{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.1784121320249777, "eval_steps": 25, "global_step": 25, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.007136485280999108, "grad_norm": NaN, "learning_rate": 3.3333333333333335e-05, "loss": 0.0, "step": 1 }, { "epoch": 0.007136485280999108, "eval_loss": NaN, "eval_runtime": 25.9135, "eval_samples_per_second": 4.554, "eval_steps_per_second": 2.277, "step": 1 }, { "epoch": 0.014272970561998216, "grad_norm": NaN, "learning_rate": 6.666666666666667e-05, "loss": 0.0, "step": 2 }, { "epoch": 0.021409455842997322, "grad_norm": NaN, "learning_rate": 0.0001, "loss": 0.0, "step": 3 }, { "epoch": 0.028545941123996433, "grad_norm": NaN, "learning_rate": 9.99524110790929e-05, "loss": 0.0, "step": 4 }, { "epoch": 0.03568242640499554, "grad_norm": NaN, "learning_rate": 9.980973490458728e-05, "loss": 0.0, "step": 5 }, { "epoch": 0.042818911685994644, "grad_norm": NaN, "learning_rate": 9.957224306869053e-05, "loss": 0.0, "step": 6 }, { "epoch": 0.049955396966993755, "grad_norm": NaN, "learning_rate": 9.924038765061042e-05, "loss": 0.0, "step": 7 }, { "epoch": 0.057091882247992866, "grad_norm": NaN, "learning_rate": 9.881480035599667e-05, "loss": 0.0, "step": 8 }, { "epoch": 0.06422836752899197, "grad_norm": NaN, "learning_rate": 9.829629131445342e-05, "loss": 0.0, "step": 9 }, { "epoch": 0.07136485280999108, "grad_norm": NaN, "learning_rate": 9.768584753741134e-05, "loss": 0.0, "step": 10 }, { "epoch": 0.07850133809099019, "grad_norm": NaN, "learning_rate": 9.698463103929542e-05, "loss": 0.0, "step": 11 }, { "epoch": 0.08563782337198929, "grad_norm": NaN, "learning_rate": 9.619397662556435e-05, "loss": 0.0, "step": 12 }, { "epoch": 0.0927743086529884, "grad_norm": NaN, "learning_rate": 9.53153893518325e-05, "loss": 0.0, "step": 13 }, { "epoch": 0.09991079393398751, "grad_norm": NaN, "learning_rate": 9.435054165891109e-05, "loss": 0.0, "step": 14 }, { "epoch": 0.10704727921498662, "grad_norm": NaN, "learning_rate": 9.330127018922194e-05, "loss": 0.0, "step": 15 }, { "epoch": 0.11418376449598573, "grad_norm": NaN, "learning_rate": 9.21695722906443e-05, "loss": 0.0, "step": 16 }, { "epoch": 0.12132024977698483, "grad_norm": NaN, "learning_rate": 9.09576022144496e-05, "loss": 0.0, "step": 17 }, { "epoch": 0.12845673505798394, "grad_norm": NaN, "learning_rate": 8.966766701456177e-05, "loss": 0.0, "step": 18 }, { "epoch": 0.13559322033898305, "grad_norm": NaN, "learning_rate": 8.83022221559489e-05, "loss": 0.0, "step": 19 }, { "epoch": 0.14272970561998216, "grad_norm": NaN, "learning_rate": 8.68638668405062e-05, "loss": 0.0, "step": 20 }, { "epoch": 0.14986619090098127, "grad_norm": NaN, "learning_rate": 8.535533905932738e-05, "loss": 0.0, "step": 21 }, { "epoch": 0.15700267618198038, "grad_norm": NaN, "learning_rate": 8.377951038078302e-05, "loss": 0.0, "step": 22 }, { "epoch": 0.1641391614629795, "grad_norm": NaN, "learning_rate": 8.213938048432697e-05, "loss": 0.0, "step": 23 }, { "epoch": 0.17127564674397858, "grad_norm": NaN, "learning_rate": 8.043807145043604e-05, "loss": 0.0, "step": 24 }, { "epoch": 0.1784121320249777, "grad_norm": NaN, "learning_rate": 7.86788218175523e-05, "loss": 0.0, "step": 25 }, { "epoch": 0.1784121320249777, "eval_loss": NaN, "eval_runtime": 26.324, "eval_samples_per_second": 4.483, "eval_steps_per_second": 2.241, "step": 25 } ], "logging_steps": 1, "max_steps": 75, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 25, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 3.53626333642752e+16, "train_batch_size": 2, "trial_name": null, "trial_params": null }