| { | |
| "best_metric": 0.6930704116821289, | |
| "best_model_checkpoint": "hBERTv2_data_aug_qnli/checkpoint-16604", | |
| "epoch": 6.0, | |
| "global_step": 99624, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.900036135870875e-05, | |
| "loss": 0.6934, | |
| "step": 16604 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6930704116821289, | |
| "eval_runtime": 6.4224, | |
| "eval_samples_per_second": 850.613, | |
| "eval_steps_per_second": 3.425, | |
| "step": 16604 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.800078294386895e-05, | |
| "loss": 0.6932, | |
| "step": 33208 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6931126117706299, | |
| "eval_runtime": 6.4301, | |
| "eval_samples_per_second": 849.596, | |
| "eval_steps_per_second": 3.421, | |
| "step": 33208 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.700114430257769e-05, | |
| "loss": 0.6932, | |
| "step": 49812 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931178569793701, | |
| "eval_runtime": 6.4408, | |
| "eval_samples_per_second": 848.182, | |
| "eval_steps_per_second": 3.416, | |
| "step": 49812 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.6001565887737896e-05, | |
| "loss": 0.6932, | |
| "step": 66416 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6931198239326477, | |
| "eval_runtime": 6.3591, | |
| "eval_samples_per_second": 859.08, | |
| "eval_steps_per_second": 3.46, | |
| "step": 66416 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.500204769934956e-05, | |
| "loss": 0.6932, | |
| "step": 83020 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.4946000366099213, | |
| "eval_loss": 0.6931706070899963, | |
| "eval_runtime": 6.3657, | |
| "eval_samples_per_second": 858.199, | |
| "eval_steps_per_second": 3.456, | |
| "step": 83020 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.400252951096122e-05, | |
| "loss": 0.6932, | |
| "step": 99624 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5053999633900788, | |
| "eval_loss": 0.6931073069572449, | |
| "eval_runtime": 6.4132, | |
| "eval_samples_per_second": 851.84, | |
| "eval_steps_per_second": 3.43, | |
| "step": 99624 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 99624, | |
| "total_flos": 3.1764944527640494e+18, | |
| "train_loss": 0.6932406767414228, | |
| "train_runtime": 86584.2008, | |
| "train_samples_per_second": 2454.553, | |
| "train_steps_per_second": 9.588 | |
| } | |
| ], | |
| "max_steps": 830200, | |
| "num_train_epochs": 50, | |
| "total_flos": 3.1764944527640494e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |