{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.016842105263157894, "eval_steps": 8, "global_step": 30, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0005614035087719298, "eval_loss": NaN, "eval_runtime": 42.8215, "eval_samples_per_second": 17.515, "eval_steps_per_second": 8.757, "step": 1 }, { "epoch": 0.0016842105263157896, "grad_norm": NaN, "learning_rate": 6e-05, "loss": 0.0, "step": 3 }, { "epoch": 0.003368421052631579, "grad_norm": NaN, "learning_rate": 0.00012, "loss": 0.0, "step": 6 }, { "epoch": 0.004491228070175438, "eval_loss": NaN, "eval_runtime": 42.3171, "eval_samples_per_second": 17.723, "eval_steps_per_second": 8.862, "step": 8 }, { "epoch": 0.0050526315789473685, "grad_norm": NaN, "learning_rate": 0.00018, "loss": 0.0, "step": 9 }, { "epoch": 0.006736842105263158, "grad_norm": NaN, "learning_rate": 0.00019510565162951537, "loss": 0.0, "step": 12 }, { "epoch": 0.008421052631578947, "grad_norm": NaN, "learning_rate": 0.00017071067811865476, "loss": 0.0, "step": 15 }, { "epoch": 0.008982456140350877, "eval_loss": NaN, "eval_runtime": 42.4456, "eval_samples_per_second": 17.67, "eval_steps_per_second": 8.835, "step": 16 }, { "epoch": 0.010105263157894737, "grad_norm": NaN, "learning_rate": 0.00013090169943749476, "loss": 0.0, "step": 18 }, { "epoch": 0.011789473684210527, "grad_norm": NaN, "learning_rate": 8.435655349597689e-05, "loss": 0.0, "step": 21 }, { "epoch": 0.013473684210526317, "grad_norm": NaN, "learning_rate": 4.12214747707527e-05, "loss": 0.0, "step": 24 }, { "epoch": 0.013473684210526317, "eval_loss": NaN, "eval_runtime": 42.4012, "eval_samples_per_second": 17.688, "eval_steps_per_second": 8.844, "step": 24 }, { "epoch": 0.015157894736842105, "grad_norm": NaN, "learning_rate": 1.0899347581163221e-05, "loss": 0.0, "step": 27 }, { "epoch": 0.016842105263157894, "grad_norm": NaN, "learning_rate": 0.0, "loss": 0.0, "step": 30 } ], "logging_steps": 3, "max_steps": 30, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 10, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1.112830925340672e+16, "train_batch_size": 2, "trial_name": null, "trial_params": null }