{ "effective_tokens_per_sec": 7613.769425389832, "epoch": 4.0, "total_flos": 3.519329208629199e+18, "train_loss": 0.7788769946752703, "train_runtime": 8944.5824, "train_samples_per_second": 6.735, "train_steps_per_second": 0.224 }