| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.029601518026565465, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007590132827324478, | |
| "eval_loss": 1.3234649896621704, | |
| "eval_runtime": 55.836, | |
| "eval_samples_per_second": 9.94, | |
| "eval_steps_per_second": 4.979, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003795066413662239, | |
| "grad_norm": 2.922577381134033, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1409, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.007590132827324478, | |
| "grad_norm": 1.7498085498809814, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7889, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.007590132827324478, | |
| "eval_loss": 0.8516691327095032, | |
| "eval_runtime": 56.4068, | |
| "eval_samples_per_second": 9.839, | |
| "eval_steps_per_second": 4.928, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.011385199240986717, | |
| "grad_norm": 2.6687936782836914, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.6899, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.015180265654648957, | |
| "grad_norm": 2.109771728515625, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.406, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.015180265654648957, | |
| "eval_loss": 0.46676212549209595, | |
| "eval_runtime": 56.5802, | |
| "eval_samples_per_second": 9.809, | |
| "eval_steps_per_second": 4.913, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.018975332068311195, | |
| "grad_norm": 2.3853554725646973, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.4133, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.022770398481973434, | |
| "grad_norm": 2.1260995864868164, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1856, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.022770398481973434, | |
| "eval_loss": 0.31681013107299805, | |
| "eval_runtime": 56.7828, | |
| "eval_samples_per_second": 9.774, | |
| "eval_steps_per_second": 4.896, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.026565464895635674, | |
| "grad_norm": 2.295457124710083, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.3922, | |
| "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": 1.1282341913690112e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |