| { | |
| "best_metric": 0.17903044819831848, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-520", | |
| "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": 2.7811906337738037, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9748, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8947368421052632, | |
| "eval_loss": 0.7215397357940674, | |
| "eval_runtime": 0.9364, | |
| "eval_samples_per_second": 142.034, | |
| "eval_steps_per_second": 18.155, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.6656582355499268, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6849, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9473684210526315, | |
| "eval_loss": 0.4107130765914917, | |
| "eval_runtime": 0.9628, | |
| "eval_samples_per_second": 138.137, | |
| "eval_steps_per_second": 17.657, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 1.416643738746643, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4838, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9473684210526315, | |
| "eval_loss": 0.2423216700553894, | |
| "eval_runtime": 0.9821, | |
| "eval_samples_per_second": 135.426, | |
| "eval_steps_per_second": 17.31, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 2.6065406799316406, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3533, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 1.5872589349746704, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2594, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9624060150375939, | |
| "eval_loss": 0.17903044819831848, | |
| "eval_runtime": 0.9895, | |
| "eval_samples_per_second": 134.417, | |
| "eval_steps_per_second": 17.181, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 2.6757798194885254, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2082, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9323308270676691, | |
| "eval_loss": 0.19146375358104706, | |
| "eval_runtime": 1.0393, | |
| "eval_samples_per_second": 127.973, | |
| "eval_steps_per_second": 16.357, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 5.998991012573242, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2159, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9097744360902256, | |
| "eval_loss": 0.3304254412651062, | |
| "eval_runtime": 0.9409, | |
| "eval_samples_per_second": 141.349, | |
| "eval_steps_per_second": 18.067, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 4.108273029327393, | |
| "learning_rate": 5e-05, | |
| "loss": 0.339, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 5.80717134475708, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5954, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6917293233082706, | |
| "eval_loss": 0.6861125230789185, | |
| "eval_runtime": 1.0533, | |
| "eval_samples_per_second": 126.266, | |
| "eval_steps_per_second": 16.139, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 4.3353142738342285, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8045, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.48872180451127817, | |
| "eval_loss": 1.0407782793045044, | |
| "eval_runtime": 1.1746, | |
| "eval_samples_per_second": 113.234, | |
| "eval_steps_per_second": 14.474, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 1.6110053062438965, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0506, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.3308270676691729, | |
| "eval_loss": 1.1016151905059814, | |
| "eval_runtime": 1.2259, | |
| "eval_samples_per_second": 108.493, | |
| "eval_steps_per_second": 13.868, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 0.6255662441253662, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1038, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 6.890110969543457, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1299, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.3458646616541353, | |
| "eval_loss": 1.0934619903564453, | |
| "eval_runtime": 1.1913, | |
| "eval_samples_per_second": 111.642, | |
| "eval_steps_per_second": 14.27, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1300, | |
| "total_flos": 8.772706474360013e+17, | |
| "train_loss": 0.6310529987628644, | |
| "train_runtime": 186.1113, | |
| "train_samples_per_second": 55.558, | |
| "train_steps_per_second": 6.985 | |
| } | |
| ], | |
| "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 | |
| } | |