| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0008680781617576847, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.7361563235153692e-05, | |
| "eval_loss": 3.2406625747680664, | |
| "eval_runtime": 3009.3145, | |
| "eval_samples_per_second": 8.059, | |
| "eval_steps_per_second": 4.029, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 8.680781617576846e-05, | |
| "grad_norm": 55.02191162109375, | |
| "learning_rate": 5e-05, | |
| "loss": 12.0913, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00017361563235153692, | |
| "grad_norm": 51.31801223754883, | |
| "learning_rate": 0.0001, | |
| "loss": 8.5732, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00017361563235153692, | |
| "eval_loss": 1.781690239906311, | |
| "eval_runtime": 1754.7844, | |
| "eval_samples_per_second": 13.821, | |
| "eval_steps_per_second": 6.91, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0002604234485273054, | |
| "grad_norm": 23.189556121826172, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 6.8911, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00034723126470307384, | |
| "grad_norm": 23.172544479370117, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 6.525, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00034723126470307384, | |
| "eval_loss": 1.3506044149398804, | |
| "eval_runtime": 1565.3062, | |
| "eval_samples_per_second": 15.493, | |
| "eval_steps_per_second": 7.747, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00043403908087884236, | |
| "grad_norm": 28.07756805419922, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 5.277, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0005208468970546108, | |
| "grad_norm": 25.659969329833984, | |
| "learning_rate": 5e-05, | |
| "loss": 5.2087, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0005208468970546108, | |
| "eval_loss": 1.267523169517517, | |
| "eval_runtime": 1566.1685, | |
| "eval_samples_per_second": 15.485, | |
| "eval_steps_per_second": 7.742, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0006076547132303793, | |
| "grad_norm": 17.720985412597656, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 4.6799, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0006944625294061477, | |
| "grad_norm": 18.21421241760254, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 5.3353, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0006944625294061477, | |
| "eval_loss": 1.2463383674621582, | |
| "eval_runtime": 1559.5002, | |
| "eval_samples_per_second": 15.551, | |
| "eval_steps_per_second": 7.776, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0007812703455819162, | |
| "grad_norm": 20.94264793395996, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 5.2961, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0008680781617576847, | |
| "grad_norm": 17.284198760986328, | |
| "learning_rate": 0.0, | |
| "loss": 4.389, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0008680781617576847, | |
| "eval_loss": 1.2408928871154785, | |
| "eval_runtime": 1554.7039, | |
| "eval_samples_per_second": 15.599, | |
| "eval_steps_per_second": 7.8, | |
| "step": 50 | |
| } | |
| ], | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8763509991014400.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |