{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.0035622050048980318, "eval_steps": 3, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.00035622050048980317, "grad_norm": NaN, "learning_rate": 2e-05, "loss": 0.0, "step": 1 }, { "epoch": 0.00035622050048980317, "eval_loss": NaN, "eval_runtime": 39.4869, "eval_samples_per_second": 29.934, "eval_steps_per_second": 14.967, "step": 1 }, { "epoch": 0.0007124410009796063, "grad_norm": NaN, "learning_rate": 4e-05, "loss": 0.0, "step": 2 }, { "epoch": 0.0010686615014694097, "grad_norm": NaN, "learning_rate": 6e-05, "loss": 0.0, "step": 3 }, { "epoch": 0.0010686615014694097, "eval_loss": NaN, "eval_runtime": 38.8037, "eval_samples_per_second": 30.461, "eval_steps_per_second": 15.231, "step": 3 }, { "epoch": 0.0014248820019592127, "grad_norm": NaN, "learning_rate": 8e-05, "loss": 0.0, "step": 4 }, { "epoch": 0.0017811025024490159, "grad_norm": NaN, "learning_rate": 0.0001, "loss": 0.0, "step": 5 }, { "epoch": 0.0021373230029388193, "grad_norm": NaN, "learning_rate": 0.00012, "loss": 0.0, "step": 6 }, { "epoch": 0.0021373230029388193, "eval_loss": NaN, "eval_runtime": 38.8187, "eval_samples_per_second": 30.449, "eval_steps_per_second": 15.225, "step": 6 }, { "epoch": 0.002493543503428622, "grad_norm": NaN, "learning_rate": 0.00014, "loss": 0.0, "step": 7 }, { "epoch": 0.0028497640039184253, "grad_norm": NaN, "learning_rate": 0.00016, "loss": 0.0, "step": 8 }, { "epoch": 0.0032059845044082286, "grad_norm": NaN, "learning_rate": 0.00018, "loss": 0.0, "step": 9 }, { "epoch": 0.0032059845044082286, "eval_loss": NaN, "eval_runtime": 38.813, "eval_samples_per_second": 30.454, "eval_steps_per_second": 15.227, "step": 9 }, { "epoch": 0.0035622050048980318, "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": 1391446091366400.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }