{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.0018535681186283596, "eval_steps": 3, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.00018535681186283596, "grad_norm": NaN, "learning_rate": 2e-05, "loss": 0.0, "step": 1 }, { "epoch": 0.00018535681186283596, "eval_loss": NaN, "eval_runtime": 46.0307, "eval_samples_per_second": 49.358, "eval_steps_per_second": 24.679, "step": 1 }, { "epoch": 0.0003707136237256719, "grad_norm": NaN, "learning_rate": 4e-05, "loss": 0.0, "step": 2 }, { "epoch": 0.0005560704355885078, "grad_norm": NaN, "learning_rate": 6e-05, "loss": 0.0, "step": 3 }, { "epoch": 0.0005560704355885078, "eval_loss": NaN, "eval_runtime": 45.1878, "eval_samples_per_second": 50.279, "eval_steps_per_second": 25.14, "step": 3 }, { "epoch": 0.0007414272474513438, "grad_norm": NaN, "learning_rate": 8e-05, "loss": 0.0, "step": 4 }, { "epoch": 0.0009267840593141798, "grad_norm": NaN, "learning_rate": 0.0001, "loss": 0.0, "step": 5 }, { "epoch": 0.0011121408711770157, "grad_norm": NaN, "learning_rate": 0.00012, "loss": 0.0, "step": 6 }, { "epoch": 0.0011121408711770157, "eval_loss": NaN, "eval_runtime": 45.1378, "eval_samples_per_second": 50.335, "eval_steps_per_second": 25.167, "step": 6 }, { "epoch": 0.0012974976830398518, "grad_norm": NaN, "learning_rate": 0.00014, "loss": 0.0, "step": 7 }, { "epoch": 0.0014828544949026877, "grad_norm": NaN, "learning_rate": 0.00016, "loss": 0.0, "step": 8 }, { "epoch": 0.0016682113067655235, "grad_norm": NaN, "learning_rate": 0.00018, "loss": 0.0, "step": 9 }, { "epoch": 0.0016682113067655235, "eval_loss": NaN, "eval_runtime": 44.8166, "eval_samples_per_second": 50.696, "eval_steps_per_second": 25.348, "step": 9 }, { "epoch": 0.0018535681186283596, "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": 389157910216704.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }