{ "achieved_tflops_per_gpu": 0.005772279796030858, "achieved_tflops_per_gpu_theoretical": 1158.6633604855235, "epoch": 7.0, "loss_nan_ranks": 0, "loss_rank_avg": 0.12985655665397644, "mfu_percent": 0.00040793496791737514, "mfu_percent_theoretical": 81.88433643007234, "total_flos": 1219521762033664.0, "train_loss": 0.10701951415567831, "train_runtime": 13204.5072, "train_samples_per_second": 5.464, "train_steps_per_second": 0.342, "valid_targets_mean": 5961.6, "valid_targets_min": 2002 }