| { | |
| "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": 4.376054763793945, | |
| "learning_rate": 9.001071811361202e-06, | |
| "loss": 2.0864, | |
| "step": 1866 | |
| }, | |
| { | |
| "epoch": 5.992773986350863, | |
| "eval_accuracy": 0.08030173987103054, | |
| "eval_loss": 2.9935145378112793, | |
| "eval_runtime": 33.1671, | |
| "eval_samples_per_second": 247.806, | |
| "eval_steps_per_second": 12.392, | |
| "step": 1866 | |
| }, | |
| { | |
| "epoch": 11.985547972701726, | |
| "grad_norm": 8.190367698669434, | |
| "learning_rate": 8.001607717041802e-06, | |
| "loss": 1.9035, | |
| "step": 3732 | |
| }, | |
| { | |
| "epoch": 11.985547972701726, | |
| "eval_accuracy": 0.08632437036135783, | |
| "eval_loss": 3.162896156311035, | |
| "eval_runtime": 33.0213, | |
| "eval_samples_per_second": 248.9, | |
| "eval_steps_per_second": 12.447, | |
| "step": 3732 | |
| }, | |
| { | |
| "epoch": 17.97832195905259, | |
| "grad_norm": 6.543500900268555, | |
| "learning_rate": 7.002143622722402e-06, | |
| "loss": 1.779, | |
| "step": 5598 | |
| }, | |
| { | |
| "epoch": 17.97832195905259, | |
| "eval_accuracy": 0.08695299509267146, | |
| "eval_loss": 3.2063722610473633, | |
| "eval_runtime": 32.9425, | |
| "eval_samples_per_second": 249.495, | |
| "eval_steps_per_second": 12.476, | |
| "step": 5598 | |
| }, | |
| { | |
| "epoch": 23.97109594540345, | |
| "grad_norm": 3.8219590187072754, | |
| "learning_rate": 6.002679528403002e-06, | |
| "loss": 1.7158, | |
| "step": 7464 | |
| }, | |
| { | |
| "epoch": 23.97109594540345, | |
| "eval_accuracy": 0.08541185059009612, | |
| "eval_loss": 3.44173526763916, | |
| "eval_runtime": 32.9094, | |
| "eval_samples_per_second": 249.746, | |
| "eval_steps_per_second": 12.489, | |
| "step": 7464 | |
| }, | |
| { | |
| "epoch": 29.963869931754317, | |
| "grad_norm": 1.9236102104187012, | |
| "learning_rate": 5.003215434083602e-06, | |
| "loss": 1.6832, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 29.963869931754317, | |
| "eval_accuracy": 0.08446283002798394, | |
| "eval_loss": 3.4987587928771973, | |
| "eval_runtime": 33.086, | |
| "eval_samples_per_second": 248.414, | |
| "eval_steps_per_second": 12.422, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 35.95664391810518, | |
| "grad_norm": 4.399581432342529, | |
| "learning_rate": 4.003751339764202e-06, | |
| "loss": 1.6554, | |
| "step": 11196 | |
| }, | |
| { | |
| "epoch": 35.95664391810518, | |
| "eval_accuracy": 0.0833434724419029, | |
| "eval_loss": 3.5538032054901123, | |
| "eval_runtime": 32.9857, | |
| "eval_samples_per_second": 249.168, | |
| "eval_steps_per_second": 12.46, | |
| "step": 11196 | |
| }, | |
| { | |
| "epoch": 41.94941790445604, | |
| "grad_norm": 4.802687168121338, | |
| "learning_rate": 3.004287245444802e-06, | |
| "loss": 1.6498, | |
| "step": 13062 | |
| }, | |
| { | |
| "epoch": 41.94941790445604, | |
| "eval_accuracy": 0.081918203465837, | |
| "eval_loss": 3.681871175765991, | |
| "eval_runtime": 33.1684, | |
| "eval_samples_per_second": 247.796, | |
| "eval_steps_per_second": 12.391, | |
| "step": 13062 | |
| }, | |
| { | |
| "epoch": 47.9421918908069, | |
| "grad_norm": 2.0320215225219727, | |
| "learning_rate": 2.004287245444802e-06, | |
| "loss": 1.6335, | |
| "step": 14928 | |
| }, | |
| { | |
| "epoch": 47.9421918908069, | |
| "eval_accuracy": 0.08087966905949628, | |
| "eval_loss": 3.7695610523223877, | |
| "eval_runtime": 32.9816, | |
| "eval_samples_per_second": 249.2, | |
| "eval_steps_per_second": 12.461, | |
| "step": 14928 | |
| }, | |
| { | |
| "epoch": 53.934965877157765, | |
| "grad_norm": 1.8332667350769043, | |
| "learning_rate": 1.0042872454448018e-06, | |
| "loss": 1.6339, | |
| "step": 16794 | |
| }, | |
| { | |
| "epoch": 53.934965877157765, | |
| "eval_accuracy": 0.07985561909396925, | |
| "eval_loss": 3.8098247051239014, | |
| "eval_runtime": 33.0186, | |
| "eval_samples_per_second": 248.92, | |
| "eval_steps_per_second": 12.448, | |
| "step": 16794 | |
| }, | |
| { | |
| "epoch": 59.927739863508634, | |
| "grad_norm": 2.1056318283081055, | |
| "learning_rate": 5.359056806002144e-09, | |
| "loss": 1.6264, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 59.927739863508634, | |
| "eval_accuracy": 0.07892687674899623, | |
| "eval_loss": 3.8913698196411133, | |
| "eval_runtime": 32.9542, | |
| "eval_samples_per_second": 249.407, | |
| "eval_steps_per_second": 12.472, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 59.927739863508634, | |
| "step": 18660, | |
| "total_flos": 1.4754079994623488e+18, | |
| "train_loss": 1.7366778243904073, | |
| "train_runtime": 33663.6695, | |
| "train_samples_per_second": 110.959, | |
| "train_steps_per_second": 0.554 | |
| } | |
| ], | |
| "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": 1.4754079994623488e+18, | |
| "train_batch_size": 25, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |