| { | |
| "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": 4.094143390655518, | |
| "learning_rate": 9.001071811361202e-06, | |
| "loss": 2.0921, | |
| "step": 1866 | |
| }, | |
| { | |
| "epoch": 5.992773986350863, | |
| "eval_accuracy": 0.07738167660299307, | |
| "eval_loss": 3.0240695476531982, | |
| "eval_runtime": 33.2741, | |
| "eval_samples_per_second": 247.009, | |
| "eval_steps_per_second": 12.352, | |
| "step": 1866 | |
| }, | |
| { | |
| "epoch": 11.985547972701726, | |
| "grad_norm": 7.4762701988220215, | |
| "learning_rate": 8.001607717041802e-06, | |
| "loss": 1.9094, | |
| "step": 3732 | |
| }, | |
| { | |
| "epoch": 11.985547972701726, | |
| "eval_accuracy": 0.08157926755079693, | |
| "eval_loss": 3.0886166095733643, | |
| "eval_runtime": 33.0139, | |
| "eval_samples_per_second": 248.955, | |
| "eval_steps_per_second": 12.449, | |
| "step": 3732 | |
| }, | |
| { | |
| "epoch": 17.97832195905259, | |
| "grad_norm": 5.10331916809082, | |
| "learning_rate": 7.002143622722402e-06, | |
| "loss": 1.7854, | |
| "step": 5598 | |
| }, | |
| { | |
| "epoch": 17.97832195905259, | |
| "eval_accuracy": 0.08026118343675225, | |
| "eval_loss": 3.2710988521575928, | |
| "eval_runtime": 32.8416, | |
| "eval_samples_per_second": 250.262, | |
| "eval_steps_per_second": 12.515, | |
| "step": 5598 | |
| }, | |
| { | |
| "epoch": 23.97109594540345, | |
| "grad_norm": 4.933435440063477, | |
| "learning_rate": 6.002679528403002e-06, | |
| "loss": 1.7194, | |
| "step": 7464 | |
| }, | |
| { | |
| "epoch": 23.97109594540345, | |
| "eval_accuracy": 0.07872003893417691, | |
| "eval_loss": 3.4031662940979004, | |
| "eval_runtime": 32.7142, | |
| "eval_samples_per_second": 251.236, | |
| "eval_steps_per_second": 12.563, | |
| "step": 7464 | |
| }, | |
| { | |
| "epoch": 29.963869931754317, | |
| "grad_norm": 2.231235980987549, | |
| "learning_rate": 5.003215434083602e-06, | |
| "loss": 1.6865, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 29.963869931754317, | |
| "eval_accuracy": 0.07784401995376566, | |
| "eval_loss": 3.391934871673584, | |
| "eval_runtime": 32.8923, | |
| "eval_samples_per_second": 249.876, | |
| "eval_steps_per_second": 12.495, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 35.95664391810518, | |
| "grad_norm": 3.9303555488586426, | |
| "learning_rate": 4.003751339764202e-06, | |
| "loss": 1.658, | |
| "step": 11196 | |
| }, | |
| { | |
| "epoch": 35.95664391810518, | |
| "eval_accuracy": 0.07665166078598369, | |
| "eval_loss": 3.444946527481079, | |
| "eval_runtime": 32.9777, | |
| "eval_samples_per_second": 249.229, | |
| "eval_steps_per_second": 12.463, | |
| "step": 11196 | |
| }, | |
| { | |
| "epoch": 41.94941790445604, | |
| "grad_norm": 5.232691764831543, | |
| "learning_rate": 3.004287245444802e-06, | |
| "loss": 1.6521, | |
| "step": 13062 | |
| }, | |
| { | |
| "epoch": 41.94941790445604, | |
| "eval_accuracy": 0.07527853579684703, | |
| "eval_loss": 3.606778383255005, | |
| "eval_runtime": 33.1429, | |
| "eval_samples_per_second": 247.987, | |
| "eval_steps_per_second": 12.401, | |
| "step": 13062 | |
| }, | |
| { | |
| "epoch": 47.9421918908069, | |
| "grad_norm": 1.497389793395996, | |
| "learning_rate": 2.004823151125402e-06, | |
| "loss": 1.6356, | |
| "step": 14928 | |
| }, | |
| { | |
| "epoch": 47.9421918908069, | |
| "eval_accuracy": 0.07429431804355761, | |
| "eval_loss": 3.637636184692383, | |
| "eval_runtime": 33.1865, | |
| "eval_samples_per_second": 247.661, | |
| "eval_steps_per_second": 12.385, | |
| "step": 14928 | |
| }, | |
| { | |
| "epoch": 53.934965877157765, | |
| "grad_norm": 1.6502782106399536, | |
| "learning_rate": 1.0048231511254019e-06, | |
| "loss": 1.6355, | |
| "step": 16794 | |
| }, | |
| { | |
| "epoch": 53.934965877157765, | |
| "eval_accuracy": 0.07394789850076382, | |
| "eval_loss": 3.7658591270446777, | |
| "eval_runtime": 32.8283, | |
| "eval_samples_per_second": 250.363, | |
| "eval_steps_per_second": 12.52, | |
| "step": 16794 | |
| }, | |
| { | |
| "epoch": 59.927739863508634, | |
| "grad_norm": 1.5192711353302002, | |
| "learning_rate": 5.359056806002144e-09, | |
| "loss": 1.6276, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 59.927739863508634, | |
| "eval_accuracy": 0.07340309040029201, | |
| "eval_loss": 3.821331739425659, | |
| "eval_runtime": 32.9669, | |
| "eval_samples_per_second": 249.311, | |
| "eval_steps_per_second": 12.467, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 59.927739863508634, | |
| "step": 18660, | |
| "total_flos": 1.4754079994623488e+18, | |
| "train_loss": 1.7401779452831927, | |
| "train_runtime": 33551.227, | |
| "train_samples_per_second": 111.331, | |
| "train_steps_per_second": 0.556 | |
| } | |
| ], | |
| "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 | |
| } | |