| { | |
| "best_metric": 0.7163974642753601, | |
| "best_model_checkpoint": "final_models/glue_models/structroberta_s2_final//finetune/mnli/checkpoint-6400", | |
| "epoch": 3.879907621247113, | |
| "global_step": 8400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "eval_accuracy": 0.5423651337623596, | |
| "eval_loss": 0.9394116997718811, | |
| "eval_runtime": 12.766, | |
| "eval_samples_per_second": 514.022, | |
| "eval_steps_per_second": 64.312, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.884526558891455e-05, | |
| "loss": 0.9831, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_accuracy": 0.6263334155082703, | |
| "eval_loss": 0.8243611454963684, | |
| "eval_runtime": 12.7598, | |
| "eval_samples_per_second": 514.27, | |
| "eval_steps_per_second": 64.343, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.7690531177829104e-05, | |
| "loss": 0.8453, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_accuracy": 0.6548308730125427, | |
| "eval_loss": 0.7803521156311035, | |
| "eval_runtime": 12.7994, | |
| "eval_samples_per_second": 512.682, | |
| "eval_steps_per_second": 64.144, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.653579676674365e-05, | |
| "loss": 0.792, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_accuracy": 0.6679366230964661, | |
| "eval_loss": 0.7586068511009216, | |
| "eval_runtime": 12.7729, | |
| "eval_samples_per_second": 513.744, | |
| "eval_steps_per_second": 64.277, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 4.53810623556582e-05, | |
| "loss": 0.762, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.6755562424659729, | |
| "eval_loss": 0.7486647963523865, | |
| "eval_runtime": 12.7741, | |
| "eval_samples_per_second": 513.696, | |
| "eval_steps_per_second": 64.271, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_accuracy": 0.6807375550270081, | |
| "eval_loss": 0.7358043789863586, | |
| "eval_runtime": 12.787, | |
| "eval_samples_per_second": 513.179, | |
| "eval_steps_per_second": 64.206, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.422632794457275e-05, | |
| "loss": 0.7076, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_accuracy": 0.6892715692520142, | |
| "eval_loss": 0.739337146282196, | |
| "eval_runtime": 12.7687, | |
| "eval_samples_per_second": 513.914, | |
| "eval_steps_per_second": 64.298, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 4.30715935334873e-05, | |
| "loss": 0.67, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "eval_accuracy": 0.6915574669837952, | |
| "eval_loss": 0.7226890921592712, | |
| "eval_runtime": 12.7849, | |
| "eval_samples_per_second": 513.26, | |
| "eval_steps_per_second": 64.216, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 4.1916859122401844e-05, | |
| "loss": 0.6666, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_accuracy": 0.7013105750083923, | |
| "eval_loss": 0.7045397162437439, | |
| "eval_runtime": 12.7756, | |
| "eval_samples_per_second": 513.634, | |
| "eval_steps_per_second": 64.263, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 4.07621247113164e-05, | |
| "loss": 0.6618, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.7007009983062744, | |
| "eval_loss": 0.7101899981498718, | |
| "eval_runtime": 12.7944, | |
| "eval_samples_per_second": 512.881, | |
| "eval_steps_per_second": 64.169, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "eval_accuracy": 0.7025297284126282, | |
| "eval_loss": 0.7343056201934814, | |
| "eval_runtime": 12.7896, | |
| "eval_samples_per_second": 513.073, | |
| "eval_steps_per_second": 64.193, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 3.960739030023095e-05, | |
| "loss": 0.6262, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_accuracy": 0.7081682682037354, | |
| "eval_loss": 0.7158433794975281, | |
| "eval_runtime": 12.7759, | |
| "eval_samples_per_second": 513.625, | |
| "eval_steps_per_second": 64.262, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 3.84526558891455e-05, | |
| "loss": 0.5627, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_accuracy": 0.7121304273605347, | |
| "eval_loss": 0.7221695184707642, | |
| "eval_runtime": 12.7709, | |
| "eval_samples_per_second": 513.824, | |
| "eval_steps_per_second": 64.287, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 3.729792147806005e-05, | |
| "loss": 0.5606, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "eval_accuracy": 0.7043583989143372, | |
| "eval_loss": 0.7297407388687134, | |
| "eval_runtime": 12.7744, | |
| "eval_samples_per_second": 513.683, | |
| "eval_steps_per_second": 64.269, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 3.61431870669746e-05, | |
| "loss": 0.5698, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_accuracy": 0.7063395380973816, | |
| "eval_loss": 0.7177144885063171, | |
| "eval_runtime": 12.7828, | |
| "eval_samples_per_second": 513.345, | |
| "eval_steps_per_second": 64.227, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.7163974642753601, | |
| "eval_loss": 0.7138766646385193, | |
| "eval_runtime": 12.7766, | |
| "eval_samples_per_second": 513.596, | |
| "eval_steps_per_second": 64.258, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.498845265588915e-05, | |
| "loss": 0.5694, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "eval_accuracy": 0.7109113335609436, | |
| "eval_loss": 0.7803912162780762, | |
| "eval_runtime": 12.7706, | |
| "eval_samples_per_second": 513.836, | |
| "eval_steps_per_second": 64.288, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.38337182448037e-05, | |
| "loss": 0.4589, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_accuracy": 0.7136543989181519, | |
| "eval_loss": 0.7615983486175537, | |
| "eval_runtime": 12.7834, | |
| "eval_samples_per_second": 513.321, | |
| "eval_steps_per_second": 64.224, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 3.2678983833718243e-05, | |
| "loss": 0.4606, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "eval_accuracy": 0.7084730267524719, | |
| "eval_loss": 0.7663705945014954, | |
| "eval_runtime": 12.7769, | |
| "eval_samples_per_second": 513.583, | |
| "eval_steps_per_second": 64.257, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 3.1524249422632794e-05, | |
| "loss": 0.4687, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_accuracy": 0.7130448222160339, | |
| "eval_loss": 0.7735299468040466, | |
| "eval_runtime": 12.7981, | |
| "eval_samples_per_second": 512.732, | |
| "eval_steps_per_second": 64.15, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "eval_accuracy": 0.7145687341690063, | |
| "eval_loss": 0.7440381050109863, | |
| "eval_runtime": 12.7849, | |
| "eval_samples_per_second": 513.26, | |
| "eval_steps_per_second": 64.216, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "step": 8400, | |
| "total_flos": 8.411222207808e+16, | |
| "train_loss": 0.639952149164109, | |
| "train_runtime": 4648.7272, | |
| "train_samples_per_second": 558.82, | |
| "train_steps_per_second": 4.657 | |
| } | |
| ], | |
| "max_steps": 21650, | |
| "num_train_epochs": 10, | |
| "total_flos": 8.411222207808e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |