| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7912957467853611, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01582591493570722, | |
| "eval_loss": 2.9574100971221924, | |
| "eval_runtime": 22.3481, | |
| "eval_samples_per_second": 19.062, | |
| "eval_steps_per_second": 2.416, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1582591493570722, | |
| "grad_norm": 18.0383358001709, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 25.6469, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1582591493570722, | |
| "eval_loss": 0.5446298718452454, | |
| "eval_runtime": 22.2504, | |
| "eval_samples_per_second": 19.146, | |
| "eval_steps_per_second": 2.427, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3165182987141444, | |
| "grad_norm": 7.299452781677246, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 8.6131, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3165182987141444, | |
| "eval_loss": 0.5305290222167969, | |
| "eval_runtime": 22.3962, | |
| "eval_samples_per_second": 19.021, | |
| "eval_steps_per_second": 2.411, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.47477744807121663, | |
| "grad_norm": 3.938899040222168, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 8.4817, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.47477744807121663, | |
| "eval_loss": 0.5269880890846252, | |
| "eval_runtime": 22.2504, | |
| "eval_samples_per_second": 19.146, | |
| "eval_steps_per_second": 2.427, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6330365974282888, | |
| "grad_norm": 5.748447418212891, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 8.4035, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6330365974282888, | |
| "eval_loss": 0.5215941071510315, | |
| "eval_runtime": 22.3805, | |
| "eval_samples_per_second": 19.034, | |
| "eval_steps_per_second": 2.413, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.7912957467853611, | |
| "grad_norm": 6.016201972961426, | |
| "learning_rate": 0.0, | |
| "loss": 8.3438, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7912957467853611, | |
| "eval_loss": 0.5206525325775146, | |
| "eval_runtime": 22.3482, | |
| "eval_samples_per_second": 19.062, | |
| "eval_steps_per_second": 2.416, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "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": 1.402137439371264e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |