| { | |
| "best_metric": 0.6491296227815271, | |
| "best_model_checkpoint": "./CARES/checkpoints/roberta-stratified/run-0/checkpoint-3266", | |
| "epoch": 23.163120567375888, | |
| "global_step": 3266, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 0.3462072014808655, | |
| "eval_macro_f1": 0.0007381889763779527, | |
| "eval_macro_precision": 0.0625, | |
| "eval_macro_recall": 0.0003712871287128713, | |
| "eval_micro_f1": 0.0039447731755424065, | |
| "eval_micro_precision": 1.0, | |
| "eval_micro_recall": 0.001976284584980237, | |
| "eval_runtime": 2.8803, | |
| "eval_samples_per_second": 337.116, | |
| "eval_steps_per_second": 10.763, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 0.26299169659614563, | |
| "eval_macro_f1": 0.03133608815426997, | |
| "eval_macro_precision": 0.051470588235294115, | |
| "eval_macro_recall": 0.022524752475247524, | |
| "eval_micro_f1": 0.20931569867740082, | |
| "eval_micro_precision": 0.8235294117647058, | |
| "eval_micro_recall": 0.11989459815546773, | |
| "eval_runtime": 2.7712, | |
| "eval_samples_per_second": 350.396, | |
| "eval_steps_per_second": 11.187, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_loss": 0.2253277599811554, | |
| "eval_macro_f1": 0.06167745523885891, | |
| "eval_macro_precision": 0.16397973284200829, | |
| "eval_macro_recall": 0.06444494429328504, | |
| "eval_micro_f1": 0.43727272727272726, | |
| "eval_micro_precision": 0.7052785923753666, | |
| "eval_micro_recall": 0.3168642951251647, | |
| "eval_runtime": 2.7754, | |
| "eval_samples_per_second": 349.86, | |
| "eval_steps_per_second": 11.17, | |
| "step": 426 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 1.858287403029302e-05, | |
| "loss": 0.3344, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "eval_loss": 0.19568884372711182, | |
| "eval_macro_f1": 0.19533405609363325, | |
| "eval_macro_precision": 0.3105566636324695, | |
| "eval_macro_recall": 0.1838257881936085, | |
| "eval_micro_f1": 0.6221161495624502, | |
| "eval_micro_precision": 0.785140562248996, | |
| "eval_micro_recall": 0.5151515151515151, | |
| "eval_runtime": 2.775, | |
| "eval_samples_per_second": 349.909, | |
| "eval_steps_per_second": 11.171, | |
| "step": 568 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "eval_loss": 0.16411657631397247, | |
| "eval_macro_f1": 0.2724936969098707, | |
| "eval_macro_precision": 0.3798299682715689, | |
| "eval_macro_recall": 0.24456296636870606, | |
| "eval_micro_f1": 0.6832491255343957, | |
| "eval_micro_precision": 0.833175355450237, | |
| "eval_micro_recall": 0.5790513833992095, | |
| "eval_runtime": 2.7735, | |
| "eval_samples_per_second": 350.097, | |
| "eval_steps_per_second": 11.177, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "eval_loss": 0.15434952080249786, | |
| "eval_macro_f1": 0.33512731711181853, | |
| "eval_macro_precision": 0.42023284164245034, | |
| "eval_macro_recall": 0.31527094611085327, | |
| "eval_micro_f1": 0.7207672445592034, | |
| "eval_micro_precision": 0.8189438390611903, | |
| "eval_micro_recall": 0.6436100131752306, | |
| "eval_runtime": 2.7765, | |
| "eval_samples_per_second": 349.72, | |
| "eval_steps_per_second": 11.165, | |
| "step": 852 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "eval_loss": 0.14012500643730164, | |
| "eval_macro_f1": 0.3798067714294707, | |
| "eval_macro_precision": 0.5469114307081249, | |
| "eval_macro_recall": 0.3317304014614234, | |
| "eval_micro_f1": 0.7399624765478424, | |
| "eval_micro_precision": 0.8596338273757629, | |
| "eval_micro_recall": 0.6495388669301713, | |
| "eval_runtime": 2.774, | |
| "eval_samples_per_second": 350.033, | |
| "eval_steps_per_second": 11.175, | |
| "step": 994 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 3.0452387239092086e-05, | |
| "loss": 0.1403, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "eval_loss": 0.1425222009420395, | |
| "eval_macro_f1": 0.4279380705267567, | |
| "eval_macro_precision": 0.603624913309443, | |
| "eval_macro_recall": 0.3703454980587147, | |
| "eval_micro_f1": 0.7544954128440367, | |
| "eval_micro_precision": 0.851698425849213, | |
| "eval_micro_recall": 0.6772068511198946, | |
| "eval_runtime": 2.7729, | |
| "eval_samples_per_second": 350.179, | |
| "eval_steps_per_second": 11.18, | |
| "step": 1136 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "eval_loss": 0.12965930998325348, | |
| "eval_macro_f1": 0.516908938978679, | |
| "eval_macro_precision": 0.6952135806026781, | |
| "eval_macro_recall": 0.4441065897269686, | |
| "eval_micro_f1": 0.7703488372093023, | |
| "eval_micro_precision": 0.8589951377633711, | |
| "eval_micro_recall": 0.6982872200263505, | |
| "eval_runtime": 2.7725, | |
| "eval_samples_per_second": 350.226, | |
| "eval_steps_per_second": 11.181, | |
| "step": 1278 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "eval_loss": 0.12852537631988525, | |
| "eval_macro_f1": 0.5388984906238257, | |
| "eval_macro_precision": 0.7130990451123878, | |
| "eval_macro_recall": 0.4711434333008039, | |
| "eval_micro_f1": 0.7730547550432276, | |
| "eval_micro_precision": 0.8529411764705882, | |
| "eval_micro_recall": 0.7068511198945981, | |
| "eval_runtime": 2.7737, | |
| "eval_samples_per_second": 350.077, | |
| "eval_steps_per_second": 11.176, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 10.64, | |
| "learning_rate": 2.929008238263819e-05, | |
| "loss": 0.0558, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "eval_loss": 0.13507980108261108, | |
| "eval_macro_f1": 0.5653547754610186, | |
| "eval_macro_precision": 0.697025101599949, | |
| "eval_macro_recall": 0.49809191604893205, | |
| "eval_micro_f1": 0.7797808412866738, | |
| "eval_micro_precision": 0.8413424866514111, | |
| "eval_micro_recall": 0.7266139657444005, | |
| "eval_runtime": 2.7736, | |
| "eval_samples_per_second": 350.083, | |
| "eval_steps_per_second": 11.177, | |
| "step": 1562 | |
| }, | |
| { | |
| "epoch": 12.09, | |
| "eval_loss": 0.1355280876159668, | |
| "eval_macro_f1": 0.5794821565833095, | |
| "eval_macro_precision": 0.6818697605893614, | |
| "eval_macro_recall": 0.5236900724860433, | |
| "eval_micro_f1": 0.7849877236057524, | |
| "eval_micro_precision": 0.8394598649662416, | |
| "eval_micro_recall": 0.7371541501976284, | |
| "eval_runtime": 2.7759, | |
| "eval_samples_per_second": 349.801, | |
| "eval_steps_per_second": 11.168, | |
| "step": 1704 | |
| }, | |
| { | |
| "epoch": 13.09, | |
| "eval_loss": 0.13659390807151794, | |
| "eval_macro_f1": 0.5891915194817472, | |
| "eval_macro_precision": 0.69343533514632, | |
| "eval_macro_recall": 0.5302292848265341, | |
| "eval_micro_f1": 0.7858642407277815, | |
| "eval_micro_precision": 0.8380597014925373, | |
| "eval_micro_recall": 0.7397891963109354, | |
| "eval_runtime": 2.7739, | |
| "eval_samples_per_second": 350.043, | |
| "eval_steps_per_second": 11.175, | |
| "step": 1846 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "eval_loss": 0.13488434255123138, | |
| "eval_macro_f1": 0.5900914578255574, | |
| "eval_macro_precision": 0.6881746937300715, | |
| "eval_macro_recall": 0.5327336547593478, | |
| "eval_micro_f1": 0.7906326459279972, | |
| "eval_micro_precision": 0.842144452717796, | |
| "eval_micro_recall": 0.7450592885375494, | |
| "eval_runtime": 2.772, | |
| "eval_samples_per_second": 350.292, | |
| "eval_steps_per_second": 11.183, | |
| "step": 1988 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "learning_rate": 2.8127777526184293e-05, | |
| "loss": 0.0229, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "eval_loss": 0.14160068333148956, | |
| "eval_macro_f1": 0.5920064296505824, | |
| "eval_macro_precision": 0.6848089552512611, | |
| "eval_macro_recall": 0.5382146732964487, | |
| "eval_micro_f1": 0.7904066736183525, | |
| "eval_micro_precision": 0.8366445916114791, | |
| "eval_micro_recall": 0.7490118577075099, | |
| "eval_runtime": 2.7738, | |
| "eval_samples_per_second": 350.061, | |
| "eval_steps_per_second": 11.176, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 16.11, | |
| "eval_loss": 0.14674818515777588, | |
| "eval_macro_f1": 0.5928173166365653, | |
| "eval_macro_precision": 0.6894481117094009, | |
| "eval_macro_recall": 0.5355030217646541, | |
| "eval_micro_f1": 0.7838118298166724, | |
| "eval_micro_precision": 0.8252002913328478, | |
| "eval_micro_recall": 0.7463768115942029, | |
| "eval_runtime": 2.7753, | |
| "eval_samples_per_second": 349.871, | |
| "eval_steps_per_second": 11.17, | |
| "step": 2272 | |
| }, | |
| { | |
| "epoch": 17.12, | |
| "eval_loss": 0.15975715219974518, | |
| "eval_macro_f1": 0.6260279452973108, | |
| "eval_macro_precision": 0.7038398537082291, | |
| "eval_macro_recall": 0.5866152515992036, | |
| "eval_micro_f1": 0.7704974271012006, | |
| "eval_micro_precision": 0.80386542591267, | |
| "eval_micro_recall": 0.7397891963109354, | |
| "eval_runtime": 2.773, | |
| "eval_samples_per_second": 350.168, | |
| "eval_steps_per_second": 11.179, | |
| "step": 2414 | |
| }, | |
| { | |
| "epoch": 17.73, | |
| "learning_rate": 2.6965472669730396e-05, | |
| "loss": 0.0137, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 18.13, | |
| "eval_loss": 0.15081025660037994, | |
| "eval_macro_f1": 0.6271302596876451, | |
| "eval_macro_precision": 0.7314709578369518, | |
| "eval_macro_recall": 0.5704921787075566, | |
| "eval_micro_f1": 0.7809989521480964, | |
| "eval_micro_precision": 0.8312267657992565, | |
| "eval_micro_recall": 0.7364953886693018, | |
| "eval_runtime": 2.7721, | |
| "eval_samples_per_second": 350.282, | |
| "eval_steps_per_second": 11.183, | |
| "step": 2556 | |
| }, | |
| { | |
| "epoch": 19.13, | |
| "eval_loss": 0.14972703158855438, | |
| "eval_macro_f1": 0.5953077373311189, | |
| "eval_macro_precision": 0.6771131830689183, | |
| "eval_macro_recall": 0.5419002355296083, | |
| "eval_micro_f1": 0.7900552486187846, | |
| "eval_micro_precision": 0.8301886792452831, | |
| "eval_micro_recall": 0.7536231884057971, | |
| "eval_runtime": 2.7725, | |
| "eval_samples_per_second": 350.229, | |
| "eval_steps_per_second": 11.181, | |
| "step": 2698 | |
| }, | |
| { | |
| "epoch": 20.14, | |
| "eval_loss": 0.16402311623096466, | |
| "eval_macro_f1": 0.5892530505072842, | |
| "eval_macro_precision": 0.6727972959139897, | |
| "eval_macro_recall": 0.5367187180351005, | |
| "eval_micro_f1": 0.779319916724497, | |
| "eval_micro_precision": 0.8233137829912024, | |
| "eval_micro_recall": 0.7397891963109354, | |
| "eval_runtime": 2.7746, | |
| "eval_samples_per_second": 349.963, | |
| "eval_steps_per_second": 11.173, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 21.15, | |
| "eval_loss": 0.16335928440093994, | |
| "eval_macro_f1": 0.6332863501223459, | |
| "eval_macro_precision": 0.7578187629295322, | |
| "eval_macro_recall": 0.5685989504723161, | |
| "eval_micro_f1": 0.7924791086350974, | |
| "eval_micro_precision": 0.8404726735598228, | |
| "eval_micro_recall": 0.7496706192358367, | |
| "eval_runtime": 2.7725, | |
| "eval_samples_per_second": 350.229, | |
| "eval_steps_per_second": 11.181, | |
| "step": 2982 | |
| }, | |
| { | |
| "epoch": 21.28, | |
| "learning_rate": 2.58031678132765e-05, | |
| "loss": 0.011, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 22.16, | |
| "eval_loss": 0.16375195980072021, | |
| "eval_macro_f1": 0.6343242676719218, | |
| "eval_macro_precision": 0.7473720060268463, | |
| "eval_macro_recall": 0.5714537925969617, | |
| "eval_micro_f1": 0.7909878682842287, | |
| "eval_micro_precision": 0.8346744696415508, | |
| "eval_micro_recall": 0.7516469038208169, | |
| "eval_runtime": 2.7735, | |
| "eval_samples_per_second": 350.093, | |
| "eval_steps_per_second": 11.177, | |
| "step": 3124 | |
| }, | |
| { | |
| "epoch": 23.16, | |
| "eval_loss": 0.16090567409992218, | |
| "eval_macro_f1": 0.6491296227815271, | |
| "eval_macro_precision": 0.7487080551943797, | |
| "eval_macro_recall": 0.5927079340096836, | |
| "eval_micro_f1": 0.7961299239806496, | |
| "eval_micro_precision": 0.8372093023255814, | |
| "eval_micro_recall": 0.758893280632411, | |
| "eval_runtime": 2.7735, | |
| "eval_samples_per_second": 350.094, | |
| "eval_steps_per_second": 11.177, | |
| "step": 3266 | |
| } | |
| ], | |
| "max_steps": 14100, | |
| "num_train_epochs": 100, | |
| "total_flos": 1.2446273222900736e+16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "adam_epsilon": 6.356433345691159e-10, | |
| "learning_rate": 3.084757089028641e-05, | |
| "per_device_eval_batch_size": 32, | |
| "per_device_train_batch_size": 16, | |
| "seed": 326, | |
| "warmup_steps": 830, | |
| "weight_decay": 0.0009910374448883887 | |
| } | |
| } | |