| { | |
| "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": 7.660736560821533, | |
| "learning_rate": 9.001071811361202e-06, | |
| "loss": 2.077, | |
| "step": 1866 | |
| }, | |
| { | |
| "epoch": 5.992773986350863, | |
| "eval_accuracy": 0.07665166078598369, | |
| "eval_loss": 3.0592968463897705, | |
| "eval_runtime": 32.9334, | |
| "eval_samples_per_second": 249.564, | |
| "eval_steps_per_second": 12.48, | |
| "step": 1866 | |
| }, | |
| { | |
| "epoch": 11.985547972701726, | |
| "grad_norm": 8.02175521850586, | |
| "learning_rate": 8.0021436227224e-06, | |
| "loss": 1.8747, | |
| "step": 3732 | |
| }, | |
| { | |
| "epoch": 11.985547972701726, | |
| "eval_accuracy": 0.07878087358559435, | |
| "eval_loss": 3.1968600749969482, | |
| "eval_runtime": 33.6195, | |
| "eval_samples_per_second": 244.471, | |
| "eval_steps_per_second": 12.225, | |
| "step": 3732 | |
| }, | |
| { | |
| "epoch": 17.97832195905259, | |
| "grad_norm": 7.991235733032227, | |
| "learning_rate": 7.002143622722402e-06, | |
| "loss": 1.7613, | |
| "step": 5598 | |
| }, | |
| { | |
| "epoch": 17.97832195905259, | |
| "eval_accuracy": 0.07819280528855903, | |
| "eval_loss": 3.2275424003601074, | |
| "eval_runtime": 33.0021, | |
| "eval_samples_per_second": 249.045, | |
| "eval_steps_per_second": 12.454, | |
| "step": 5598 | |
| }, | |
| { | |
| "epoch": 23.97109594540345, | |
| "grad_norm": 4.696638584136963, | |
| "learning_rate": 6.002679528403002e-06, | |
| "loss": 1.703, | |
| "step": 7464 | |
| }, | |
| { | |
| "epoch": 23.97109594540345, | |
| "eval_accuracy": 0.07881129091130308, | |
| "eval_loss": 3.367746114730835, | |
| "eval_runtime": 33.0083, | |
| "eval_samples_per_second": 248.998, | |
| "eval_steps_per_second": 12.451, | |
| "step": 7464 | |
| }, | |
| { | |
| "epoch": 29.963869931754317, | |
| "grad_norm": 2.74943208694458, | |
| "learning_rate": 5.0037513397642025e-06, | |
| "loss": 1.676, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 29.963869931754317, | |
| "eval_accuracy": 0.07835503102567222, | |
| "eval_loss": 3.4368343353271484, | |
| "eval_runtime": 32.8486, | |
| "eval_samples_per_second": 250.208, | |
| "eval_steps_per_second": 12.512, | |
| "step": 9330 | |
| }, | |
| { | |
| "epoch": 35.95664391810518, | |
| "grad_norm": 4.24375057220459, | |
| "learning_rate": 4.003751339764202e-06, | |
| "loss": 1.6495, | |
| "step": 11196 | |
| }, | |
| { | |
| "epoch": 35.95664391810518, | |
| "eval_accuracy": 0.07829419637425478, | |
| "eval_loss": 3.5519556999206543, | |
| "eval_runtime": 32.9918, | |
| "eval_samples_per_second": 249.123, | |
| "eval_steps_per_second": 12.458, | |
| "step": 11196 | |
| }, | |
| { | |
| "epoch": 41.94941790445604, | |
| "grad_norm": 3.4726574420928955, | |
| "learning_rate": 3.004287245444802e-06, | |
| "loss": 1.6449, | |
| "step": 13062 | |
| }, | |
| { | |
| "epoch": 41.94941790445604, | |
| "eval_accuracy": 0.07814645507795526, | |
| "eval_loss": 3.556215763092041, | |
| "eval_runtime": 32.6442, | |
| "eval_samples_per_second": 251.776, | |
| "eval_steps_per_second": 12.59, | |
| "step": 13062 | |
| }, | |
| { | |
| "epoch": 47.9421918908069, | |
| "grad_norm": 3.2187790870666504, | |
| "learning_rate": 2.004287245444802e-06, | |
| "loss": 1.6293, | |
| "step": 14928 | |
| }, | |
| { | |
| "epoch": 47.9421918908069, | |
| "eval_accuracy": 0.07750334590582796, | |
| "eval_loss": 3.621753215789795, | |
| "eval_runtime": 32.8748, | |
| "eval_samples_per_second": 250.009, | |
| "eval_steps_per_second": 12.502, | |
| "step": 14928 | |
| }, | |
| { | |
| "epoch": 53.934965877157765, | |
| "grad_norm": 1.5753388404846191, | |
| "learning_rate": 1.0048231511254019e-06, | |
| "loss": 1.6301, | |
| "step": 16794 | |
| }, | |
| { | |
| "epoch": 53.934965877157765, | |
| "eval_accuracy": 0.07704370631734057, | |
| "eval_loss": 3.7435097694396973, | |
| "eval_runtime": 32.6254, | |
| "eval_samples_per_second": 251.92, | |
| "eval_steps_per_second": 12.598, | |
| "step": 16794 | |
| }, | |
| { | |
| "epoch": 59.927739863508634, | |
| "grad_norm": 1.8606178760528564, | |
| "learning_rate": 5.359056806002144e-09, | |
| "loss": 1.6232, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 59.927739863508634, | |
| "eval_accuracy": 0.07645698990144786, | |
| "eval_loss": 3.775934934616089, | |
| "eval_runtime": 32.9764, | |
| "eval_samples_per_second": 249.239, | |
| "eval_steps_per_second": 12.463, | |
| "step": 18660 | |
| }, | |
| { | |
| "epoch": 59.927739863508634, | |
| "step": 18660, | |
| "total_flos": 1.4754079994623488e+18, | |
| "train_loss": 1.726910066757938, | |
| "train_runtime": 33547.8653, | |
| "train_samples_per_second": 111.342, | |
| "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 | |
| } | |