| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.10460251046025104, | |
| "eval_steps": 300, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.017433751743375175, | |
| "grad_norm": 11.72121810913086, | |
| "learning_rate": 9.966527196652718e-07, | |
| "loss": 10.429, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.03486750348675035, | |
| "grad_norm": 7.926635265350342, | |
| "learning_rate": 9.933054393305438e-07, | |
| "loss": 9.5621, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.05230125523012552, | |
| "grad_norm": 14.6965970993042, | |
| "learning_rate": 9.898884239888422e-07, | |
| "loss": 8.9643, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0697350069735007, | |
| "grad_norm": 19.375049591064453, | |
| "learning_rate": 9.864714086471408e-07, | |
| "loss": 8.4059, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08716875871687588, | |
| "grad_norm": 18.03301429748535, | |
| "learning_rate": 9.833333333333332e-07, | |
| "loss": 7.612, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.10460251046025104, | |
| "grad_norm": 10.96057415008545, | |
| "learning_rate": 9.798465829846583e-07, | |
| "loss": 6.8137, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.10460251046025104, | |
| "eval_gen_len": 10.4739, | |
| "eval_loss": 6.225385665893555, | |
| "eval_rouge1": 6.5635, | |
| "eval_rouge2": 1.1205, | |
| "eval_rougeL": 5.1919, | |
| "eval_rougeLsum": 5.8048, | |
| "eval_runtime": 493.488, | |
| "eval_samples_per_second": 11.036, | |
| "eval_steps_per_second": 0.087, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 14340, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 300, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5845995749376000.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |