| { | |
| "best_metric": 0.5105542540550232, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-1170", | |
| "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": 7.213888168334961, | |
| "learning_rate": 5e-05, | |
| "loss": 1.094, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.518796992481203, | |
| "eval_loss": 1.0090891122817993, | |
| "eval_runtime": 1.1391, | |
| "eval_samples_per_second": 116.756, | |
| "eval_steps_per_second": 14.924, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 2.6927850246429443, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6090225563909775, | |
| "eval_loss": 0.880857527256012, | |
| "eval_runtime": 1.158, | |
| "eval_samples_per_second": 114.856, | |
| "eval_steps_per_second": 14.681, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 6.594575881958008, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8928, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6390977443609023, | |
| "eval_loss": 0.8694148063659668, | |
| "eval_runtime": 1.1236, | |
| "eval_samples_per_second": 118.367, | |
| "eval_steps_per_second": 15.13, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 7.34794807434082, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7855, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 4.2354817390441895, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7179, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7142857142857143, | |
| "eval_loss": 0.6974284052848816, | |
| "eval_runtime": 1.1571, | |
| "eval_samples_per_second": 114.947, | |
| "eval_steps_per_second": 14.692, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 5.142866611480713, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6783, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7518796992481203, | |
| "eval_loss": 0.669804573059082, | |
| "eval_runtime": 1.0102, | |
| "eval_samples_per_second": 131.662, | |
| "eval_steps_per_second": 16.829, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 6.277930736541748, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7052, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7518796992481203, | |
| "eval_loss": 0.6211526989936829, | |
| "eval_runtime": 1.1627, | |
| "eval_samples_per_second": 114.39, | |
| "eval_steps_per_second": 14.621, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 7.9038262367248535, | |
| "learning_rate": 5e-05, | |
| "loss": 0.636, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 15.989542007446289, | |
| "learning_rate": 5e-05, | |
| "loss": 0.608, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.706766917293233, | |
| "eval_loss": 0.7636638283729553, | |
| "eval_runtime": 1.1935, | |
| "eval_samples_per_second": 111.438, | |
| "eval_steps_per_second": 14.244, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 5.844383716583252, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6457, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8195488721804511, | |
| "eval_loss": 0.5296058654785156, | |
| "eval_runtime": 1.1689, | |
| "eval_samples_per_second": 113.785, | |
| "eval_steps_per_second": 14.544, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 21.204557418823242, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5747, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7969924812030075, | |
| "eval_loss": 0.5105542540550232, | |
| "eval_runtime": 1.037, | |
| "eval_samples_per_second": 128.258, | |
| "eval_steps_per_second": 16.394, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 15.454485893249512, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5939, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 12.411216735839844, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5294, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6917293233082706, | |
| "eval_loss": 0.7639761567115784, | |
| "eval_runtime": 1.0303, | |
| "eval_samples_per_second": 129.085, | |
| "eval_steps_per_second": 16.5, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1300, | |
| "total_flos": 8.772706474360013e+17, | |
| "train_loss": 0.7277990575937124, | |
| "train_runtime": 205.3588, | |
| "train_samples_per_second": 50.351, | |
| "train_steps_per_second": 6.33 | |
| } | |
| ], | |
| "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 | |
| } | |