| { |
| "best_global_step": 1218, |
| "best_metric": 0.7279029462738301, |
| "best_model_checkpoint": "./deberta-unfair-tos/checkpoint-1218", |
| "epoch": 14.0, |
| "eval_steps": 500, |
| "global_step": 1218, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.28901734104046245, |
| "grad_norm": 0.6572251319885254, |
| "learning_rate": 3.6641221374045806e-06, |
| "loss": 0.324, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.5780346820809249, |
| "grad_norm": 5.866250038146973, |
| "learning_rate": 7.480916030534352e-06, |
| "loss": 0.351, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.8670520231213873, |
| "grad_norm": 7.3032636642456055, |
| "learning_rate": 1.1297709923664125e-05, |
| "loss": 0.3958, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_f1_macro": 0.09475873757916975, |
| "eval_f1_micro": 0.0742146320893949, |
| "eval_loss": 0.288751482963562, |
| "eval_precision_micro": 0.039163328882955054, |
| "eval_recall_micro": 0.7068273092369478, |
| "eval_runtime": 17.1509, |
| "eval_samples_per_second": 132.646, |
| "eval_steps_per_second": 4.198, |
| "step": 87 |
| }, |
| { |
| "epoch": 1.1502890173410405, |
| "grad_norm": 2.369174003601074, |
| "learning_rate": 1.5114503816793895e-05, |
| "loss": 0.2875, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.439306358381503, |
| "grad_norm": 1.5161077976226807, |
| "learning_rate": 1.8931297709923668e-05, |
| "loss": 0.1816, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.7283236994219653, |
| "grad_norm": 1.0002554655075073, |
| "learning_rate": 1.9988401693084502e-05, |
| "loss": 0.1207, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_f1_macro": 0.26524636829070025, |
| "eval_f1_micro": 0.25287356321839083, |
| "eval_loss": 0.08097408711910248, |
| "eval_precision_micro": 0.14534534534534535, |
| "eval_recall_micro": 0.9718875502008032, |
| "eval_runtime": 17.2746, |
| "eval_samples_per_second": 131.696, |
| "eval_steps_per_second": 4.168, |
| "step": 174 |
| }, |
| { |
| "epoch": 2.0115606936416186, |
| "grad_norm": 0.8926426768302917, |
| "learning_rate": 1.9933871110628215e-05, |
| "loss": 0.0923, |
| "step": 175 |
| }, |
| { |
| "epoch": 2.300578034682081, |
| "grad_norm": 10.878843307495117, |
| "learning_rate": 1.9834897861561572e-05, |
| "loss": 0.0774, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.5895953757225434, |
| "grad_norm": 1.0657168626785278, |
| "learning_rate": 1.9691924737527694e-05, |
| "loss": 0.0676, |
| "step": 225 |
| }, |
| { |
| "epoch": 2.878612716763006, |
| "grad_norm": 4.809388160705566, |
| "learning_rate": 1.9505591379087126e-05, |
| "loss": 0.049, |
| "step": 250 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_f1_macro": 0.4353178441624277, |
| "eval_f1_micro": 0.44317096466093603, |
| "eval_loss": 0.07703664153814316, |
| "eval_precision_micro": 0.2907268170426065, |
| "eval_recall_micro": 0.9317269076305221, |
| "eval_runtime": 17.4749, |
| "eval_samples_per_second": 130.187, |
| "eval_steps_per_second": 4.12, |
| "step": 261 |
| }, |
| { |
| "epoch": 3.161849710982659, |
| "grad_norm": 1.4025094509124756, |
| "learning_rate": 1.927673141406083e-05, |
| "loss": 0.0538, |
| "step": 275 |
| }, |
| { |
| "epoch": 3.4508670520231215, |
| "grad_norm": 0.3441365957260132, |
| "learning_rate": 1.9006368728002864e-05, |
| "loss": 0.0507, |
| "step": 300 |
| }, |
| { |
| "epoch": 3.739884393063584, |
| "grad_norm": 0.5084119439125061, |
| "learning_rate": 1.8695712883488243e-05, |
| "loss": 0.0253, |
| "step": 325 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_f1_macro": 0.49704492410886936, |
| "eval_f1_micro": 0.46774193548387094, |
| "eval_loss": 0.09584943950176239, |
| "eval_precision_micro": 0.3122476446837147, |
| "eval_recall_micro": 0.9317269076305221, |
| "eval_runtime": 17.0721, |
| "eval_samples_per_second": 133.258, |
| "eval_steps_per_second": 4.217, |
| "step": 348 |
| }, |
| { |
| "epoch": 4.023121387283237, |
| "grad_norm": 0.3451331555843353, |
| "learning_rate": 1.8346153708709267e-05, |
| "loss": 0.0369, |
| "step": 350 |
| }, |
| { |
| "epoch": 4.312138728323699, |
| "grad_norm": 0.7666098475456238, |
| "learning_rate": 1.795925507959027e-05, |
| "loss": 0.0188, |
| "step": 375 |
| }, |
| { |
| "epoch": 4.601156069364162, |
| "grad_norm": 0.3931030333042145, |
| "learning_rate": 1.7536747923238566e-05, |
| "loss": 0.0175, |
| "step": 400 |
| }, |
| { |
| "epoch": 4.890173410404624, |
| "grad_norm": 10.357611656188965, |
| "learning_rate": 1.7080522474033194e-05, |
| "loss": 0.0277, |
| "step": 425 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_f1_macro": 0.5729573660442645, |
| "eval_f1_micro": 0.5513905683192262, |
| "eval_loss": 0.10878393799066544, |
| "eval_precision_micro": 0.3944636678200692, |
| "eval_recall_micro": 0.9156626506024096, |
| "eval_runtime": 17.1098, |
| "eval_samples_per_second": 132.965, |
| "eval_steps_per_second": 4.208, |
| "step": 435 |
| }, |
| { |
| "epoch": 5.173410404624278, |
| "grad_norm": 0.42035040259361267, |
| "learning_rate": 1.659261981699653e-05, |
| "loss": 0.018, |
| "step": 450 |
| }, |
| { |
| "epoch": 5.46242774566474, |
| "grad_norm": 4.960078716278076, |
| "learning_rate": 1.6075222756282513e-05, |
| "loss": 0.0152, |
| "step": 475 |
| }, |
| { |
| "epoch": 5.7514450867052025, |
| "grad_norm": 0.6047903299331665, |
| "learning_rate": 1.5530646049634473e-05, |
| "loss": 0.0121, |
| "step": 500 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_f1_macro": 0.6328988151219024, |
| "eval_f1_micro": 0.6419019316493314, |
| "eval_loss": 0.16698166728019714, |
| "eval_precision_micro": 0.5094339622641509, |
| "eval_recall_micro": 0.8674698795180723, |
| "eval_runtime": 16.9188, |
| "eval_samples_per_second": 134.466, |
| "eval_steps_per_second": 4.256, |
| "step": 522 |
| }, |
| { |
| "epoch": 6.034682080924855, |
| "grad_norm": 0.10740824788808823, |
| "learning_rate": 1.4961326052502146e-05, |
| "loss": 0.0135, |
| "step": 525 |
| }, |
| { |
| "epoch": 6.323699421965318, |
| "grad_norm": 2.375861406326294, |
| "learning_rate": 1.4369809818148586e-05, |
| "loss": 0.0088, |
| "step": 550 |
| }, |
| { |
| "epoch": 6.61271676300578, |
| "grad_norm": 0.30369603633880615, |
| "learning_rate": 1.3758743702511495e-05, |
| "loss": 0.0109, |
| "step": 575 |
| }, |
| { |
| "epoch": 6.901734104046243, |
| "grad_norm": 0.2703251540660858, |
| "learning_rate": 1.3130861524799091e-05, |
| "loss": 0.0111, |
| "step": 600 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_f1_macro": 0.674210169675286, |
| "eval_f1_micro": 0.6728682170542636, |
| "eval_loss": 0.13765300810337067, |
| "eval_precision_micro": 0.547979797979798, |
| "eval_recall_micro": 0.8714859437751004, |
| "eval_runtime": 16.9576, |
| "eval_samples_per_second": 134.158, |
| "eval_steps_per_second": 4.246, |
| "step": 609 |
| }, |
| { |
| "epoch": 7.184971098265896, |
| "grad_norm": 0.16049759089946747, |
| "learning_rate": 1.248897233678826e-05, |
| "loss": 0.0065, |
| "step": 625 |
| }, |
| { |
| "epoch": 7.473988439306359, |
| "grad_norm": 0.30845990777015686, |
| "learning_rate": 1.1835947855543356e-05, |
| "loss": 0.0074, |
| "step": 650 |
| }, |
| { |
| "epoch": 7.763005780346821, |
| "grad_norm": 0.40923434495925903, |
| "learning_rate": 1.1174709615779827e-05, |
| "loss": 0.0082, |
| "step": 675 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_f1_macro": 0.6818995873907134, |
| "eval_f1_micro": 0.6787330316742082, |
| "eval_loss": 0.14643684029579163, |
| "eval_precision_micro": 0.5434782608695652, |
| "eval_recall_micro": 0.9036144578313253, |
| "eval_runtime": 16.8897, |
| "eval_samples_per_second": 134.697, |
| "eval_steps_per_second": 4.263, |
| "step": 696 |
| }, |
| { |
| "epoch": 8.046242774566474, |
| "grad_norm": 0.188787579536438, |
| "learning_rate": 1.0508215899350986e-05, |
| "loss": 0.0063, |
| "step": 700 |
| }, |
| { |
| "epoch": 8.335260115606937, |
| "grad_norm": 0.4512380361557007, |
| "learning_rate": 9.839448500333616e-06, |
| "loss": 0.0061, |
| "step": 725 |
| }, |
| { |
| "epoch": 8.624277456647398, |
| "grad_norm": 0.191133514046669, |
| "learning_rate": 9.17139938492331e-06, |
| "loss": 0.0049, |
| "step": 750 |
| }, |
| { |
| "epoch": 8.913294797687861, |
| "grad_norm": 0.6083097457885742, |
| "learning_rate": 8.507057305821294e-06, |
| "loss": 0.0043, |
| "step": 775 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_f1_macro": 0.6575205455489375, |
| "eval_f1_micro": 0.6676737160120846, |
| "eval_loss": 0.15850034356117249, |
| "eval_precision_micro": 0.5351089588377724, |
| "eval_recall_micro": 0.8875502008032129, |
| "eval_runtime": 17.1028, |
| "eval_samples_per_second": 133.019, |
| "eval_steps_per_second": 4.21, |
| "step": 783 |
| }, |
| { |
| "epoch": 9.196531791907514, |
| "grad_norm": 0.09172528237104416, |
| "learning_rate": 7.849394430997941e-06, |
| "loss": 0.0037, |
| "step": 800 |
| }, |
| { |
| "epoch": 9.485549132947977, |
| "grad_norm": 0.0930006131529808, |
| "learning_rate": 7.201353046653892e-06, |
| "loss": 0.0033, |
| "step": 825 |
| }, |
| { |
| "epoch": 9.77456647398844, |
| "grad_norm": 0.25650352239608765, |
| "learning_rate": 6.565832393867808e-06, |
| "loss": 0.0054, |
| "step": 850 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_f1_macro": 0.7011244720560289, |
| "eval_f1_micro": 0.7041322314049587, |
| "eval_loss": 0.22890231013298035, |
| "eval_precision_micro": 0.598314606741573, |
| "eval_recall_micro": 0.8554216867469879, |
| "eval_runtime": 17.0241, |
| "eval_samples_per_second": 133.634, |
| "eval_steps_per_second": 4.229, |
| "step": 870 |
| }, |
| { |
| "epoch": 10.057803468208093, |
| "grad_norm": 0.20074431598186493, |
| "learning_rate": 5.945675697821627e-06, |
| "loss": 0.0041, |
| "step": 875 |
| }, |
| { |
| "epoch": 10.346820809248555, |
| "grad_norm": 0.06646797806024551, |
| "learning_rate": 5.343657447632593e-06, |
| "loss": 0.0027, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.635838150289018, |
| "grad_norm": 0.10714161396026611, |
| "learning_rate": 4.7624709837002794e-06, |
| "loss": 0.0025, |
| "step": 925 |
| }, |
| { |
| "epoch": 10.92485549132948, |
| "grad_norm": 0.14236567914485931, |
| "learning_rate": 4.204716448100967e-06, |
| "loss": 0.0038, |
| "step": 950 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_f1_macro": 0.7042070354121188, |
| "eval_f1_micro": 0.7103109656301145, |
| "eval_loss": 0.2040930837392807, |
| "eval_precision_micro": 0.5994475138121547, |
| "eval_recall_micro": 0.8714859437751004, |
| "eval_runtime": 17.0325, |
| "eval_samples_per_second": 133.568, |
| "eval_steps_per_second": 4.227, |
| "step": 957 |
| }, |
| { |
| "epoch": 11.208092485549132, |
| "grad_norm": 0.20406219363212585, |
| "learning_rate": 3.6728891519374467e-06, |
| "loss": 0.0028, |
| "step": 975 |
| }, |
| { |
| "epoch": 11.497109826589595, |
| "grad_norm": 0.26664334535598755, |
| "learning_rate": 3.1693684116870915e-06, |
| "loss": 0.0027, |
| "step": 1000 |
| }, |
| { |
| "epoch": 11.786127167630058, |
| "grad_norm": 0.09276499599218369, |
| "learning_rate": 2.6964069044927044e-06, |
| "loss": 0.0024, |
| "step": 1025 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_f1_macro": 0.7031684663931863, |
| "eval_f1_micro": 0.7166666666666667, |
| "eval_loss": 0.225614532828331, |
| "eval_precision_micro": 0.6125356125356125, |
| "eval_recall_micro": 0.8634538152610441, |
| "eval_runtime": 17.0969, |
| "eval_samples_per_second": 133.065, |
| "eval_steps_per_second": 4.211, |
| "step": 1044 |
| }, |
| { |
| "epoch": 12.06936416184971, |
| "grad_norm": 0.1458662748336792, |
| "learning_rate": 2.2561205900189064e-06, |
| "loss": 0.0027, |
| "step": 1050 |
| }, |
| { |
| "epoch": 12.358381502890174, |
| "grad_norm": 0.1117081567645073, |
| "learning_rate": 1.8504792439623475e-06, |
| "loss": 0.0025, |
| "step": 1075 |
| }, |
| { |
| "epoch": 12.647398843930636, |
| "grad_norm": 0.17292848229408264, |
| "learning_rate": 1.4812976455672278e-06, |
| "loss": 0.0025, |
| "step": 1100 |
| }, |
| { |
| "epoch": 12.936416184971097, |
| "grad_norm": 0.37239304184913635, |
| "learning_rate": 1.1502274585720252e-06, |
| "loss": 0.0021, |
| "step": 1125 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_f1_macro": 0.7020439217567145, |
| "eval_f1_micro": 0.7174280879864636, |
| "eval_loss": 0.22631782293319702, |
| "eval_precision_micro": 0.6198830409356725, |
| "eval_recall_micro": 0.8514056224899599, |
| "eval_runtime": 17.1562, |
| "eval_samples_per_second": 132.605, |
| "eval_steps_per_second": 4.197, |
| "step": 1131 |
| }, |
| { |
| "epoch": 13.219653179190752, |
| "grad_norm": 0.051143016666173935, |
| "learning_rate": 8.587498419108009e-07, |
| "loss": 0.0021, |
| "step": 1150 |
| }, |
| { |
| "epoch": 13.508670520231213, |
| "grad_norm": 0.20565341413021088, |
| "learning_rate": 6.08168823227695e-07, |
| "loss": 0.0028, |
| "step": 1175 |
| }, |
| { |
| "epoch": 13.797687861271676, |
| "grad_norm": 0.015616578981280327, |
| "learning_rate": 3.996054648505965e-07, |
| "loss": 0.0021, |
| "step": 1200 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_f1_macro": 0.710205871606475, |
| "eval_f1_micro": 0.7279029462738301, |
| "eval_loss": 0.23643752932548523, |
| "eval_precision_micro": 0.6402439024390244, |
| "eval_recall_micro": 0.8433734939759037, |
| "eval_runtime": 16.9864, |
| "eval_samples_per_second": 133.93, |
| "eval_steps_per_second": 4.239, |
| "step": 1218 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 1305, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 15, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 3, |
| "early_stopping_threshold": 0.0 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 0 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.374646862938112e+16, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
