{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.00021329920546045967, "eval_steps": 3, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 2.1329920546045965e-05, "grad_norm": NaN, "learning_rate": 2e-05, "loss": 0.0, "step": 1 }, { "epoch": 2.1329920546045965e-05, "eval_loss": NaN, "eval_runtime": 1810.8524, "eval_samples_per_second": 10.901, "eval_steps_per_second": 5.45, "step": 1 }, { "epoch": 4.265984109209193e-05, "grad_norm": NaN, "learning_rate": 4e-05, "loss": 0.0, "step": 2 }, { "epoch": 6.39897616381379e-05, "grad_norm": NaN, "learning_rate": 6e-05, "loss": 0.0, "step": 3 }, { "epoch": 6.39897616381379e-05, "eval_loss": NaN, "eval_runtime": 1807.2251, "eval_samples_per_second": 10.923, "eval_steps_per_second": 5.461, "step": 3 }, { "epoch": 8.531968218418386e-05, "grad_norm": NaN, "learning_rate": 8e-05, "loss": 0.0, "step": 4 }, { "epoch": 0.00010664960273022984, "grad_norm": NaN, "learning_rate": 0.0001, "loss": 0.0, "step": 5 }, { "epoch": 0.0001279795232762758, "grad_norm": NaN, "learning_rate": 0.00012, "loss": 0.0, "step": 6 }, { "epoch": 0.0001279795232762758, "eval_loss": NaN, "eval_runtime": 1809.9333, "eval_samples_per_second": 10.906, "eval_steps_per_second": 5.453, "step": 6 }, { "epoch": 0.00014930944382232175, "grad_norm": NaN, "learning_rate": 0.00014, "loss": 0.0, "step": 7 }, { "epoch": 0.00017063936436836772, "grad_norm": NaN, "learning_rate": 0.00016, "loss": 0.0, "step": 8 }, { "epoch": 0.00019196928491441368, "grad_norm": NaN, "learning_rate": 0.00018, "loss": 0.0, "step": 9 }, { "epoch": 0.00019196928491441368, "eval_loss": NaN, "eval_runtime": 1811.3211, "eval_samples_per_second": 10.898, "eval_steps_per_second": 5.449, "step": 9 }, { "epoch": 0.00021329920546045967, "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": 6315562638508032.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }