{ "achieved_tflops_per_gpu": 0.002469331217094423, "achieved_tflops_per_gpu_theoretical": 1243.3609745561027, "epoch": 7.0, "loss_nan_ranks": 0, "loss_rank_avg": 0.002985658124089241, "mfu_percent": 0.00017451104007734438, "mfu_percent_theoretical": 87.87003353753376, "total_flos": 331629159579648.0, "train_loss": 0.006270654577846554, "train_runtime": 8393.6988, "train_samples_per_second": 5.86, "train_steps_per_second": 0.367 }