| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0205607476635514, | |
| "eval_steps": 7, | |
| "global_step": 34, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.029906542056074768, | |
| "eval_loss": 3.908674716949463, | |
| "eval_runtime": 11.8982, | |
| "eval_samples_per_second": 18.995, | |
| "eval_steps_per_second": 2.437, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.20934579439252338, | |
| "eval_loss": 0.4576127827167511, | |
| "eval_runtime": 11.8715, | |
| "eval_samples_per_second": 19.037, | |
| "eval_steps_per_second": 2.443, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.29906542056074764, | |
| "grad_norm": 29.873859405517578, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 27.8654, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.41869158878504675, | |
| "eval_loss": 9.885928193398286e-06, | |
| "eval_runtime": 11.905, | |
| "eval_samples_per_second": 18.984, | |
| "eval_steps_per_second": 2.436, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5981308411214953, | |
| "grad_norm": 0.007271628361195326, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.0024, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6280373831775701, | |
| "eval_loss": 3.436322458583163e-06, | |
| "eval_runtime": 11.8841, | |
| "eval_samples_per_second": 19.017, | |
| "eval_steps_per_second": 2.44, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8373831775700935, | |
| "eval_loss": 3.368011675775051e-06, | |
| "eval_runtime": 11.9911, | |
| "eval_samples_per_second": 18.847, | |
| "eval_steps_per_second": 2.418, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.897196261682243, | |
| "grad_norm": 0.006284361705183983, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0001, | |
| "step": 30 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 34, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.53452363555799e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |