{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.0010375865736297373, "eval_steps": 3, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.00010375865736297373, "grad_norm": NaN, "learning_rate": 2e-05, "loss": 0.0, "step": 1 }, { "epoch": 0.00010375865736297373, "eval_loss": NaN, "eval_runtime": 101.3259, "eval_samples_per_second": 40.049, "eval_steps_per_second": 20.024, "step": 1 }, { "epoch": 0.00020751731472594745, "grad_norm": NaN, "learning_rate": 4e-05, "loss": 0.0, "step": 2 }, { "epoch": 0.00031127597208892115, "grad_norm": NaN, "learning_rate": 6e-05, "loss": 0.0, "step": 3 }, { "epoch": 0.00031127597208892115, "eval_loss": NaN, "eval_runtime": 104.7739, "eval_samples_per_second": 38.731, "eval_steps_per_second": 19.366, "step": 3 }, { "epoch": 0.0004150346294518949, "grad_norm": NaN, "learning_rate": 8e-05, "loss": 0.0, "step": 4 }, { "epoch": 0.0005187932868148687, "grad_norm": NaN, "learning_rate": 0.0001, "loss": 0.0, "step": 5 }, { "epoch": 0.0006225519441778423, "grad_norm": NaN, "learning_rate": 0.00012, "loss": 0.0, "step": 6 }, { "epoch": 0.0006225519441778423, "eval_loss": NaN, "eval_runtime": 104.3311, "eval_samples_per_second": 38.895, "eval_steps_per_second": 19.448, "step": 6 }, { "epoch": 0.000726310601540816, "grad_norm": NaN, "learning_rate": 0.00014, "loss": 0.0, "step": 7 }, { "epoch": 0.0008300692589037898, "grad_norm": NaN, "learning_rate": 0.00016, "loss": 0.0, "step": 8 }, { "epoch": 0.0009338279162667636, "grad_norm": NaN, "learning_rate": 0.00018, "loss": 0.0, "step": 9 }, { "epoch": 0.0009338279162667636, "eval_loss": NaN, "eval_runtime": 100.5985, "eval_samples_per_second": 40.339, "eval_steps_per_second": 20.169, "step": 9 }, { "epoch": 0.0010375865736297373, "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": 429235545047040.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }