{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.008458646616541353, "eval_steps": 8, "global_step": 30, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.00028195488721804513, "eval_loss": NaN, "eval_runtime": 107.5821, "eval_samples_per_second": 20.821, "eval_steps_per_second": 10.411, "step": 1 }, { "epoch": 0.0008458646616541353, "grad_norm": NaN, "learning_rate": 6e-05, "loss": 0.0, "step": 3 }, { "epoch": 0.0016917293233082707, "grad_norm": NaN, "learning_rate": 0.00012, "loss": 0.0, "step": 6 }, { "epoch": 0.002255639097744361, "eval_loss": NaN, "eval_runtime": 106.8864, "eval_samples_per_second": 20.957, "eval_steps_per_second": 10.478, "step": 8 }, { "epoch": 0.0025375939849624062, "grad_norm": NaN, "learning_rate": 0.00018, "loss": 0.0, "step": 9 }, { "epoch": 0.0033834586466165413, "grad_norm": NaN, "learning_rate": 0.00019510565162951537, "loss": 0.0, "step": 12 }, { "epoch": 0.0042293233082706765, "grad_norm": NaN, "learning_rate": 0.00017071067811865476, "loss": 0.0, "step": 15 }, { "epoch": 0.004511278195488722, "eval_loss": NaN, "eval_runtime": 106.9043, "eval_samples_per_second": 20.953, "eval_steps_per_second": 10.477, "step": 16 }, { "epoch": 0.0050751879699248124, "grad_norm": NaN, "learning_rate": 0.00013090169943749476, "loss": 0.0, "step": 18 }, { "epoch": 0.0059210526315789476, "grad_norm": NaN, "learning_rate": 8.435655349597689e-05, "loss": 0.0, "step": 21 }, { "epoch": 0.006766917293233083, "grad_norm": NaN, "learning_rate": 4.12214747707527e-05, "loss": 0.0, "step": 24 }, { "epoch": 0.006766917293233083, "eval_loss": NaN, "eval_runtime": 106.9395, "eval_samples_per_second": 20.946, "eval_steps_per_second": 10.473, "step": 24 }, { "epoch": 0.007612781954887218, "grad_norm": NaN, "learning_rate": 1.0899347581163221e-05, "loss": 0.0, "step": 27 }, { "epoch": 0.008458646616541353, "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": 3602647256924160.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }