| { | |
| "best_metric": 0.8339622641509433, | |
| "best_model_checkpoint": "final_models/structroberta_sx2_final/finetune/sst2/checkpoint-2000", | |
| "epoch": 9.47867298578199, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.7992125749588013, | |
| "eval_f1": 0.7976190476190474, | |
| "eval_loss": 0.42879804968833923, | |
| "eval_mcc": 0.598425075943802, | |
| "eval_runtime": 1.0717, | |
| "eval_samples_per_second": 474.004, | |
| "eval_steps_per_second": 59.717, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 4.407582938388626e-05, | |
| "loss": 0.3876, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.789370059967041, | |
| "eval_f1": 0.7897838899803538, | |
| "eval_loss": 0.4634992778301239, | |
| "eval_mcc": 0.5787935859950727, | |
| "eval_runtime": 1.069, | |
| "eval_samples_per_second": 475.219, | |
| "eval_steps_per_second": 59.87, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 3.815165876777251e-05, | |
| "loss": 0.2086, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_accuracy": 0.8149606585502625, | |
| "eval_f1": 0.8, | |
| "eval_loss": 0.5669229626655579, | |
| "eval_mcc": 0.6361381051113675, | |
| "eval_runtime": 1.0678, | |
| "eval_samples_per_second": 475.734, | |
| "eval_steps_per_second": 59.935, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 3.222748815165877e-05, | |
| "loss": 0.1375, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "eval_accuracy": 0.8090550899505615, | |
| "eval_f1": 0.8079207920792079, | |
| "eval_loss": 0.594048023223877, | |
| "eval_mcc": 0.6181031878719743, | |
| "eval_runtime": 1.0784, | |
| "eval_samples_per_second": 471.09, | |
| "eval_steps_per_second": 59.35, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 2.6303317535545023e-05, | |
| "loss": 0.091, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "eval_accuracy": 0.8267716765403748, | |
| "eval_f1": 0.8339622641509433, | |
| "eval_loss": 0.6517126560211182, | |
| "eval_mcc": 0.6566023271980082, | |
| "eval_runtime": 1.0643, | |
| "eval_samples_per_second": 477.324, | |
| "eval_steps_per_second": 60.135, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "eval_accuracy": 0.8188976645469666, | |
| "eval_f1": 0.8167330677290837, | |
| "eval_loss": 0.7473486661911011, | |
| "eval_mcc": 0.6378463127603581, | |
| "eval_runtime": 1.0657, | |
| "eval_samples_per_second": 476.704, | |
| "eval_steps_per_second": 60.057, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 2.037914691943128e-05, | |
| "loss": 0.0604, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "eval_accuracy": 0.8031495809555054, | |
| "eval_f1": 0.8062015503875969, | |
| "eval_loss": 0.8508022427558899, | |
| "eval_mcc": 0.6068244673039802, | |
| "eval_runtime": 1.069, | |
| "eval_samples_per_second": 475.198, | |
| "eval_steps_per_second": 59.867, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 1.4454976303317535e-05, | |
| "loss": 0.0387, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "eval_accuracy": 0.8149606585502625, | |
| "eval_f1": 0.8081632653061225, | |
| "eval_loss": 0.9328407049179077, | |
| "eval_mcc": 0.6310776847569037, | |
| "eval_runtime": 1.0678, | |
| "eval_samples_per_second": 475.739, | |
| "eval_steps_per_second": 59.936, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 8.530805687203793e-06, | |
| "loss": 0.0251, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "eval_accuracy": 0.8169291615486145, | |
| "eval_f1": 0.8194174757281554, | |
| "eval_loss": 1.0624468326568604, | |
| "eval_mcc": 0.6343018738159408, | |
| "eval_runtime": 1.0698, | |
| "eval_samples_per_second": 474.877, | |
| "eval_steps_per_second": 59.827, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "learning_rate": 2.6066350710900472e-06, | |
| "loss": 0.0161, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "eval_accuracy": 0.8188976645469666, | |
| "eval_f1": 0.8174603174603174, | |
| "eval_loss": 1.1392195224761963, | |
| "eval_mcc": 0.6377982061833372, | |
| "eval_runtime": 1.0675, | |
| "eval_samples_per_second": 475.89, | |
| "eval_steps_per_second": 59.955, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "step": 4000, | |
| "total_flos": 4.387755827574374e+16, | |
| "train_loss": 0.12062864470481872, | |
| "train_runtime": 2195.6071, | |
| "train_samples_per_second": 230.132, | |
| "train_steps_per_second": 1.922 | |
| } | |
| ], | |
| "max_steps": 4220, | |
| "num_train_epochs": 10, | |
| "total_flos": 4.387755827574374e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |