| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 59.927739863508634, | |
| "eval_steps": 1866, | |
| "global_step": 18660, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.992773986350863, | |
| "grad_norm": 4.623592853546143, | |
| "learning_rate": 9.001071811361202e-06, | |
| "loss": 2.0366, | |
| "step": 1866 | |
| }, | |
| { | |
| "epoch": 5.992773986350863, | |
| "eval_accuracy": 0.32132862878695706, | |
| "eval_loss": 2.3591790199279785, | |
| "eval_runtime": 26.4016, | |
| "eval_samples_per_second": 311.307, | |
| "eval_steps_per_second": 15.567, | |
| "step": 1866 | |
| }, | |
| { | |
| "epoch": 11.985547972701726, | |
| "grad_norm": 4.165753364562988, | |
| "learning_rate": 8.001607717041802e-06, | |
| "loss": 1.8711, | |
| "step": 3732 | |
| }, | |
| { | |
| "epoch": 11.985547972701726, | |
| "eval_accuracy": 0.31104757269740846, | |
| "eval_loss": 2.4432082176208496, | |
| "eval_runtime": 26.6876, | |
| "eval_samples_per_second": 307.971, | |
| "eval_steps_per_second": 15.4, | |
| "step": 3732 | |
| }, | |
| { | |
| "epoch": 17.97832195905259, | |
| "grad_norm": 3.7397663593292236, | |
| "learning_rate": 7.002143622722402e-06, | |
| "loss": 1.7987, | |
| "step": 5598 | |
| }, | |
| { | |
| "epoch": 17.97832195905259, | |
| "eval_accuracy": 0.3014154195563126, | |
| "eval_loss": 2.5181965827941895, | |
| "eval_runtime": 26.5667, | |
| "eval_samples_per_second": 309.372, | |
| "eval_steps_per_second": 15.47, | |
| "step": 5598 | |
| }, | |
| { | |
| "epoch": 23.97109594540345, | |
| "grad_norm": 3.9918863773345947, | |
| "learning_rate": 6.002679528403002e-06, | |
| "loss": 1.7585, | |
| "step": 7464 | |
| }, | |
| { | |
| "epoch": 23.97109594540345, | |
| "eval_accuracy": 0.2936792797177272, | |
| "eval_loss": 2.556520938873291, | |
| "eval_runtime": 26.5054, | |
| "eval_samples_per_second": 310.088, | |
| "eval_steps_per_second": 15.506, | |
| "step": 7464 | |
| }, | |
| { | |
| "epoch": 29.963869931754317, | |
| "grad_norm": 2.9927406311035156, | |
| "learning_rate": 5.003215434083602e-06, | |
| "loss": 1.7381, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 29.963869931754317, | |
| "eval_accuracy": 0.2894756053047816, | |
| "eval_loss": 2.597092628479004, | |
| "eval_runtime": 26.5051, | |
| "eval_samples_per_second": 310.092, | |
| "eval_steps_per_second": 15.506, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 35.95664391810518, | |
| "grad_norm": 2.9211699962615967, | |
| "learning_rate": 4.003751339764202e-06, | |
| "loss": 1.7139, | |
| "step": 11196 | |
| }, | |
| { | |
| "epoch": 35.95664391810518, | |
| "eval_accuracy": 0.28486839437076694, | |
| "eval_loss": 2.6405980587005615, | |
| "eval_runtime": 26.4167, | |
| "eval_samples_per_second": 311.128, | |
| "eval_steps_per_second": 15.558, | |
| "step": 11196 | |
| }, | |
| { | |
| "epoch": 41.94941790445604, | |
| "grad_norm": 3.3739333152770996, | |
| "learning_rate": 3.004287245444802e-06, | |
| "loss": 1.7102, | |
| "step": 13062 | |
| }, | |
| { | |
| "epoch": 41.94941790445604, | |
| "eval_accuracy": 0.2814732414440408, | |
| "eval_loss": 2.67031192779541, | |
| "eval_runtime": 26.3808, | |
| "eval_samples_per_second": 311.552, | |
| "eval_steps_per_second": 15.579, | |
| "step": 13062 | |
| }, | |
| { | |
| "epoch": 47.9421918908069, | |
| "grad_norm": 3.284862995147705, | |
| "learning_rate": 2.0053590568060023e-06, | |
| "loss": 1.6951, | |
| "step": 14928 | |
| }, | |
| { | |
| "epoch": 47.9421918908069, | |
| "eval_accuracy": 0.27834894756053047, | |
| "eval_loss": 2.6752700805664062, | |
| "eval_runtime": 26.5225, | |
| "eval_samples_per_second": 309.888, | |
| "eval_steps_per_second": 15.496, | |
| "step": 14928 | |
| }, | |
| { | |
| "epoch": 53.934965877157765, | |
| "grad_norm": 2.573415994644165, | |
| "learning_rate": 1.005359056806002e-06, | |
| "loss": 1.6954, | |
| "step": 16794 | |
| }, | |
| { | |
| "epoch": 53.934965877157765, | |
| "eval_accuracy": 0.2760811669438023, | |
| "eval_loss": 2.6846840381622314, | |
| "eval_runtime": 26.5482, | |
| "eval_samples_per_second": 309.588, | |
| "eval_steps_per_second": 15.481, | |
| "step": 16794 | |
| }, | |
| { | |
| "epoch": 59.927739863508634, | |
| "grad_norm": 5.332968235015869, | |
| "learning_rate": 5.359056806002144e-09, | |
| "loss": 1.6888, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 59.927739863508634, | |
| "eval_accuracy": 0.27407227156588393, | |
| "eval_loss": 2.6999714374542236, | |
| "eval_runtime": 26.3708, | |
| "eval_samples_per_second": 311.67, | |
| "eval_steps_per_second": 15.585, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 59.927739863508634, | |
| "step": 18660, | |
| "total_flos": 9.793402242125722e+17, | |
| "train_loss": 1.7706359928699424, | |
| "train_runtime": 26526.4476, | |
| "train_samples_per_second": 140.814, | |
| "train_steps_per_second": 0.703 | |
| } | |
| ], | |
| "logging_steps": 1866, | |
| "max_steps": 18660, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 60, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.793402242125722e+17, | |
| "train_batch_size": 25, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |