| 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 | |
| 0,2.698207,-1.000000,0.000600,4.494938,1254.051147,6532.428934,7.251200,960.745605,3.987392,231.429344,15.948480,14.672000,36700.875000 | |
| 1,7.925948,-1.000000,0.000600,19.026079,342.769135,23899.468111,0.280864,102.697758,0.481152,209.203171,15.925248,11.102272,50076.875000 | |