| { |
| "best_metric": 0.19915683567523956, |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-clinc-oos-plus/run-0/checkpoint-1071", |
| "epoch": 8.925, |
| "eval_steps": 119, |
| "global_step": 1071, |
| "is_hyper_param_search": true, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.9916666666666667, |
| "grad_norm": 0.7661037445068359, |
| "learning_rate": 1.77962962962963e-05, |
| "loss": 0.6655, |
| "step": 119 |
| }, |
| { |
| "epoch": 0.9916666666666667, |
| "eval_accuracy": 0.05709677419354839, |
| "eval_loss": 0.5611550807952881, |
| "eval_runtime": 5.2298, |
| "eval_samples_per_second": 592.757, |
| "eval_steps_per_second": 4.78, |
| "step": 119 |
| }, |
| { |
| "epoch": 1.9833333333333334, |
| "grad_norm": 1.0290801525115967, |
| "learning_rate": 1.5592592592592593e-05, |
| "loss": 0.5116, |
| "step": 238 |
| }, |
| { |
| "epoch": 1.9833333333333334, |
| "eval_accuracy": 0.2, |
| "eval_loss": 0.42120361328125, |
| "eval_runtime": 5.3353, |
| "eval_samples_per_second": 581.039, |
| "eval_steps_per_second": 4.686, |
| "step": 238 |
| }, |
| { |
| "epoch": 2.975, |
| "grad_norm": 0.851043164730072, |
| "learning_rate": 1.338888888888889e-05, |
| "loss": 0.402, |
| "step": 357 |
| }, |
| { |
| "epoch": 2.975, |
| "eval_accuracy": 0.41838709677419356, |
| "eval_loss": 0.3284984827041626, |
| "eval_runtime": 5.283, |
| "eval_samples_per_second": 586.785, |
| "eval_steps_per_second": 4.732, |
| "step": 357 |
| }, |
| { |
| "epoch": 3.966666666666667, |
| "grad_norm": 0.7865561246871948, |
| "learning_rate": 1.1185185185185185e-05, |
| "loss": 0.3261, |
| "step": 476 |
| }, |
| { |
| "epoch": 3.966666666666667, |
| "eval_accuracy": 0.5587096774193548, |
| "eval_loss": 0.2734394669532776, |
| "eval_runtime": 5.255, |
| "eval_samples_per_second": 589.91, |
| "eval_steps_per_second": 4.757, |
| "step": 476 |
| }, |
| { |
| "epoch": 4.958333333333333, |
| "grad_norm": 0.7745070457458496, |
| "learning_rate": 8.981481481481483e-06, |
| "loss": 0.282, |
| "step": 595 |
| }, |
| { |
| "epoch": 4.958333333333333, |
| "eval_accuracy": 0.6261290322580645, |
| "eval_loss": 0.23937486112117767, |
| "eval_runtime": 5.2704, |
| "eval_samples_per_second": 588.194, |
| "eval_steps_per_second": 4.743, |
| "step": 595 |
| }, |
| { |
| "epoch": 5.95, |
| "grad_norm": 0.9656067490577698, |
| "learning_rate": 6.777777777777779e-06, |
| "loss": 0.2523, |
| "step": 714 |
| }, |
| { |
| "epoch": 5.95, |
| "eval_accuracy": 0.6803225806451613, |
| "eval_loss": 0.21992823481559753, |
| "eval_runtime": 5.241, |
| "eval_samples_per_second": 591.487, |
| "eval_steps_per_second": 4.77, |
| "step": 714 |
| }, |
| { |
| "epoch": 6.941666666666666, |
| "grad_norm": 0.6306410431861877, |
| "learning_rate": 4.5740740740740745e-06, |
| "loss": 0.235, |
| "step": 833 |
| }, |
| { |
| "epoch": 6.941666666666666, |
| "eval_accuracy": 0.6967741935483871, |
| "eval_loss": 0.20852065086364746, |
| "eval_runtime": 5.2518, |
| "eval_samples_per_second": 590.271, |
| "eval_steps_per_second": 4.76, |
| "step": 833 |
| }, |
| { |
| "epoch": 7.933333333333334, |
| "grad_norm": 0.6096643805503845, |
| "learning_rate": 2.3703703703703707e-06, |
| "loss": 0.2228, |
| "step": 952 |
| }, |
| { |
| "epoch": 7.933333333333334, |
| "eval_accuracy": 0.7103225806451613, |
| "eval_loss": 0.20159533619880676, |
| "eval_runtime": 5.2662, |
| "eval_samples_per_second": 588.662, |
| "eval_steps_per_second": 4.747, |
| "step": 952 |
| }, |
| { |
| "epoch": 8.925, |
| "grad_norm": 0.5338801145553589, |
| "learning_rate": 1.6666666666666668e-07, |
| "loss": 0.2174, |
| "step": 1071 |
| }, |
| { |
| "epoch": 8.925, |
| "eval_accuracy": 0.714516129032258, |
| "eval_loss": 0.19915683567523956, |
| "eval_runtime": 5.2762, |
| "eval_samples_per_second": 587.546, |
| "eval_steps_per_second": 4.738, |
| "step": 1071 |
| } |
| ], |
| "logging_steps": 119, |
| "max_steps": 1080, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9, |
| "save_steps": 119, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 816805015151832.0, |
| "train_batch_size": 128, |
| "trial_name": null, |
| "trial_params": { |
| "alpha": 0.7937035517473829, |
| "num_train_epochs": 9, |
| "temperature": 13 |
| } |
| } |
|
|
