{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.0032131094867057595, "eval_steps": 3, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.00032131094867057596, "grad_norm": NaN, "learning_rate": 2e-05, "loss": 0.0, "step": 1 }, { "epoch": 0.00032131094867057596, "eval_loss": NaN, "eval_runtime": 50.984, "eval_samples_per_second": 25.714, "eval_steps_per_second": 12.867, "step": 1 }, { "epoch": 0.0006426218973411519, "grad_norm": NaN, "learning_rate": 4e-05, "loss": 0.0, "step": 2 }, { "epoch": 0.0009639328460117278, "grad_norm": NaN, "learning_rate": 6e-05, "loss": 0.0, "step": 3 }, { "epoch": 0.0009639328460117278, "eval_loss": NaN, "eval_runtime": 48.8915, "eval_samples_per_second": 26.814, "eval_steps_per_second": 13.417, "step": 3 }, { "epoch": 0.0012852437946823038, "grad_norm": NaN, "learning_rate": 8e-05, "loss": 0.0, "step": 4 }, { "epoch": 0.0016065547433528797, "grad_norm": NaN, "learning_rate": 0.0001, "loss": 0.0, "step": 5 }, { "epoch": 0.0019278656920234556, "grad_norm": NaN, "learning_rate": 0.00012, "loss": 0.0, "step": 6 }, { "epoch": 0.0019278656920234556, "eval_loss": NaN, "eval_runtime": 49.3381, "eval_samples_per_second": 26.572, "eval_steps_per_second": 13.296, "step": 6 }, { "epoch": 0.0022491766406940315, "grad_norm": NaN, "learning_rate": 0.00014, "loss": 0.0, "step": 7 }, { "epoch": 0.0025704875893646077, "grad_norm": NaN, "learning_rate": 0.00016, "loss": 0.3096, "step": 8 }, { "epoch": 0.0028917985380351834, "grad_norm": NaN, "learning_rate": 0.00018, "loss": 0.0, "step": 9 }, { "epoch": 0.0028917985380351834, "eval_loss": NaN, "eval_runtime": 48.9544, "eval_samples_per_second": 26.78, "eval_steps_per_second": 13.4, "step": 9 }, { "epoch": 0.0032131094867057595, "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": 2035917678182400.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }