| { | |
| "best_metric": 0.9381032586097717, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-1300", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 1300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.513921022415161, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 1.1019, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3007518796992481, | |
| "eval_loss": 1.0955818891525269, | |
| "eval_runtime": 0.9308, | |
| "eval_samples_per_second": 142.882, | |
| "eval_steps_per_second": 18.263, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.6020777225494385, | |
| "learning_rate": 4.230769230769231e-05, | |
| "loss": 1.0934, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5413533834586466, | |
| "eval_loss": 1.0218424797058105, | |
| "eval_runtime": 0.9394, | |
| "eval_samples_per_second": 141.582, | |
| "eval_steps_per_second": 18.097, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 1.3475981950759888, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 1.0541, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5037593984962406, | |
| "eval_loss": 0.9863384366035461, | |
| "eval_runtime": 0.961, | |
| "eval_samples_per_second": 138.393, | |
| "eval_steps_per_second": 17.689, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 1.3711971044540405, | |
| "learning_rate": 3.461538461538462e-05, | |
| "loss": 0.9991, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 0.6074744462966919, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.9911, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5338345864661654, | |
| "eval_loss": 0.9627529382705688, | |
| "eval_runtime": 0.9661, | |
| "eval_samples_per_second": 137.661, | |
| "eval_steps_per_second": 17.596, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 0.4636833667755127, | |
| "learning_rate": 2.6923076923076923e-05, | |
| "loss": 0.9776, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5413533834586466, | |
| "eval_loss": 0.9637588858604431, | |
| "eval_runtime": 0.96, | |
| "eval_samples_per_second": 138.544, | |
| "eval_steps_per_second": 17.709, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 1.6218713521957397, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.9852, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5413533834586466, | |
| "eval_loss": 0.9598925113677979, | |
| "eval_runtime": 0.956, | |
| "eval_samples_per_second": 139.127, | |
| "eval_steps_per_second": 17.783, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 1.7121293544769287, | |
| "learning_rate": 1.923076923076923e-05, | |
| "loss": 0.9753, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 1.5765596628189087, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.9666, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5639097744360902, | |
| "eval_loss": 0.9482960104942322, | |
| "eval_runtime": 0.9859, | |
| "eval_samples_per_second": 134.905, | |
| "eval_steps_per_second": 17.244, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 1.4977715015411377, | |
| "learning_rate": 1.153846153846154e-05, | |
| "loss": 0.9528, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5413533834586466, | |
| "eval_loss": 0.9446954727172852, | |
| "eval_runtime": 0.9431, | |
| "eval_samples_per_second": 141.029, | |
| "eval_steps_per_second": 18.026, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 1.6241708993911743, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.9583, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.5338345864661654, | |
| "eval_loss": 0.9394896030426025, | |
| "eval_runtime": 0.9745, | |
| "eval_samples_per_second": 136.479, | |
| "eval_steps_per_second": 17.445, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 0.5392513871192932, | |
| "learning_rate": 3.846153846153847e-06, | |
| "loss": 0.9421, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.028888702392578, | |
| "learning_rate": 0.0, | |
| "loss": 0.9401, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5413533834586466, | |
| "eval_loss": 0.9381032586097717, | |
| "eval_runtime": 0.9652, | |
| "eval_samples_per_second": 137.798, | |
| "eval_steps_per_second": 17.613, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1300, | |
| "total_flos": 8.772706474360013e+17, | |
| "train_loss": 0.9952024547870343, | |
| "train_runtime": 174.8367, | |
| "train_samples_per_second": 59.141, | |
| "train_steps_per_second": 7.436 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 8.772706474360013e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |