{"loss": 2.66, "learning_rate": 2.9999999999999997e-05, "epoch": 10.0, "step": 10}
{"train_runtime": 10.4951, "train_samples_per_second": 0.953, "train_steps_per_second": 0.953, "total_flos": 203106570731520.0, "train_loss": 2.660002326965332, "epoch": 10.0, "step": 10}