| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.057803468208092484, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0011560693641618498, | |
| "eval_loss": 4.957520008087158, | |
| "eval_runtime": 11.1595, | |
| "eval_samples_per_second": 32.708, | |
| "eval_steps_per_second": 16.399, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011560693641618497, | |
| "grad_norm": 21.333415985107422, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 16.8186, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.015028901734104046, | |
| "eval_loss": 2.6783394813537598, | |
| "eval_runtime": 10.0406, | |
| "eval_samples_per_second": 36.353, | |
| "eval_steps_per_second": 18.226, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.023121387283236993, | |
| "grad_norm": 25.738534927368164, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 10.5807, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03005780346820809, | |
| "eval_loss": 2.4230074882507324, | |
| "eval_runtime": 10.084, | |
| "eval_samples_per_second": 36.196, | |
| "eval_steps_per_second": 18.148, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03468208092485549, | |
| "grad_norm": 31.6507511138916, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 9.2891, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04508670520231214, | |
| "eval_loss": 2.3214426040649414, | |
| "eval_runtime": 10.0582, | |
| "eval_samples_per_second": 36.289, | |
| "eval_steps_per_second": 18.194, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.046242774566473986, | |
| "grad_norm": 21.361957550048828, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 9.5961, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.057803468208092484, | |
| "grad_norm": 29.381607055664062, | |
| "learning_rate": 0.0, | |
| "loss": 8.722, | |
| "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": 8407261858234368.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |