| { | |
| "best_metric": 0.6895588636398315, | |
| "best_model_checkpoint": "hBERTv1_new_pretrain_rte/checkpoint-140", | |
| "epoch": 12.0, | |
| "global_step": 240, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.9200000000000004e-05, | |
| "loss": 0.7407, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.7002341151237488, | |
| "eval_runtime": 0.5049, | |
| "eval_samples_per_second": 548.575, | |
| "eval_steps_per_second": 5.941, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.8400000000000005e-05, | |
| "loss": 0.7061, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.7245458960533142, | |
| "eval_runtime": 0.508, | |
| "eval_samples_per_second": 545.223, | |
| "eval_steps_per_second": 5.905, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.76e-05, | |
| "loss": 0.7102, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6948612332344055, | |
| "eval_runtime": 0.504, | |
| "eval_samples_per_second": 549.561, | |
| "eval_steps_per_second": 5.952, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.680000000000001e-05, | |
| "loss": 0.703, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.6950586438179016, | |
| "eval_runtime": 0.5087, | |
| "eval_samples_per_second": 544.51, | |
| "eval_steps_per_second": 5.897, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.7097, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.6973713636398315, | |
| "eval_runtime": 0.5068, | |
| "eval_samples_per_second": 546.522, | |
| "eval_steps_per_second": 5.919, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.52e-05, | |
| "loss": 0.7006, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.7052543759346008, | |
| "eval_runtime": 0.5068, | |
| "eval_samples_per_second": 546.614, | |
| "eval_steps_per_second": 5.92, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.44e-05, | |
| "loss": 0.6986, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5306859205776173, | |
| "eval_loss": 0.6895588636398315, | |
| "eval_runtime": 0.5057, | |
| "eval_samples_per_second": 547.783, | |
| "eval_steps_per_second": 5.933, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 3.3600000000000004e-05, | |
| "loss": 0.6935, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.7711036205291748, | |
| "eval_runtime": 0.5097, | |
| "eval_samples_per_second": 543.49, | |
| "eval_steps_per_second": 5.886, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 3.28e-05, | |
| "loss": 0.6109, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.4981949458483754, | |
| "eval_loss": 0.8443405032157898, | |
| "eval_runtime": 0.5053, | |
| "eval_samples_per_second": 548.178, | |
| "eval_steps_per_second": 5.937, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.469, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5126353790613718, | |
| "eval_loss": 1.0369467735290527, | |
| "eval_runtime": 0.5078, | |
| "eval_samples_per_second": 545.529, | |
| "eval_steps_per_second": 5.908, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.1200000000000006e-05, | |
| "loss": 0.3028, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.5234657039711191, | |
| "eval_loss": 1.1621102094650269, | |
| "eval_runtime": 0.512, | |
| "eval_samples_per_second": 541.059, | |
| "eval_steps_per_second": 5.86, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 3.0400000000000004e-05, | |
| "loss": 0.2155, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5379061371841155, | |
| "eval_loss": 1.2095954418182373, | |
| "eval_runtime": 0.5057, | |
| "eval_samples_per_second": 547.72, | |
| "eval_steps_per_second": 5.932, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 240, | |
| "total_flos": 4425131514396672.0, | |
| "train_loss": 0.6050535062948863, | |
| "train_runtime": 211.7637, | |
| "train_samples_per_second": 587.919, | |
| "train_steps_per_second": 4.722 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 50, | |
| "total_flos": 4425131514396672.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |