{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.00013869962169678183, "eval_steps": 3, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.3869962169678183e-05, "grad_norm": NaN, "learning_rate": 2e-05, "loss": 0.0, "step": 1 }, { "epoch": 1.3869962169678183e-05, "eval_loss": NaN, "eval_runtime": 581.7099, "eval_samples_per_second": 52.188, "eval_steps_per_second": 26.094, "step": 1 }, { "epoch": 2.7739924339356365e-05, "grad_norm": NaN, "learning_rate": 4e-05, "loss": 0.0, "step": 2 }, { "epoch": 4.1609886509034544e-05, "grad_norm": NaN, "learning_rate": 6e-05, "loss": 0.0, "step": 3 }, { "epoch": 4.1609886509034544e-05, "eval_loss": NaN, "eval_runtime": 577.9325, "eval_samples_per_second": 52.529, "eval_steps_per_second": 26.264, "step": 3 }, { "epoch": 5.547984867871273e-05, "grad_norm": NaN, "learning_rate": 8e-05, "loss": 0.0, "step": 4 }, { "epoch": 6.934981084839092e-05, "grad_norm": NaN, "learning_rate": 0.0001, "loss": 0.0, "step": 5 }, { "epoch": 8.321977301806909e-05, "grad_norm": NaN, "learning_rate": 0.00012, "loss": 0.0, "step": 6 }, { "epoch": 8.321977301806909e-05, "eval_loss": NaN, "eval_runtime": 580.1344, "eval_samples_per_second": 52.329, "eval_steps_per_second": 26.165, "step": 6 }, { "epoch": 9.708973518774727e-05, "grad_norm": NaN, "learning_rate": 0.00014, "loss": 0.0, "step": 7 }, { "epoch": 0.00011095969735742546, "grad_norm": NaN, "learning_rate": 0.00016, "loss": 0.0, "step": 8 }, { "epoch": 0.00012482965952710363, "grad_norm": NaN, "learning_rate": 0.00018, "loss": 0.0, "step": 9 }, { "epoch": 0.00012482965952710363, "eval_loss": NaN, "eval_runtime": 580.9924, "eval_samples_per_second": 52.252, "eval_steps_per_second": 26.126, "step": 9 }, { "epoch": 0.00013869962169678183, "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 }