{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.014749262536873156, "eval_steps": 3, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0014749262536873156, "grad_norm": NaN, "learning_rate": 2e-05, "loss": 0.0, "step": 1 }, { "epoch": 0.0014749262536873156, "eval_loss": NaN, "eval_runtime": 9.3677, "eval_samples_per_second": 30.531, "eval_steps_per_second": 15.265, "step": 1 }, { "epoch": 0.0029498525073746312, "grad_norm": NaN, "learning_rate": 4e-05, "loss": 0.0, "step": 2 }, { "epoch": 0.004424778761061947, "grad_norm": NaN, "learning_rate": 6e-05, "loss": 0.0, "step": 3 }, { "epoch": 0.004424778761061947, "eval_loss": NaN, "eval_runtime": 8.8918, "eval_samples_per_second": 32.164, "eval_steps_per_second": 16.082, "step": 3 }, { "epoch": 0.0058997050147492625, "grad_norm": NaN, "learning_rate": 8e-05, "loss": 0.0, "step": 4 }, { "epoch": 0.007374631268436578, "grad_norm": NaN, "learning_rate": 0.0001, "loss": 0.0, "step": 5 }, { "epoch": 0.008849557522123894, "grad_norm": NaN, "learning_rate": 0.00012, "loss": 0.0, "step": 6 }, { "epoch": 0.008849557522123894, "eval_loss": NaN, "eval_runtime": 8.4831, "eval_samples_per_second": 33.714, "eval_steps_per_second": 16.857, "step": 6 }, { "epoch": 0.01032448377581121, "grad_norm": NaN, "learning_rate": 0.00014, "loss": 0.0, "step": 7 }, { "epoch": 0.011799410029498525, "grad_norm": NaN, "learning_rate": 0.00016, "loss": 0.0, "step": 8 }, { "epoch": 0.01327433628318584, "grad_norm": NaN, "learning_rate": 0.00018, "loss": 0.0, "step": 9 }, { "epoch": 0.01327433628318584, "eval_loss": NaN, "eval_runtime": 7.5667, "eval_samples_per_second": 37.797, "eval_steps_per_second": 18.899, "step": 9 }, { "epoch": 0.014749262536873156, "grad_norm": NaN, "learning_rate": 0.0002, "loss": 0.0, "step": 10 } ], "logging_steps": 1, "max_steps": 10, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 3, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 316387255910400.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }