{ "effective_tokens_per_sec": 731.8317898201691, "epoch": 9.0, "total_flos": 3.686068977139712e+17, "train_loss": 0.22300153877586126, "train_runtime": 2410.3131, "train_samples_per_second": 3.734, "train_steps_per_second": 0.119 }