{ "effective_tokens_per_sec": 7119.60252078194, "epoch": 4.0, "total_flos": 3.6942184710227886e+18, "train_loss": 0.7407096255137103, "train_runtime": 10071.2083, "train_samples_per_second": 5.529, "train_steps_per_second": 0.092 }