| { | |
| "best_metric": 0.8514851485148515, | |
| "best_model_checkpoint": "finetune_results/omarmomen/structformer_s1_final_with_pos/sst2/checkpoint-4000", | |
| "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": 0.7972440719604492, | |
| "eval_f1": 0.7919191919191918, | |
| "eval_loss": 0.44957396388053894, | |
| "eval_mcc": 0.5949711545663119, | |
| "eval_runtime": 0.9864, | |
| "eval_samples_per_second": 515.006, | |
| "eval_steps_per_second": 64.883, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 4.407582938388626e-05, | |
| "loss": 0.4711, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.8110235929489136, | |
| "eval_f1": 0.8195488721804512, | |
| "eval_loss": 0.4547339081764221, | |
| "eval_mcc": 0.6254766003729282, | |
| "eval_runtime": 0.9908, | |
| "eval_samples_per_second": 512.701, | |
| "eval_steps_per_second": 64.592, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 3.815165876777251e-05, | |
| "loss": 0.2815, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_accuracy": 0.8149606585502625, | |
| "eval_f1": 0.798283261802575, | |
| "eval_loss": 0.5317888259887695, | |
| "eval_mcc": 0.637652660463876, | |
| "eval_runtime": 0.9915, | |
| "eval_samples_per_second": 512.337, | |
| "eval_steps_per_second": 64.546, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 3.222748815165877e-05, | |
| "loss": 0.1564, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "eval_accuracy": 0.836614191532135, | |
| "eval_f1": 0.8400770712909442, | |
| "eval_loss": 0.5775780081748962, | |
| "eval_mcc": 0.6741723640208644, | |
| "eval_runtime": 0.9911, | |
| "eval_samples_per_second": 512.56, | |
| "eval_steps_per_second": 64.574, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 2.6303317535545023e-05, | |
| "loss": 0.0854, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "eval_accuracy": 0.8267716765403748, | |
| "eval_f1": 0.8333333333333333, | |
| "eval_loss": 0.6006239056587219, | |
| "eval_mcc": 0.6561201797627784, | |
| "eval_runtime": 0.9911, | |
| "eval_samples_per_second": 512.571, | |
| "eval_steps_per_second": 64.576, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "eval_accuracy": 0.834645688533783, | |
| "eval_f1": 0.8333333333333334, | |
| "eval_loss": 0.593051552772522, | |
| "eval_mcc": 0.6692967103749653, | |
| "eval_runtime": 0.9927, | |
| "eval_samples_per_second": 511.722, | |
| "eval_steps_per_second": 64.469, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 2.037914691943128e-05, | |
| "loss": 0.0488, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "eval_accuracy": 0.8425197005271912, | |
| "eval_f1": 0.846743295019157, | |
| "eval_loss": 0.7115373611450195, | |
| "eval_mcc": 0.6864752706187329, | |
| "eval_runtime": 0.9914, | |
| "eval_samples_per_second": 512.395, | |
| "eval_steps_per_second": 64.554, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 1.4454976303317535e-05, | |
| "loss": 0.0314, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "eval_accuracy": 0.8484252095222473, | |
| "eval_f1": 0.8487229862475442, | |
| "eval_loss": 0.8149858713150024, | |
| "eval_mcc": 0.6969083992056365, | |
| "eval_runtime": 0.9906, | |
| "eval_samples_per_second": 512.795, | |
| "eval_steps_per_second": 64.604, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 8.530805687203793e-06, | |
| "loss": 0.0174, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 8.53, | |
| "eval_accuracy": 0.834645688533783, | |
| "eval_f1": 0.8346456692913385, | |
| "eval_loss": 0.8690735101699829, | |
| "eval_mcc": 0.6693172130512284, | |
| "eval_runtime": 0.9909, | |
| "eval_samples_per_second": 512.668, | |
| "eval_steps_per_second": 64.588, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "learning_rate": 2.6066350710900472e-06, | |
| "loss": 0.0107, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "eval_accuracy": 0.8523622155189514, | |
| "eval_f1": 0.8514851485148515, | |
| "eval_loss": 0.9377376437187195, | |
| "eval_mcc": 0.7047207175597211, | |
| "eval_runtime": 0.99, | |
| "eval_samples_per_second": 513.131, | |
| "eval_steps_per_second": 64.646, | |
| "step": 4000 | |
| } | |
| ], | |
| "max_steps": 4220, | |
| "num_train_epochs": 10, | |
| "total_flos": 3.998323026461491e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |