{ "effective_tokens_per_sec": 7581.228864519554, "epoch": 6.0, "total_flos": 5.392281114922451e+18, "train_loss": 0.6896465788091076, "train_runtime": 6887.9733, "train_samples_per_second": 12.387, "train_steps_per_second": 0.413 }