908096,9988.749802,21.508704,8997.138885,0.003328,62.464226,68220.937500 |
| 1,2.769008,-1.000000,0.000012,0.369318,15397.692383,68099.555046,12.860640,6346.339447,13.852000,8952.085632,0.003232,49.240578,69148.937500 |
| 2,2.739645,-1.000000,0.000018,0.368019,15472.694336,67769.450959,13.329216,6811.885391,13.934272,8560.524719,0.003264,49.327934,69180.937500 |
| 3,2.713675,-1.000000,0.000024,0.336673,12844.812500,81634.200577,12.084192,4144.275375,13.874944,8601.305389,0.003648,49.334496,69180.937500 |
|
|
