{ "best_metric": 0.9205694310134601, "best_model_checkpoint": "./saved_models/roberta_prompt_sbdh_gpt4_v2_0/checkpoint-744", "epoch": 31.0, "eval_steps": 500, "global_step": 744, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "grad_norm": 1.8722108602523804, "learning_rate": 2.5e-06, "loss": 0.7199, "step": 24 }, { "epoch": 1.0, "eval_acc_macro": 0.07035287889134419, "eval_acc_micro": 0.15675280351047888, "eval_auc_macro": 0.663313870197172, "eval_auc_micro": 0.6559659079771284, "eval_f1_at_5": 0.2619950790125531, "eval_f1_at_8": 0.24094809864446162, "eval_f1_macro": 0.11929565026205535, "eval_f1_micro": 0.2710221285563637, "eval_loss": 0.3433417081832886, "eval_prec_at_5": 0.1684931506849315, "eval_prec_at_8": 0.14169520547945205, "eval_prec_macro": 0.11362671803403451, "eval_prec_micro": 0.18168974286521103, "eval_rec_at_5": 0.5886605783866058, "eval_rec_at_8": 0.8044140030441401, "eval_rec_macro": 0.3442251461985764, "eval_rec_micro": 0.5331674958540188, "eval_runtime": 6.2217, "eval_samples_per_second": 140.798, "eval_steps_per_second": 17.68, "step": 24 }, { "epoch": 2.0, "grad_norm": 5.086896896362305, "learning_rate": 5e-06, "loss": 0.3297, "step": 48 }, { "epoch": 2.0, "eval_acc_macro": 0.5110805606565244, "eval_acc_micro": 0.543895747599414, "eval_auc_macro": 0.9371959387131173, "eval_auc_micro": 0.9373504192626666, "eval_f1_at_5": 0.4096411096246807, "eval_f1_at_8": 0.2899510119744706, "eval_f1_macro": 0.6473524785746846, "eval_f1_micro": 0.7045757441136646, "eval_loss": 0.1864374279975891, "eval_prec_at_5": 0.2621004566210046, "eval_prec_at_8": 0.1703767123287671, "eval_prec_macro": 0.8055450466236987, "eval_prec_micro": 0.7588516746410757, "eval_rec_at_5": 0.9372146118721462, "eval_rec_at_8": 0.9724124809741247, "eval_rec_macro": 0.6042332875883025, "eval_rec_micro": 0.6575456053067448, "eval_runtime": 6.2474, "eval_samples_per_second": 140.219, "eval_steps_per_second": 17.607, "step": 48 }, { "epoch": 3.0, "grad_norm": 1.4806246757507324, "learning_rate": 7.500000000000001e-06, "loss": 0.1717, "step": 72 }, { "epoch": 3.0, "eval_acc_macro": 0.7353572773142955, "eval_acc_micro": 0.7442348008385224, "eval_auc_macro": 0.9820663738220755, "eval_auc_micro": 0.9835785967362127, "eval_f1_at_5": 0.42686614175859805, "eval_f1_at_8": 0.2922793659426448, "eval_f1_macro": 0.841220323579556, "eval_f1_micro": 0.8533653846153163, "eval_loss": 0.1064019426703453, "eval_prec_at_5": 0.2732876712328767, "eval_prec_at_8": 0.17180365296803654, "eval_prec_macro": 0.8250237186513111, "eval_prec_micro": 0.8255813953487732, "eval_rec_at_5": 0.9745053272450532, "eval_rec_at_8": 0.978310502283105, "eval_rec_macro": 0.8662646242524481, "eval_rec_micro": 0.8830845771143546, "eval_runtime": 6.1744, "eval_samples_per_second": 141.876, "eval_steps_per_second": 17.815, "step": 72 }, { "epoch": 4.0, "grad_norm": 1.224611759185791, "learning_rate": 1e-05, "loss": 0.0995, "step": 96 }, { "epoch": 4.0, "eval_acc_macro": 0.801538415253349, "eval_acc_micro": 0.8093126385808714, "eval_auc_macro": 0.989204170129788, "eval_auc_micro": 0.9911075755259093, "eval_f1_at_5": 0.4288655829720509, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.8875969325055185, "eval_f1_micro": 0.8946078431371818, "eval_loss": 0.07651617377996445, "eval_prec_at_5": 0.27465753424657535, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.881843938607854, "eval_prec_micro": 0.8816425120772237, "eval_rec_at_5": 0.9779299847792998, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.8961158899484207, "eval_rec_micro": 0.9079601990048998, "eval_runtime": 6.2178, "eval_samples_per_second": 140.886, "eval_steps_per_second": 17.691, "step": 96 }, { "epoch": 5.0, "grad_norm": 1.1437005996704102, "learning_rate": 9.722222222222223e-06, "loss": 0.0805, "step": 120 }, { "epoch": 5.0, "eval_acc_macro": 0.8121493747200438, "eval_acc_micro": 0.8208841463414008, "eval_auc_macro": 0.991491527290926, "eval_auc_micro": 0.9929716878748317, "eval_f1_at_5": 0.4291987915672543, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.8934495150714268, "eval_f1_micro": 0.9016324822100543, "eval_loss": 0.06796517223119736, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.901996461034861, "eval_prec_micro": 0.9103972950126027, "eval_rec_at_5": 0.9785007610350076, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.8880403796830154, "eval_rec_micro": 0.8930348258705727, "eval_runtime": 6.2151, "eval_samples_per_second": 140.947, "eval_steps_per_second": 17.699, "step": 120 }, { "epoch": 6.0, "grad_norm": 2.1739773750305176, "learning_rate": 9.444444444444445e-06, "loss": 0.0718, "step": 144 }, { "epoch": 6.0, "eval_acc_macro": 0.828523573268117, "eval_acc_micro": 0.8335843373493348, "eval_auc_macro": 0.9925057744832354, "eval_auc_micro": 0.9937106144896979, "eval_f1_at_5": 0.4291987915672543, "eval_f1_at_8": 0.2925113305987028, "eval_f1_macro": 0.9041659334919602, "eval_f1_micro": 0.9092402464064961, "eval_loss": 0.06353675574064255, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.17194634703196346, "eval_prec_macro": 0.8956330448729077, "eval_prec_micro": 0.9007323026850366, "eval_rec_at_5": 0.9785007610350076, "eval_rec_at_8": 0.9788812785388128, "eval_rec_macro": 0.9145914149826037, "eval_rec_micro": 0.9179104477611179, "eval_runtime": 6.2166, "eval_samples_per_second": 140.913, "eval_steps_per_second": 17.695, "step": 144 }, { "epoch": 7.0, "grad_norm": 1.558062195777893, "learning_rate": 9.166666666666666e-06, "loss": 0.0669, "step": 168 }, { "epoch": 7.0, "eval_acc_macro": 0.8323133875195107, "eval_acc_micro": 0.8400918133128661, "eval_auc_macro": 0.9934557716722295, "eval_auc_micro": 0.9948184252460742, "eval_f1_at_5": 0.4291987915672543, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9059398386358487, "eval_f1_micro": 0.913097713097637, "eval_loss": 0.06089127063751221, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9108588887690331, "eval_prec_micro": 0.9157631359465458, "eval_rec_at_5": 0.9785007610350076, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9032889716188378, "eval_rec_micro": 0.9104477611939543, "eval_runtime": 6.2143, "eval_samples_per_second": 140.965, "eval_steps_per_second": 17.701, "step": 168 }, { "epoch": 8.0, "grad_norm": 1.132032871246338, "learning_rate": 8.888888888888888e-06, "loss": 0.0611, "step": 192 }, { "epoch": 8.0, "eval_acc_macro": 0.8399125580413327, "eval_acc_micro": 0.8451563691837647, "eval_auc_macro": 0.9939843717869511, "eval_auc_micro": 0.9950275486953798, "eval_f1_at_5": 0.4291987915672543, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9107249879766437, "eval_f1_micro": 0.916081025216956, "eval_loss": 0.06228514760732651, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9075154035613321, "eval_prec_micro": 0.913437757625646, "eval_rec_at_5": 0.9785007610350076, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9155563129570495, "eval_rec_micro": 0.9187396351574694, "eval_runtime": 6.1966, "eval_samples_per_second": 141.368, "eval_steps_per_second": 17.752, "step": 192 }, { "epoch": 9.0, "grad_norm": 0.9028905034065247, "learning_rate": 8.611111111111112e-06, "loss": 0.0561, "step": 216 }, { "epoch": 9.0, "eval_acc_macro": 0.8453726969988299, "eval_acc_micro": 0.850689127105601, "eval_auc_macro": 0.9944939772465632, "eval_auc_micro": 0.9954155898635628, "eval_f1_at_5": 0.4291987915672543, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.914643073353695, "eval_f1_micro": 0.9193214729002135, "eval_loss": 0.05663854256272316, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9126789502644125, "eval_prec_micro": 0.9174236168455063, "eval_rec_at_5": 0.9785007610350076, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9185389204843853, "eval_rec_micro": 0.9212271973465239, "eval_runtime": 6.145, "eval_samples_per_second": 142.554, "eval_steps_per_second": 17.901, "step": 216 }, { "epoch": 10.0, "grad_norm": 1.0204308032989502, "learning_rate": 8.333333333333334e-06, "loss": 0.0535, "step": 240 }, { "epoch": 10.0, "eval_acc_macro": 0.8498863755077791, "eval_acc_micro": 0.858365758754797, "eval_auc_macro": 0.9946638093104078, "eval_auc_micro": 0.9956081402572352, "eval_f1_at_5": 0.4291987915672543, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.916676493616951, "eval_f1_micro": 0.9237855946397886, "eval_loss": 0.05517387017607689, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9255652453336515, "eval_prec_micro": 0.9331641285955217, "eval_rec_at_5": 0.9785007610350076, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9088352191837838, "eval_rec_micro": 0.9145936981757119, "eval_runtime": 6.1446, "eval_samples_per_second": 142.564, "eval_steps_per_second": 17.902, "step": 240 }, { "epoch": 11.0, "grad_norm": 1.0917142629623413, "learning_rate": 8.055555555555557e-06, "loss": 0.0501, "step": 264 }, { "epoch": 11.0, "eval_acc_macro": 0.8493708087290419, "eval_acc_micro": 0.8584758942456564, "eval_auc_macro": 0.9949027930572298, "eval_auc_micro": 0.9961340229150117, "eval_f1_at_5": 0.4295319911779049, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9165770410883837, "eval_f1_micro": 0.92384937238486, "eval_loss": 0.05478766933083534, "eval_prec_at_5": 0.2751141552511416, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9246420093584519, "eval_prec_micro": 0.9324324324323536, "eval_rec_at_5": 0.9790715372907153, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9096237229214146, "eval_rec_micro": 0.9154228855720634, "eval_runtime": 6.1786, "eval_samples_per_second": 141.779, "eval_steps_per_second": 17.803, "step": 264 }, { "epoch": 12.0, "grad_norm": 1.1021435260772705, "learning_rate": 7.77777777777778e-06, "loss": 0.0474, "step": 288 }, { "epoch": 12.0, "eval_acc_macro": 0.8453196602235581, "eval_acc_micro": 0.856143079315641, "eval_auc_macro": 0.9949355514759608, "eval_auc_micro": 0.9959768461938463, "eval_f1_at_5": 0.4291987915672543, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9135326281320886, "eval_f1_micro": 0.9224968579806515, "eval_loss": 0.05536971986293793, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9280234523786058, "eval_prec_micro": 0.9322607959355688, "eval_rec_at_5": 0.9785007610350076, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9026177068383121, "eval_rec_micro": 0.9129353233830089, "eval_runtime": 6.2635, "eval_samples_per_second": 139.858, "eval_steps_per_second": 17.562, "step": 288 }, { "epoch": 13.0, "grad_norm": 1.661024570465088, "learning_rate": 7.500000000000001e-06, "loss": 0.0456, "step": 312 }, { "epoch": 13.0, "eval_acc_macro": 0.8539896866112132, "eval_acc_micro": 0.8623924941359763, "eval_auc_macro": 0.9951266438356762, "eval_auc_micro": 0.9960736114541555, "eval_f1_at_5": 0.42918048633589306, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9193621316490773, "eval_f1_micro": 0.9261125104953043, "eval_loss": 0.05397089943289757, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9310933950333862, "eval_prec_micro": 0.9379251700679474, "eval_rec_at_5": 0.978310502283105, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9090157207820031, "eval_rec_micro": 0.9145936981757119, "eval_runtime": 6.188, "eval_samples_per_second": 141.564, "eval_steps_per_second": 17.776, "step": 312 }, { "epoch": 14.0, "grad_norm": 1.8844163417816162, "learning_rate": 7.222222222222223e-06, "loss": 0.0448, "step": 336 }, { "epoch": 14.0, "eval_acc_macro": 0.8471019101615411, "eval_acc_micro": 0.8584831899921143, "eval_auc_macro": 0.9954775103594954, "eval_auc_micro": 0.9964240959398153, "eval_f1_at_5": 0.4295319911779049, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9145784925942765, "eval_f1_micro": 0.9238535969708941, "eval_loss": 0.05232784524559975, "eval_prec_at_5": 0.2751141552511416, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9310832701535142, "eval_prec_micro": 0.937660119555855, "eval_rec_at_5": 0.9790715372907153, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.900954322532269, "eval_rec_micro": 0.9104477611939543, "eval_runtime": 6.196, "eval_samples_per_second": 141.382, "eval_steps_per_second": 17.753, "step": 336 }, { "epoch": 15.0, "grad_norm": 1.3576740026474, "learning_rate": 6.944444444444445e-06, "loss": 0.0425, "step": 360 }, { "epoch": 15.0, "eval_acc_macro": 0.8512337950118664, "eval_acc_micro": 0.8615863141523435, "eval_auc_macro": 0.9951475121132285, "eval_auc_micro": 0.9962107846753767, "eval_f1_at_5": 0.4295319911779049, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9173794915227328, "eval_f1_micro": 0.925647451963164, "eval_loss": 0.053657278418540955, "eval_prec_at_5": 0.2751141552511416, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9267018062985609, "eval_prec_micro": 0.9326599326598541, "eval_rec_at_5": 0.9790715372907153, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9096846119468953, "eval_rec_micro": 0.9187396351574694, "eval_runtime": 6.1516, "eval_samples_per_second": 142.402, "eval_steps_per_second": 17.882, "step": 360 }, { "epoch": 16.0, "grad_norm": 0.7812018990516663, "learning_rate": 6.666666666666667e-06, "loss": 0.0402, "step": 384 }, { "epoch": 16.0, "eval_acc_macro": 0.8482275698613361, "eval_acc_micro": 0.861502347417773, "eval_auc_macro": 0.9952181846925838, "eval_auc_micro": 0.9963574918587093, "eval_f1_at_5": 0.4298468624753246, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9150689476603097, "eval_f1_micro": 0.9255989911726837, "eval_loss": 0.055027130991220474, "eval_prec_at_5": 0.2753424657534247, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9321798187054336, "eval_prec_micro": 0.9386189258311219, "eval_rec_at_5": 0.9794520547945206, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9015047595626373, "eval_rec_micro": 0.9129353233830089, "eval_runtime": 6.1964, "eval_samples_per_second": 141.373, "eval_steps_per_second": 17.752, "step": 384 }, { "epoch": 17.0, "grad_norm": 0.6827139854431152, "learning_rate": 6.3888888888888885e-06, "loss": 0.0388, "step": 408 }, { "epoch": 17.0, "eval_acc_macro": 0.8515875486589554, "eval_acc_micro": 0.8601236476042612, "eval_auc_macro": 0.9951263473453763, "eval_auc_micro": 0.9961734953002022, "eval_f1_at_5": 0.4295136788854873, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9176542506423578, "eval_f1_micro": 0.9248026589114312, "eval_loss": 0.05326356366276741, "eval_prec_at_5": 0.2751141552511416, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9224814917404568, "eval_prec_micro": 0.9267277268941776, "eval_rec_at_5": 0.9788812785388128, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9138993691616202, "eval_rec_micro": 0.922885572139227, "eval_runtime": 6.2448, "eval_samples_per_second": 140.276, "eval_steps_per_second": 17.615, "step": 408 }, { "epoch": 18.0, "grad_norm": 2.072528123855591, "learning_rate": 6.111111111111112e-06, "loss": 0.0379, "step": 432 }, { "epoch": 18.0, "eval_acc_macro": 0.8504993062436726, "eval_acc_micro": 0.8585780525501655, "eval_auc_macro": 0.9950323486570704, "eval_auc_micro": 0.9961765247834898, "eval_f1_at_5": 0.4298468624753246, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.917328682822007, "eval_f1_micro": 0.923908523908447, "eval_loss": 0.05443555489182472, "eval_prec_at_5": 0.2753424657534247, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9227209928734593, "eval_prec_micro": 0.9266055045870787, "eval_rec_at_5": 0.9794520547945206, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9127347764028103, "eval_rec_micro": 0.9212271973465239, "eval_runtime": 6.1606, "eval_samples_per_second": 142.194, "eval_steps_per_second": 17.855, "step": 432 }, { "epoch": 19.0, "grad_norm": 1.0216673612594604, "learning_rate": 5.833333333333334e-06, "loss": 0.0367, "step": 456 }, { "epoch": 19.0, "eval_acc_macro": 0.8491462024293238, "eval_acc_micro": 0.8604832424005564, "eval_auc_macro": 0.9952033636321062, "eval_auc_micro": 0.9961533581465836, "eval_f1_at_5": 0.42918048633589306, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9162533148683757, "eval_f1_micro": 0.9250104733974928, "eval_loss": 0.053822897374629974, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9302531226804523, "eval_prec_micro": 0.9348010160879818, "eval_rec_at_5": 0.978310502283105, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9044133701637805, "eval_rec_micro": 0.9154228855720634, "eval_runtime": 6.2679, "eval_samples_per_second": 139.759, "eval_steps_per_second": 17.55, "step": 456 }, { "epoch": 20.0, "grad_norm": 0.8565398454666138, "learning_rate": 5.555555555555557e-06, "loss": 0.035, "step": 480 }, { "epoch": 20.0, "eval_acc_macro": 0.8539275510220851, "eval_acc_micro": 0.8656832298135974, "eval_auc_macro": 0.9953444620919104, "eval_auc_micro": 0.9962831804157093, "eval_f1_at_5": 0.4288472848093916, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9186680652840996, "eval_f1_micro": 0.9280066583436597, "eval_loss": 0.05500434711575508, "eval_prec_at_5": 0.27465753424657535, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9265469458942642, "eval_prec_micro": 0.9314954051795379, "eval_rec_at_5": 0.9777397260273972, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9134436001515955, "eval_rec_micro": 0.92454394693193, "eval_runtime": 6.1331, "eval_samples_per_second": 142.831, "eval_steps_per_second": 17.935, "step": 480 }, { "epoch": 21.0, "grad_norm": 0.8573027849197388, "learning_rate": 5.2777777777777785e-06, "loss": 0.034, "step": 504 }, { "epoch": 21.0, "eval_acc_macro": 0.8487139365583213, "eval_acc_micro": 0.8599071207429675, "eval_auc_macro": 0.9952593352184999, "eval_auc_micro": 0.99623096638022, "eval_f1_at_5": 0.4295136788854873, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9158985967502566, "eval_f1_micro": 0.9246774864751622, "eval_loss": 0.05316817760467529, "eval_prec_at_5": 0.2751141552511416, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9212276417335067, "eval_prec_micro": 0.9281537176273242, "eval_rec_at_5": 0.9788812785388128, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9117274622825242, "eval_rec_micro": 0.9212271973465239, "eval_runtime": 6.2044, "eval_samples_per_second": 141.191, "eval_steps_per_second": 17.729, "step": 504 }, { "epoch": 22.0, "grad_norm": 0.6982609033584595, "learning_rate": 5e-06, "loss": 0.0321, "step": 528 }, { "epoch": 22.0, "eval_acc_macro": 0.8491707056123833, "eval_acc_micro": 0.8597986057319241, "eval_auc_macro": 0.995166189612911, "eval_auc_micro": 0.9962029882110331, "eval_f1_at_5": 0.42918048633589306, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9161030038339234, "eval_f1_micro": 0.9246147438566493, "eval_loss": 0.056650497019290924, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9201734731045618, "eval_prec_micro": 0.9288702928869516, "eval_rec_at_5": 0.978310502283105, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9130204494213245, "eval_rec_micro": 0.9203980099501724, "eval_runtime": 6.2454, "eval_samples_per_second": 140.262, "eval_steps_per_second": 17.613, "step": 528 }, { "epoch": 23.0, "grad_norm": 0.7926116585731506, "learning_rate": 4.722222222222222e-06, "loss": 0.0303, "step": 552 }, { "epoch": 23.0, "eval_acc_macro": 0.8491868722119115, "eval_acc_micro": 0.8571428571427916, "eval_auc_macro": 0.9953682041367656, "eval_auc_micro": 0.9962074433335152, "eval_f1_at_5": 0.42918048633589306, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9157423579975225, "eval_f1_micro": 0.9230769230768471, "eval_loss": 0.056672144681215286, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9083006225839708, "eval_prec_micro": 0.9159183673468639, "eval_rec_at_5": 0.978310502283105, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9239908582927064, "eval_rec_micro": 0.9303482587063905, "eval_runtime": 6.2236, "eval_samples_per_second": 140.755, "eval_steps_per_second": 17.675, "step": 552 }, { "epoch": 24.0, "grad_norm": 1.125510573387146, "learning_rate": 4.444444444444444e-06, "loss": 0.0301, "step": 576 }, { "epoch": 24.0, "eval_acc_macro": 0.8485250231781508, "eval_acc_micro": 0.8570318570317904, "eval_auc_macro": 0.9952185441458874, "eval_auc_micro": 0.9962312336875689, "eval_f1_at_5": 0.42918048633589306, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9157383370149551, "eval_f1_micro": 0.923012552301178, "eval_loss": 0.06080281734466553, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9252875267591651, "eval_prec_micro": 0.9315878378377591, "eval_rec_at_5": 0.978310502283105, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9073331635314181, "eval_rec_micro": 0.9145936981757119, "eval_runtime": 6.1671, "eval_samples_per_second": 142.044, "eval_steps_per_second": 17.837, "step": 576 }, { "epoch": 25.0, "grad_norm": 1.347324013710022, "learning_rate": 4.166666666666667e-06, "loss": 0.0282, "step": 600 }, { "epoch": 25.0, "eval_acc_macro": 0.842166168753605, "eval_acc_micro": 0.8551068883609774, "eval_auc_macro": 0.9949314011679055, "eval_auc_micro": 0.9960538307103354, "eval_f1_at_5": 0.4288472848093916, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9119808094359786, "eval_f1_micro": 0.9218950064019699, "eval_loss": 0.056968480348587036, "eval_prec_at_5": 0.27465753424657535, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9426243208291692, "eval_prec_micro": 0.9498680738785444, "eval_rec_at_5": 0.9777397260273972, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.8854364498112819, "eval_rec_micro": 0.8955223880596273, "eval_runtime": 6.2218, "eval_samples_per_second": 140.795, "eval_steps_per_second": 17.68, "step": 600 }, { "epoch": 26.0, "grad_norm": 0.7992838025093079, "learning_rate": 3.88888888888889e-06, "loss": 0.0271, "step": 624 }, { "epoch": 26.0, "eval_acc_macro": 0.850448503713554, "eval_acc_micro": 0.8584544758989396, "eval_auc_macro": 0.994964877927155, "eval_auc_micro": 0.9959627234555783, "eval_f1_at_5": 0.42851407428878907, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.916988259808195, "eval_f1_micro": 0.923836969946404, "eval_loss": 0.062447551637887955, "eval_prec_at_5": 0.27442922374429224, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9094336920398504, "eval_prec_micro": 0.9174161896973902, "eval_rec_at_5": 0.9771689497716894, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9249848527276295, "eval_rec_micro": 0.9303482587063905, "eval_runtime": 6.205, "eval_samples_per_second": 141.177, "eval_steps_per_second": 17.728, "step": 624 }, { "epoch": 27.0, "grad_norm": 0.7699769735336304, "learning_rate": 3.6111111111111115e-06, "loss": 0.0264, "step": 648 }, { "epoch": 27.0, "eval_acc_macro": 0.8509862309003138, "eval_acc_micro": 0.857365549493308, "eval_auc_macro": 0.9948429133410858, "eval_auc_micro": 0.9958119175595623, "eval_f1_at_5": 0.42851407428878907, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9178276987832569, "eval_f1_micro": 0.9232060428031118, "eval_loss": 0.06019788607954979, "eval_prec_at_5": 0.27442922374429224, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9247554610597524, "eval_prec_micro": 0.9345794392522571, "eval_rec_at_5": 0.9771689497716894, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9123211173889221, "eval_rec_micro": 0.9121061359866574, "eval_runtime": 6.199, "eval_samples_per_second": 141.314, "eval_steps_per_second": 17.745, "step": 648 }, { "epoch": 28.0, "grad_norm": 0.8111199736595154, "learning_rate": 3.3333333333333333e-06, "loss": 0.0247, "step": 672 }, { "epoch": 28.0, "eval_acc_macro": 0.8539050124448745, "eval_acc_micro": 0.8599397590360798, "eval_auc_macro": 0.9949311017267249, "eval_auc_micro": 0.9958826649045767, "eval_f1_at_5": 0.4288472848093916, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9187578667846501, "eval_f1_micro": 0.9246963562752287, "eval_loss": 0.06124128773808479, "eval_prec_at_5": 0.27465753424657535, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.898554094667798, "eval_prec_micro": 0.9034810126581564, "eval_rec_at_5": 0.9777397260273972, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9403712267485004, "eval_rec_micro": 0.9469320066334206, "eval_runtime": 6.2175, "eval_samples_per_second": 140.893, "eval_steps_per_second": 17.692, "step": 672 }, { "epoch": 29.0, "grad_norm": 0.8091293573379517, "learning_rate": 3.055555555555556e-06, "loss": 0.0236, "step": 696 }, { "epoch": 29.0, "eval_acc_macro": 0.8515123078962853, "eval_acc_micro": 0.8605182926828612, "eval_auc_macro": 0.9949441815288761, "eval_auc_micro": 0.9957389426533069, "eval_f1_at_5": 0.42918048633589306, "eval_f1_at_8": 0.2925113305987028, "eval_f1_macro": 0.9176180291221252, "eval_f1_micro": 0.9250307251125829, "eval_loss": 0.06163506582379341, "eval_prec_at_5": 0.27488584474885847, "eval_prec_at_8": 0.17194634703196346, "eval_prec_macro": 0.9064412973486995, "eval_prec_micro": 0.9141700404857559, "eval_rec_at_5": 0.978310502283105, "eval_rec_at_8": 0.9788812785388128, "eval_rec_macro": 0.9296481224485932, "eval_rec_micro": 0.936152570480851, "eval_runtime": 6.2226, "eval_samples_per_second": 140.776, "eval_steps_per_second": 17.677, "step": 696 }, { "epoch": 30.0, "grad_norm": 0.7638717889785767, "learning_rate": 2.7777777777777783e-06, "loss": 0.0242, "step": 720 }, { "epoch": 30.0, "eval_acc_macro": 0.8514782533320497, "eval_acc_micro": 0.8610897927858127, "eval_auc_macro": 0.9950854271410213, "eval_auc_micro": 0.9960944168761464, "eval_f1_at_5": 0.4295136788854873, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9177801984021922, "eval_f1_micro": 0.9253608247421917, "eval_loss": 0.06092337146401405, "eval_prec_at_5": 0.2751141552511416, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9134538753515611, "eval_prec_micro": 0.920426579163173, "eval_rec_at_5": 0.9788812785388128, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9227004606245881, "eval_rec_micro": 0.9303482587063905, "eval_runtime": 6.1492, "eval_samples_per_second": 142.457, "eval_steps_per_second": 17.888, "step": 720 }, { "epoch": 31.0, "grad_norm": 0.9947636127471924, "learning_rate": 2.5e-06, "loss": 0.023, "step": 744 }, { "epoch": 31.0, "eval_acc_macro": 0.8560068706371594, "eval_acc_micro": 0.8637770897832149, "eval_auc_macro": 0.9947951862441506, "eval_auc_micro": 0.9958837341339722, "eval_f1_at_5": 0.4295136788854873, "eval_f1_at_8": 0.29274329072668936, "eval_f1_macro": 0.9205694310134601, "eval_f1_micro": 0.9269102990032453, "eval_loss": 0.06532252579927444, "eval_prec_at_5": 0.2751141552511416, "eval_prec_at_8": 0.1720890410958904, "eval_prec_macro": 0.9201882436524288, "eval_prec_micro": 0.9284525790348644, "eval_rec_at_5": 0.9788812785388128, "eval_rec_at_8": 0.9794520547945206, "eval_rec_macro": 0.9217575019308203, "eval_rec_micro": 0.9253731343282815, "eval_runtime": 6.1427, "eval_samples_per_second": 142.608, "eval_steps_per_second": 17.907, "step": 744 } ], "logging_steps": 500, "max_steps": 960, "num_input_tokens_seen": 0, "num_train_epochs": 40, "save_steps": 500, "total_flos": 2.503865168315597e+16, "train_batch_size": 32, "trial_name": null, "trial_params": null }