| { | |
| "best_metric": 0.8231664299964905, | |
| "best_model_checkpoint": "hBERTv1_data_aug_wnli/checkpoint-218", | |
| "epoch": 6.0, | |
| "global_step": 1308, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.900917431192661e-05, | |
| "loss": 0.6916, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.323943661971831, | |
| "eval_loss": 0.8231664299964905, | |
| "eval_runtime": 0.1005, | |
| "eval_samples_per_second": 706.229, | |
| "eval_steps_per_second": 9.947, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.800917431192661e-05, | |
| "loss": 0.5909, | |
| "step": 436 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.07042253521126761, | |
| "eval_loss": 2.906492233276367, | |
| "eval_runtime": 0.101, | |
| "eval_samples_per_second": 703.289, | |
| "eval_steps_per_second": 9.905, | |
| "step": 436 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7009174311926604e-05, | |
| "loss": 0.3754, | |
| "step": 654 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.08450704225352113, | |
| "eval_loss": 4.767050266265869, | |
| "eval_runtime": 0.1009, | |
| "eval_samples_per_second": 703.829, | |
| "eval_steps_per_second": 9.913, | |
| "step": 654 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600917431192661e-05, | |
| "loss": 0.2639, | |
| "step": 872 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.11267605633802817, | |
| "eval_loss": 5.692210674285889, | |
| "eval_runtime": 0.1019, | |
| "eval_samples_per_second": 696.536, | |
| "eval_steps_per_second": 9.81, | |
| "step": 872 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.50091743119266e-05, | |
| "loss": 0.1921, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.08450704225352113, | |
| "eval_loss": 5.994777202606201, | |
| "eval_runtime": 0.1009, | |
| "eval_samples_per_second": 703.865, | |
| "eval_steps_per_second": 9.914, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.400917431192661e-05, | |
| "loss": 0.1317, | |
| "step": 1308 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.09859154929577464, | |
| "eval_loss": 6.7443647384643555, | |
| "eval_runtime": 0.1002, | |
| "eval_samples_per_second": 708.652, | |
| "eval_steps_per_second": 9.981, | |
| "step": 1308 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 1308, | |
| "total_flos": 4.216767430577357e+16, | |
| "train_loss": 0.3742816163859236, | |
| "train_runtime": 1346.1035, | |
| "train_samples_per_second": 2066.706, | |
| "train_steps_per_second": 8.097 | |
| } | |
| ], | |
| "max_steps": 10900, | |
| "num_train_epochs": 50, | |
| "total_flos": 4.216767430577357e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |