{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.4547392354696604, "eval_steps": 40, "global_step": 200, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.002273696177348302, "eval_loss": NaN, "eval_runtime": 89.7585, "eval_samples_per_second": 8.255, "eval_steps_per_second": 4.133, "step": 1 }, { "epoch": 0.09094784709393207, "grad_norm": NaN, "learning_rate": 1e-05, "loss": 0.0, "step": 40 }, { "epoch": 0.09094784709393207, "eval_loss": NaN, "eval_runtime": 88.4448, "eval_samples_per_second": 8.378, "eval_steps_per_second": 4.195, "step": 40 }, { "epoch": 0.18189569418786414, "grad_norm": NaN, "learning_rate": 2e-05, "loss": 0.0, "step": 80 }, { "epoch": 0.18189569418786414, "eval_loss": NaN, "eval_runtime": 88.3357, "eval_samples_per_second": 8.388, "eval_steps_per_second": 4.2, "step": 80 }, { "epoch": 0.2728435412817962, "grad_norm": NaN, "learning_rate": 1.5000000000000002e-05, "loss": 0.0, "step": 120 }, { "epoch": 0.2728435412817962, "eval_loss": NaN, "eval_runtime": 88.487, "eval_samples_per_second": 8.374, "eval_steps_per_second": 4.193, "step": 120 }, { "epoch": 0.3637913883757283, "grad_norm": NaN, "learning_rate": 5.000000000000003e-06, "loss": 0.0, "step": 160 }, { "epoch": 0.3637913883757283, "eval_loss": NaN, "eval_runtime": 88.4091, "eval_samples_per_second": 8.381, "eval_steps_per_second": 4.196, "step": 160 }, { "epoch": 0.4547392354696604, "grad_norm": NaN, "learning_rate": 0.0, "loss": 0.0, "step": 200 }, { "epoch": 0.4547392354696604, "eval_loss": NaN, "eval_runtime": 88.3863, "eval_samples_per_second": 8.384, "eval_steps_per_second": 4.197, "step": 200 } ], "logging_steps": 40, "max_steps": 200, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 40, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 2.8735810005211546e+17, "train_batch_size": 2, "trial_name": null, "trial_params": null }