{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.0161976949434119, "eval_steps": 10, "global_step": 39, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0004153255113695359, "eval_loss": NaN, "eval_runtime": 45.2758, "eval_samples_per_second": 22.396, "eval_steps_per_second": 11.198, "step": 1 }, { "epoch": 0.0020766275568476792, "grad_norm": NaN, "learning_rate": 5e-05, "loss": 0.0, "step": 5 }, { "epoch": 0.0041532551136953585, "grad_norm": NaN, "learning_rate": 0.0001, "loss": 0.0, "step": 10 }, { "epoch": 0.0041532551136953585, "eval_loss": NaN, "eval_runtime": 43.7873, "eval_samples_per_second": 23.157, "eval_steps_per_second": 11.579, "step": 10 }, { "epoch": 0.006229882670543038, "grad_norm": NaN, "learning_rate": 9.619397662556435e-05, "loss": 0.0, "step": 15 }, { "epoch": 0.008306510227390717, "grad_norm": NaN, "learning_rate": 8.535533905932738e-05, "loss": 0.0, "step": 20 }, { "epoch": 0.008306510227390717, "eval_loss": NaN, "eval_runtime": 43.782, "eval_samples_per_second": 23.16, "eval_steps_per_second": 11.58, "step": 20 }, { "epoch": 0.010383137784238397, "grad_norm": NaN, "learning_rate": 6.91341716182545e-05, "loss": 0.0, "step": 25 }, { "epoch": 0.012459765341086075, "grad_norm": NaN, "learning_rate": 5e-05, "loss": 0.0, "step": 30 }, { "epoch": 0.012459765341086075, "eval_loss": NaN, "eval_runtime": 43.7965, "eval_samples_per_second": 23.153, "eval_steps_per_second": 11.576, "step": 30 }, { "epoch": 0.014536392897933756, "grad_norm": NaN, "learning_rate": 3.086582838174551e-05, "loss": 0.0, "step": 35 } ], "logging_steps": 5, "max_steps": 50, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 13, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 2343054917763072.0, "train_batch_size": 2, "trial_name": null, "trial_params": null }