{ "best_metric": null, "best_model_checkpoint": null, "best_supernet_model_checkpoint": null, "epoch": 3.0, "global_step": 16599, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "compression_loss": 0.0, "epoch": 0.09, "learning_rate": 2.993368837199307e-05, "loss": 3.708, "step": 500 }, { "compression_loss": 0.0, "epoch": 0.18, "learning_rate": 2.973374457836494e-05, "loss": 1.3139, "step": 1000 }, { "compression_loss": 0.0, "epoch": 0.27, "learning_rate": 2.940195540458019e-05, "loss": 1.0383, "step": 1500 }, { "compression_loss": 0.0, "epoch": 0.36, "learning_rate": 2.894128987657224e-05, "loss": 0.8724, "step": 2000 }, { "compression_loss": 0.0, "epoch": 0.45, "learning_rate": 2.8357162317742997e-05, "loss": 0.787, "step": 2500 }, { "compression_loss": 0.0, "epoch": 0.54, "learning_rate": 2.765246037034805e-05, "loss": 0.7197, "step": 3000 }, { "compression_loss": 0.0, "epoch": 0.63, "learning_rate": 2.6834537480471538e-05, "loss": 0.6765, "step": 3500 }, { "compression_loss": 0.0, "epoch": 0.72, "learning_rate": 2.5910712857007108e-05, "loss": 0.6851, "step": 4000 }, { "compression_loss": 0.0, "epoch": 0.81, "learning_rate": 2.4889253373866598e-05, "loss": 0.6264, "step": 4500 }, { "compression_loss": 0.0, "epoch": 0.9, "learning_rate": 2.3781601339877912e-05, "loss": 0.5946, "step": 5000 }, { "compression_loss": 0.0, "epoch": 0.99, "learning_rate": 2.259323244741251e-05, "loss": 0.5737, "step": 5500 }, { "Minimum SubNet": "OrderedDict([(, {0: 576, 1: 448, 2: 448, 3: 448, 4: 512, 5: 384, 6: 448, 7: 448, 8: 512, 9: 384, 10: 192, 11: 320, 12: 1169, 13: 1170, 14: 1284, 15: 1218, 16: 1205, 17: 1130, 18: 1030, 19: 976, 20: 794, 21: 401, 22: 404, 23: 422})])", "epoch": 1.0, "eval_exact_match": 79.17691579943235, "eval_f1": 86.96854919071785, "eval_runtime": 29.6736, "eval_samples_per_second": 363.421, "eval_steps_per_second": 2.865, "step": 5533 }, { "SuperNet": "OrderedDict([(, {0: 768, 1: 768, 2: 768, 3: 768, 4: 768, 5: 768, 6: 768, 7: 768, 8: 768, 9: 768, 10: 768, 11: 768, 12: 3072, 13: 3072, 14: 3072, 15: 3072, 16: 3072, 17: 3072, 18: 3072, 19: 3072, 20: 3072, 21: 3072, 22: 3072, 23: 3072})])", "epoch": 1.0, "eval_exact_match": 81.60832544938505, "eval_f1": 88.68083548824183, "eval_runtime": 44.3132, "eval_samples_per_second": 243.358, "eval_steps_per_second": 1.918, "step": 5533 }, { "compression_loss": 0.0, "epoch": 1.08, "learning_rate": 2.1339488339585353e-05, "loss": 0.4132, "step": 6000 }, { "compression_loss": 0.0, "epoch": 1.17, "learning_rate": 2.0026566916852832e-05, "loss": 0.3938, "step": 6500 }, { "compression_loss": 0.0, "epoch": 1.27, "learning_rate": 1.8668665100812266e-05, "loss": 0.3756, "step": 7000 }, { "compression_loss": 0.0, "epoch": 1.36, "learning_rate": 1.727793411888579e-05, "loss": 0.3776, "step": 7500 }, { "compression_loss": 0.0, "epoch": 1.45, "learning_rate": 1.586681897132654e-05, "loss": 0.3699, "step": 8000 }, { "compression_loss": 0.0, "epoch": 1.54, "learning_rate": 1.4447947066739665e-05, "loss": 0.3632, "step": 8500 }, { "compression_loss": 0.0, "epoch": 1.63, "learning_rate": 1.303401522531631e-05, "loss": 0.3636, "step": 9000 }, { "compression_loss": 0.0, "epoch": 1.72, "learning_rate": 1.1640442839182642e-05, "loss": 0.3505, "step": 9500 }, { "compression_loss": 0.0, "epoch": 1.81, "learning_rate": 1.0274119051724544e-05, "loss": 0.3461, "step": 10000 }, { "compression_loss": 0.0, "epoch": 1.9, "learning_rate": 8.950084959658292e-06, "loss": 0.355, "step": 10500 }, { "compression_loss": 0.0, "epoch": 1.99, "learning_rate": 7.680188724005017e-06, "loss": 0.343, "step": 11000 }, { "Minimum SubNet": "OrderedDict([(, {0: 576, 1: 448, 2: 448, 3: 448, 4: 512, 5: 384, 6: 448, 7: 448, 8: 512, 9: 384, 10: 192, 11: 320, 12: 1169, 13: 1170, 14: 1284, 15: 1218, 16: 1205, 17: 1130, 18: 1030, 19: 976, 20: 794, 21: 401, 22: 404, 23: 422})])", "epoch": 2.0, "eval_exact_match": 79.94323557237465, "eval_f1": 87.58166030437053, "eval_runtime": 29.4984, "eval_samples_per_second": 365.579, "eval_steps_per_second": 2.882, "step": 11066 }, { "SuperNet": "OrderedDict([(, {0: 768, 1: 768, 2: 768, 3: 768, 4: 768, 5: 768, 6: 768, 7: 768, 8: 768, 9: 768, 10: 768, 11: 768, 12: 3072, 13: 3072, 14: 3072, 15: 3072, 16: 3072, 17: 3072, 18: 3072, 19: 3072, 20: 3072, 21: 3072, 22: 3072, 23: 3072})])", "epoch": 2.0, "eval_exact_match": 82.07190160832545, "eval_f1": 89.27809806276005, "eval_runtime": 44.3723, "eval_samples_per_second": 243.034, "eval_steps_per_second": 1.916, "step": 11066 }, { "compression_loss": 0.0, "epoch": 2.08, "learning_rate": 6.480466652465469e-06, "loss": 0.2873, "step": 11500 }, { "compression_loss": 0.0, "epoch": 2.17, "learning_rate": 5.354199358345275e-06, "loss": 0.2772, "step": 12000 }, { "compression_loss": 0.0, "epoch": 2.26, "learning_rate": 4.311911058558205e-06, "loss": 0.2712, "step": 12500 }, { "compression_loss": 0.0, "epoch": 2.35, "learning_rate": 3.365265461285558e-06, "loss": 0.2707, "step": 13000 }, { "compression_loss": 0.0, "epoch": 2.44, "learning_rate": 2.522733654646333e-06, "loss": 0.2727, "step": 13500 }, { "compression_loss": 0.0, "epoch": 2.53, "learning_rate": 1.7918550612054763e-06, "loss": 0.2702, "step": 14000 }, { "compression_loss": 0.0, "epoch": 2.62, "learning_rate": 1.1791699712122217e-06, "loss": 0.2712, "step": 14500 }, { "compression_loss": 0.0, "epoch": 2.71, "learning_rate": 6.901610166055744e-07, "loss": 0.2654, "step": 15000 }, { "compression_loss": 0.0, "epoch": 2.8, "learning_rate": 3.292041095087001e-07, "loss": 0.2701, "step": 15500 }, { "compression_loss": 0.0, "epoch": 2.89, "learning_rate": 9.985605052251767e-08, "loss": 0.2709, "step": 16000 }, { "compression_loss": 0.0, "epoch": 2.98, "learning_rate": 3.2506247875463413e-09, "loss": 0.2662, "step": 16500 }, { "Minimum SubNet": "OrderedDict([(, {0: 576, 1: 448, 2: 448, 3: 448, 4: 512, 5: 384, 6: 448, 7: 448, 8: 512, 9: 384, 10: 192, 11: 320, 12: 1169, 13: 1170, 14: 1284, 15: 1218, 16: 1205, 17: 1130, 18: 1030, 19: 976, 20: 794, 21: 401, 22: 404, 23: 422})])", "epoch": 3.0, "eval_exact_match": 80.33112582781457, "eval_f1": 87.82976478725165, "eval_runtime": 28.6692, "eval_samples_per_second": 376.153, "eval_steps_per_second": 2.965, "step": 16599 }, { "SuperNet": "OrderedDict([(, {0: 768, 1: 768, 2: 768, 3: 768, 4: 768, 5: 768, 6: 768, 7: 768, 8: 768, 9: 768, 10: 768, 11: 768, 12: 3072, 13: 3072, 14: 3072, 15: 3072, 16: 3072, 17: 3072, 18: 3072, 19: 3072, 20: 3072, 21: 3072, 22: 3072, 23: 3072})])", "epoch": 3.0, "eval_exact_match": 82.18543046357615, "eval_f1": 89.41932504869663, "eval_runtime": 44.3404, "eval_samples_per_second": 243.21, "eval_steps_per_second": 1.917, "step": 16599 }, { "epoch": 3.0, "step": 16599, "total_flos": 5.204482670991974e+16, "train_loss": 0.563158837906804, "train_runtime": 32896.2087, "train_samples_per_second": 8.073, "train_steps_per_second": 0.505 } ], "max_steps": 16599, "min_subnet_acc": null, "min_subnet_best_acc": null, "num_train_epochs": 3, "supernet_acc": null, "supernet_best_acc": null, "total_flos": 5.204482670991974e+16, "trial_name": null, "trial_params": null }