| { | |
| "best_metric": 0.9861495731704059, | |
| "best_model_checkpoint": "/home3/s5431786/nlp-final-project/results/roberta-large-e-snli-classification-nli_explanation-base-b16/checkpoint-7200", | |
| "epoch": 0.2213420316868593, | |
| "global_step": 7600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 7.765482430596001e-07, | |
| "loss": 1.0726, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "eval_accuracy": 0.7982117455801666, | |
| "eval_f1": 0.7974911320075254, | |
| "eval_loss": 0.8326770067214966, | |
| "eval_runtime": 12.8669, | |
| "eval_samples_per_second": 764.91, | |
| "eval_steps_per_second": 47.875, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.5530964861192002e-06, | |
| "loss": 0.3233, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "eval_accuracy": 0.9642349116033326, | |
| "eval_f1": 0.9642718853707861, | |
| "eval_loss": 0.14329373836517334, | |
| "eval_runtime": 12.933, | |
| "eval_samples_per_second": 760.998, | |
| "eval_steps_per_second": 47.63, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.3296447291788007e-06, | |
| "loss": 0.173, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_accuracy": 0.9762243446453973, | |
| "eval_f1": 0.9761597449883214, | |
| "eval_loss": 0.1054357960820198, | |
| "eval_runtime": 12.9969, | |
| "eval_samples_per_second": 757.258, | |
| "eval_steps_per_second": 47.396, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 3.1061929722384003e-06, | |
| "loss": 0.1452, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "eval_accuracy": 0.9799837431416378, | |
| "eval_f1": 0.9799316333368723, | |
| "eval_loss": 0.09648650884628296, | |
| "eval_runtime": 13.0065, | |
| "eval_samples_per_second": 756.7, | |
| "eval_steps_per_second": 47.361, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 3.882741215298001e-06, | |
| "loss": 0.1449, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_accuracy": 0.9813046128835603, | |
| "eval_f1": 0.9812341372941731, | |
| "eval_loss": 0.09331633150577545, | |
| "eval_runtime": 12.921, | |
| "eval_samples_per_second": 761.706, | |
| "eval_steps_per_second": 47.674, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.659289458357601e-06, | |
| "loss": 0.1303, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_accuracy": 0.9815078236130867, | |
| "eval_f1": 0.9814520761782233, | |
| "eval_loss": 0.10890379548072815, | |
| "eval_runtime": 12.8683, | |
| "eval_samples_per_second": 764.824, | |
| "eval_steps_per_second": 47.87, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 5.435837701417202e-06, | |
| "loss": 0.1372, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.9832351148140622, | |
| "eval_f1": 0.983181371856266, | |
| "eval_loss": 0.07784133404493332, | |
| "eval_runtime": 12.8322, | |
| "eval_samples_per_second": 766.977, | |
| "eval_steps_per_second": 48.004, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 6.212385944476801e-06, | |
| "loss": 0.1229, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_accuracy": 0.9820158504369031, | |
| "eval_f1": 0.9819483437268689, | |
| "eval_loss": 0.10309657454490662, | |
| "eval_runtime": 13.2533, | |
| "eval_samples_per_second": 742.608, | |
| "eval_steps_per_second": 46.479, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 6.988934187536401e-06, | |
| "loss": 0.1291, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.9842511684616948, | |
| "eval_f1": 0.9842093647131546, | |
| "eval_loss": 0.08313809335231781, | |
| "eval_runtime": 13.0051, | |
| "eval_samples_per_second": 756.778, | |
| "eval_steps_per_second": 47.366, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 7.765482430596002e-06, | |
| "loss": 0.1049, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.9824222718959561, | |
| "eval_f1": 0.9823847600112674, | |
| "eval_loss": 0.0907953679561615, | |
| "eval_runtime": 12.984, | |
| "eval_samples_per_second": 758.01, | |
| "eval_steps_per_second": 47.443, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 8.542030673655602e-06, | |
| "loss": 0.1243, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_accuracy": 0.9823206665311929, | |
| "eval_f1": 0.9822362137371193, | |
| "eval_loss": 0.0925893783569336, | |
| "eval_runtime": 12.9285, | |
| "eval_samples_per_second": 761.262, | |
| "eval_steps_per_second": 47.647, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.318578916715203e-06, | |
| "loss": 0.1291, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_accuracy": 0.9855720382036172, | |
| "eval_f1": 0.9855364167633462, | |
| "eval_loss": 0.08309133350849152, | |
| "eval_runtime": 12.8482, | |
| "eval_samples_per_second": 766.023, | |
| "eval_steps_per_second": 47.945, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.994992693419992e-06, | |
| "loss": 0.1148, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_accuracy": 0.983946352367405, | |
| "eval_f1": 0.9839146445196306, | |
| "eval_loss": 0.08415436744689941, | |
| "eval_runtime": 12.8411, | |
| "eval_samples_per_second": 766.443, | |
| "eval_steps_per_second": 47.971, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.954116721338281e-06, | |
| "loss": 0.1256, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_accuracy": 0.9836415362731152, | |
| "eval_f1": 0.9836219614739408, | |
| "eval_loss": 0.084382563829422, | |
| "eval_runtime": 13.2723, | |
| "eval_samples_per_second": 741.546, | |
| "eval_steps_per_second": 46.413, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.91324074925657e-06, | |
| "loss": 0.1207, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_accuracy": 0.9829302987197724, | |
| "eval_f1": 0.9829157443588743, | |
| "eval_loss": 0.0957166850566864, | |
| "eval_runtime": 12.9479, | |
| "eval_samples_per_second": 760.125, | |
| "eval_steps_per_second": 47.575, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.872364777174857e-06, | |
| "loss": 0.1162, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_accuracy": 0.9859784596626702, | |
| "eval_f1": 0.9859267481206228, | |
| "eval_loss": 0.0752706453204155, | |
| "eval_runtime": 13.005, | |
| "eval_samples_per_second": 756.786, | |
| "eval_steps_per_second": 47.366, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.831488805093147e-06, | |
| "loss": 0.095, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.9844543791912213, | |
| "eval_f1": 0.9844184876885574, | |
| "eval_loss": 0.09373413771390915, | |
| "eval_runtime": 13.0171, | |
| "eval_samples_per_second": 756.082, | |
| "eval_steps_per_second": 47.322, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 9.790612833011435e-06, | |
| "loss": 0.1069, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_accuracy": 0.9861816703921967, | |
| "eval_f1": 0.9861495731704059, | |
| "eval_loss": 0.07330357283353806, | |
| "eval_runtime": 12.8964, | |
| "eval_samples_per_second": 763.158, | |
| "eval_steps_per_second": 47.765, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.749736860929725e-06, | |
| "loss": 0.1054, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_accuracy": 0.9853688274740906, | |
| "eval_f1": 0.9853406697769191, | |
| "eval_loss": 0.07271973788738251, | |
| "eval_runtime": 12.8137, | |
| "eval_samples_per_second": 768.085, | |
| "eval_steps_per_second": 48.074, | |
| "step": 7600 | |
| } | |
| ], | |
| "max_steps": 103008, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.4913660628723872e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |