{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.06279069767441861, "eval_steps": 9, "global_step": 27, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.002325581395348837, "eval_loss": 5.767578601837158, "eval_runtime": 10.4189, "eval_samples_per_second": 69.489, "eval_steps_per_second": 8.734, "step": 1 }, { "epoch": 0.0069767441860465115, "grad_norm": 0.7177151441574097, "learning_rate": 3e-05, "loss": 5.7725, "step": 3 }, { "epoch": 0.013953488372093023, "grad_norm": 0.6705913543701172, "learning_rate": 6e-05, "loss": 5.6508, "step": 6 }, { "epoch": 0.020930232558139535, "grad_norm": 0.9152357578277588, "learning_rate": 9e-05, "loss": 5.7661, "step": 9 }, { "epoch": 0.020930232558139535, "eval_loss": 5.740058898925781, "eval_runtime": 10.4015, "eval_samples_per_second": 69.605, "eval_steps_per_second": 8.749, "step": 9 }, { "epoch": 0.027906976744186046, "grad_norm": 0.8683664798736572, "learning_rate": 9.987820251299122e-05, "loss": 5.7731, "step": 12 }, { "epoch": 0.03488372093023256, "grad_norm": 1.1061254739761353, "learning_rate": 9.924038765061042e-05, "loss": 5.4242, "step": 15 }, { "epoch": 0.04186046511627907, "grad_norm": 0.99778813123703, "learning_rate": 9.806308479691595e-05, "loss": 5.2717, "step": 18 }, { "epoch": 0.04186046511627907, "eval_loss": 5.513608932495117, "eval_runtime": 10.4282, "eval_samples_per_second": 69.427, "eval_steps_per_second": 8.726, "step": 18 }, { "epoch": 0.04883720930232558, "grad_norm": 1.3651623725891113, "learning_rate": 9.635919272833938e-05, "loss": 5.5186, "step": 21 }, { "epoch": 0.05581395348837209, "grad_norm": 1.554046630859375, "learning_rate": 9.414737964294636e-05, "loss": 5.3002, "step": 24 }, { "epoch": 0.06279069767441861, "grad_norm": 1.4339934587478638, "learning_rate": 9.145187862775209e-05, "loss": 5.1694, "step": 27 }, { "epoch": 0.06279069767441861, "eval_loss": 5.061354637145996, "eval_runtime": 10.4152, "eval_samples_per_second": 69.514, "eval_steps_per_second": 8.737, "step": 27 } ], "logging_steps": 3, "max_steps": 100, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 9, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 1693262398095360.0, "train_batch_size": 8, "trial_name": null, "trial_params": null }