{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.3280032800328003, "eval_steps": 10, "global_step": 50, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.006560065600656006, "eval_loss": NaN, "eval_runtime": 23.3316, "eval_samples_per_second": 44.018, "eval_steps_per_second": 5.529, "step": 1 }, { "epoch": 0.06560065600656007, "grad_norm": NaN, "learning_rate": 9.330127018922194e-05, "loss": 0.0, "step": 10 }, { "epoch": 0.06560065600656007, "eval_loss": NaN, "eval_runtime": 22.0132, "eval_samples_per_second": 46.654, "eval_steps_per_second": 5.86, "step": 10 }, { "epoch": 0.13120131201312013, "grad_norm": NaN, "learning_rate": 6.91341716182545e-05, "loss": 0.0, "step": 20 }, { "epoch": 0.13120131201312013, "eval_loss": NaN, "eval_runtime": 22.0176, "eval_samples_per_second": 46.644, "eval_steps_per_second": 5.859, "step": 20 }, { "epoch": 0.1968019680196802, "grad_norm": NaN, "learning_rate": 3.705904774487396e-05, "loss": 0.0, "step": 30 }, { "epoch": 0.1968019680196802, "eval_loss": NaN, "eval_runtime": 22.0626, "eval_samples_per_second": 46.549, "eval_steps_per_second": 5.847, "step": 30 }, { "epoch": 0.26240262402624026, "grad_norm": NaN, "learning_rate": 1.0332332985438248e-05, "loss": 0.0, "step": 40 }, { "epoch": 0.26240262402624026, "eval_loss": NaN, "eval_runtime": 22.1093, "eval_samples_per_second": 46.451, "eval_steps_per_second": 5.835, "step": 40 }, { "epoch": 0.3280032800328003, "grad_norm": NaN, "learning_rate": 0.0, "loss": 0.0, "step": 50 }, { "epoch": 0.3280032800328003, "eval_loss": NaN, "eval_runtime": 22.0006, "eval_samples_per_second": 46.681, "eval_steps_per_second": 5.863, "step": 50 } ], "logging_steps": 10, "max_steps": 50, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 13, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1.42461062086656e+16, "train_batch_size": 8, "trial_name": null, "trial_params": null }