{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.00046108447067502766, "eval_steps": 3, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 4.610844706750277e-05, "grad_norm": NaN, "learning_rate": 2e-05, "loss": 0.3648, "step": 1 }, { "epoch": 4.610844706750277e-05, "eval_loss": NaN, "eval_runtime": 192.3196, "eval_samples_per_second": 47.483, "eval_steps_per_second": 23.742, "step": 1 }, { "epoch": 9.221689413500554e-05, "grad_norm": NaN, "learning_rate": 4e-05, "loss": 0.0, "step": 2 }, { "epoch": 0.0001383253412025083, "grad_norm": NaN, "learning_rate": 6e-05, "loss": 0.0, "step": 3 }, { "epoch": 0.0001383253412025083, "eval_loss": NaN, "eval_runtime": 189.3675, "eval_samples_per_second": 48.224, "eval_steps_per_second": 24.112, "step": 3 }, { "epoch": 0.00018443378827001107, "grad_norm": NaN, "learning_rate": 8e-05, "loss": 0.0, "step": 4 }, { "epoch": 0.00023054223533751383, "grad_norm": NaN, "learning_rate": 0.0001, "loss": 0.0, "step": 5 }, { "epoch": 0.0002766506824050166, "grad_norm": NaN, "learning_rate": 0.00012, "loss": 0.0, "step": 6 }, { "epoch": 0.0002766506824050166, "eval_loss": NaN, "eval_runtime": 190.0376, "eval_samples_per_second": 48.054, "eval_steps_per_second": 24.027, "step": 6 }, { "epoch": 0.00032275912947251934, "grad_norm": NaN, "learning_rate": 0.00014, "loss": 0.0, "step": 7 }, { "epoch": 0.00036886757654002215, "grad_norm": NaN, "learning_rate": 0.00016, "loss": 0.0, "step": 8 }, { "epoch": 0.0004149760236075249, "grad_norm": NaN, "learning_rate": 0.00018, "loss": 0.0, "step": 9 }, { "epoch": 0.0004149760236075249, "eval_loss": NaN, "eval_runtime": 190.522, "eval_samples_per_second": 47.931, "eval_steps_per_second": 23.966, "step": 9 }, { "epoch": 0.00046108447067502766, "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": 97989722112000.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }