{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.0019504900606277327, "eval_steps": 8, "global_step": 30, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 6.501633535425775e-05, "eval_loss": NaN, "eval_runtime": 372.3048, "eval_samples_per_second": 17.397, "eval_steps_per_second": 8.7, "step": 1 }, { "epoch": 0.00019504900606277329, "grad_norm": NaN, "learning_rate": 6e-05, "loss": 0.0, "step": 3 }, { "epoch": 0.00039009801212554657, "grad_norm": NaN, "learning_rate": 0.00012, "loss": 0.0, "step": 6 }, { "epoch": 0.000520130682834062, "eval_loss": NaN, "eval_runtime": 370.7167, "eval_samples_per_second": 17.472, "eval_steps_per_second": 8.737, "step": 8 }, { "epoch": 0.0005851470181883198, "grad_norm": NaN, "learning_rate": 0.00018, "loss": 0.0, "step": 9 }, { "epoch": 0.0007801960242510931, "grad_norm": NaN, "learning_rate": 0.00019510565162951537, "loss": 0.0, "step": 12 }, { "epoch": 0.0009752450303138663, "grad_norm": NaN, "learning_rate": 0.00017071067811865476, "loss": 0.0, "step": 15 }, { "epoch": 0.001040261365668124, "eval_loss": NaN, "eval_runtime": 370.7384, "eval_samples_per_second": 17.471, "eval_steps_per_second": 8.737, "step": 16 }, { "epoch": 0.0011702940363766397, "grad_norm": NaN, "learning_rate": 0.00013090169943749476, "loss": 0.0, "step": 18 }, { "epoch": 0.0013653430424394129, "grad_norm": NaN, "learning_rate": 8.435655349597689e-05, "loss": 0.0, "step": 21 }, { "epoch": 0.0015603920485021863, "grad_norm": NaN, "learning_rate": 4.12214747707527e-05, "loss": 0.0, "step": 24 }, { "epoch": 0.0015603920485021863, "eval_loss": NaN, "eval_runtime": 370.8397, "eval_samples_per_second": 17.466, "eval_steps_per_second": 8.734, "step": 24 }, { "epoch": 0.0017554410545649595, "grad_norm": NaN, "learning_rate": 1.0899347581163221e-05, "loss": 0.0, "step": 27 }, { "epoch": 0.0019504900606277327, "grad_norm": NaN, "learning_rate": 0.0, "loss": 0.0, "step": 30 } ], "logging_steps": 3, "max_steps": 30, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 10, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 1.112830925340672e+16, "train_batch_size": 2, "trial_name": null, "trial_params": null }