| { | |
| "best_metric": 0.6127391457557678, | |
| "best_model_checkpoint": "mobilebert_add_GLUE_Experiment_cola/checkpoint-335", | |
| "epoch": 10.0, | |
| "global_step": 670, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6126, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6183119416236877, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.2731, | |
| "eval_samples_per_second": 458.836, | |
| "eval_steps_per_second": 3.959, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6078, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.6178781390190125, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.2536, | |
| "eval_samples_per_second": 462.812, | |
| "eval_steps_per_second": 3.994, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.6072, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.6183475255966187, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.2545, | |
| "eval_samples_per_second": 462.63, | |
| "eval_steps_per_second": 3.992, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.6062, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.6163871884346008, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.2538, | |
| "eval_samples_per_second": 462.764, | |
| "eval_steps_per_second": 3.993, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.601, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.6127391457557678, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.2614, | |
| "eval_samples_per_second": 461.209, | |
| "eval_steps_per_second": 3.98, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.5928, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.6148296594619751, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.256, | |
| "eval_samples_per_second": 462.326, | |
| "eval_steps_per_second": 3.989, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.588, | |
| "step": 469 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.622357964515686, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 2.2548, | |
| "eval_samples_per_second": 462.562, | |
| "eval_steps_per_second": 3.991, | |
| "step": 469 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.582, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.6174187064170837, | |
| "eval_matthews_correlation": 0.00286100001416597, | |
| "eval_runtime": 2.2525, | |
| "eval_samples_per_second": 463.04, | |
| "eval_steps_per_second": 3.996, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.5807, | |
| "step": 603 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.6301043033599854, | |
| "eval_matthews_correlation": 0.00286100001416597, | |
| "eval_runtime": 2.2518, | |
| "eval_samples_per_second": 463.189, | |
| "eval_steps_per_second": 3.997, | |
| "step": 603 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.5743, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.6156460642814636, | |
| "eval_matthews_correlation": 0.0437601222642778, | |
| "eval_runtime": 2.2578, | |
| "eval_samples_per_second": 461.958, | |
| "eval_steps_per_second": 3.986, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 670, | |
| "total_flos": 2529416970240000.0, | |
| "train_loss": 0.5952504741611765, | |
| "train_runtime": 701.2614, | |
| "train_samples_per_second": 609.687, | |
| "train_steps_per_second": 4.777 | |
| } | |
| ], | |
| "max_steps": 3350, | |
| "num_train_epochs": 50, | |
| "total_flos": 2529416970240000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |