diff --git "a/trainer_state.json" "b/trainer_state.json" --- "a/trainer_state.json" +++ "b/trainer_state.json" @@ -1,10 +1,10 @@ { - "best_global_step": 2900, - "best_metric": 0.6237816764132553, + "best_global_step": 3100, + "best_metric": 0.5907692307692308, "best_model_checkpoint": null, - "epoch": 32.38277511961722, + "epoch": 27.273764258555133, "eval_steps": 100, - "global_step": 3400, + "global_step": 3600, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, @@ -12,12629 +12,15747 @@ { "epoch": 0, "step": 0, - "train/classification_loss": 0.6900129914283752, - "train/contrastive_loss": 9.194344520568848, - "train/negative_loss": 9.194210052490234, - "train/num_negatives": 42, - "train/num_positives": 8, - "train/positive_loss": 0.00013464232324622571, - "train/total_loss": 2.5288820266723633 + "train/classification_loss": 0.6873060464859009, + "train/contrastive_loss": 9.88529109954834, + "train/negative_loss": 9.885237693786621, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 5.318886542227119e-05, + "train/total_loss": 2.6643643379211426 }, { "epoch": 0, "step": 0, - "train/classification_loss": 0.6949148774147034, - "train/contrastive_loss": 8.761173248291016, - "train/negative_loss": 8.760587692260742, - "train/num_negatives": 32, - "train/num_positives": 24, - "train/positive_loss": 0.0005851813475601375, - "train/total_loss": 2.4471495151519775 + "train/classification_loss": 0.6811691522598267, + "train/contrastive_loss": 9.81307601928711, + "train/negative_loss": 9.813041687011719, + "train/num_negatives": 38, + "train/num_positives": 14, + "train/positive_loss": 3.443512105150148e-05, + "train/total_loss": 2.643784523010254 }, { - "epoch": 0.4784688995215311, - "grad_norm": 21.042522430419922, + "epoch": 0.38022813688212925, + "grad_norm": 20.09731101989746, "learning_rate": 9.4e-06, - "loss": 3.5789, + "loss": 3.9048, "step": 50 }, { - "epoch": 0.4784688995215311, + "epoch": 0.38022813688212925, "step": 50, - "train/classification_loss": 0.6594263315200806, - "train/contrastive_loss": 2.1301491260528564, - "train/negative_loss": 2.0000417232513428, - "train/num_negatives": 40, + "train/classification_loss": 0.6536898016929626, + "train/contrastive_loss": 2.600647211074829, + "train/negative_loss": 2.1319706439971924, + "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.13010750710964203, - "train/total_loss": 1.085456132888794 + "train/positive_loss": 0.46867647767066956, + "train/total_loss": 1.1738193035125732 }, { - "epoch": 0.4784688995215311, + "epoch": 0.38022813688212925, "step": 50, - "train/classification_loss": 0.6591920256614685, - "train/contrastive_loss": 2.1461284160614014, - "train/negative_loss": 1.8941457271575928, - "train/num_negatives": 38, - "train/num_positives": 10, - "train/positive_loss": 0.2519826889038086, - "train/total_loss": 1.0884177684783936 + "train/classification_loss": 0.6653611660003662, + "train/contrastive_loss": 2.8467068672180176, + "train/negative_loss": 2.6595070362091064, + "train/num_negatives": 40, + "train/num_positives": 14, + "train/positive_loss": 0.18719978630542755, + "train/total_loss": 1.2347025871276855 }, { - "epoch": 0.9569377990430622, - "grad_norm": 18.264829635620117, + "epoch": 0.7604562737642585, + "grad_norm": 16.864822387695312, "learning_rate": 1.94e-05, - "loss": 1.8421, + "loss": 1.9478, "step": 100 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, + "step": 100, + "train/classification_loss": 0.5244528651237488, + "train/contrastive_loss": 2.3768680095672607, + "train/negative_loss": 2.142180919647217, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.23468711972236633, + "train/total_loss": 0.9998264312744141 + }, + { + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.4930008053779602, - "train/contrastive_loss": 1.3434940576553345, - "train/negative_loss": 1.2266253232955933, + "train/classification_loss": 0.5354360342025757, + "train/contrastive_loss": 1.9692317247390747, + "train/negative_loss": 1.8838703632354736, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.11686870455741882, - "train/total_loss": 0.7616996169090271 + "train/num_positives": 6, + "train/positive_loss": 0.08536132425069809, + "train/total_loss": 0.9292824268341064 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.47979944944381714, - "train/contrastive_loss": 1.1162347793579102, - "train/negative_loss": 0.769817054271698, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.34641775488853455, - "train/total_loss": 0.703046441078186 + "train/classification_loss": 0.5053771734237671, + "train/contrastive_loss": 2.6655991077423096, + "train/negative_loss": 2.5037496089935303, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.16184940934181213, + "train/total_loss": 1.038496971130371 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.4981132447719574, - "train/contrastive_loss": 1.5237772464752197, - "train/negative_loss": 1.4966115951538086, - "train/num_negatives": 44, + "train/classification_loss": 0.5328295230865479, + "train/contrastive_loss": 2.116457462310791, + "train/negative_loss": 2.068239688873291, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.04821772873401642, + "train/total_loss": 0.956121027469635 + }, + { + "epoch": 0.7604562737642585, + "step": 100, + "train/classification_loss": 0.510808527469635, + "train/contrastive_loss": 1.6296155452728271, + "train/negative_loss": 1.6142444610595703, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.015371120534837246, + "train/total_loss": 0.8367316722869873 + }, + { + "epoch": 0.7604562737642585, + "step": 100, + "train/classification_loss": 0.5183000564575195, + "train/contrastive_loss": 1.937532901763916, + "train/negative_loss": 1.7127246856689453, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.027165664359927177, - "train/total_loss": 0.8028687238693237 + "train/positive_loss": 0.2248082160949707, + "train/total_loss": 0.9058066606521606 + }, + { + "epoch": 0.7604562737642585, + "step": 100, + "train/classification_loss": 0.5386996865272522, + "train/contrastive_loss": 2.7680375576019287, + "train/negative_loss": 2.652676582336426, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.11536108702421188, + "train/total_loss": 1.092307209968567 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.49897071719169617, - "train/contrastive_loss": 2.2621116638183594, - "train/negative_loss": 2.0091710090637207, + "train/classification_loss": 0.5218415856361389, + "train/contrastive_loss": 2.6983625888824463, + "train/negative_loss": 2.299687385559082, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 0.2529405653476715, - "train/total_loss": 0.9513930678367615 + "train/positive_loss": 0.39867523312568665, + "train/total_loss": 1.061514139175415 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.4825623035430908, - "train/contrastive_loss": 1.6604968309402466, - "train/negative_loss": 1.4046852588653564, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 0.2558116018772125, - "train/total_loss": 0.8146616816520691 + "train/classification_loss": 0.5185121297836304, + "train/contrastive_loss": 1.8179523944854736, + "train/negative_loss": 1.7571269273757935, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.06082548201084137, + "train/total_loss": 0.8821026086807251 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.4876653552055359, - "train/contrastive_loss": 2.3859760761260986, - "train/negative_loss": 1.6625949144363403, - "train/num_negatives": 32, + "train/classification_loss": 0.5215716361999512, + "train/contrastive_loss": 2.848592758178711, + "train/negative_loss": 2.364572763442993, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.4840199947357178, + "train/total_loss": 1.0912902355194092 + }, + { + "epoch": 0.7604562737642585, + "step": 100, + "train/classification_loss": 0.518670916557312, + "train/contrastive_loss": 3.8867077827453613, + "train/negative_loss": 3.869272470474243, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 0.7233811020851135, - "train/total_loss": 0.9648605585098267 + "train/positive_loss": 0.017435358837246895, + "train/total_loss": 1.2960124015808105 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.49910062551498413, - "train/contrastive_loss": 2.3952174186706543, - "train/negative_loss": 2.2775075435638428, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.11770983785390854, - "train/total_loss": 0.978144109249115 + "train/classification_loss": 0.5266229510307312, + "train/contrastive_loss": 3.3826851844787598, + "train/negative_loss": 3.2217044830322266, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.16098065674304962, + "train/total_loss": 1.203160047531128 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.5102370977401733, - "train/contrastive_loss": 1.333275318145752, - "train/negative_loss": 1.241356372833252, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.09191899001598358, - "train/total_loss": 0.7768921852111816 + "train/classification_loss": 0.5248523950576782, + "train/contrastive_loss": 1.8465431928634644, + "train/negative_loss": 1.5474315881729126, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.29911160469055176, + "train/total_loss": 0.8941610455513 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.5128220915794373, - "train/contrastive_loss": 2.3772125244140625, - "train/negative_loss": 1.541087031364441, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.836125373840332, - "train/total_loss": 0.9882645606994629 + "train/classification_loss": 0.5384407043457031, + "train/contrastive_loss": 3.009988307952881, + "train/negative_loss": 2.078011989593506, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 0.9319764375686646, + "train/total_loss": 1.1404383182525635 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.5104720592498779, - "train/contrastive_loss": 2.1815783977508545, - "train/negative_loss": 1.3918211460113525, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 0.7897571921348572, - "train/total_loss": 0.9467877149581909 + "train/classification_loss": 0.5157962441444397, + "train/contrastive_loss": 3.2546281814575195, + "train/negative_loss": 3.2008092403411865, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.0538189634680748, + "train/total_loss": 1.1667218208312988 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.4990098178386688, - "train/contrastive_loss": 2.5549020767211914, - "train/negative_loss": 2.535061836242676, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.01984027400612831, - "train/total_loss": 1.0099902153015137 + "train/classification_loss": 0.5171838402748108, + "train/contrastive_loss": 1.9717735052108765, + "train/negative_loss": 1.7365847826004028, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.23518876731395721, + "train/total_loss": 0.9115385413169861 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.4887517988681793, - "train/contrastive_loss": 1.7596021890640259, - "train/negative_loss": 1.5423253774642944, + "train/classification_loss": 0.5265250205993652, + "train/contrastive_loss": 2.083327293395996, + "train/negative_loss": 1.846558690071106, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 0.21727681159973145, - "train/total_loss": 0.8406722545623779 + "train/positive_loss": 0.23676852881908417, + "train/total_loss": 0.9431904554367065 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.48168495297431946, - "train/contrastive_loss": 2.4231743812561035, - "train/negative_loss": 1.933935523033142, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.4892389476299286, - "train/total_loss": 0.9663197994232178 + "train/classification_loss": 0.5085857510566711, + "train/contrastive_loss": 2.695046901702881, + "train/negative_loss": 2.656100034713745, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.03894675523042679, + "train/total_loss": 1.0475951433181763 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.4779914915561676, - "train/contrastive_loss": 1.1248044967651367, - "train/negative_loss": 1.1038047075271606, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.02099980041384697, - "train/total_loss": 0.7029523849487305 + "train/classification_loss": 0.5178019404411316, + "train/contrastive_loss": 3.5475518703460693, + "train/negative_loss": 3.527902603149414, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.01964922808110714, + "train/total_loss": 1.2273123264312744 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.51288902759552, - "train/contrastive_loss": 1.4161746501922607, - "train/negative_loss": 1.4080833196640015, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.008091272786259651, - "train/total_loss": 0.7961239814758301 + "train/classification_loss": 0.5271193981170654, + "train/contrastive_loss": 3.572105646133423, + "train/negative_loss": 2.9463143348693848, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.6257913708686829, + "train/total_loss": 1.241540551185608 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.4913043975830078, - "train/contrastive_loss": 1.4710538387298584, - "train/negative_loss": 1.4276061058044434, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.04344775900244713, - "train/total_loss": 0.7855151891708374 + "train/classification_loss": 0.5302030444145203, + "train/contrastive_loss": 2.121208429336548, + "train/negative_loss": 2.0719375610351562, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.04927096888422966, + "train/total_loss": 0.9544447660446167 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.4946765899658203, - "train/contrastive_loss": 2.800309181213379, - "train/negative_loss": 2.1733720302581787, - "train/num_negatives": 46, + "train/classification_loss": 0.5271154046058655, + "train/contrastive_loss": 2.9539458751678467, + "train/negative_loss": 2.7680017948150635, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.6269370913505554, - "train/total_loss": 1.0547385215759277 + "train/positive_loss": 0.18594402074813843, + "train/total_loss": 1.1179046630859375 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.4985974133014679, - "train/contrastive_loss": 2.4790587425231934, - "train/negative_loss": 2.3625760078430176, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.11648285388946533, - "train/total_loss": 0.9944092035293579 + "train/classification_loss": 0.5330164432525635, + "train/contrastive_loss": 2.1491127014160156, + "train/negative_loss": 2.133575201034546, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.015537448227405548, + "train/total_loss": 0.9628390073776245 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.5010672807693481, - "train/contrastive_loss": 2.1310417652130127, - "train/negative_loss": 1.913866639137268, + "train/classification_loss": 0.5302675366401672, + "train/contrastive_loss": 1.6204469203948975, + "train/negative_loss": 1.606399416923523, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.21717506647109985, - "train/total_loss": 0.9272756576538086 + "train/num_positives": 8, + "train/positive_loss": 0.014047529548406601, + "train/total_loss": 0.8543568849563599 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.506515622138977, - "train/contrastive_loss": 2.602304458618164, - "train/negative_loss": 1.6453665494918823, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.9569379091262817, - "train/total_loss": 1.0269765853881836 + "train/classification_loss": 0.5127363801002502, + "train/contrastive_loss": 3.6104438304901123, + "train/negative_loss": 3.4512603282928467, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.15918344259262085, + "train/total_loss": 1.2348251342773438 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.49669748544692993, - "train/contrastive_loss": 1.8191921710968018, - "train/negative_loss": 1.5493022203445435, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.2698900103569031, - "train/total_loss": 0.8605359196662903 + "train/classification_loss": 0.514090359210968, + "train/contrastive_loss": 3.596312999725342, + "train/negative_loss": 3.4683454036712646, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.12796764075756073, + "train/total_loss": 1.2333528995513916 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.48844069242477417, - "train/contrastive_loss": 2.409947395324707, - "train/negative_loss": 2.0951945781707764, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.31475287675857544, - "train/total_loss": 0.9704301357269287 + "train/classification_loss": 0.5109968781471252, + "train/contrastive_loss": 4.206634521484375, + "train/negative_loss": 4.10882568359375, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.09780905395746231, + "train/total_loss": 1.3523237705230713 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.49638253450393677, - "train/contrastive_loss": 2.3796138763427734, - "train/negative_loss": 2.1489663124084473, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 0.23064753413200378, - "train/total_loss": 0.9723052978515625 + "train/classification_loss": 0.5175558924674988, + "train/contrastive_loss": 2.12876558303833, + "train/negative_loss": 2.128297805786133, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.00046788668259978294, + "train/total_loss": 0.9433090090751648 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.48586687445640564, - "train/contrastive_loss": 2.26029372215271, - "train/negative_loss": 2.165158987045288, - "train/num_negatives": 42, + "train/classification_loss": 0.5331130623817444, + "train/contrastive_loss": 1.9256881475448608, + "train/negative_loss": 1.910994052886963, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.09513482451438904, - "train/total_loss": 0.9379256367683411 + "train/positive_loss": 0.01469406858086586, + "train/total_loss": 0.9182506799697876 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.49845436215400696, - "train/contrastive_loss": 2.0704336166381836, - "train/negative_loss": 1.1410434246063232, - "train/num_negatives": 48, + "train/classification_loss": 0.5323390960693359, + "train/contrastive_loss": 2.4557032585144043, + "train/negative_loss": 2.4130325317382812, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.9293901920318604, - "train/total_loss": 0.9125410914421082 + "train/positive_loss": 0.04267076402902603, + "train/total_loss": 1.023479700088501 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.4814828932285309, - "train/contrastive_loss": 1.6405832767486572, - "train/negative_loss": 1.5608952045440674, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.07968801259994507, - "train/total_loss": 0.8095995187759399 + "train/classification_loss": 0.5137200951576233, + "train/contrastive_loss": 3.1115944385528564, + "train/negative_loss": 2.912454605102539, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.19913983345031738, + "train/total_loss": 1.1360390186309814 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.49443626403808594, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.49443626403808594 - }, - { - "epoch": 0.9569377990430622, - "eval_exact_match_accuracy": 0.0, - "eval_hamming_loss": 0.08302842668167745, - "eval_loss": 0.8925831913948059, - "eval_macro_f1": 0.0029411764705882353, - "eval_macro_precision": 0.0392156862745098, - "eval_macro_recall": 0.0015278838808250575, - "eval_micro_f1": 0.013377926421404682, - "eval_micro_precision": 0.6666666666666666, - "eval_micro_recall": 0.006756756756756757, - "eval_runtime": 1.0666, - "eval_samples_per_second": 195.952, - "eval_steps_per_second": 25.314, + "train/classification_loss": 0.5359312891960144, + "train/contrastive_loss": 2.2100601196289062, + "train/negative_loss": 1.983330488204956, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.22672957181930542, + "train/total_loss": 0.9779433012008667 + }, + { + "epoch": 0.7604562737642585, + "step": 100, + "train/classification_loss": 0.5138111114501953, + "train/contrastive_loss": 2.355898141860962, + "train/negative_loss": 2.165081024169922, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.19081711769104004, + "train/total_loss": 0.9849907159805298 + }, + { + "epoch": 0.7604562737642585, + "eval_exact_match_accuracy": 0.06870229007633588, + "eval_hamming_loss": 0.08868432869330939, + "eval_loss": 1.0477241277694702, + "eval_macro_f1": 0.028261583963194215, + "eval_macro_precision": 0.030112044817927168, + "eval_macro_recall": 0.026625386996904025, + "eval_micro_f1": 0.1787941787941788, + "eval_micro_precision": 0.48314606741573035, + "eval_micro_recall": 0.1096938775510204, + "eval_runtime": 1.3109, + "eval_samples_per_second": 199.859, + "eval_steps_per_second": 25.173, "step": 100 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.5208109021186829, - "train/contrastive_loss": 1.4805831909179688, - "train/negative_loss": 1.0015603303909302, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.47902292013168335, - "train/total_loss": 0.8169275522232056 + "train/classification_loss": 0.5223861336708069, + "train/contrastive_loss": 1.9081828594207764, + "train/negative_loss": 1.572178602218628, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.33600425720214844, + "train/total_loss": 0.9040226936340332 }, { - "epoch": 0.9569377990430622, + "epoch": 0.7604562737642585, "step": 100, - "train/classification_loss": 0.4976999759674072, - "train/contrastive_loss": 1.3909823894500732, - "train/negative_loss": 0.7713649272918701, - "train/num_negatives": 32, - "train/num_positives": 24, - "train/positive_loss": 0.6196174025535583, - "train/total_loss": 0.775896430015564 + "train/classification_loss": 0.5333789587020874, + "train/contrastive_loss": 1.7644151449203491, + "train/negative_loss": 1.6611601114273071, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.1032550036907196, + "train/total_loss": 0.8862619996070862 }, { - "epoch": 1.430622009569378, - "grad_norm": 16.709274291992188, - "learning_rate": 1.9982061068702293e-05, - "loss": 1.5305, + "epoch": 1.1368821292775666, + "grad_norm": 19.18036651611328, + "learning_rate": 1.9985735963581184e-05, + "loss": 1.591, "step": 150 }, { - "epoch": 1.430622009569378, + "epoch": 1.1368821292775666, "step": 150, - "train/classification_loss": 0.41379594802856445, - "train/contrastive_loss": 2.1727852821350098, - "train/negative_loss": 1.4437021017074585, - "train/num_negatives": 40, - "train/num_positives": 6, - "train/positive_loss": 0.7290831804275513, - "train/total_loss": 0.8483530282974243 + "train/classification_loss": 0.42077043652534485, + "train/contrastive_loss": 1.8482379913330078, + "train/negative_loss": 1.552756905555725, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 0.2954811453819275, + "train/total_loss": 0.7904180288314819 }, { - "epoch": 1.430622009569378, + "epoch": 1.1368821292775666, "step": 150, - "train/classification_loss": 0.37484118342399597, - "train/contrastive_loss": 1.5184831619262695, - "train/negative_loss": 1.1229286193847656, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.3955545425415039, - "train/total_loss": 0.6785378456115723 + "train/classification_loss": 0.402802973985672, + "train/contrastive_loss": 1.3415889739990234, + "train/negative_loss": 1.311341643333435, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.030247293412685394, + "train/total_loss": 0.6711207628250122 }, { - "epoch": 1.9090909090909092, - "grad_norm": 14.846596717834473, - "learning_rate": 1.996297709923664e-05, - "loss": 1.3499, + "epoch": 1.5171102661596958, + "grad_norm": 28.48988151550293, + "learning_rate": 1.997056145675266e-05, + "loss": 1.3625, "step": 200 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.31925636529922485, - "train/contrastive_loss": 2.2801711559295654, - "train/negative_loss": 2.1692323684692383, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.11093886941671371, - "train/total_loss": 0.7752906084060669 + "train/classification_loss": 0.32900288701057434, + "train/contrastive_loss": 1.1824053525924683, + "train/negative_loss": 0.9582441449165344, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.22416117787361145, + "train/total_loss": 0.5654839277267456 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.28812292218208313, - "train/contrastive_loss": 1.6087501049041748, - "train/negative_loss": 1.5858157873153687, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.022934362292289734, - "train/total_loss": 0.6098729372024536 + "train/classification_loss": 0.36390191316604614, + "train/contrastive_loss": 2.3821651935577393, + "train/negative_loss": 2.36857271194458, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.013592593371868134, + "train/total_loss": 0.840334951877594 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.3198684751987457, - "train/contrastive_loss": 1.445984125137329, - "train/negative_loss": 1.3667182922363281, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.0792657881975174, - "train/total_loss": 0.6090652942657471 + "train/classification_loss": 0.31767538189888, + "train/contrastive_loss": 1.8386619091033936, + "train/negative_loss": 1.7620638608932495, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.07659809291362762, + "train/total_loss": 0.6854077577590942 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.3194320499897003, - "train/contrastive_loss": 2.8762643337249756, - "train/negative_loss": 2.5007057189941406, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.3755585849285126, - "train/total_loss": 0.894684910774231 + "train/classification_loss": 0.3647700548171997, + "train/contrastive_loss": 1.7594399452209473, + "train/negative_loss": 1.3987864255905151, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.3606535792350769, + "train/total_loss": 0.7166580557823181 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.29062312841415405, - "train/contrastive_loss": 1.9086765050888062, - "train/negative_loss": 1.7826333045959473, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 0.12604321539402008, - "train/total_loss": 0.6723583936691284 + "train/classification_loss": 0.3093717396259308, + "train/contrastive_loss": 0.7735762000083923, + "train/negative_loss": 0.7734593152999878, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.00011689532402670011, + "train/total_loss": 0.46408697962760925 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.3190228343009949, - "train/contrastive_loss": 4.001949787139893, - "train/negative_loss": 2.836303234100342, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.1656465530395508, - "train/total_loss": 1.1194127798080444 + "train/classification_loss": 0.34929248690605164, + "train/contrastive_loss": 1.4811104536056519, + "train/negative_loss": 1.4718199968338013, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.009290502406656742, + "train/total_loss": 0.6455146074295044 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.3398359715938568, - "train/contrastive_loss": 2.6246445178985596, - "train/negative_loss": 2.5920016765594482, + "train/classification_loss": 0.36613598465919495, + "train/contrastive_loss": 2.7939929962158203, + "train/negative_loss": 2.585693359375, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.03264281153678894, - "train/total_loss": 0.864764928817749 + "train/num_positives": 8, + "train/positive_loss": 0.20829962193965912, + "train/total_loss": 0.9249346256256104 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.35659971833229065, - "train/contrastive_loss": 2.2124276161193848, - "train/negative_loss": 2.1679036617279053, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.04452385753393173, - "train/total_loss": 0.799085259437561 + "train/classification_loss": 0.3568226397037506, + "train/contrastive_loss": 2.4076409339904785, + "train/negative_loss": 1.525605320930481, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.8820356726646423, + "train/total_loss": 0.8383508324623108 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.34821346402168274, - "train/contrastive_loss": 2.296016216278076, - "train/negative_loss": 1.8387569189071655, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.4572594165802002, - "train/total_loss": 0.8074166774749756 + "train/classification_loss": 0.3294413089752197, + "train/contrastive_loss": 0.8395364284515381, + "train/negative_loss": 0.8260493874549866, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.013487019576132298, + "train/total_loss": 0.4973486065864563 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.34547069668769836, - "train/contrastive_loss": 2.447352409362793, - "train/negative_loss": 1.5934425592422485, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 0.8539099097251892, - "train/total_loss": 0.8349411487579346 + "train/classification_loss": 0.34261563420295715, + "train/contrastive_loss": 3.0652570724487305, + "train/negative_loss": 2.4933691024780273, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.5718878507614136, + "train/total_loss": 0.9556670188903809 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.337612122297287, - "train/contrastive_loss": 3.261737108230591, - "train/negative_loss": 3.2172765731811523, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.044460564851760864, - "train/total_loss": 0.9899595975875854 + "train/classification_loss": 0.332425057888031, + "train/contrastive_loss": 3.9916396141052246, + "train/negative_loss": 3.988046169281006, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.0035935600753873587, + "train/total_loss": 1.1307530403137207 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.3123414218425751, - "train/contrastive_loss": 1.4541531801223755, - "train/negative_loss": 1.4199692010879517, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.03418395295739174, - "train/total_loss": 0.6031720638275146 + "train/classification_loss": 0.35452327132225037, + "train/contrastive_loss": 3.530172824859619, + "train/negative_loss": 2.8470876216888428, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.6830852031707764, + "train/total_loss": 1.0605578422546387 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.30551061034202576, - "train/contrastive_loss": 3.13470721244812, - "train/negative_loss": 2.8316569328308105, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.3030502498149872, - "train/total_loss": 0.9324520826339722 + "train/classification_loss": 0.34451693296432495, + "train/contrastive_loss": 2.537248134613037, + "train/negative_loss": 1.986166000366211, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.551082193851471, + "train/total_loss": 0.8519665598869324 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.2891707718372345, - "train/contrastive_loss": 1.3312711715698242, - "train/negative_loss": 1.3292616605758667, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.0020095123909413815, - "train/total_loss": 0.5554250478744507 + "train/classification_loss": 0.3750525712966919, + "train/contrastive_loss": 4.9051408767700195, + "train/negative_loss": 1.7746000289916992, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 3.1305408477783203, + "train/total_loss": 1.3560807704925537 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.35660141706466675, - "train/contrastive_loss": 1.451112985610962, - "train/negative_loss": 1.4425424337387085, + "train/classification_loss": 0.3379116952419281, + "train/contrastive_loss": 2.6428987979888916, + "train/negative_loss": 2.6106889247894287, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.008570589125156403, - "train/total_loss": 0.6468240022659302 + "train/positive_loss": 0.03220982477068901, + "train/total_loss": 0.866491436958313 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.3162277340888977, - "train/contrastive_loss": 1.7718696594238281, - "train/negative_loss": 1.763313889503479, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.008555769920349121, - "train/total_loss": 0.6706016659736633 + "train/classification_loss": 0.3371126651763916, + "train/contrastive_loss": 1.1804876327514648, + "train/negative_loss": 1.1787278652191162, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0017597917467355728, + "train/total_loss": 0.5732101798057556 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.32514408230781555, - "train/contrastive_loss": 3.159630298614502, - "train/negative_loss": 2.8428165912628174, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.3168136179447174, - "train/total_loss": 0.9570701122283936 + "train/classification_loss": 0.356777161359787, + "train/contrastive_loss": 1.3955410718917847, + "train/negative_loss": 1.3798600435256958, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.015681061893701553, + "train/total_loss": 0.6358853578567505 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.3323442339897156, - "train/contrastive_loss": 3.6162667274475098, - "train/negative_loss": 3.55619740486145, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.0600692480802536, - "train/total_loss": 1.0555975437164307 + "train/classification_loss": 0.32341086864471436, + "train/contrastive_loss": 2.155562162399292, + "train/negative_loss": 2.086655616760254, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.06890648603439331, + "train/total_loss": 0.7545232772827148 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.3296099603176117, - "train/contrastive_loss": 2.3942296504974365, - "train/negative_loss": 2.155193328857422, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.23903624713420868, - "train/total_loss": 0.8084558844566345 + "train/classification_loss": 0.3432146906852722, + "train/contrastive_loss": 2.8056225776672363, + "train/negative_loss": 2.7048938274383545, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.1007288247346878, + "train/total_loss": 0.9043391942977905 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.33709946274757385, - "train/contrastive_loss": 3.315659999847412, - "train/negative_loss": 2.9742629528045654, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.34139713644981384, - "train/total_loss": 1.0002315044403076 + "train/classification_loss": 0.3556811511516571, + "train/contrastive_loss": 2.9596049785614014, + "train/negative_loss": 2.557649612426758, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.40195536613464355, + "train/total_loss": 0.9476021528244019 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.3299678862094879, - "train/contrastive_loss": 3.4666600227355957, - "train/negative_loss": 2.2976880073547363, - "train/num_negatives": 44, + "train/classification_loss": 0.3746018707752228, + "train/contrastive_loss": 1.8797647953033447, + "train/negative_loss": 1.8710613250732422, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.008703440427780151, + "train/total_loss": 0.7505548000335693 + }, + { + "epoch": 1.5171102661596958, + "step": 200, + "train/classification_loss": 0.36530470848083496, + "train/contrastive_loss": 2.200697660446167, + "train/negative_loss": 2.084299325942993, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.1689718961715698, - "train/total_loss": 1.0232999324798584 + "train/positive_loss": 0.11639823764562607, + "train/total_loss": 0.8054442405700684 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.30479738116264343, - "train/contrastive_loss": 3.6985039710998535, - "train/negative_loss": 3.4958419799804688, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.20266187191009521, - "train/total_loss": 1.0444982051849365 + "train/classification_loss": 0.35285484790802, + "train/contrastive_loss": 2.247840166091919, + "train/negative_loss": 2.2334108352661133, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.014429239556193352, + "train/total_loss": 0.8024228811264038 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.31925779581069946, - "train/contrastive_loss": 2.8711631298065186, - "train/negative_loss": 2.678241014480591, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 0.1929221898317337, - "train/total_loss": 0.8934904336929321 + "train/classification_loss": 0.36981505155563354, + "train/contrastive_loss": 1.6797798871994019, + "train/negative_loss": 1.6776148080825806, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 0.0021651226561516523, + "train/total_loss": 0.7057710289955139 + }, + { + "epoch": 1.5171102661596958, + "step": 200, + "train/classification_loss": 0.31847327947616577, + "train/contrastive_loss": 4.027771949768066, + "train/negative_loss": 3.697258710861206, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.3305131494998932, + "train/total_loss": 1.1240277290344238 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.29547131061553955, - "train/contrastive_loss": 2.1167469024658203, - "train/negative_loss": 2.1081838607788086, + "train/classification_loss": 0.33062344789505005, + "train/contrastive_loss": 2.8953511714935303, + "train/negative_loss": 2.7430531978607178, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.008562983013689518, - "train/total_loss": 0.7188206911087036 + "train/positive_loss": 0.15229807794094086, + "train/total_loss": 0.909693717956543 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.32729917764663696, - "train/contrastive_loss": 2.0170092582702637, - "train/negative_loss": 1.738647699356079, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.27836158871650696, - "train/total_loss": 0.7307010293006897 + "train/classification_loss": 0.3354160487651825, + "train/contrastive_loss": 3.913938283920288, + "train/negative_loss": 3.8658928871154785, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.0480453185737133, + "train/total_loss": 1.1182037591934204 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.3033764660358429, - "train/contrastive_loss": 2.3427703380584717, - "train/negative_loss": 2.112016201019287, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.23075424134731293, - "train/total_loss": 0.7719305753707886 + "train/classification_loss": 0.33298665285110474, + "train/contrastive_loss": 1.8301345109939575, + "train/negative_loss": 1.830007553100586, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.00012696595513261855, + "train/total_loss": 0.6990135908126831 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.3208795487880707, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.3208795487880707 - }, - { - "epoch": 1.9090909090909092, - "eval_exact_match_accuracy": 0.0, - "eval_hamming_loss": 0.08330987897551365, - "eval_loss": 0.8202692866325378, - "eval_macro_f1": 0.0, - "eval_macro_precision": 0.0, - "eval_macro_recall": 0.0, - "eval_micro_f1": 0.0, - "eval_micro_precision": 0.0, - "eval_micro_recall": 0.0, - "eval_runtime": 1.0447, - "eval_samples_per_second": 200.064, - "eval_steps_per_second": 25.846, - "step": 200 + "train/classification_loss": 0.380653440952301, + "train/contrastive_loss": 1.3187235593795776, + "train/negative_loss": 0.9294828772544861, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 0.38924065232276917, + "train/total_loss": 0.6443981528282166 + }, + { + "epoch": 1.5171102661596958, + "step": 200, + "train/classification_loss": 0.36259859800338745, + "train/contrastive_loss": 1.697849154472351, + "train/negative_loss": 1.56440007686615, + "train/num_negatives": 46, + "train/num_positives": 8, + "train/positive_loss": 0.13344910740852356, + "train/total_loss": 0.7021684646606445 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.3305688798427582, - "train/contrastive_loss": 0.8643814325332642, - "train/negative_loss": 0.6721360087394714, + "train/classification_loss": 0.3377118408679962, + "train/contrastive_loss": 2.374089479446411, + "train/negative_loss": 1.9461544752120972, "train/num_negatives": 40, "train/num_positives": 16, - "train/positive_loss": 0.1922454535961151, - "train/total_loss": 0.5034451484680176 + "train/positive_loss": 0.4279349446296692, + "train/total_loss": 0.8125297427177429 }, { - "epoch": 1.9090909090909092, + "epoch": 1.5171102661596958, "step": 200, - "train/classification_loss": 0.32394587993621826, - "train/contrastive_loss": 2.567824363708496, - "train/negative_loss": 1.3685816526412964, - "train/num_negatives": 44, + "train/classification_loss": 0.37744829058647156, + "train/contrastive_loss": 2.163613796234131, + "train/negative_loss": 1.860403060913086, + "train/num_negatives": 42, "train/num_positives": 12, - "train/positive_loss": 1.1992425918579102, - "train/total_loss": 0.8375107645988464 - }, - { - "epoch": 2.382775119617225, - "grad_norm": 19.6639347076416, - "learning_rate": 1.9943893129770995e-05, - "loss": 1.2147, - "step": 250 + "train/positive_loss": 0.3032108247280121, + "train/total_loss": 0.8101710677146912 }, { - "epoch": 2.382775119617225, - "step": 250, - "train/classification_loss": 0.27953842282295227, - "train/contrastive_loss": 2.651360511779785, - "train/negative_loss": 1.5591284036636353, + "epoch": 1.5171102661596958, + "step": 200, + "train/classification_loss": 0.32499366998672485, + "train/contrastive_loss": 3.2692909240722656, + "train/negative_loss": 1.5369350910186768, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 1.7323558330535889, + "train/total_loss": 0.978851854801178 + }, + { + "epoch": 1.5171102661596958, + "eval_exact_match_accuracy": 0.08396946564885496, + "eval_hamming_loss": 0.07970363718006286, + "eval_loss": 0.8193506598472595, + "eval_macro_f1": 0.03577106518282989, + "eval_macro_precision": 0.049944506104328525, + "eval_macro_recall": 0.02786377708978328, + "eval_micro_f1": 0.20224719101123595, + "eval_micro_precision": 0.8490566037735849, + "eval_micro_recall": 0.11479591836734694, + "eval_runtime": 1.3192, + "eval_samples_per_second": 198.603, + "eval_steps_per_second": 25.015, + "step": 200 + }, + { + "epoch": 1.5171102661596958, + "step": 200, + "train/classification_loss": 0.350442111492157, + "train/contrastive_loss": 1.9061377048492432, + "train/negative_loss": 1.6962623596191406, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.20987534523010254, + "train/total_loss": 0.7316696643829346 + }, + { + "epoch": 1.5171102661596958, + "step": 200, + "train/classification_loss": 0.37069669365882874, + "train/contrastive_loss": 1.7428470849990845, + "train/negative_loss": 1.0726768970489502, "train/num_negatives": 52, "train/num_positives": 4, - "train/positive_loss": 1.0922319889068604, - "train/total_loss": 0.8098105192184448 + "train/positive_loss": 0.6701701879501343, + "train/total_loss": 0.7192661166191101 }, { - "epoch": 2.382775119617225, + "epoch": 1.8973384030418252, + "grad_norm": 20.050065994262695, + "learning_rate": 1.995538694992413e-05, + "loss": 1.2864, + "step": 250 + }, + { + "epoch": 1.8973384030418252, "step": 250, - "train/classification_loss": 0.27353185415267944, - "train/contrastive_loss": 1.47084379196167, - "train/negative_loss": 1.268683910369873, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.2021598219871521, - "train/total_loss": 0.5677006244659424 + "train/classification_loss": 0.3013644218444824, + "train/contrastive_loss": 1.8454833030700684, + "train/negative_loss": 1.0752781629562378, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.7702051997184753, + "train/total_loss": 0.6704610586166382 + }, + { + "epoch": 1.8973384030418252, + "step": 250, + "train/classification_loss": 0.3350011706352234, + "train/contrastive_loss": 1.5934147834777832, + "train/negative_loss": 1.323768138885498, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.26964661478996277, + "train/total_loss": 0.653684139251709 }, { - "epoch": 2.861244019138756, - "grad_norm": 18.294469833374023, - "learning_rate": 1.9924809160305346e-05, - "loss": 1.1328, + "epoch": 2.273764258555133, + "grad_norm": 14.1633939743042, + "learning_rate": 1.99402124430956e-05, + "loss": 1.1979, "step": 300 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.26580262184143066, - "train/contrastive_loss": 1.5227141380310059, - "train/negative_loss": 1.427420973777771, + "train/classification_loss": 0.25491616129875183, + "train/contrastive_loss": 1.5536291599273682, + "train/negative_loss": 1.533349871635437, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.020279258489608765, + "train/total_loss": 0.5656419992446899 + }, + { + "epoch": 2.273764258555133, + "step": 300, + "train/classification_loss": 0.29954853653907776, + "train/contrastive_loss": 2.599911689758301, + "train/negative_loss": 2.5576469898223877, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.09529320150613785, - "train/total_loss": 0.5703454613685608 + "train/num_positives": 6, + "train/positive_loss": 0.042264729738235474, + "train/total_loss": 0.8195308446884155 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.2227538675069809, - "train/contrastive_loss": 1.7876951694488525, - "train/negative_loss": 1.4747518301010132, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.31294330954551697, - "train/total_loss": 0.5802928805351257 + "train/classification_loss": 0.24273481965065002, + "train/contrastive_loss": 2.6730997562408447, + "train/negative_loss": 2.667206048965454, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.0058936551213264465, + "train/total_loss": 0.7773548364639282 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.2613501250743866, - "train/contrastive_loss": 1.9958046674728394, - "train/negative_loss": 1.8843146562576294, - "train/num_negatives": 44, + "train/classification_loss": 0.31181463599205017, + "train/contrastive_loss": 1.8219068050384521, + "train/negative_loss": 1.6200464963912964, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.201860249042511, + "train/total_loss": 0.6761959791183472 + }, + { + "epoch": 2.273764258555133, + "step": 300, + "train/classification_loss": 0.23089000582695007, + "train/contrastive_loss": 1.2907836437225342, + "train/negative_loss": 1.290390968322754, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.0003927172801923007, + "train/total_loss": 0.48904675245285034 + }, + { + "epoch": 2.273764258555133, + "step": 300, + "train/classification_loss": 0.29317009449005127, + "train/contrastive_loss": 1.674552083015442, + "train/negative_loss": 1.668002963066101, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.11149000376462936, - "train/total_loss": 0.6605110764503479 + "train/positive_loss": 0.0065490929409861565, + "train/total_loss": 0.6280804872512817 + }, + { + "epoch": 2.273764258555133, + "step": 300, + "train/classification_loss": 0.3076646625995636, + "train/contrastive_loss": 2.9504120349884033, + "train/negative_loss": 2.5791590213775635, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.37125295400619507, + "train/total_loss": 0.8977470397949219 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.25792014598846436, - "train/contrastive_loss": 3.114227771759033, - "train/negative_loss": 2.221250534057617, + "train/classification_loss": 0.29427385330200195, + "train/contrastive_loss": 3.0223915576934814, + "train/negative_loss": 2.748037338256836, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 0.8929773569107056, - "train/total_loss": 0.8807657361030579 + "train/positive_loss": 0.27435430884361267, + "train/total_loss": 0.8987521529197693 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.22696107625961304, - "train/contrastive_loss": 2.8760719299316406, - "train/negative_loss": 2.26186466217041, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 0.61420738697052, - "train/total_loss": 0.8021754622459412 + "train/classification_loss": 0.2589974105358124, + "train/contrastive_loss": 0.8304426074028015, + "train/negative_loss": 0.8259152173995972, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.004527402576059103, + "train/total_loss": 0.4250859320163727 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.2651028335094452, - "train/contrastive_loss": 2.950491189956665, - "train/negative_loss": 2.059633255004883, - "train/num_negatives": 32, + "train/classification_loss": 0.27714458107948303, + "train/contrastive_loss": 3.083094596862793, + "train/negative_loss": 2.773099422454834, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.30999520421028137, + "train/total_loss": 0.893763542175293 + }, + { + "epoch": 2.273764258555133, + "step": 300, + "train/classification_loss": 0.26757410168647766, + "train/contrastive_loss": 4.473361968994141, + "train/negative_loss": 4.472601890563965, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 0.8908578753471375, - "train/total_loss": 0.8552011251449585 + "train/positive_loss": 0.0007601076504215598, + "train/total_loss": 1.1622464656829834 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.2920515537261963, - "train/contrastive_loss": 2.7261154651641846, - "train/negative_loss": 2.702221632003784, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.023893848061561584, - "train/total_loss": 0.8372746706008911 + "train/classification_loss": 0.2899538278579712, + "train/contrastive_loss": 3.362851858139038, + "train/negative_loss": 3.308614492416382, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.05423736199736595, + "train/total_loss": 0.9625242352485657 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.3137778043746948, - "train/contrastive_loss": 1.5224679708480835, - "train/negative_loss": 1.5168673992156982, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.005600549280643463, - "train/total_loss": 0.6182714104652405 + "train/classification_loss": 0.28810834884643555, + "train/contrastive_loss": 3.006155014038086, + "train/negative_loss": 2.068119764328003, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.9380353093147278, + "train/total_loss": 0.8893393874168396 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.28973501920700073, - "train/contrastive_loss": 4.043308258056641, - "train/negative_loss": 2.8897502422332764, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 1.1535577774047852, - "train/total_loss": 1.0983966588974 + "train/classification_loss": 0.3197099566459656, + "train/contrastive_loss": 4.641449451446533, + "train/negative_loss": 2.166569948196411, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.474879503250122, + "train/total_loss": 1.247999906539917 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.29699039459228516, - "train/contrastive_loss": 3.7632551193237305, - "train/negative_loss": 2.7729926109313965, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 0.9902624487876892, - "train/total_loss": 1.0496413707733154 + "train/classification_loss": 0.28049102425575256, + "train/contrastive_loss": 3.283708095550537, + "train/negative_loss": 3.2608258724212646, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.02288232371211052, + "train/total_loss": 0.9372326135635376 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.2868160903453827, - "train/contrastive_loss": 2.7892770767211914, - "train/negative_loss": 2.786298990249634, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.0029780613258481026, - "train/total_loss": 0.8446714878082275 + "train/classification_loss": 0.2707643210887909, + "train/contrastive_loss": 1.6367957592010498, + "train/negative_loss": 1.6343282461166382, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0024674904998391867, + "train/total_loss": 0.5981234908103943 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.25049999356269836, - "train/contrastive_loss": 2.381927490234375, - "train/negative_loss": 2.3081092834472656, + "train/classification_loss": 0.2928057610988617, + "train/contrastive_loss": 1.5255415439605713, + "train/negative_loss": 1.512730360031128, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 0.07381822913885117, - "train/total_loss": 0.7268854975700378 + "train/positive_loss": 0.012811150401830673, + "train/total_loss": 0.5979140996932983 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.24846747517585754, - "train/contrastive_loss": 2.0870747566223145, - "train/negative_loss": 1.714982271194458, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.37209245562553406, - "train/total_loss": 0.6658824682235718 + "train/classification_loss": 0.2547549605369568, + "train/contrastive_loss": 2.4734504222869873, + "train/negative_loss": 2.3607325553894043, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.1127178892493248, + "train/total_loss": 0.7494450807571411 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.22282133996486664, - "train/contrastive_loss": 1.3425726890563965, - "train/negative_loss": 1.2993876934051514, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.0431850403547287, - "train/total_loss": 0.4913358688354492 + "train/classification_loss": 0.28654780983924866, + "train/contrastive_loss": 4.171542644500732, + "train/negative_loss": 4.114980697631836, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.05656217411160469, + "train/total_loss": 1.1208564043045044 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.3073972761631012, - "train/contrastive_loss": 1.7427904605865479, - "train/negative_loss": 1.7407640218734741, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.002026432193815708, - "train/total_loss": 0.6559553742408752 + "train/classification_loss": 0.29737022519111633, + "train/contrastive_loss": 3.2169342041015625, + "train/negative_loss": 2.6175084114074707, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.5994256734848022, + "train/total_loss": 0.9407570362091064 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.25809600949287415, - "train/contrastive_loss": 1.3023488521575928, - "train/negative_loss": 1.2532306909561157, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.049118172377347946, - "train/total_loss": 0.5185657739639282 + "train/classification_loss": 0.3295590281486511, + "train/contrastive_loss": 2.036266565322876, + "train/negative_loss": 2.026873826980591, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.009392729960381985, + "train/total_loss": 0.7368123531341553 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.2740459144115448, - "train/contrastive_loss": 4.054697036743164, - "train/negative_loss": 3.1549251079559326, - "train/num_negatives": 46, + "train/classification_loss": 0.31136491894721985, + "train/contrastive_loss": 2.412642240524292, + "train/negative_loss": 2.1899948120117188, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.8997721076011658, - "train/total_loss": 1.084985375404358 + "train/positive_loss": 0.22264735400676727, + "train/total_loss": 0.7938933372497559 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.2879021167755127, - "train/contrastive_loss": 2.8365402221679688, - "train/negative_loss": 2.6929051876068115, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.14363506436347961, - "train/total_loss": 0.8552101850509644 + "train/classification_loss": 0.29065966606140137, + "train/contrastive_loss": 2.265434980392456, + "train/negative_loss": 2.216989040374756, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.04844602569937706, + "train/total_loss": 0.7437466382980347 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.26835277676582336, - "train/contrastive_loss": 2.350980281829834, - "train/negative_loss": 2.2986252307891846, + "train/classification_loss": 0.30250295996665955, + "train/contrastive_loss": 1.790692687034607, + "train/negative_loss": 1.7903856039047241, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.052355166524648666, - "train/total_loss": 0.7385488748550415 + "train/num_positives": 8, + "train/positive_loss": 0.0003070286475121975, + "train/total_loss": 0.6606414914131165 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.29024118185043335, - "train/contrastive_loss": 4.434307098388672, - "train/negative_loss": 2.704580545425415, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 1.7297264337539673, - "train/total_loss": 1.1771025657653809 + "train/classification_loss": 0.24409355223178864, + "train/contrastive_loss": 4.328821659088135, + "train/negative_loss": 4.065413475036621, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.26340824365615845, + "train/total_loss": 1.1098579168319702 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.27999335527420044, - "train/contrastive_loss": 3.5914034843444824, - "train/negative_loss": 1.9110057353973389, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 1.680397868156433, - "train/total_loss": 0.9982740879058838 + "train/classification_loss": 0.26633745431900024, + "train/contrastive_loss": 3.708638906478882, + "train/negative_loss": 3.6673285961151123, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.04131036251783371, + "train/total_loss": 1.0080652236938477 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.24752801656723022, - "train/contrastive_loss": 2.314065456390381, - "train/negative_loss": 1.7939671277999878, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.5200983285903931, - "train/total_loss": 0.7103410959243774 + "train/classification_loss": 0.2741924524307251, + "train/contrastive_loss": 3.186722993850708, + "train/negative_loss": 3.1098177433013916, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.07690516114234924, + "train/total_loss": 0.9115370512008667 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.26041120290756226, - "train/contrastive_loss": 3.4482243061065674, - "train/negative_loss": 2.1723415851593018, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.2758827209472656, - "train/total_loss": 0.9500560760498047 + "train/classification_loss": 0.2638273537158966, + "train/contrastive_loss": 2.5175156593322754, + "train/negative_loss": 2.517482042312622, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 3.361758354003541e-05, + "train/total_loss": 0.767330527305603 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.22988000512123108, - "train/contrastive_loss": 1.649794101715088, - "train/negative_loss": 1.5332409143447876, - "train/num_negatives": 42, + "train/classification_loss": 0.33007335662841797, + "train/contrastive_loss": 0.9988154172897339, + "train/negative_loss": 0.9478541612625122, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.11655324697494507, - "train/total_loss": 0.5598388314247131 + "train/positive_loss": 0.05096126347780228, + "train/total_loss": 0.5298364162445068 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.2718926966190338, - "train/contrastive_loss": 3.006974697113037, - "train/negative_loss": 1.8715862035751343, - "train/num_negatives": 48, + "train/classification_loss": 0.3013105094432831, + "train/contrastive_loss": 1.8227550983428955, + "train/negative_loss": 1.7811139822006226, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 1.1353884935379028, - "train/total_loss": 0.8732876777648926 + "train/positive_loss": 0.04164108633995056, + "train/total_loss": 0.6658615469932556 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.2438923865556717, - "train/contrastive_loss": 1.988685965538025, - "train/negative_loss": 1.8516954183578491, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.13699057698249817, - "train/total_loss": 0.6416295766830444 + "train/classification_loss": 0.27776476740837097, + "train/contrastive_loss": 2.5091753005981445, + "train/negative_loss": 2.1760003566741943, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.33317500352859497, + "train/total_loss": 0.7795997858047485 + }, + { + "epoch": 2.273764258555133, + "step": 300, + "train/classification_loss": 0.32927829027175903, + "train/contrastive_loss": 1.9010207653045654, + "train/negative_loss": 1.8903042078018188, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.0107165751978755, + "train/total_loss": 0.7094824314117432 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.246324822306633, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.246324822306633 - }, - { - "epoch": 2.861244019138756, - "eval_exact_match_accuracy": 0.009569377990430622, - "eval_hamming_loss": 0.08274697438784126, - "eval_loss": 0.7837793231010437, - "eval_macro_f1": 0.0029784065524944155, + "train/classification_loss": 0.252132773399353, + "train/contrastive_loss": 3.9680917263031006, + "train/negative_loss": 3.402106761932373, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.5659850239753723, + "train/total_loss": 1.0457510948181152 + }, + { + "epoch": 2.273764258555133, + "eval_exact_match_accuracy": 0.019083969465648856, + "eval_hamming_loss": 0.08576560395150426, + "eval_loss": 0.8085073828697205, + "eval_macro_f1": 0.011204481792717087, "eval_macro_precision": 0.058823529411764705, - "eval_macro_recall": 0.0015278838808250575, - "eval_micro_f1": 0.013422818791946308, + "eval_macro_recall": 0.006191950464396284, + "eval_micro_f1": 0.04975124378109453, "eval_micro_precision": 1.0, - "eval_micro_recall": 0.006756756756756757, - "eval_runtime": 1.0566, - "eval_samples_per_second": 197.796, - "eval_steps_per_second": 25.553, + "eval_micro_recall": 0.025510204081632654, + "eval_runtime": 1.323, + "eval_samples_per_second": 198.037, + "eval_steps_per_second": 24.944, "step": 300 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.2793967127799988, - "train/contrastive_loss": 1.38784921169281, - "train/negative_loss": 1.262190818786621, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.12565839290618896, - "train/total_loss": 0.5569665431976318 + "train/classification_loss": 0.2768913805484772, + "train/contrastive_loss": 1.7243705987930298, + "train/negative_loss": 1.3448032140731812, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.37956735491752625, + "train/total_loss": 0.6217654943466187 }, { - "epoch": 2.861244019138756, + "epoch": 2.273764258555133, "step": 300, - "train/classification_loss": 0.25937533378601074, - "train/contrastive_loss": 1.5829660892486572, - "train/negative_loss": 1.0275195837020874, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.555446445941925, - "train/total_loss": 0.5759685635566711 + "train/classification_loss": 0.28591808676719666, + "train/contrastive_loss": 1.6158452033996582, + "train/negative_loss": 1.438856601715088, + "train/num_negatives": 38, + "train/num_positives": 18, + "train/positive_loss": 0.1769886016845703, + "train/total_loss": 0.6090871095657349 }, { - "epoch": 3.334928229665072, - "grad_norm": 13.323820114135742, - "learning_rate": 1.9905725190839694e-05, - "loss": 1.0738, + "epoch": 2.653992395437262, + "grad_norm": 17.926849365234375, + "learning_rate": 1.9925037936267073e-05, + "loss": 1.1031, "step": 350 }, { - "epoch": 3.334928229665072, + "epoch": 2.653992395437262, "step": 350, - "train/classification_loss": 0.24265973269939423, - "train/contrastive_loss": 1.4494377374649048, - "train/negative_loss": 0.9286813735961914, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.5207563638687134, - "train/total_loss": 0.5325472950935364 + "train/classification_loss": 0.27148982882499695, + "train/contrastive_loss": 1.5277045965194702, + "train/negative_loss": 1.1311314105987549, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.3965732157230377, + "train/total_loss": 0.5770307779312134 }, { - "epoch": 3.334928229665072, + "epoch": 2.653992395437262, "step": 350, - "train/classification_loss": 0.2474333494901657, - "train/contrastive_loss": 0.9493862986564636, - "train/negative_loss": 0.8313264846801758, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.11805981397628784, - "train/total_loss": 0.4373106062412262 + "train/classification_loss": 0.24656108021736145, + "train/contrastive_loss": 0.9015072584152222, + "train/negative_loss": 0.8753058910369873, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.026201386004686356, + "train/total_loss": 0.42686253786087036 }, { - "epoch": 3.813397129186603, - "grad_norm": 20.207334518432617, - "learning_rate": 1.988664122137405e-05, - "loss": 1.0661, + "epoch": 3.03041825095057, + "grad_norm": 14.439035415649414, + "learning_rate": 1.9909863429438544e-05, + "loss": 1.0846, "step": 400 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.24026766419410706, - "train/contrastive_loss": 1.0785106420516968, - "train/negative_loss": 1.0225943326950073, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.05591635778546333, - "train/total_loss": 0.45596981048583984 + "train/classification_loss": 0.2197924256324768, + "train/contrastive_loss": 1.827135443687439, + "train/negative_loss": 1.676537036895752, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.15059837698936462, + "train/total_loss": 0.5852195024490356 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.19365742802619934, - "train/contrastive_loss": 1.6367528438568115, - "train/negative_loss": 1.0576921701431274, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.5790607333183289, - "train/total_loss": 0.5210080146789551 + "train/classification_loss": 0.2729213535785675, + "train/contrastive_loss": 2.514535903930664, + "train/negative_loss": 2.5107390880584717, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.0037967036478221416, + "train/total_loss": 0.7758285999298096 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.23477545380592346, - "train/contrastive_loss": 1.8640332221984863, - "train/negative_loss": 1.8396815061569214, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.0243516992777586, - "train/total_loss": 0.6075820922851562 + "train/classification_loss": 0.20918740332126617, + "train/contrastive_loss": 2.411865711212158, + "train/negative_loss": 2.4015095233917236, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.01035609096288681, + "train/total_loss": 0.6915605664253235 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.23291240632534027, - "train/contrastive_loss": 3.058412551879883, - "train/negative_loss": 1.7860167026519775, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.2723957300186157, - "train/total_loss": 0.8445949554443359 + "train/classification_loss": 0.2887383997440338, + "train/contrastive_loss": 1.7028381824493408, + "train/negative_loss": 1.2514948844909668, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.45134326815605164, + "train/total_loss": 0.6293060779571533 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.19753392040729523, - "train/contrastive_loss": 2.3293113708496094, - "train/negative_loss": 1.5836740732192993, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 0.7456374168395996, - "train/total_loss": 0.6633961796760559 + "train/classification_loss": 0.1964968740940094, + "train/contrastive_loss": 1.0105160474777222, + "train/negative_loss": 1.0102107524871826, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.0003053471154998988, + "train/total_loss": 0.39860010147094727 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.23879612982273102, - "train/contrastive_loss": 2.953056573867798, - "train/negative_loss": 1.7694884538650513, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.1835681200027466, - "train/total_loss": 0.8294074535369873 + "train/classification_loss": 0.2658403813838959, + "train/contrastive_loss": 1.416798710823059, + "train/negative_loss": 1.3710590600967407, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.04573960229754448, + "train/total_loss": 0.5492001175880432 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.27191653847694397, - "train/contrastive_loss": 2.927450180053711, - "train/negative_loss": 2.9058797359466553, + "train/classification_loss": 0.2826295793056488, + "train/contrastive_loss": 2.7708487510681152, + "train/negative_loss": 2.5189695358276367, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.02157042920589447, - "train/total_loss": 0.8574066162109375 + "train/num_positives": 8, + "train/positive_loss": 0.25187909603118896, + "train/total_loss": 0.8367993831634521 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.29890307784080505, - "train/contrastive_loss": 1.2137832641601562, - "train/negative_loss": 1.203925371170044, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.009857850149273872, - "train/total_loss": 0.5416597127914429 + "train/classification_loss": 0.2719978094100952, + "train/contrastive_loss": 2.236053228378296, + "train/negative_loss": 1.6411713361740112, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.5948818325996399, + "train/total_loss": 0.7192084789276123 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.26937535405158997, - "train/contrastive_loss": 3.118445873260498, - "train/negative_loss": 1.8302675485610962, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 1.2881784439086914, - "train/total_loss": 0.8930644989013672 + "train/classification_loss": 0.22791478037834167, + "train/contrastive_loss": 1.136169672012329, + "train/negative_loss": 1.0827893018722534, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.053380317986011505, + "train/total_loss": 0.45514869689941406 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.2732482850551605, - "train/contrastive_loss": 2.3592722415924072, - "train/negative_loss": 1.5629132986068726, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 0.7963589429855347, - "train/total_loss": 0.7451027631759644 + "train/classification_loss": 0.25101974606513977, + "train/contrastive_loss": 2.1472620964050293, + "train/negative_loss": 1.9587390422821045, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.18852299451828003, + "train/total_loss": 0.6804721355438232 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.2706783711910248, - "train/contrastive_loss": 2.413907289505005, - "train/negative_loss": 2.403463125228882, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.010444121435284615, - "train/total_loss": 0.7534598112106323 + "train/classification_loss": 0.23771263659000397, + "train/contrastive_loss": 3.5465471744537354, + "train/negative_loss": 3.533109426498413, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.013437860645353794, + "train/total_loss": 0.9470220804214478 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.234036386013031, - "train/contrastive_loss": 2.452155828475952, - "train/negative_loss": 1.562475562095642, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.8896802663803101, - "train/total_loss": 0.7244675159454346 + "train/classification_loss": 0.2612927556037903, + "train/contrastive_loss": 2.8373072147369385, + "train/negative_loss": 2.753218650817871, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.08408848941326141, + "train/total_loss": 0.828754186630249 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.22426529228687286, - "train/contrastive_loss": 3.0510401725769043, - "train/negative_loss": 2.3909056186676025, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.6601346135139465, - "train/total_loss": 0.8344733119010925 + "train/classification_loss": 0.2673112154006958, + "train/contrastive_loss": 2.715165376663208, + "train/negative_loss": 1.5933972597122192, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.1217681169509888, + "train/total_loss": 0.8103442788124084 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.19531863927841187, - "train/contrastive_loss": 1.144457221031189, - "train/negative_loss": 1.059010624885559, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.0854465514421463, - "train/total_loss": 0.4242100715637207 + "train/classification_loss": 0.2995271682739258, + "train/contrastive_loss": 4.1230316162109375, + "train/negative_loss": 2.0183169841766357, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.1047146320343018, + "train/total_loss": 1.124133586883545 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.28563353419303894, - "train/contrastive_loss": 1.1974560022354126, - "train/negative_loss": 1.1945608854293823, + "train/classification_loss": 0.2555828094482422, + "train/contrastive_loss": 2.5197737216949463, + "train/negative_loss": 2.392738103866577, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.0028951456770300865, - "train/total_loss": 0.525124728679657 - }, - { - "epoch": 3.813397129186603, - "step": 400, - "train/classification_loss": 0.23241135478019714, - "train/contrastive_loss": 1.7288637161254883, - "train/negative_loss": 1.1297426223754883, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.5991210341453552, - "train/total_loss": 0.5781841278076172 + "train/positive_loss": 0.1270356923341751, + "train/total_loss": 0.7595375776290894 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.25422608852386475, - "train/contrastive_loss": 3.673543930053711, - "train/negative_loss": 2.546977996826172, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.1265660524368286, - "train/total_loss": 0.9889348745346069 + "train/classification_loss": 0.2464260458946228, + "train/contrastive_loss": 1.4977697134017944, + "train/negative_loss": 1.2982738018035889, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.19949588179588318, + "train/total_loss": 0.5459799766540527 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.26892220973968506, - "train/contrastive_loss": 2.5951058864593506, - "train/negative_loss": 2.041205644607544, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.5539002418518066, - "train/total_loss": 0.787943422794342 + "train/classification_loss": 0.2679178714752197, + "train/contrastive_loss": 1.3014880418777466, + "train/negative_loss": 1.2821574211120605, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0193305816501379, + "train/total_loss": 0.5282155275344849 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.244903564453125, - "train/contrastive_loss": 2.976461172103882, - "train/negative_loss": 2.400601863861084, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.5758593082427979, - "train/total_loss": 0.8401958346366882 + "train/classification_loss": 0.22863072156906128, + "train/contrastive_loss": 1.8977104425430298, + "train/negative_loss": 1.7217440605163574, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.17596638202667236, + "train/total_loss": 0.6081727743148804 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.26987412571907043, - "train/contrastive_loss": 3.8584249019622803, - "train/negative_loss": 2.106626272201538, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 1.7517986297607422, - "train/total_loss": 1.041559100151062 + "train/classification_loss": 0.26536592841148376, + "train/contrastive_loss": 4.00694465637207, + "train/negative_loss": 3.8788981437683105, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.1280466616153717, + "train/total_loss": 1.0667548179626465 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.2545139193534851, - "train/contrastive_loss": 3.0485129356384277, - "train/negative_loss": 1.8705531358718872, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 1.177959680557251, - "train/total_loss": 0.8642165064811707 + "train/classification_loss": 0.27599576115608215, + "train/contrastive_loss": 2.5086002349853516, + "train/negative_loss": 2.1880860328674316, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.3205142021179199, + "train/total_loss": 0.777715802192688 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.2228371948003769, - "train/contrastive_loss": 2.610086679458618, - "train/negative_loss": 2.0621488094329834, + "train/classification_loss": 0.31455114483833313, + "train/contrastive_loss": 1.6933870315551758, + "train/negative_loss": 1.680220603942871, "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.5479379296302795, - "train/total_loss": 0.7448545694351196 + "train/num_positives": 6, + "train/positive_loss": 0.013166392222046852, + "train/total_loss": 0.6532285213470459 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.23414386808872223, - "train/contrastive_loss": 3.88481068611145, - "train/negative_loss": 2.4395415782928467, + "train/classification_loss": 0.2924763262271881, + "train/contrastive_loss": 2.9441070556640625, + "train/negative_loss": 1.69832444190979, "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.4452691078186035, - "train/total_loss": 1.011106014251709 + "train/num_positives": 10, + "train/positive_loss": 1.2457826137542725, + "train/total_loss": 0.8812977075576782 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.2001403272151947, - "train/contrastive_loss": 1.8014850616455078, - "train/negative_loss": 1.1020820140838623, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.6994031071662903, - "train/total_loss": 0.5604373216629028 + "train/classification_loss": 0.26980483531951904, + "train/contrastive_loss": 1.9046413898468018, + "train/negative_loss": 1.7639403343200684, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.1407010555267334, + "train/total_loss": 0.6507331132888794 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.250838965177536, - "train/contrastive_loss": 2.55708646774292, - "train/negative_loss": 1.3373082876205444, - "train/num_negatives": 48, + "train/classification_loss": 0.2837485671043396, + "train/contrastive_loss": 1.1330324411392212, + "train/negative_loss": 1.1328229904174805, + "train/num_negatives": 44, "train/num_positives": 8, - "train/positive_loss": 1.219778299331665, - "train/total_loss": 0.7622562646865845 + "train/positive_loss": 0.00020948713063262403, + "train/total_loss": 0.5103550553321838 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.22004178166389465, - "train/contrastive_loss": 2.3861351013183594, - "train/negative_loss": 1.8388175964355469, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.5473175048828125, - "train/total_loss": 0.6972688436508179 + "train/classification_loss": 0.21235892176628113, + "train/contrastive_loss": 4.289538860321045, + "train/negative_loss": 3.9962024688720703, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.2933363914489746, + "train/total_loss": 1.0702667236328125 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.2078346610069275, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.2078346610069275 - }, - { - "epoch": 3.813397129186603, - "eval_exact_match_accuracy": 0.06220095693779904, - "eval_hamming_loss": 0.07824373768646214, - "eval_loss": 0.7320138812065125, - "eval_macro_f1": 0.022291021671826627, - "eval_macro_precision": 0.058823529411764705, - "eval_macro_recall": 0.013750954927425516, - "eval_micro_f1": 0.11464968152866242, - "eval_micro_precision": 1.0, - "eval_micro_recall": 0.060810810810810814, - "eval_runtime": 1.0633, - "eval_samples_per_second": 196.567, - "eval_steps_per_second": 25.394, - "step": 400 + "train/classification_loss": 0.24071219563484192, + "train/contrastive_loss": 3.569736957550049, + "train/negative_loss": 3.5517826080322266, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.017954308539628983, + "train/total_loss": 0.9546595811843872 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.25990116596221924, - "train/contrastive_loss": 1.650044322013855, - "train/negative_loss": 1.4213740825653076, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.22867022454738617, - "train/total_loss": 0.5899100303649902 + "train/classification_loss": 0.2475460320711136, + "train/contrastive_loss": 2.6243603229522705, + "train/negative_loss": 2.23184871673584, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.39251163601875305, + "train/total_loss": 0.7724180817604065 }, { - "epoch": 3.813397129186603, + "epoch": 3.03041825095057, "step": 400, - "train/classification_loss": 0.24804633855819702, - "train/contrastive_loss": 1.0883100032806396, - "train/negative_loss": 0.7743624448776245, - "train/num_negatives": 32, - "train/num_positives": 18, - "train/positive_loss": 0.31394752860069275, - "train/total_loss": 0.46570834517478943 + "train/classification_loss": 0.24072837829589844, + "train/contrastive_loss": 2.8207128047943115, + "train/negative_loss": 2.8207080364227295, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 4.768382950715022e-06, + "train/total_loss": 0.8048709630966187 + }, + { + "epoch": 3.03041825095057, + "step": 400, + "train/classification_loss": 0.3069027364253998, + "train/contrastive_loss": 1.2622120380401611, + "train/negative_loss": 0.8352072834968567, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 0.42700469493865967, + "train/total_loss": 0.5593451261520386 + }, + { + "epoch": 3.03041825095057, + "step": 400, + "train/classification_loss": 0.28020113706588745, + "train/contrastive_loss": 1.1650441884994507, + "train/negative_loss": 1.1491775512695312, + "train/num_negatives": 46, + "train/num_positives": 8, + "train/positive_loss": 0.015866611152887344, + "train/total_loss": 0.5132099986076355 + }, + { + "epoch": 3.03041825095057, + "step": 400, + "train/classification_loss": 0.24875307083129883, + "train/contrastive_loss": 2.5042314529418945, + "train/negative_loss": 2.199629545211792, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.3046019375324249, + "train/total_loss": 0.7495993971824646 + }, + { + "epoch": 3.03041825095057, + "step": 400, + "train/classification_loss": 0.30241549015045166, + "train/contrastive_loss": 1.6356182098388672, + "train/negative_loss": 1.5786916017532349, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.056926559656858444, + "train/total_loss": 0.6295391321182251 + }, + { + "epoch": 3.03041825095057, + "step": 400, + "train/classification_loss": 0.22708582878112793, + "train/contrastive_loss": 3.3374059200286865, + "train/negative_loss": 3.025468349456787, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.3119376301765442, + "train/total_loss": 0.8945670127868652 + }, + { + "epoch": 3.03041825095057, + "eval_exact_match_accuracy": 0.07633587786259542, + "eval_hamming_loss": 0.08015267175572519, + "eval_loss": 0.7248373031616211, + "eval_macro_f1": 0.0324846356453029, + "eval_macro_precision": 0.05580693815987933, + "eval_macro_recall": 0.022910216718266253, + "eval_micro_f1": 0.1716937354988399, + "eval_micro_precision": 0.9487179487179487, + "eval_micro_recall": 0.09438775510204081, + "eval_runtime": 1.3171, + "eval_samples_per_second": 198.918, + "eval_steps_per_second": 25.055, + "step": 400 + }, + { + "epoch": 3.03041825095057, + "step": 400, + "train/classification_loss": 0.2691384553909302, + "train/contrastive_loss": 0.6949138045310974, + "train/negative_loss": 0.604699969291687, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.0902138501405716, + "train/total_loss": 0.4081212282180786 + }, + { + "epoch": 3.03041825095057, + "step": 400, + "train/classification_loss": 0.2863709628582001, + "train/contrastive_loss": 1.532965064048767, + "train/negative_loss": 1.0573481321334839, + "train/num_negatives": 38, + "train/num_positives": 14, + "train/positive_loss": 0.4756169617176056, + "train/total_loss": 0.5929639935493469 }, { - "epoch": 4.287081339712919, - "grad_norm": 17.198307037353516, - "learning_rate": 1.9867557251908396e-05, - "loss": 1.0018, + "epoch": 3.4106463878326996, + "grad_norm": 19.33768081665039, + "learning_rate": 1.9894688922610016e-05, + "loss": 1.0003, "step": 450 }, { - "epoch": 4.287081339712919, + "epoch": 3.4106463878326996, "step": 450, - "train/classification_loss": 0.20112133026123047, - "train/contrastive_loss": 2.0845022201538086, - "train/negative_loss": 2.08074688911438, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 0.0037552255671471357, - "train/total_loss": 0.6180217862129211 + "train/classification_loss": 0.2645235061645508, + "train/contrastive_loss": 2.3655872344970703, + "train/negative_loss": 1.4515283107757568, + "train/num_negatives": 52, + "train/num_positives": 2, + "train/positive_loss": 0.914059042930603, + "train/total_loss": 0.7376409769058228 }, { - "epoch": 4.287081339712919, + "epoch": 3.4106463878326996, "step": 450, - "train/classification_loss": 0.20793287456035614, - "train/contrastive_loss": 0.8527883887290955, - "train/negative_loss": 0.7434472441673279, + "train/classification_loss": 0.2725066840648651, + "train/contrastive_loss": 4.0382866859436035, + "train/negative_loss": 1.718237280845642, "train/num_negatives": 52, "train/num_positives": 4, - "train/positive_loss": 0.10934113711118698, - "train/total_loss": 0.3784905672073364 + "train/positive_loss": 2.320049285888672, + "train/total_loss": 1.0801640748977661 }, { - "epoch": 4.76555023923445, - "grad_norm": 12.959186553955078, - "learning_rate": 1.984847328244275e-05, - "loss": 0.9702, + "epoch": 3.790874524714829, + "grad_norm": 16.187469482421875, + "learning_rate": 1.987951441578149e-05, + "loss": 1.0796, "step": 500 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, + "step": 500, + "train/classification_loss": 0.19835269451141357, + "train/contrastive_loss": 1.069898009300232, + "train/negative_loss": 1.0382381677627563, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.03165983781218529, + "train/total_loss": 0.41233229637145996 + }, + { + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.22847090661525726, - "train/contrastive_loss": 1.507367491722107, - "train/negative_loss": 1.1273188591003418, + "train/classification_loss": 0.26089248061180115, + "train/contrastive_loss": 2.223613739013672, + "train/negative_loss": 2.155305862426758, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.38004863262176514, - "train/total_loss": 0.5299444198608398 + "train/num_positives": 6, + "train/positive_loss": 0.06830792874097824, + "train/total_loss": 0.705615222454071 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.18183214962482452, - "train/contrastive_loss": 1.9184844493865967, - "train/negative_loss": 1.3726434707641602, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.5458409190177917, - "train/total_loss": 0.5655290484428406 + "train/classification_loss": 0.18719376623630524, + "train/contrastive_loss": 1.8583571910858154, + "train/negative_loss": 1.782490849494934, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.07586639374494553, + "train/total_loss": 0.5588651895523071 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.2178204506635666, - "train/contrastive_loss": 2.1548452377319336, - "train/negative_loss": 1.9722398519515991, - "train/num_negatives": 44, + "train/classification_loss": 0.2759978473186493, + "train/contrastive_loss": 2.4656786918640137, + "train/negative_loss": 2.2771129608154297, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.18856573104858398, + "train/total_loss": 0.7691335678100586 + }, + { + "epoch": 3.790874524714829, + "step": 500, + "train/classification_loss": 0.179106667637825, + "train/contrastive_loss": 1.5297706127166748, + "train/negative_loss": 1.5061531066894531, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.023617543280124664, + "train/total_loss": 0.48506081104278564 + }, + { + "epoch": 3.790874524714829, + "step": 500, + "train/classification_loss": 0.2519179582595825, + "train/contrastive_loss": 2.007237434387207, + "train/negative_loss": 1.6196955442428589, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.18260541558265686, - "train/total_loss": 0.6487895250320435 + "train/positive_loss": 0.3875420093536377, + "train/total_loss": 0.6533654928207397 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.2218342423439026, - "train/contrastive_loss": 4.259952545166016, - "train/negative_loss": 2.4111897945404053, + "train/classification_loss": 0.2672218978404999, + "train/contrastive_loss": 2.6387410163879395, + "train/negative_loss": 2.325331211090088, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.313409686088562, + "train/total_loss": 0.7949701547622681 + }, + { + "epoch": 3.790874524714829, + "step": 500, + "train/classification_loss": 0.2604474425315857, + "train/contrastive_loss": 2.6393988132476807, + "train/negative_loss": 2.6073074340820312, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.8487627506256104, - "train/total_loss": 1.0738247632980347 + "train/positive_loss": 0.03209148719906807, + "train/total_loss": 0.7883272171020508 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.1825561374425888, - "train/contrastive_loss": 2.402792453765869, - "train/negative_loss": 1.5025832653045654, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 0.9002091288566589, - "train/total_loss": 0.663114607334137 + "train/classification_loss": 0.20990930497646332, + "train/contrastive_loss": 0.988922655582428, + "train/negative_loss": 0.9502810835838318, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.038641564548015594, + "train/total_loss": 0.4076938331127167 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.22165803611278534, - "train/contrastive_loss": 3.0863237380981445, - "train/negative_loss": 1.6543729305267334, - "train/num_negatives": 32, + "train/classification_loss": 0.23646770417690277, + "train/contrastive_loss": 1.8454017639160156, + "train/negative_loss": 1.6590354442596436, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.1863662600517273, + "train/total_loss": 0.605548083782196 + }, + { + "epoch": 3.790874524714829, + "step": 500, + "train/classification_loss": 0.22533586621284485, + "train/contrastive_loss": 3.1734423637390137, + "train/negative_loss": 3.009831666946411, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.4319509267807007, - "train/total_loss": 0.8389227986335754 + "train/positive_loss": 0.16361072659492493, + "train/total_loss": 0.8600243330001831 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.25726717710494995, - "train/contrastive_loss": 2.4816088676452637, - "train/negative_loss": 2.4768500328063965, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.004758821334689856, - "train/total_loss": 0.7535889148712158 + "train/classification_loss": 0.24323587119579315, + "train/contrastive_loss": 2.3215436935424805, + "train/negative_loss": 2.156306028366089, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.1652376651763916, + "train/total_loss": 0.7075446248054504 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.2950655519962311, - "train/contrastive_loss": 1.5261887311935425, - "train/negative_loss": 1.3152823448181152, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.21090643107891083, - "train/total_loss": 0.6003032922744751 + "train/classification_loss": 0.25515371561050415, + "train/contrastive_loss": 1.3358643054962158, + "train/negative_loss": 1.228835940361023, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.10702838748693466, + "train/total_loss": 0.5223265886306763 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.24993778765201569, - "train/contrastive_loss": 2.1548750400543213, - "train/negative_loss": 1.7250356674194336, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.4298393130302429, - "train/total_loss": 0.6809127926826477 + "train/classification_loss": 0.2863117754459381, + "train/contrastive_loss": 3.22023868560791, + "train/negative_loss": 1.9634218215942383, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.2568169832229614, + "train/total_loss": 0.9303594827651978 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.2667790651321411, - "train/contrastive_loss": 3.984083414077759, - "train/negative_loss": 2.2195374965667725, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 1.7645459175109863, - "train/total_loss": 1.0635957717895508 + "train/classification_loss": 0.24230530858039856, + "train/contrastive_loss": 2.6409153938293457, + "train/negative_loss": 2.581348419189453, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.059566907584667206, + "train/total_loss": 0.7704883813858032 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.25508782267570496, - "train/contrastive_loss": 2.4588260650634766, - "train/negative_loss": 2.168625831604004, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.2902001142501831, - "train/total_loss": 0.7468530535697937 + "train/classification_loss": 0.23097071051597595, + "train/contrastive_loss": 1.606810212135315, + "train/negative_loss": 1.5222221612930298, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.08458807319402695, + "train/total_loss": 0.5523327589035034 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.21748694777488708, - "train/contrastive_loss": 1.5446568727493286, - "train/negative_loss": 1.3703566789627075, + "train/classification_loss": 0.2497035712003708, + "train/contrastive_loss": 1.0026277303695679, + "train/negative_loss": 0.963001549243927, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 0.17430022358894348, - "train/total_loss": 0.5264183282852173 + "train/positive_loss": 0.039626143872737885, + "train/total_loss": 0.45022910833358765 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.21222342550754547, - "train/contrastive_loss": 3.061692476272583, - "train/negative_loss": 2.308593273162842, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.7530991435050964, - "train/total_loss": 0.8245618939399719 + "train/classification_loss": 0.21453770995140076, + "train/contrastive_loss": 2.6063873767852783, + "train/negative_loss": 2.5971081256866455, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.009279265068471432, + "train/total_loss": 0.735815167427063 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.1752244234085083, - "train/contrastive_loss": 0.8701037764549255, - "train/negative_loss": 0.8461488485336304, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.023954937234520912, - "train/total_loss": 0.34924519062042236 + "train/classification_loss": 0.25398415327072144, + "train/contrastive_loss": 3.6879971027374268, + "train/negative_loss": 3.096141815185547, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.5918552875518799, + "train/total_loss": 0.9915835857391357 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.2685397267341614, - "train/contrastive_loss": 1.1335504055023193, - "train/negative_loss": 1.1067736148834229, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.026776744052767754, - "train/total_loss": 0.49524980783462524 + "train/classification_loss": 0.26311734318733215, + "train/contrastive_loss": 4.304393768310547, + "train/negative_loss": 2.68188738822937, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.6225062608718872, + "train/total_loss": 1.1239961385726929 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.21363481879234314, - "train/contrastive_loss": 1.0658785104751587, - "train/negative_loss": 0.9016512632369995, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.1642272025346756, - "train/total_loss": 0.42681050300598145 + "train/classification_loss": 0.30699679255485535, + "train/contrastive_loss": 1.7111941576004028, + "train/negative_loss": 1.6831047534942627, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.028089385479688644, + "train/total_loss": 0.6492356061935425 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.2435130476951599, - "train/contrastive_loss": 3.971268653869629, - "train/negative_loss": 2.5900228023529053, - "train/num_negatives": 46, + "train/classification_loss": 0.28276219964027405, + "train/contrastive_loss": 2.576589584350586, + "train/negative_loss": 1.983477234840393, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.3812459707260132, - "train/total_loss": 1.0377668142318726 + "train/positive_loss": 0.5931122303009033, + "train/total_loss": 0.7980800867080688 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.25954005122184753, - "train/contrastive_loss": 2.0660617351531982, - "train/negative_loss": 1.431864619255066, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.6341971158981323, - "train/total_loss": 0.6727523803710938 + "train/classification_loss": 0.25770390033721924, + "train/contrastive_loss": 2.087583065032959, + "train/negative_loss": 1.8427098989486694, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.2448732703924179, + "train/total_loss": 0.6752204895019531 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.2290964424610138, - "train/contrastive_loss": 3.2305779457092285, - "train/negative_loss": 2.327533483505249, + "train/classification_loss": 0.2688325345516205, + "train/contrastive_loss": 0.8113923072814941, + "train/negative_loss": 0.8103097081184387, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.9030444025993347, - "train/total_loss": 0.8752120733261108 + "train/num_positives": 8, + "train/positive_loss": 0.0010825962526723742, + "train/total_loss": 0.4311109781265259 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.2573116421699524, - "train/contrastive_loss": 3.364183187484741, - "train/negative_loss": 2.6627418994903564, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.70144122838974, - "train/total_loss": 0.9301483035087585 + "train/classification_loss": 0.19385488331317902, + "train/contrastive_loss": 3.4519248008728027, + "train/negative_loss": 3.213092088699341, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.23883265256881714, + "train/total_loss": 0.8842398524284363 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.24555742740631104, - "train/contrastive_loss": 2.8220205307006836, - "train/negative_loss": 1.4727319478988647, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 1.3492884635925293, - "train/total_loss": 0.8099615573883057 + "train/classification_loss": 0.228322371840477, + "train/contrastive_loss": 3.739466428756714, + "train/negative_loss": 3.638273000717163, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.1011933758854866, + "train/total_loss": 0.976215660572052 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.21403758227825165, - "train/contrastive_loss": 2.731072187423706, - "train/negative_loss": 1.6464836597442627, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.0845885276794434, - "train/total_loss": 0.760252058506012 + "train/classification_loss": 0.23144786059856415, + "train/contrastive_loss": 2.589219570159912, + "train/negative_loss": 1.9653313159942627, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.6238881945610046, + "train/total_loss": 0.7492917776107788 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.22283515334129333, - "train/contrastive_loss": 4.1764116287231445, - "train/negative_loss": 1.8596839904785156, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.316727876663208, - "train/total_loss": 1.0581175088882446 + "train/classification_loss": 0.22241143882274628, + "train/contrastive_loss": 2.9310951232910156, + "train/negative_loss": 2.9310877323150635, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 7.391003236989491e-06, + "train/total_loss": 0.8086304664611816 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.18104305863380432, - "train/contrastive_loss": 1.425093173980713, - "train/negative_loss": 0.8760972619056702, - "train/num_negatives": 42, + "train/classification_loss": 0.28593602776527405, + "train/contrastive_loss": 1.2006244659423828, + "train/negative_loss": 1.0355539321899414, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.5489959716796875, - "train/total_loss": 0.46606171131134033 + "train/positive_loss": 0.1650705635547638, + "train/total_loss": 0.526060938835144 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.23428753018379211, - "train/contrastive_loss": 2.614992618560791, - "train/negative_loss": 1.8184722661972046, - "train/num_negatives": 48, + "train/classification_loss": 0.2659342885017395, + "train/contrastive_loss": 1.104332447052002, + "train/negative_loss": 1.0312639474868774, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.796520471572876, - "train/total_loss": 0.7572860717773438 + "train/positive_loss": 0.07306844741106033, + "train/total_loss": 0.48680078983306885 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.20737527310848236, - "train/contrastive_loss": 2.0745599269866943, - "train/negative_loss": 1.5692590475082397, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.5053008794784546, - "train/total_loss": 0.6222872734069824 + "train/classification_loss": 0.23223736882209778, + "train/contrastive_loss": 2.9046688079833984, + "train/negative_loss": 2.4338555335998535, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.4708133637905121, + "train/total_loss": 0.8131711483001709 + }, + { + "epoch": 3.790874524714829, + "step": 500, + "train/classification_loss": 0.28851571679115295, + "train/contrastive_loss": 1.46825110912323, + "train/negative_loss": 1.4547431468963623, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.013507934287190437, + "train/total_loss": 0.5821659564971924 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.17832092940807343, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.17832092940807343 - }, - { - "epoch": 4.76555023923445, - "eval_exact_match_accuracy": 0.08133971291866028, - "eval_hamming_loss": 0.07655502392344497, - "eval_loss": 0.7196096777915955, - "eval_macro_f1": 0.028555111364934323, - "eval_macro_precision": 0.05656108597285068, - "eval_macro_recall": 0.019098548510313215, - "eval_micro_f1": 0.15527950310559005, - "eval_micro_precision": 0.9615384615384616, - "eval_micro_recall": 0.08445945945945946, - "eval_runtime": 1.0623, - "eval_samples_per_second": 196.737, - "eval_steps_per_second": 25.416, + "train/classification_loss": 0.21237283945083618, + "train/contrastive_loss": 3.739514112472534, + "train/negative_loss": 3.677900791168213, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.06161336973309517, + "train/total_loss": 0.9602756500244141 + }, + { + "epoch": 3.790874524714829, + "eval_exact_match_accuracy": 0.0916030534351145, + "eval_hamming_loss": 0.07903008531656938, + "eval_loss": 0.7000317573547363, + "eval_macro_f1": 0.03619909502262444, + "eval_macro_precision": 0.053921568627450976, + "eval_macro_recall": 0.02724458204334365, + "eval_micro_f1": 0.2, + "eval_micro_precision": 0.9166666666666666, + "eval_micro_recall": 0.11224489795918367, + "eval_runtime": 1.3188, + "eval_samples_per_second": 198.661, + "eval_steps_per_second": 25.022, "step": 500 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.23324115574359894, - "train/contrastive_loss": 0.9287371635437012, - "train/negative_loss": 0.8608346581459045, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.06790249794721603, - "train/total_loss": 0.41898858547210693 + "train/classification_loss": 0.1933455467224121, + "train/contrastive_loss": 1.0451794862747192, + "train/negative_loss": 0.2984260618686676, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.746753454208374, + "train/total_loss": 0.40238144993782043 }, { - "epoch": 4.76555023923445, + "epoch": 3.790874524714829, "step": 500, - "train/classification_loss": 0.21934470534324646, - "train/contrastive_loss": 1.1871473789215088, - "train/negative_loss": 0.75278240442276, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.4343649446964264, - "train/total_loss": 0.45677417516708374 + "train/classification_loss": 0.25328299403190613, + "train/contrastive_loss": 1.7287596464157104, + "train/negative_loss": 1.0786031484603882, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.6501564979553223, + "train/total_loss": 0.5990349054336548 }, { - "epoch": 5.239234449760765, - "grad_norm": 14.581633567810059, - "learning_rate": 1.9829389312977102e-05, - "loss": 0.9493, + "epoch": 4.167300380228137, + "grad_norm": 15.410737037658691, + "learning_rate": 1.9864339908952962e-05, + "loss": 0.9571, "step": 550 }, { - "epoch": 5.239234449760765, + "epoch": 4.167300380228137, "step": 550, - "train/classification_loss": 0.20794972777366638, - "train/contrastive_loss": 0.5813038945198059, - "train/negative_loss": 0.43719276785850525, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.14411114156246185, - "train/total_loss": 0.324210524559021 + "train/classification_loss": 0.28464433550834656, + "train/contrastive_loss": 1.4266362190246582, + "train/negative_loss": 0.6432851552963257, + "train/num_negatives": 42, + "train/num_positives": 4, + "train/positive_loss": 0.7833510637283325, + "train/total_loss": 0.5699715614318848 }, { - "epoch": 5.239234449760765, + "epoch": 4.167300380228137, "step": 550, - "train/classification_loss": 0.2817191481590271, - "train/contrastive_loss": 1.3667455911636353, - "train/negative_loss": 1.1708356142044067, + "train/classification_loss": 0.24933688342571259, + "train/contrastive_loss": 1.3782861232757568, + "train/negative_loss": 1.1738717555999756, "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.1959100067615509, - "train/total_loss": 0.5550682544708252 + "train/num_positives": 10, + "train/positive_loss": 0.20441436767578125, + "train/total_loss": 0.5249941349029541 }, { - "epoch": 5.717703349282297, - "grad_norm": 12.362120628356934, - "learning_rate": 1.981030534351145e-05, - "loss": 0.9244, + "epoch": 4.547528517110266, + "grad_norm": 19.841901779174805, + "learning_rate": 1.9849165402124433e-05, + "loss": 0.9287, "step": 600 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.21299529075622559, - "train/contrastive_loss": 2.4085195064544678, - "train/negative_loss": 1.962404727935791, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.44611480832099915, - "train/total_loss": 0.6946991682052612 + "train/classification_loss": 0.185683935880661, + "train/contrastive_loss": 1.538334608078003, + "train/negative_loss": 1.1002206802368164, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.4381139874458313, + "train/total_loss": 0.4933508634567261 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.16070562601089478, - "train/contrastive_loss": 1.7069010734558105, - "train/negative_loss": 1.6807260513305664, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.026175059378147125, - "train/total_loss": 0.50208580493927 + "train/classification_loss": 0.24872033298015594, + "train/contrastive_loss": 1.711256980895996, + "train/negative_loss": 1.6210346221923828, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.0902223140001297, + "train/total_loss": 0.5909717082977295 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.20884037017822266, - "train/contrastive_loss": 2.0338776111602783, - "train/negative_loss": 1.7709299325942993, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.262947678565979, - "train/total_loss": 0.6156159043312073 + "train/classification_loss": 0.17727485299110413, + "train/contrastive_loss": 1.2633171081542969, + "train/negative_loss": 1.224433183670044, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.03888390585780144, + "train/total_loss": 0.42993828654289246 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.2041846215724945, - "train/contrastive_loss": 1.9516278505325317, - "train/negative_loss": 1.6729663610458374, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.27866148948669434, - "train/total_loss": 0.5945101976394653 + "train/classification_loss": 0.257461279630661, + "train/contrastive_loss": 1.6987884044647217, + "train/negative_loss": 1.6839889287948608, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.014799481257796288, + "train/total_loss": 0.5972189903259277 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.16137540340423584, - "train/contrastive_loss": 1.4728519916534424, - "train/negative_loss": 1.0795598030090332, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 0.39329221844673157, - "train/total_loss": 0.45594581961631775 + "train/classification_loss": 0.16349440813064575, + "train/contrastive_loss": 0.6415857076644897, + "train/negative_loss": 0.6400108337402344, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.0015748686855658889, + "train/total_loss": 0.2918115556240082 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.20636917650699615, - "train/contrastive_loss": 3.0711922645568848, - "train/negative_loss": 1.7805665731430054, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.290625810623169, - "train/total_loss": 0.820607602596283 + "train/classification_loss": 0.24354785680770874, + "train/contrastive_loss": 1.7640540599822998, + "train/negative_loss": 1.3532617092132568, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.4107923209667206, + "train/total_loss": 0.5963586568832397 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.24891847372055054, - "train/contrastive_loss": 2.116351366043091, - "train/negative_loss": 2.1145589351654053, + "train/classification_loss": 0.25700539350509644, + "train/contrastive_loss": 2.471395492553711, + "train/negative_loss": 1.8592246770858765, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.001792325871065259, - "train/total_loss": 0.6721887588500977 + "train/num_positives": 8, + "train/positive_loss": 0.6121706962585449, + "train/total_loss": 0.7512844800949097 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.28080955147743225, - "train/contrastive_loss": 2.7266814708709717, - "train/negative_loss": 2.538490056991577, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.18819139897823334, - "train/total_loss": 0.8261458873748779 + "train/classification_loss": 0.24706271290779114, + "train/contrastive_loss": 1.0804160833358765, + "train/negative_loss": 1.010697364807129, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.06971872597932816, + "train/total_loss": 0.463145911693573 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.23736874759197235, - "train/contrastive_loss": 1.485619306564331, - "train/negative_loss": 1.179334282875061, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.3062850832939148, - "train/total_loss": 0.5344926118850708 + "train/classification_loss": 0.19560402631759644, + "train/contrastive_loss": 1.1551380157470703, + "train/negative_loss": 1.052696943283081, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.10244111716747284, + "train/total_loss": 0.4266316294670105 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.24924132227897644, - "train/contrastive_loss": 2.4429941177368164, - "train/negative_loss": 1.2157313823699951, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 1.2272627353668213, - "train/total_loss": 0.7378401756286621 + "train/classification_loss": 0.22719134390354156, + "train/contrastive_loss": 3.022037982940674, + "train/negative_loss": 1.8730491399765015, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.1489887237548828, + "train/total_loss": 0.8315989375114441 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.2507771849632263, - "train/contrastive_loss": 2.858355760574341, - "train/negative_loss": 2.8529763221740723, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.005379475187510252, - "train/total_loss": 0.8224483728408813 + "train/classification_loss": 0.20888745784759521, + "train/contrastive_loss": 3.309751033782959, + "train/negative_loss": 3.2066681385040283, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.10308297723531723, + "train/total_loss": 0.8708376884460449 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.20686233043670654, - "train/contrastive_loss": 0.835037350654602, - "train/negative_loss": 0.7904272079467773, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.04461011290550232, - "train/total_loss": 0.37386980652809143 + "train/classification_loss": 0.23102213442325592, + "train/contrastive_loss": 1.8752875328063965, + "train/negative_loss": 1.396403431892395, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.47888410091400146, + "train/total_loss": 0.606079638004303 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.20057865977287292, - "train/contrastive_loss": 3.650618553161621, - "train/negative_loss": 3.2791390419006348, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.37147951126098633, - "train/total_loss": 0.9307023286819458 + "train/classification_loss": 0.24396635591983795, + "train/contrastive_loss": 1.5731885433197021, + "train/negative_loss": 1.1039327383041382, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.46925583481788635, + "train/total_loss": 0.5586040616035461 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.15776968002319336, - "train/contrastive_loss": 0.3513735830783844, - "train/negative_loss": 0.3506568372249603, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.0007167591829784214, - "train/total_loss": 0.22804439067840576 + "train/classification_loss": 0.27164798974990845, + "train/contrastive_loss": 3.178654670715332, + "train/negative_loss": 1.4206693172454834, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.7579853534698486, + "train/total_loss": 0.9073789119720459 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.25645092129707336, - "train/contrastive_loss": 1.0674442052841187, - "train/negative_loss": 1.0158092975616455, + "train/classification_loss": 0.226993128657341, + "train/contrastive_loss": 2.143115520477295, + "train/negative_loss": 1.791688084602356, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.05163488909602165, - "train/total_loss": 0.4699397683143616 - }, - { - "epoch": 5.717703349282297, - "step": 600, - "train/classification_loss": 0.1933647096157074, - "train/contrastive_loss": 1.094408631324768, - "train/negative_loss": 1.0473837852478027, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.04702489823102951, - "train/total_loss": 0.412246435880661 + "train/positive_loss": 0.35142743587493896, + "train/total_loss": 0.6556162238121033 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.237483412027359, - "train/contrastive_loss": 3.6170272827148438, - "train/negative_loss": 2.35722017288208, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.2598069906234741, - "train/total_loss": 0.9608888626098633 + "train/classification_loss": 0.2216033786535263, + "train/contrastive_loss": 1.6987159252166748, + "train/negative_loss": 1.056557059288025, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.6421588063240051, + "train/total_loss": 0.5613465905189514 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.2450028657913208, - "train/contrastive_loss": 3.0960988998413086, - "train/negative_loss": 2.630944013595581, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.46515485644340515, - "train/total_loss": 0.8642226457595825 + "train/classification_loss": 0.23610864579677582, + "train/contrastive_loss": 1.4451138973236084, + "train/negative_loss": 1.1651251316070557, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.27998876571655273, + "train/total_loss": 0.5251314043998718 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.22213687002658844, - "train/contrastive_loss": 4.382094383239746, - "train/negative_loss": 2.0995733737945557, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 2.2825210094451904, - "train/total_loss": 1.0985558032989502 + "train/classification_loss": 0.20923550426959991, + "train/contrastive_loss": 2.875916004180908, + "train/negative_loss": 2.8702402114868164, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.005675807129591703, + "train/total_loss": 0.7844187021255493 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.24791254103183746, - "train/contrastive_loss": 2.5727779865264893, - "train/negative_loss": 2.467876672744751, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.10490133613348007, - "train/total_loss": 0.762468159198761 + "train/classification_loss": 0.24676227569580078, + "train/contrastive_loss": 2.3954010009765625, + "train/negative_loss": 1.945774793624878, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.4496261775493622, + "train/total_loss": 0.7258424758911133 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.2286001741886139, - "train/contrastive_loss": 2.734487771987915, - "train/negative_loss": 1.5274473428726196, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 1.2070404291152954, - "train/total_loss": 0.7754977941513062 + "train/classification_loss": 0.2552326023578644, + "train/contrastive_loss": 2.8672783374786377, + "train/negative_loss": 2.0444040298461914, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.8228743076324463, + "train/total_loss": 0.8286882638931274 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.19981487095355988, - "train/contrastive_loss": 4.169658660888672, - "train/negative_loss": 4.1311354637146, + "train/classification_loss": 0.30126380920410156, + "train/contrastive_loss": 1.4485678672790527, + "train/negative_loss": 1.396661639213562, "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.03852330893278122, - "train/total_loss": 1.033746600151062 + "train/num_positives": 6, + "train/positive_loss": 0.05190623551607132, + "train/total_loss": 0.5909774303436279 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.2136196345090866, - "train/contrastive_loss": 3.3502092361450195, - "train/negative_loss": 2.2201130390167236, + "train/classification_loss": 0.2776874601840973, + "train/contrastive_loss": 2.1049394607543945, + "train/negative_loss": 1.0887320041656494, "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.130096197128296, - "train/total_loss": 0.8836615085601807 + "train/num_positives": 10, + "train/positive_loss": 1.0162075757980347, + "train/total_loss": 0.6986753940582275 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.16550250351428986, - "train/contrastive_loss": 1.8104902505874634, - "train/negative_loss": 1.7759180068969727, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.03457220643758774, - "train/total_loss": 0.5276005864143372 + "train/classification_loss": 0.2546650469303131, + "train/contrastive_loss": 2.1004350185394287, + "train/negative_loss": 1.5180795192718506, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.5823554992675781, + "train/total_loss": 0.6747520565986633 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.22221581637859344, - "train/contrastive_loss": 2.863647937774658, - "train/negative_loss": 1.4989546537399292, - "train/num_negatives": 48, + "train/classification_loss": 0.25882866978645325, + "train/contrastive_loss": 1.127711296081543, + "train/negative_loss": 1.1275767087936401, + "train/num_negatives": 44, "train/num_positives": 8, - "train/positive_loss": 1.3646931648254395, - "train/total_loss": 0.7949454188346863 + "train/positive_loss": 0.00013455995940603316, + "train/total_loss": 0.4843709468841553 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.19517862796783447, - "train/contrastive_loss": 3.459982395172119, - "train/negative_loss": 1.4788957834243774, + "train/classification_loss": 0.17988984286785126, + "train/contrastive_loss": 4.3857502937316895, + "train/negative_loss": 3.8434712886810303, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.5422790050506592, + "train/total_loss": 1.0570398569107056 + }, + { + "epoch": 4.547528517110266, + "step": 600, + "train/classification_loss": 0.22225694358348846, + "train/contrastive_loss": 2.9959278106689453, + "train/negative_loss": 2.794717311859131, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.2012103945016861, + "train/total_loss": 0.8214425444602966 + }, + { + "epoch": 4.547528517110266, + "step": 600, + "train/classification_loss": 0.21943682432174683, + "train/contrastive_loss": 2.9918437004089355, + "train/negative_loss": 1.8983039855957031, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.0935395956039429, + "train/total_loss": 0.8178055882453918 + }, + { + "epoch": 4.547528517110266, + "step": 600, + "train/classification_loss": 0.21373173594474792, + "train/contrastive_loss": 2.927711248397827, + "train/negative_loss": 2.927708387374878, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 2.861027041944908e-06, + "train/total_loss": 0.7992739677429199 + }, + { + "epoch": 4.547528517110266, + "step": 600, + "train/classification_loss": 0.27115222811698914, + "train/contrastive_loss": 1.3276827335357666, + "train/negative_loss": 0.4148215651512146, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 0.9128612279891968, + "train/total_loss": 0.5366888046264648 + }, + { + "epoch": 4.547528517110266, + "step": 600, + "train/classification_loss": 0.249916210770607, + "train/contrastive_loss": 1.1438508033752441, + "train/negative_loss": 1.1014806032180786, "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.9810867309570312, - "train/total_loss": 0.8871751427650452 + "train/num_positives": 8, + "train/positive_loss": 0.04237024486064911, + "train/total_loss": 0.4786863923072815 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.15537132322788239, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.15537132322788239 - }, - { - "epoch": 5.717703349282297, - "eval_exact_match_accuracy": 0.20574162679425836, - "eval_hamming_loss": 0.06754855052068674, - "eval_loss": 0.7004619240760803, - "eval_macro_f1": 0.08215879179345492, - "eval_macro_precision": 0.10514705882352941, - "eval_macro_recall": 0.06837280366692132, - "eval_micro_f1": 0.3548387096774194, - "eval_micro_precision": 0.868421052631579, - "eval_micro_recall": 0.22297297297297297, - "eval_runtime": 1.0574, - "eval_samples_per_second": 197.663, - "eval_steps_per_second": 25.535, - "step": 600 + "train/classification_loss": 0.2278408408164978, + "train/contrastive_loss": 2.2376232147216797, + "train/negative_loss": 1.6010509729385376, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.6365721225738525, + "train/total_loss": 0.6753654479980469 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.21773792803287506, - "train/contrastive_loss": 0.4755852222442627, - "train/negative_loss": 0.4385237693786621, + "train/classification_loss": 0.2769981622695923, + "train/contrastive_loss": 1.325269103050232, + "train/negative_loss": 1.1748706102371216, "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.03706144914031029, - "train/total_loss": 0.31285497546195984 + "train/num_positives": 12, + "train/positive_loss": 0.15039852261543274, + "train/total_loss": 0.5420520305633545 }, { - "epoch": 5.717703349282297, + "epoch": 4.547528517110266, "step": 600, - "train/classification_loss": 0.173377126455307, - "train/contrastive_loss": 1.432297706604004, - "train/negative_loss": 1.3938711881637573, + "train/classification_loss": 0.19886130094528198, + "train/contrastive_loss": 2.788313865661621, + "train/negative_loss": 2.139673948287964, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.6486400365829468, + "train/total_loss": 0.7565240859985352 + }, + { + "epoch": 4.547528517110266, + "eval_exact_match_accuracy": 0.13358778625954199, + "eval_hamming_loss": 0.07453973955994611, + "eval_loss": 0.6485733389854431, + "eval_macro_f1": 0.05949510058304588, + "eval_macro_precision": 0.10327609304591402, + "eval_macro_recall": 0.047213622291021676, + "eval_micro_f1": 0.2905982905982906, + "eval_micro_precision": 0.8947368421052632, + "eval_micro_recall": 0.17346938775510204, + "eval_runtime": 1.3131, + "eval_samples_per_second": 199.524, + "eval_steps_per_second": 25.131, + "step": 600 + }, + { + "epoch": 4.547528517110266, + "step": 600, + "train/classification_loss": 0.25481849908828735, + "train/contrastive_loss": 1.6434886455535889, + "train/negative_loss": 1.0503907203674316, "train/num_negatives": 48, "train/num_positives": 8, - "train/positive_loss": 0.0384264774620533, - "train/total_loss": 0.4598366618156433 + "train/positive_loss": 0.593097984790802, + "train/total_loss": 0.5835162401199341 + }, + { + "epoch": 4.547528517110266, + "step": 600, + "train/classification_loss": 0.21765415370464325, + "train/contrastive_loss": 0.725735068321228, + "train/negative_loss": 0.6822137236595154, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.04352132976055145, + "train/total_loss": 0.3628011643886566 }, { - "epoch": 6.1913875598086126, - "grad_norm": 12.937857627868652, - "learning_rate": 1.9791221374045804e-05, - "loss": 0.914, + "epoch": 4.927756653992396, + "grad_norm": 20.596277236938477, + "learning_rate": 1.9833990895295905e-05, + "loss": 0.9626, "step": 650 }, { - "epoch": 6.1913875598086126, + "epoch": 4.927756653992396, "step": 650, - "train/classification_loss": 0.1873951256275177, - "train/contrastive_loss": 0.8535038232803345, - "train/negative_loss": 0.8528820276260376, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.0006217751069925725, - "train/total_loss": 0.3580958843231201 + "train/classification_loss": 0.21791037917137146, + "train/contrastive_loss": 0.9815923571586609, + "train/negative_loss": 0.828101634979248, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 0.15349072217941284, + "train/total_loss": 0.4142288565635681 }, { - "epoch": 6.1913875598086126, + "epoch": 4.927756653992396, "step": 650, - "train/classification_loss": 0.23962441086769104, - "train/contrastive_loss": 0.9358694553375244, - "train/negative_loss": 0.7900848388671875, + "train/classification_loss": 0.22267268598079681, + "train/contrastive_loss": 1.393779993057251, + "train/negative_loss": 1.1434943675994873, "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.1457846462726593, - "train/total_loss": 0.4267982840538025 + "train/num_positives": 14, + "train/positive_loss": 0.2502855956554413, + "train/total_loss": 0.5014286637306213 }, { - "epoch": 6.669856459330144, - "grad_norm": 11.751221656799316, - "learning_rate": 1.9772137404580155e-05, - "loss": 0.8352, + "epoch": 5.304182509505703, + "grad_norm": 10.824745178222656, + "learning_rate": 1.9818816388467376e-05, + "loss": 0.8822, "step": 700 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.19808214902877808, - "train/contrastive_loss": 1.4838237762451172, - "train/negative_loss": 1.353387713432312, + "train/classification_loss": 0.16740599274635315, + "train/contrastive_loss": 0.47180429100990295, + "train/negative_loss": 0.45720967650413513, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.014594613574445248, + "train/total_loss": 0.26176685094833374 + }, + { + "epoch": 5.304182509505703, + "step": 700, + "train/classification_loss": 0.2460470199584961, + "train/contrastive_loss": 2.50142765045166, + "train/negative_loss": 2.466494083404541, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.13043607771396637, - "train/total_loss": 0.494846910238266 + "train/num_positives": 6, + "train/positive_loss": 0.03493361920118332, + "train/total_loss": 0.746332585811615 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.15566247701644897, - "train/contrastive_loss": 1.3841538429260254, - "train/negative_loss": 1.1927660703659058, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.19138774275779724, - "train/total_loss": 0.4324932396411896 + "train/classification_loss": 0.1621931940317154, + "train/contrastive_loss": 1.8035231828689575, + "train/negative_loss": 1.6096158027648926, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.19390742480754852, + "train/total_loss": 0.5228978395462036 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.1954330950975418, - "train/contrastive_loss": 1.9036831855773926, - "train/negative_loss": 1.726672887802124, - "train/num_negatives": 44, + "train/classification_loss": 0.2525585889816284, + "train/contrastive_loss": 1.2175300121307373, + "train/negative_loss": 1.049878478050232, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.16765154898166656, + "train/total_loss": 0.49606460332870483 + }, + { + "epoch": 5.304182509505703, + "step": 700, + "train/classification_loss": 0.1503552943468094, + "train/contrastive_loss": 1.081328272819519, + "train/negative_loss": 1.0790566205978394, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.002271606819704175, + "train/total_loss": 0.3666209578514099 + }, + { + "epoch": 5.304182509505703, + "step": 700, + "train/classification_loss": 0.23685042560100555, + "train/contrastive_loss": 1.5698126554489136, + "train/negative_loss": 1.4415899515151978, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.17701032757759094, - "train/total_loss": 0.5761697292327881 + "train/positive_loss": 0.12822268903255463, + "train/total_loss": 0.5508129596710205 + }, + { + "epoch": 5.304182509505703, + "step": 700, + "train/classification_loss": 0.2536981403827667, + "train/contrastive_loss": 2.839479923248291, + "train/negative_loss": 2.230193614959717, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.6092862486839294, + "train/total_loss": 0.8215941190719604 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.1981566846370697, - "train/contrastive_loss": 3.553668975830078, - "train/negative_loss": 1.6961627006530762, + "train/classification_loss": 0.24230743944644928, + "train/contrastive_loss": 2.1068456172943115, + "train/negative_loss": 1.5413496494293213, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.8575061559677124, - "train/total_loss": 0.9088904857635498 + "train/positive_loss": 0.5654959678649902, + "train/total_loss": 0.6636765599250793 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.1535646766424179, - "train/contrastive_loss": 2.7808213233947754, - "train/negative_loss": 1.4051653146743774, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.375656008720398, - "train/total_loss": 0.7097289562225342 + "train/classification_loss": 0.17380210757255554, + "train/contrastive_loss": 0.6029179096221924, + "train/negative_loss": 0.6020911931991577, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.0008267341763712466, + "train/total_loss": 0.294385701417923 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.1992603987455368, - "train/contrastive_loss": 3.022217273712158, - "train/negative_loss": 1.4569721221923828, - "train/num_negatives": 32, + "train/classification_loss": 0.21845218539237976, + "train/contrastive_loss": 2.335630178451538, + "train/negative_loss": 1.926637053489685, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.408993124961853, + "train/total_loss": 0.6855782270431519 + }, + { + "epoch": 5.304182509505703, + "step": 700, + "train/classification_loss": 0.19999970495700836, + "train/contrastive_loss": 3.807934284210205, + "train/negative_loss": 3.7758233547210693, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.5652450323104858, - "train/total_loss": 0.8037039041519165 + "train/positive_loss": 0.032110970467329025, + "train/total_loss": 0.9615865349769592 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.24662023782730103, - "train/contrastive_loss": 2.239626884460449, - "train/negative_loss": 2.1966521739959717, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.04297459125518799, - "train/total_loss": 0.6945456266403198 + "train/classification_loss": 0.21862217783927917, + "train/contrastive_loss": 1.8575116395950317, + "train/negative_loss": 1.4479016065597534, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.4096100330352783, + "train/total_loss": 0.5901244878768921 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.2719389498233795, - "train/contrastive_loss": 0.8679707050323486, - "train/negative_loss": 0.8618390560150146, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.006131673231720924, - "train/total_loss": 0.4455330967903137 + "train/classification_loss": 0.2401118278503418, + "train/contrastive_loss": 2.5254836082458496, + "train/negative_loss": 1.2321728467941284, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.2933108806610107, + "train/total_loss": 0.7452085614204407 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.22478683292865753, - "train/contrastive_loss": 1.3405802249908447, - "train/negative_loss": 1.1273008584976196, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.21327941119670868, - "train/total_loss": 0.49290287494659424 + "train/classification_loss": 0.26577168703079224, + "train/contrastive_loss": 3.4778504371643066, + "train/negative_loss": 1.467635154724121, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.0102152824401855, + "train/total_loss": 0.9613417983055115 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.2528555989265442, - "train/contrastive_loss": 3.6077165603637695, - "train/negative_loss": 1.3248255252838135, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.282891035079956, - "train/total_loss": 0.9743989109992981 + "train/classification_loss": 0.22342434525489807, + "train/contrastive_loss": 2.7095389366149902, + "train/negative_loss": 2.5343079566955566, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.1752309650182724, + "train/total_loss": 0.7653321027755737 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.24026359617710114, - "train/contrastive_loss": 2.0456738471984863, - "train/negative_loss": 2.0308609008789062, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.014812852256000042, - "train/total_loss": 0.6493983864784241 + "train/classification_loss": 0.21240782737731934, + "train/contrastive_loss": 1.1563222408294678, + "train/negative_loss": 0.9630736112594604, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.19324862957000732, + "train/total_loss": 0.4436722993850708 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.19811920821666718, - "train/contrastive_loss": 1.3686950206756592, - "train/negative_loss": 0.9168319702148438, + "train/classification_loss": 0.23012249171733856, + "train/contrastive_loss": 1.1739046573638916, + "train/negative_loss": 1.094364881515503, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 0.45186305046081543, - "train/total_loss": 0.4718582034111023 + "train/positive_loss": 0.07953979074954987, + "train/total_loss": 0.46490341424942017 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.19179439544677734, - "train/contrastive_loss": 3.1437082290649414, - "train/negative_loss": 2.888381242752075, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.25532692670822144, - "train/total_loss": 0.8205360770225525 + "train/classification_loss": 0.19938194751739502, + "train/contrastive_loss": 2.071166515350342, + "train/negative_loss": 2.040794610977173, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.030371814966201782, + "train/total_loss": 0.6136152744293213 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.13946370780467987, - "train/contrastive_loss": 0.43017011880874634, - "train/negative_loss": 0.4271286725997925, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.0030414315406233072, - "train/total_loss": 0.22549772262573242 + "train/classification_loss": 0.24658146500587463, + "train/contrastive_loss": 3.4629082679748535, + "train/negative_loss": 2.698740005493164, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.7641682624816895, + "train/total_loss": 0.939163088798523 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.24283167719841003, - "train/contrastive_loss": 0.9508154988288879, - "train/negative_loss": 0.8505736589431763, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.10024185478687286, - "train/total_loss": 0.4329947829246521 + "train/classification_loss": 0.2515937089920044, + "train/contrastive_loss": 3.0552425384521484, + "train/negative_loss": 1.9398469924926758, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.1153955459594727, + "train/total_loss": 0.862642228603363 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.18183135986328125, - "train/contrastive_loss": 1.3545801639556885, - "train/negative_loss": 1.0999059677124023, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.2546742260456085, - "train/total_loss": 0.4527474045753479 + "train/classification_loss": 0.2904978394508362, + "train/contrastive_loss": 1.2467948198318481, + "train/negative_loss": 1.1264547109603882, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.12034012377262115, + "train/total_loss": 0.5398567914962769 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.2295791059732437, - "train/contrastive_loss": 3.5949010848999023, - "train/negative_loss": 2.2340500354766846, - "train/num_negatives": 46, + "train/classification_loss": 0.2681194841861725, + "train/contrastive_loss": 2.109994888305664, + "train/negative_loss": 1.0723707675933838, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.3608509302139282, - "train/total_loss": 0.9485593438148499 + "train/positive_loss": 1.0376242399215698, + "train/total_loss": 0.6901184320449829 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.2419673651456833, - "train/contrastive_loss": 3.1831541061401367, - "train/negative_loss": 1.8178821802139282, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.365272045135498, - "train/total_loss": 0.8785982131958008 + "train/classification_loss": 0.24389852583408356, + "train/contrastive_loss": 1.5237219333648682, + "train/negative_loss": 1.2158764600753784, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.30784550309181213, + "train/total_loss": 0.5486429333686829 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.21365386247634888, - "train/contrastive_loss": 3.216569423675537, - "train/negative_loss": 2.1245827674865723, + "train/classification_loss": 0.24989525973796844, + "train/contrastive_loss": 1.2151169776916504, + "train/negative_loss": 1.2150499820709229, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.0919865369796753, - "train/total_loss": 0.8569677472114563 + "train/num_positives": 8, + "train/positive_loss": 6.704490806441754e-05, + "train/total_loss": 0.4929186701774597 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.23446451127529144, - "train/contrastive_loss": 1.5658801794052124, - "train/negative_loss": 1.1991602182388306, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.36671996116638184, - "train/total_loss": 0.5476405620574951 + "train/classification_loss": 0.17094166576862335, + "train/contrastive_loss": 4.2648515701293945, + "train/negative_loss": 3.9105000495910645, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.3543517589569092, + "train/total_loss": 1.023911952972412 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.22219513356685638, - "train/contrastive_loss": 2.551480531692505, - "train/negative_loss": 1.6017118692398071, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.9497686624526978, - "train/total_loss": 0.7324912548065186 + "train/classification_loss": 0.2129494696855545, + "train/contrastive_loss": 3.6451425552368164, + "train/negative_loss": 3.472891092300415, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.17225144803524017, + "train/total_loss": 0.9419779777526855 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.19044440984725952, - "train/contrastive_loss": 2.588310718536377, - "train/negative_loss": 2.035419225692749, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.5528916120529175, - "train/total_loss": 0.7081065773963928 + "train/classification_loss": 0.21260009706020355, + "train/contrastive_loss": 1.8191502094268799, + "train/negative_loss": 1.3813613653182983, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.43778878450393677, + "train/total_loss": 0.5764301419258118 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.20443850755691528, - "train/contrastive_loss": 3.5193724632263184, - "train/negative_loss": 2.0765984058380127, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.4427740573883057, - "train/total_loss": 0.9083130359649658 + "train/classification_loss": 0.20052579045295715, + "train/contrastive_loss": 2.823437213897705, + "train/negative_loss": 2.8234310150146484, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 6.198902156029362e-06, + "train/total_loss": 0.7652132511138916 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.1478935033082962, - "train/contrastive_loss": 0.8822575211524963, - "train/negative_loss": 0.7638904452323914, - "train/num_negatives": 42, + "train/classification_loss": 0.2563076615333557, + "train/contrastive_loss": 1.366358757019043, + "train/negative_loss": 0.6672197580337524, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.11836707592010498, - "train/total_loss": 0.3243449926376343 + "train/positive_loss": 0.6991390585899353, + "train/total_loss": 0.5295794010162354 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.2158610075712204, - "train/contrastive_loss": 2.2921366691589355, - "train/negative_loss": 1.7014920711517334, - "train/num_negatives": 48, + "train/classification_loss": 0.23983754217624664, + "train/contrastive_loss": 0.5010166168212891, + "train/negative_loss": 0.4815311133861542, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.5906447172164917, - "train/total_loss": 0.6742883324623108 + "train/positive_loss": 0.01948552206158638, + "train/total_loss": 0.3400408625602722 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.18910111486911774, - "train/contrastive_loss": 2.8582773208618164, - "train/negative_loss": 1.4808449745178223, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.3774323463439941, - "train/total_loss": 0.7607565522193909 + "train/classification_loss": 0.21285730600357056, + "train/contrastive_loss": 1.8287352323532104, + "train/negative_loss": 1.280645728111267, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.5480895042419434, + "train/total_loss": 0.5786043405532837 + }, + { + "epoch": 5.304182509505703, + "step": 700, + "train/classification_loss": 0.2660568058490753, + "train/contrastive_loss": 1.7839137315750122, + "train/negative_loss": 1.7389439344406128, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.04496984928846359, + "train/total_loss": 0.6228395700454712 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.13482233881950378, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.13482233881950378 - }, - { - "epoch": 6.669856459330144, - "eval_exact_match_accuracy": 0.19617224880382775, - "eval_hamming_loss": 0.06726709822685055, - "eval_loss": 0.6485421657562256, - "eval_macro_f1": 0.08392364579419553, - "eval_macro_precision": 0.10491916064671483, - "eval_macro_recall": 0.07000254647313471, - "eval_micro_f1": 0.3523035230352303, - "eval_micro_precision": 0.8904109589041096, - "eval_micro_recall": 0.2195945945945946, - "eval_runtime": 1.0593, - "eval_samples_per_second": 197.298, - "eval_steps_per_second": 25.488, + "train/classification_loss": 0.18642298877239227, + "train/contrastive_loss": 2.9369001388549805, + "train/negative_loss": 1.571377158164978, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 1.3655229806900024, + "train/total_loss": 0.7738030552864075 + }, + { + "epoch": 5.304182509505703, + "eval_exact_match_accuracy": 0.17557251908396945, + "eval_hamming_loss": 0.0709474629546475, + "eval_loss": 0.6408491134643555, + "eval_macro_f1": 0.08100302267799775, + "eval_macro_precision": 0.16194404591104733, + "eval_macro_recall": 0.06709891894635335, + "eval_micro_f1": 0.35772357723577236, + "eval_micro_precision": 0.88, + "eval_micro_recall": 0.22448979591836735, + "eval_runtime": 1.3127, + "eval_samples_per_second": 199.582, + "eval_steps_per_second": 25.138, "step": 700 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.19836746156215668, - "train/contrastive_loss": 0.5568075776100159, - "train/negative_loss": 0.4947497546672821, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.06205780431628227, - "train/total_loss": 0.3097289800643921 + "train/classification_loss": 0.22138828039169312, + "train/contrastive_loss": 1.2366139888763428, + "train/negative_loss": 1.0508748292922974, + "train/num_negatives": 42, + "train/num_positives": 6, + "train/positive_loss": 0.18573912978172302, + "train/total_loss": 0.46871107816696167 }, { - "epoch": 6.669856459330144, + "epoch": 5.304182509505703, "step": 700, - "train/classification_loss": 0.19153456389904022, - "train/contrastive_loss": 0.8537465333938599, - "train/negative_loss": 0.7354013919830322, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.11834516376256943, - "train/total_loss": 0.3622838854789734 + "train/classification_loss": 0.2496449202299118, + "train/contrastive_loss": 1.3672740459442139, + "train/negative_loss": 0.8236318230628967, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 0.5436421632766724, + "train/total_loss": 0.5230997204780579 }, { - "epoch": 7.143540669856459, - "grad_norm": 9.678147315979004, - "learning_rate": 1.9753053435114507e-05, - "loss": 0.8356, + "epoch": 5.684410646387833, + "grad_norm": 12.202975273132324, + "learning_rate": 1.9803641881638848e-05, + "loss": 0.8944, "step": 750 }, { - "epoch": 7.143540669856459, + "epoch": 5.684410646387833, "step": 750, - "train/classification_loss": 0.2157241255044937, - "train/contrastive_loss": 1.9817616939544678, - "train/negative_loss": 1.2062450647354126, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.7755166888237, - "train/total_loss": 0.612076461315155 + "train/classification_loss": 0.24177974462509155, + "train/contrastive_loss": 1.2769036293029785, + "train/negative_loss": 1.0918139219284058, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 0.18508975207805634, + "train/total_loss": 0.4971604645252228 }, { - "epoch": 7.143540669856459, + "epoch": 5.684410646387833, "step": 750, - "train/classification_loss": 0.18445050716400146, - "train/contrastive_loss": 0.5103529095649719, - "train/negative_loss": 0.43424680829048157, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.07610610872507095, - "train/total_loss": 0.2865210771560669 + "train/classification_loss": 0.18661455810070038, + "train/contrastive_loss": 1.194482684135437, + "train/negative_loss": 0.9874798655509949, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.20700283348560333, + "train/total_loss": 0.42551109194755554 }, { - "epoch": 7.62200956937799, - "grad_norm": 13.529181480407715, - "learning_rate": 1.9733969465648858e-05, - "loss": 0.7691, + "epoch": 6.06083650190114, + "grad_norm": 14.363480567932129, + "learning_rate": 1.978846737481032e-05, + "loss": 0.8525, "step": 800 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.18574850261211395, - "train/contrastive_loss": 1.6601858139038086, - "train/negative_loss": 1.1218105554580688, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.5383752584457397, - "train/total_loss": 0.5177856683731079 + "train/classification_loss": 0.15009111166000366, + "train/contrastive_loss": 0.4403425455093384, + "train/negative_loss": 0.40895330905914307, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.03138924017548561, + "train/total_loss": 0.23815962672233582 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.1530434936285019, - "train/contrastive_loss": 1.4249517917633057, - "train/negative_loss": 0.9086145162582397, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.5163372755050659, - "train/total_loss": 0.4380338788032532 + "train/classification_loss": 0.23929211497306824, + "train/contrastive_loss": 2.415578603744507, + "train/negative_loss": 2.308985710144043, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.10659293830394745, + "train/total_loss": 0.7224078178405762 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.18271176517009735, - "train/contrastive_loss": 1.44862699508667, - "train/negative_loss": 1.175005555152893, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.2736213803291321, - "train/total_loss": 0.47243714332580566 + "train/classification_loss": 0.14979572594165802, + "train/contrastive_loss": 2.710944652557373, + "train/negative_loss": 2.6382393836975098, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.07270531356334686, + "train/total_loss": 0.6919846534729004 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.19325460493564606, - "train/contrastive_loss": 2.8367390632629395, - "train/negative_loss": 1.764556884765625, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.072182059288025, - "train/total_loss": 0.7606024146080017 + "train/classification_loss": 0.24520717561244965, + "train/contrastive_loss": 1.662591576576233, + "train/negative_loss": 1.5968942642211914, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.0656973347067833, + "train/total_loss": 0.5777254700660706 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.1514885276556015, - "train/contrastive_loss": 2.668860912322998, - "train/negative_loss": 1.2913126945495605, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.377548098564148, - "train/total_loss": 0.6852607131004333 + "train/classification_loss": 0.13543274998664856, + "train/contrastive_loss": 1.495301604270935, + "train/negative_loss": 1.4479866027832031, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.04731502756476402, + "train/total_loss": 0.4344930648803711 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.18670423328876495, - "train/contrastive_loss": 3.7518134117126465, - "train/negative_loss": 1.5469133853912354, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 2.204900026321411, - "train/total_loss": 0.937066912651062 + "train/classification_loss": 0.22985397279262543, + "train/contrastive_loss": 1.594689965248108, + "train/negative_loss": 1.5694082975387573, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.025281716138124466, + "train/total_loss": 0.5487919449806213 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.22201229631900787, - "train/contrastive_loss": 2.3016610145568848, - "train/negative_loss": 2.2953782081604004, + "train/classification_loss": 0.25779759883880615, + "train/contrastive_loss": 3.255129337310791, + "train/negative_loss": 2.4857804775238037, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.006282904651015997, - "train/total_loss": 0.6823444962501526 + "train/num_positives": 8, + "train/positive_loss": 0.7693488001823425, + "train/total_loss": 0.9088234901428223 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.26980793476104736, - "train/contrastive_loss": 1.4521459341049194, - "train/negative_loss": 1.449941635131836, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.002204321091994643, - "train/total_loss": 0.5602371692657471 + "train/classification_loss": 0.23209209740161896, + "train/contrastive_loss": 1.8948516845703125, + "train/negative_loss": 1.8053616285324097, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.08949006348848343, + "train/total_loss": 0.6110624670982361 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.21358992159366608, - "train/contrastive_loss": 1.7351572513580322, - "train/negative_loss": 1.2202502489089966, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.5149069428443909, - "train/total_loss": 0.5606213808059692 + "train/classification_loss": 0.15626738965511322, + "train/contrastive_loss": 0.3257325291633606, + "train/negative_loss": 0.31238555908203125, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.01334697101265192, + "train/total_loss": 0.22141389548778534 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.23452545702457428, - "train/contrastive_loss": 3.5243372917175293, - "train/negative_loss": 1.56524658203125, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 1.9590907096862793, - "train/total_loss": 0.9393929243087769 + "train/classification_loss": 0.21623654663562775, + "train/contrastive_loss": 2.447456121444702, + "train/negative_loss": 2.1979589462280273, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.24949707090854645, + "train/total_loss": 0.705727756023407 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.2336571216583252, - "train/contrastive_loss": 2.6681463718414307, - "train/negative_loss": 2.534600019454956, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.13354641199111938, - "train/total_loss": 0.7672864198684692 + "train/classification_loss": 0.1892853081226349, + "train/contrastive_loss": 4.69804048538208, + "train/negative_loss": 4.696638107299805, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.001402342226356268, + "train/total_loss": 1.1288933753967285 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.19253571331501007, - "train/contrastive_loss": 1.5781105756759644, - "train/negative_loss": 1.037731409072876, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.5403791666030884, - "train/total_loss": 0.5081578493118286 + "train/classification_loss": 0.2120504081249237, + "train/contrastive_loss": 2.253072500228882, + "train/negative_loss": 1.6282747983932495, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.6247977614402771, + "train/total_loss": 0.6626648902893066 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.18409515917301178, - "train/contrastive_loss": 2.93182110786438, - "train/negative_loss": 2.295199155807495, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.63662189245224, - "train/total_loss": 0.7704593539237976 + "train/classification_loss": 0.2362944632768631, + "train/contrastive_loss": 2.5803818702697754, + "train/negative_loss": 1.3812488317489624, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.1991331577301025, + "train/total_loss": 0.7523708343505859 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.13012896478176117, - "train/contrastive_loss": 0.3074164390563965, - "train/negative_loss": 0.30208536982536316, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.005331079941242933, - "train/total_loss": 0.19161225855350494 + "train/classification_loss": 0.25892767310142517, + "train/contrastive_loss": 4.295989990234375, + "train/negative_loss": 1.7422785758972168, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.5537116527557373, + "train/total_loss": 1.1181256771087646 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.22780188918113708, - "train/contrastive_loss": 1.2508461475372314, - "train/negative_loss": 0.9440591335296631, + "train/classification_loss": 0.2231304943561554, + "train/contrastive_loss": 2.621795415878296, + "train/negative_loss": 2.247373580932617, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.30678701400756836, - "train/total_loss": 0.4779711365699768 - }, - { - "epoch": 7.62200956937799, - "step": 800, - "train/classification_loss": 0.1675064116716385, - "train/contrastive_loss": 1.2626363039016724, - "train/negative_loss": 0.9452417492866516, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.31739455461502075, - "train/total_loss": 0.42003369331359863 + "train/positive_loss": 0.37442177534103394, + "train/total_loss": 0.7474895715713501 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.22568191587924957, - "train/contrastive_loss": 3.950946807861328, - "train/negative_loss": 2.2581143379211426, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.692832589149475, - "train/total_loss": 1.015871286392212 + "train/classification_loss": 0.20236508548259735, + "train/contrastive_loss": 1.4289568662643433, + "train/negative_loss": 1.418973684310913, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.009983235038816929, + "train/total_loss": 0.48815643787384033 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.2326086014509201, - "train/contrastive_loss": 2.3070178031921387, - "train/negative_loss": 1.5653910636901855, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.7416266202926636, - "train/total_loss": 0.6940121650695801 + "train/classification_loss": 0.22263778746128082, + "train/contrastive_loss": 1.2341525554656982, + "train/negative_loss": 0.9818772077560425, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.25227534770965576, + "train/total_loss": 0.46946829557418823 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.20981167256832123, - "train/contrastive_loss": 3.7888951301574707, - "train/negative_loss": 1.6633203029632568, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 2.125574827194214, - "train/total_loss": 0.9675907492637634 + "train/classification_loss": 0.19112440943717957, + "train/contrastive_loss": 3.028393507003784, + "train/negative_loss": 2.460374593734741, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.5680189728736877, + "train/total_loss": 0.7968031167984009 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.23912391066551208, - "train/contrastive_loss": 2.786156415939331, - "train/negative_loss": 2.287961483001709, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.49819499254226685, - "train/total_loss": 0.7963552474975586 + "train/classification_loss": 0.24955323338508606, + "train/contrastive_loss": 4.228336811065674, + "train/negative_loss": 3.178739309310913, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.0495975017547607, + "train/total_loss": 1.0952205657958984 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.21387852728366852, - "train/contrastive_loss": 2.0762758255004883, - "train/negative_loss": 1.2216238975524902, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.8546520471572876, - "train/total_loss": 0.6291337013244629 + "train/classification_loss": 0.2462989091873169, + "train/contrastive_loss": 3.512939929962158, + "train/negative_loss": 2.521088123321533, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.9918516874313354, + "train/total_loss": 0.9488869309425354 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.1889273226261139, - "train/contrastive_loss": 2.4016144275665283, - "train/negative_loss": 1.5914461612701416, + "train/classification_loss": 0.290688693523407, + "train/contrastive_loss": 1.209981918334961, + "train/negative_loss": 1.1767854690551758, "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.8101682662963867, - "train/total_loss": 0.6692502498626709 + "train/num_positives": 6, + "train/positive_loss": 0.03319641202688217, + "train/total_loss": 0.5326851010322571 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.20116224884986877, - "train/contrastive_loss": 3.7637922763824463, - "train/negative_loss": 1.619835615158081, + "train/classification_loss": 0.2602737247943878, + "train/contrastive_loss": 1.8127105236053467, + "train/negative_loss": 1.0463355779647827, "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.1439566612243652, - "train/total_loss": 0.9539207220077515 + "train/num_positives": 10, + "train/positive_loss": 0.7663750052452087, + "train/total_loss": 0.6228158473968506 + }, + { + "epoch": 6.06083650190114, + "step": 800, + "train/classification_loss": 0.23717240989208221, + "train/contrastive_loss": 1.9448328018188477, + "train/negative_loss": 1.1196272373199463, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.8252055048942566, + "train/total_loss": 0.6261389851570129 + }, + { + "epoch": 6.06083650190114, + "step": 800, + "train/classification_loss": 0.23953090608119965, + "train/contrastive_loss": 1.0600738525390625, + "train/negative_loss": 1.0600546598434448, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 1.9193179468857124e-05, + "train/total_loss": 0.45154568552970886 + }, + { + "epoch": 6.06083650190114, + "step": 800, + "train/classification_loss": 0.15944793820381165, + "train/contrastive_loss": 3.9818954467773438, + "train/negative_loss": 3.489140510559082, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.49275487661361694, + "train/total_loss": 0.955826997756958 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.13507309556007385, - "train/contrastive_loss": 1.113762378692627, - "train/negative_loss": 0.5808771252632141, + "train/classification_loss": 0.21026791632175446, + "train/contrastive_loss": 3.846649169921875, + "train/negative_loss": 3.703850269317627, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.5328852534294128, - "train/total_loss": 0.3578255772590637 + "train/positive_loss": 0.1427990198135376, + "train/total_loss": 0.9795977473258972 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.20268304646015167, - "train/contrastive_loss": 2.1444191932678223, - "train/negative_loss": 1.5575807094573975, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.5868385434150696, - "train/total_loss": 0.6315668821334839 + "train/classification_loss": 0.20258766412734985, + "train/contrastive_loss": 2.031541585922241, + "train/negative_loss": 1.6109451055526733, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.42059656977653503, + "train/total_loss": 0.608896017074585 + }, + { + "epoch": 6.06083650190114, + "step": 800, + "train/classification_loss": 0.1878986358642578, + "train/contrastive_loss": 2.299082040786743, + "train/negative_loss": 2.2990615367889404, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 2.0504208805505186e-05, + "train/total_loss": 0.6477150917053223 + }, + { + "epoch": 6.06083650190114, + "step": 800, + "train/classification_loss": 0.2493591010570526, + "train/contrastive_loss": 1.344496726989746, + "train/negative_loss": 0.9628952145576477, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 0.381601482629776, + "train/total_loss": 0.5182584524154663 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.18610766530036926, - "train/contrastive_loss": 2.4628398418426514, - "train/negative_loss": 1.323961615562439, + "train/classification_loss": 0.233664408326149, + "train/contrastive_loss": 0.6864070296287537, + "train/negative_loss": 0.5658330321311951, "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.1388782262802124, - "train/total_loss": 0.678675651550293 + "train/num_positives": 8, + "train/positive_loss": 0.12057401984930038, + "train/total_loss": 0.3709458112716675 + }, + { + "epoch": 6.06083650190114, + "step": 800, + "train/classification_loss": 0.19909174740314484, + "train/contrastive_loss": 2.4142704010009766, + "train/negative_loss": 1.882295846939087, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.5319746732711792, + "train/total_loss": 0.6819458603858948 + }, + { + "epoch": 6.06083650190114, + "step": 800, + "train/classification_loss": 0.2593294680118561, + "train/contrastive_loss": 1.4210054874420166, + "train/negative_loss": 1.4174134731292725, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.0035920317750424147, + "train/total_loss": 0.5435305833816528 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.12191065400838852, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.12191065400838852 - }, - { - "epoch": 7.62200956937799, - "eval_exact_match_accuracy": 0.24880382775119617, - "eval_hamming_loss": 0.06501547987616099, - "eval_loss": 0.6544973850250244, - "eval_macro_f1": 0.12830173433669548, - "eval_macro_precision": 0.2015979484492731, - "eval_macro_recall": 0.1009676597911892, - "eval_micro_f1": 0.40616966580976865, - "eval_micro_precision": 0.8494623655913979, - "eval_micro_recall": 0.2668918918918919, - "eval_runtime": 1.0612, - "eval_samples_per_second": 196.952, - "eval_steps_per_second": 25.444, + "train/classification_loss": 0.17588947713375092, + "train/contrastive_loss": 3.8125221729278564, + "train/negative_loss": 2.415342092514038, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 1.3971800804138184, + "train/total_loss": 0.9383939504623413 + }, + { + "epoch": 6.06083650190114, + "eval_exact_match_accuracy": 0.183206106870229, + "eval_hamming_loss": 0.07072294566681635, + "eval_loss": 0.6751867532730103, + "eval_macro_f1": 0.0857366237330904, + "eval_macro_precision": 0.1569704804998923, + "eval_macro_recall": 0.07297492767598843, + "eval_micro_f1": 0.37623762376237624, + "eval_micro_precision": 0.8407079646017699, + "eval_micro_recall": 0.2423469387755102, + "eval_runtime": 1.3167, + "eval_samples_per_second": 198.984, + "eval_steps_per_second": 25.063, "step": 800 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.20908351242542267, - "train/contrastive_loss": 0.7401354312896729, - "train/negative_loss": 0.620380699634552, + "train/classification_loss": 0.17596741020679474, + "train/contrastive_loss": 1.2591993808746338, + "train/negative_loss": 1.0091285705566406, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.11975475400686264, - "train/total_loss": 0.3571106195449829 + "train/positive_loss": 0.25007086992263794, + "train/total_loss": 0.4278072714805603 }, { - "epoch": 7.62200956937799, + "epoch": 6.06083650190114, "step": 800, - "train/classification_loss": 0.19263923168182373, - "train/contrastive_loss": 0.864037275314331, - "train/negative_loss": 0.8467162847518921, - "train/num_negatives": 48, + "train/classification_loss": 0.2718932330608368, + "train/contrastive_loss": 1.046523094177246, + "train/negative_loss": 0.9183290004730225, + "train/num_negatives": 42, "train/num_positives": 8, - "train/positive_loss": 0.01732099987566471, - "train/total_loss": 0.36544668674468994 + "train/positive_loss": 0.12819403409957886, + "train/total_loss": 0.4811978340148926 }, { - "epoch": 8.095693779904305, - "grad_norm": 14.349428176879883, - "learning_rate": 1.9714885496183205e-05, - "loss": 0.8073, + "epoch": 6.44106463878327, + "grad_norm": 7.1285552978515625, + "learning_rate": 1.9773292867981794e-05, + "loss": 0.8011, "step": 850 }, { - "epoch": 8.095693779904305, + "epoch": 6.44106463878327, "step": 850, - "train/classification_loss": 0.21733205020427704, - "train/contrastive_loss": 0.8519055247306824, - "train/negative_loss": 0.7477812767028809, - "train/num_negatives": 52, - "train/num_positives": 4, - "train/positive_loss": 0.10412423312664032, - "train/total_loss": 0.38771316409111023 + "train/classification_loss": 0.19517402350902557, + "train/contrastive_loss": 0.992882490158081, + "train/negative_loss": 0.5848174095153809, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.4080650806427002, + "train/total_loss": 0.39375051856040955 }, { - "epoch": 8.095693779904305, + "epoch": 6.44106463878327, "step": 850, - "train/classification_loss": 0.22415895760059357, - "train/contrastive_loss": 0.737302303314209, - "train/negative_loss": 0.5600704550743103, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.17723187804222107, - "train/total_loss": 0.37161940336227417 + "train/classification_loss": 0.2521856725215912, + "train/contrastive_loss": 0.9317941665649414, + "train/negative_loss": 0.876724898815155, + "train/num_negatives": 46, + "train/num_positives": 8, + "train/positive_loss": 0.055069293826818466, + "train/total_loss": 0.43854451179504395 }, { - "epoch": 8.574162679425838, - "grad_norm": 11.02156925201416, - "learning_rate": 1.969580152671756e-05, - "loss": 0.7258, + "epoch": 6.821292775665399, + "grad_norm": 15.67064380645752, + "learning_rate": 1.9758118361153265e-05, + "loss": 0.8107, "step": 900 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, + "step": 900, + "train/classification_loss": 0.14200833439826965, + "train/contrastive_loss": 0.6040069460868835, + "train/negative_loss": 0.4661591947078705, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.13784773647785187, + "train/total_loss": 0.26280972361564636 + }, + { + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.17929725348949432, - "train/contrastive_loss": 2.1669445037841797, - "train/negative_loss": 1.6006252765655518, + "train/classification_loss": 0.2333807647228241, + "train/contrastive_loss": 1.6134933233261108, + "train/negative_loss": 1.5398399829864502, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.5663193464279175, - "train/total_loss": 0.6126861572265625 + "train/num_positives": 6, + "train/positive_loss": 0.07365336269140244, + "train/total_loss": 0.5560794472694397 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.16097275912761688, - "train/contrastive_loss": 2.6076440811157227, - "train/negative_loss": 1.7354910373687744, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.8721529245376587, - "train/total_loss": 0.6825016140937805 + "train/classification_loss": 0.14471642673015594, + "train/contrastive_loss": 1.6488196849822998, + "train/negative_loss": 1.4792921543121338, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.16952753067016602, + "train/total_loss": 0.47448039054870605 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.17017687857151031, - "train/contrastive_loss": 0.9796109199523926, - "train/negative_loss": 0.8671532869338989, - "train/num_negatives": 44, + "train/classification_loss": 0.2529566287994385, + "train/contrastive_loss": 1.548437476158142, + "train/negative_loss": 0.9623008966445923, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.5861365795135498, + "train/total_loss": 0.5626441240310669 + }, + { + "epoch": 6.821292775665399, + "step": 900, + "train/classification_loss": 0.12462147325277328, + "train/contrastive_loss": 0.459652841091156, + "train/negative_loss": 0.4592992961406708, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.0003535559808369726, + "train/total_loss": 0.21655204892158508 + }, + { + "epoch": 6.821292775665399, + "step": 900, + "train/classification_loss": 0.22391144931316376, + "train/contrastive_loss": 1.5608726739883423, + "train/negative_loss": 1.55425226688385, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.11245766282081604, - "train/total_loss": 0.3660990595817566 + "train/positive_loss": 0.006620408967137337, + "train/total_loss": 0.5360859632492065 + }, + { + "epoch": 6.821292775665399, + "step": 900, + "train/classification_loss": 0.2511955499649048, + "train/contrastive_loss": 3.5467796325683594, + "train/negative_loss": 2.313225746154785, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.2335540056228638, + "train/total_loss": 0.9605515003204346 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.18489161133766174, - "train/contrastive_loss": 3.100346565246582, - "train/negative_loss": 1.6984999179840088, + "train/classification_loss": 0.23277756571769714, + "train/contrastive_loss": 2.3796637058258057, + "train/negative_loss": 1.4776890277862549, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.4018465280532837, - "train/total_loss": 0.8049609661102295 + "train/positive_loss": 0.9019746780395508, + "train/total_loss": 0.7087103128433228 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.13669779896736145, - "train/contrastive_loss": 2.752277374267578, - "train/negative_loss": 0.999453067779541, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.7528241872787476, - "train/total_loss": 0.6871533393859863 + "train/classification_loss": 0.14850184321403503, + "train/contrastive_loss": 0.5401371121406555, + "train/negative_loss": 0.5276151299476624, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.01252196915447712, + "train/total_loss": 0.2565292716026306 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.18337076902389526, - "train/contrastive_loss": 3.797107219696045, - "train/negative_loss": 2.058586597442627, - "train/num_negatives": 32, + "train/classification_loss": 0.21204017102718353, + "train/contrastive_loss": 2.8423776626586914, + "train/negative_loss": 2.670199394226074, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.17217817902565002, + "train/total_loss": 0.780515730381012 + }, + { + "epoch": 6.821292775665399, + "step": 900, + "train/classification_loss": 0.18989913165569305, + "train/contrastive_loss": 5.3741912841796875, + "train/negative_loss": 5.373791217803955, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.738520622253418, - "train/total_loss": 0.9427922368049622 + "train/positive_loss": 0.00039994725375436246, + "train/total_loss": 1.2647373676300049 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.21357081830501556, - "train/contrastive_loss": 2.0062570571899414, - "train/negative_loss": 2.0009219646453857, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.005335062742233276, - "train/total_loss": 0.6148222088813782 + "train/classification_loss": 0.20812946557998657, + "train/contrastive_loss": 3.139085292816162, + "train/negative_loss": 2.4876205921173096, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.6514647006988525, + "train/total_loss": 0.8359465599060059 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.2618056833744049, - "train/contrastive_loss": 1.747767448425293, - "train/negative_loss": 1.674180030822754, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.07358741760253906, - "train/total_loss": 0.611359179019928 + "train/classification_loss": 0.24138905107975006, + "train/contrastive_loss": 2.4470314979553223, + "train/negative_loss": 1.5703805685043335, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.8766509294509888, + "train/total_loss": 0.7307953834533691 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.18986839056015015, - "train/contrastive_loss": 1.359466314315796, - "train/negative_loss": 1.101283073425293, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.25818324089050293, - "train/total_loss": 0.4617616534233093 + "train/classification_loss": 0.24664582312107086, + "train/contrastive_loss": 4.036377429962158, + "train/negative_loss": 1.8275724649429321, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.2088050842285156, + "train/total_loss": 1.0539213418960571 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.2352956384420395, - "train/contrastive_loss": 2.083298921585083, - "train/negative_loss": 1.642700433731079, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 0.4405985474586487, - "train/total_loss": 0.6519554257392883 + "train/classification_loss": 0.2174188643693924, + "train/contrastive_loss": 3.1298577785491943, + "train/negative_loss": 2.930157423019409, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.19970032572746277, + "train/total_loss": 0.8433904051780701 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.23468036949634552, - "train/contrastive_loss": 2.6050026416778564, - "train/negative_loss": 2.429340124130249, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.17566241323947906, - "train/total_loss": 0.7556809186935425 + "train/classification_loss": 0.19125540554523468, + "train/contrastive_loss": 0.8676894903182983, + "train/negative_loss": 0.8672274351119995, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0004620542749762535, + "train/total_loss": 0.3647933006286621 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.18573540449142456, - "train/contrastive_loss": 2.277825117111206, - "train/negative_loss": 1.0931161642074585, + "train/classification_loss": 0.20735371112823486, + "train/contrastive_loss": 1.228316068649292, + "train/negative_loss": 1.0455677509307861, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 1.1847089529037476, - "train/total_loss": 0.6413004398345947 + "train/positive_loss": 0.18274828791618347, + "train/total_loss": 0.4530169367790222 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.18342682719230652, - "train/contrastive_loss": 3.6214537620544434, - "train/negative_loss": 3.0294930934906006, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.5919607281684875, - "train/total_loss": 0.9077175855636597 + "train/classification_loss": 0.18642742931842804, + "train/contrastive_loss": 2.692878484725952, + "train/negative_loss": 2.6139655113220215, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.07891305536031723, + "train/total_loss": 0.7250031232833862 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.11789260059595108, - "train/contrastive_loss": 0.6275992393493652, - "train/negative_loss": 0.5501975417137146, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.07740169018507004, - "train/total_loss": 0.24341246485710144 + "train/classification_loss": 0.24357852339744568, + "train/contrastive_loss": 2.4185757637023926, + "train/negative_loss": 1.9832916259765625, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.43528425693511963, + "train/total_loss": 0.7272936701774597 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.21845024824142456, - "train/contrastive_loss": 0.9018421173095703, - "train/negative_loss": 0.7851783633232117, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.11666377633810043, - "train/total_loss": 0.3988186717033386 + "train/classification_loss": 0.24625980854034424, + "train/contrastive_loss": 3.8102169036865234, + "train/negative_loss": 2.3975534439086914, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.4126633405685425, + "train/total_loss": 1.008303165435791 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.1569637656211853, - "train/contrastive_loss": 1.2951709032058716, - "train/negative_loss": 0.8997073769569397, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.3954634964466095, - "train/total_loss": 0.4159979522228241 + "train/classification_loss": 0.28147438168525696, + "train/contrastive_loss": 0.9859337210655212, + "train/negative_loss": 0.9597668051719666, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.026166899129748344, + "train/total_loss": 0.47866111993789673 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.21739326417446136, - "train/contrastive_loss": 3.739473819732666, - "train/negative_loss": 2.2859113216400146, - "train/num_negatives": 46, + "train/classification_loss": 0.2627241313457489, + "train/contrastive_loss": 1.5247864723205566, + "train/negative_loss": 0.8660334348678589, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.453562617301941, - "train/total_loss": 0.9652880430221558 + "train/positive_loss": 0.658752977848053, + "train/total_loss": 0.5676814317703247 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.23579387366771698, - "train/contrastive_loss": 2.6309587955474854, - "train/negative_loss": 1.8713855743408203, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.759573221206665, - "train/total_loss": 0.7619856595993042 + "train/classification_loss": 0.2328360676765442, + "train/contrastive_loss": 1.3365968465805054, + "train/negative_loss": 1.302452802658081, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.03414405137300491, + "train/total_loss": 0.5001554489135742 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.19933868944644928, - "train/contrastive_loss": 3.2679734230041504, - "train/negative_loss": 1.9103063344955444, + "train/classification_loss": 0.23645280301570892, + "train/contrastive_loss": 1.2493042945861816, + "train/negative_loss": 1.2492973804473877, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.3576669692993164, - "train/total_loss": 0.8529333472251892 + "train/num_positives": 8, + "train/positive_loss": 6.943985681573395e-06, + "train/total_loss": 0.48631367087364197 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.22426901757717133, - "train/contrastive_loss": 1.8000414371490479, - "train/negative_loss": 1.4806671142578125, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.3193742632865906, - "train/total_loss": 0.584277331829071 + "train/classification_loss": 0.1606534868478775, + "train/contrastive_loss": 5.86124324798584, + "train/negative_loss": 4.694880962371826, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 1.1663624048233032, + "train/total_loss": 1.332902193069458 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.2002340853214264, - "train/contrastive_loss": 1.8053568601608276, - "train/negative_loss": 0.8547818064689636, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.950575053691864, - "train/total_loss": 0.5613054633140564 + "train/classification_loss": 0.20818360149860382, + "train/contrastive_loss": 3.2521426677703857, + "train/negative_loss": 3.0393099784851074, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.21283259987831116, + "train/total_loss": 0.8586121201515198 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.18360592424869537, - "train/contrastive_loss": 2.9231441020965576, - "train/negative_loss": 2.3557090759277344, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.5674350261688232, - "train/total_loss": 0.7682347297668457 + "train/classification_loss": 0.200105220079422, + "train/contrastive_loss": 2.3706798553466797, + "train/negative_loss": 1.9312899112701416, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.4393899142742157, + "train/total_loss": 0.6742411851882935 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.19627521932125092, - "train/contrastive_loss": 3.401052236557007, - "train/negative_loss": 1.993591547012329, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.4074606895446777, - "train/total_loss": 0.8764857053756714 + "train/classification_loss": 0.18218202888965607, + "train/contrastive_loss": 2.6530959606170654, + "train/negative_loss": 2.6530842781066895, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 1.1563368389033712e-05, + "train/total_loss": 0.7128012180328369 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.1327422708272934, - "train/contrastive_loss": 2.4151759147644043, - "train/negative_loss": 1.0470236539840698, - "train/num_negatives": 42, + "train/classification_loss": 0.24004633724689484, + "train/contrastive_loss": 1.0881969928741455, + "train/negative_loss": 0.38205966353416443, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 1.368152379989624, - "train/total_loss": 0.6157774329185486 + "train/positive_loss": 0.7061373591423035, + "train/total_loss": 0.4576857388019562 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.1901172548532486, - "train/contrastive_loss": 2.157074451446533, - "train/negative_loss": 1.5336318016052246, - "train/num_negatives": 48, + "train/classification_loss": 0.22105562686920166, + "train/contrastive_loss": 0.6681925654411316, + "train/negative_loss": 0.5807611346244812, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.6234425902366638, - "train/total_loss": 0.621532142162323 + "train/positive_loss": 0.08743143826723099, + "train/total_loss": 0.354694128036499 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.1803000420331955, - "train/contrastive_loss": 2.243943452835083, - "train/negative_loss": 1.4003387689590454, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.8436047434806824, - "train/total_loss": 0.6290887594223022 + "train/classification_loss": 0.1977541595697403, + "train/contrastive_loss": 2.3878707885742188, + "train/negative_loss": 1.7448351383209229, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.6430356502532959, + "train/total_loss": 0.6753283143043518 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.10478032380342484, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.10478032380342484 - }, - { - "epoch": 8.574162679425838, - "eval_exact_match_accuracy": 0.2966507177033493, - "eval_hamming_loss": 0.06079369546861807, - "eval_loss": 0.6525943279266357, - "eval_macro_f1": 0.15581638905904552, - "eval_macro_precision": 0.20672111761955836, - "eval_macro_recall": 0.12879551820728288, - "eval_micro_f1": 0.47317073170731705, - "eval_micro_precision": 0.8508771929824561, - "eval_micro_recall": 0.3277027027027027, - "eval_runtime": 1.0564, - "eval_samples_per_second": 197.843, - "eval_steps_per_second": 25.559, + "train/classification_loss": 0.24618221819400787, + "train/contrastive_loss": 1.4384888410568237, + "train/negative_loss": 1.3930977582931519, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.04539113864302635, + "train/total_loss": 0.5338799953460693 + }, + { + "epoch": 6.821292775665399, + "step": 900, + "train/classification_loss": 0.1633346676826477, + "train/contrastive_loss": 1.5251208543777466, + "train/negative_loss": 1.5033804178237915, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.02174045331776142, + "train/total_loss": 0.4683588445186615 + }, + { + "epoch": 6.821292775665399, + "eval_exact_match_accuracy": 0.20229007633587787, + "eval_hamming_loss": 0.0698248765154917, + "eval_loss": 0.650576651096344, + "eval_macro_f1": 0.09497402132529206, + "eval_macro_precision": 0.1490092743289674, + "eval_macro_recall": 0.07926711668273867, + "eval_micro_f1": 0.3937621832358674, + "eval_micro_precision": 0.8347107438016529, + "eval_micro_recall": 0.2576530612244898, + "eval_runtime": 1.3214, + "eval_samples_per_second": 198.269, + "eval_steps_per_second": 24.973, "step": 900 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.21837973594665527, - "train/contrastive_loss": 0.6792857646942139, - "train/negative_loss": 0.6423966884613037, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.03688909113407135, - "train/total_loss": 0.354236900806427 + "train/classification_loss": 0.17967765033245087, + "train/contrastive_loss": 0.6434458494186401, + "train/negative_loss": 0.23951822519302368, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.40392765402793884, + "train/total_loss": 0.3083668351173401 }, { - "epoch": 8.574162679425838, + "epoch": 6.821292775665399, "step": 900, - "train/classification_loss": 0.18556059896945953, - "train/contrastive_loss": 1.065626621246338, - "train/negative_loss": 0.891682505607605, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.1739441603422165, - "train/total_loss": 0.39868593215942383 + "train/classification_loss": 0.21401585638523102, + "train/contrastive_loss": 0.9666712880134583, + "train/negative_loss": 0.5709455013275146, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.3957257866859436, + "train/total_loss": 0.4073501229286194 }, { - "epoch": 9.047846889952153, - "grad_norm": 11.439697265625, - "learning_rate": 1.967671755725191e-05, - "loss": 0.7376, + "epoch": 7.197718631178708, + "grad_norm": 12.999037742614746, + "learning_rate": 1.9742943854324737e-05, + "loss": 0.7775, "step": 950 }, { - "epoch": 9.047846889952153, + "epoch": 7.197718631178708, "step": 950, - "train/classification_loss": 0.18386907875537872, - "train/contrastive_loss": 0.6875880360603333, - "train/negative_loss": 0.4792012572288513, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 0.20838679373264313, - "train/total_loss": 0.3213866949081421 + "train/classification_loss": 0.20943096280097961, + "train/contrastive_loss": 0.9891229867935181, + "train/negative_loss": 0.960711658000946, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.028411345556378365, + "train/total_loss": 0.4072555601596832 }, { - "epoch": 9.047846889952153, + "epoch": 7.197718631178708, "step": 950, - "train/classification_loss": 0.187240868806839, - "train/contrastive_loss": 0.6511194705963135, - "train/negative_loss": 0.634975790977478, - "train/num_negatives": 32, - "train/num_positives": 24, - "train/positive_loss": 0.016143664717674255, - "train/total_loss": 0.31746476888656616 + "train/classification_loss": 0.25477170944213867, + "train/contrastive_loss": 0.9611056447029114, + "train/negative_loss": 0.4958164393901825, + "train/num_negatives": 40, + "train/num_positives": 8, + "train/positive_loss": 0.4652892053127289, + "train/total_loss": 0.4469928443431854 }, { - "epoch": 9.526315789473685, - "grad_norm": 18.221525192260742, - "learning_rate": 1.965763358778626e-05, - "loss": 0.6766, + "epoch": 7.577946768060836, + "grad_norm": 15.076811790466309, + "learning_rate": 1.9727769347496208e-05, + "loss": 0.7608, "step": 1000 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.1729215383529663, - "train/contrastive_loss": 2.8441033363342285, - "train/negative_loss": 1.9266600608825684, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.9174433946609497, - "train/total_loss": 0.7417421936988831 + "train/classification_loss": 0.1289159506559372, + "train/contrastive_loss": 0.5733219385147095, + "train/negative_loss": 0.5439671874046326, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.02935473807156086, + "train/total_loss": 0.24358034133911133 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.13645710051059723, - "train/contrastive_loss": 1.2893476486206055, - "train/negative_loss": 0.9585465788841248, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.3308010697364807, - "train/total_loss": 0.3943266272544861 + "train/classification_loss": 0.22799119353294373, + "train/contrastive_loss": 1.8253757953643799, + "train/negative_loss": 1.8012479543685913, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.02412780188024044, + "train/total_loss": 0.5930663347244263 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.171706885099411, - "train/contrastive_loss": 1.96231210231781, - "train/negative_loss": 1.1993398666381836, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.7629722356796265, - "train/total_loss": 0.5641692876815796 + "train/classification_loss": 0.13783158361911774, + "train/contrastive_loss": 2.6482205390930176, + "train/negative_loss": 2.624173164367676, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.024047449231147766, + "train/total_loss": 0.667475700378418 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.1761072725057602, - "train/contrastive_loss": 3.3688437938690186, - "train/negative_loss": 1.4506144523620605, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.918229341506958, - "train/total_loss": 0.8498760461807251 + "train/classification_loss": 0.24348831176757812, + "train/contrastive_loss": 1.4508249759674072, + "train/negative_loss": 1.4046658277511597, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.04615914449095726, + "train/total_loss": 0.5336533188819885 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.12876953184604645, - "train/contrastive_loss": 2.8737897872924805, - "train/negative_loss": 1.0348875522613525, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.838902235031128, - "train/total_loss": 0.7035275101661682 + "train/classification_loss": 0.11849958449602127, + "train/contrastive_loss": 0.9546734094619751, + "train/negative_loss": 0.954458475112915, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.0002149249194189906, + "train/total_loss": 0.30943426489830017 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.169353187084198, - "train/contrastive_loss": 2.6646556854248047, - "train/negative_loss": 1.4619190692901611, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.202736496925354, - "train/total_loss": 0.7022843360900879 + "train/classification_loss": 0.22186420857906342, + "train/contrastive_loss": 1.7089600563049316, + "train/negative_loss": 1.7054134607315063, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.0035466270055621862, + "train/total_loss": 0.563656210899353 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.2035907357931137, - "train/contrastive_loss": 1.7520946264266968, - "train/negative_loss": 1.7324743270874023, + "train/classification_loss": 0.24348554015159607, + "train/contrastive_loss": 3.3030858039855957, + "train/negative_loss": 2.24501895904541, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.01962033286690712, - "train/total_loss": 0.5540096759796143 + "train/num_positives": 8, + "train/positive_loss": 1.058066725730896, + "train/total_loss": 0.9041026830673218 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.25763842463493347, - "train/contrastive_loss": 1.9812527894973755, - "train/negative_loss": 1.9242794513702393, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.05697333812713623, - "train/total_loss": 0.653889000415802 + "train/classification_loss": 0.22570711374282837, + "train/contrastive_loss": 1.6925597190856934, + "train/negative_loss": 1.127984642982483, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.5645751357078552, + "train/total_loss": 0.564219057559967 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.18436577916145325, - "train/contrastive_loss": 1.030785083770752, - "train/negative_loss": 0.850085437297821, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.1806996762752533, - "train/total_loss": 0.3905227780342102 + "train/classification_loss": 0.1349203735589981, + "train/contrastive_loss": 0.14933709800243378, + "train/negative_loss": 0.1493080109357834, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 2.908822352765128e-05, + "train/total_loss": 0.16478779911994934 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.21492846310138702, - "train/contrastive_loss": 3.6369478702545166, - "train/negative_loss": 1.41361403465271, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.2233338356018066, - "train/total_loss": 0.9423180222511292 + "train/classification_loss": 0.20240287482738495, + "train/contrastive_loss": 2.668659210205078, + "train/negative_loss": 2.652769088745117, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.015890056267380714, + "train/total_loss": 0.7361347675323486 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.22714455425739288, - "train/contrastive_loss": 2.890570640563965, - "train/negative_loss": 2.7555301189422607, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.13504059612751007, - "train/total_loss": 0.8052586913108826 + "train/classification_loss": 0.1831376701593399, + "train/contrastive_loss": 5.051712989807129, + "train/negative_loss": 5.048658847808838, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.003054049564525485, + "train/total_loss": 1.1934802532196045 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.1726914346218109, - "train/contrastive_loss": 0.8908134698867798, - "train/negative_loss": 0.5843527317047119, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.30646073818206787, - "train/total_loss": 0.35085412859916687 + "train/classification_loss": 0.20074355602264404, + "train/contrastive_loss": 4.433694362640381, + "train/negative_loss": 4.179481506347656, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.2542129158973694, + "train/total_loss": 1.0874824523925781 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.17412365972995758, - "train/contrastive_loss": 3.2342848777770996, - "train/negative_loss": 2.6617648601531982, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.5725200772285461, - "train/total_loss": 0.8209806084632874 + "train/classification_loss": 0.23724988102912903, + "train/contrastive_loss": 2.71942138671875, + "train/negative_loss": 1.8057583570480347, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.9136630296707153, + "train/total_loss": 0.7811341285705566 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.10565164685249329, - "train/contrastive_loss": 0.43753185868263245, - "train/negative_loss": 0.43717774748802185, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.0003540970792528242, - "train/total_loss": 0.19315803050994873 + "train/classification_loss": 0.23611994087696075, + "train/contrastive_loss": 3.4884390830993652, + "train/negative_loss": 2.151172161102295, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.3372670412063599, + "train/total_loss": 0.9338077306747437 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.21303632855415344, - "train/contrastive_loss": 1.2228420972824097, - "train/negative_loss": 0.8490359783172607, + "train/classification_loss": 0.2124759703874588, + "train/contrastive_loss": 3.64603590965271, + "train/negative_loss": 3.6041364669799805, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.37380608916282654, - "train/total_loss": 0.4576047658920288 + "train/positive_loss": 0.04189945012331009, + "train/total_loss": 0.9416831731796265 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.14537955820560455, - "train/contrastive_loss": 1.5707788467407227, - "train/negative_loss": 1.0724871158599854, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.4982917308807373, - "train/total_loss": 0.4595353603363037 + "train/classification_loss": 0.1829582154750824, + "train/contrastive_loss": 1.4599394798278809, + "train/negative_loss": 1.459488034248352, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0004514587053563446, + "train/total_loss": 0.47494611144065857 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.22066007554531097, - "train/contrastive_loss": 3.4562950134277344, - "train/negative_loss": 1.9845813512802124, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.4717135429382324, - "train/total_loss": 0.911919116973877 + "train/classification_loss": 0.20088103413581848, + "train/contrastive_loss": 1.1316457986831665, + "train/negative_loss": 1.0825669765472412, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0490788035094738, + "train/total_loss": 0.4272102117538452 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.23375697433948517, - "train/contrastive_loss": 2.8525259494781494, - "train/negative_loss": 1.976617455482483, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.8759084939956665, - "train/total_loss": 0.8042621612548828 + "train/classification_loss": 0.18234311044216156, + "train/contrastive_loss": 2.623671054840088, + "train/negative_loss": 2.0149190425872803, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.6087521314620972, + "train/total_loss": 0.7070773243904114 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.1994798630475998, - "train/contrastive_loss": 3.3053414821624756, - "train/negative_loss": 2.0525028705596924, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.2528386116027832, - "train/total_loss": 0.860548198223114 + "train/classification_loss": 0.2491370439529419, + "train/contrastive_loss": 3.319690704345703, + "train/negative_loss": 3.1465697288513184, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.17312106490135193, + "train/total_loss": 0.9130752086639404 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.23415248095989227, - "train/contrastive_loss": 2.0373647212982178, - "train/negative_loss": 1.7933008670806885, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.24406389892101288, - "train/total_loss": 0.6416254043579102 + "train/classification_loss": 0.24315166473388672, + "train/contrastive_loss": 3.118285655975342, + "train/negative_loss": 2.222135066986084, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.8961504697799683, + "train/total_loss": 0.8668088316917419 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.19440925121307373, - "train/contrastive_loss": 1.2693939208984375, - "train/negative_loss": 0.6354436278343201, - "train/num_negatives": 44, + "train/classification_loss": 0.2682984173297882, + "train/contrastive_loss": 1.016623616218567, + "train/negative_loss": 1.0072484016418457, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.009375272318720818, + "train/total_loss": 0.47162312269210815 + }, + { + "epoch": 7.577946768060836, + "step": 1000, + "train/classification_loss": 0.25375455617904663, + "train/contrastive_loss": 2.2992734909057617, + "train/negative_loss": 1.0143358707427979, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.6339502334594727, - "train/total_loss": 0.44828805327415466 + "train/positive_loss": 1.2849375009536743, + "train/total_loss": 0.7136092185974121 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.18589338660240173, - "train/contrastive_loss": 4.379177093505859, - "train/negative_loss": 3.873616933822632, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.5055599212646484, - "train/total_loss": 1.061728835105896 + "train/classification_loss": 0.2327210009098053, + "train/contrastive_loss": 1.5156731605529785, + "train/negative_loss": 1.4916613101959229, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.024011798202991486, + "train/total_loss": 0.5358556509017944 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.19963188469409943, - "train/contrastive_loss": 3.23679256439209, - "train/negative_loss": 1.9499467611312866, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.2868456840515137, - "train/total_loss": 0.8469904065132141 + "train/classification_loss": 0.22704993188381195, + "train/contrastive_loss": 1.6753370761871338, + "train/negative_loss": 1.675333023071289, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 4.053128577652387e-06, + "train/total_loss": 0.562117338180542 + }, + { + "epoch": 7.577946768060836, + "step": 1000, + "train/classification_loss": 0.15847577154636383, + "train/contrastive_loss": 6.8455491065979, + "train/negative_loss": 6.295019626617432, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.5505294799804688, + "train/total_loss": 1.5275856256484985 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.11683640629053116, - "train/contrastive_loss": 1.420976161956787, - "train/negative_loss": 1.3318644762039185, + "train/classification_loss": 0.21001209318637848, + "train/contrastive_loss": 3.3458523750305176, + "train/negative_loss": 3.2161364555358887, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.08911172300577164, - "train/total_loss": 0.40103164315223694 + "train/positive_loss": 0.12971603870391846, + "train/total_loss": 0.8791825771331787 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.17816518247127533, - "train/contrastive_loss": 1.9548791646957397, - "train/negative_loss": 1.1922357082366943, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.7626434564590454, - "train/total_loss": 0.5691410303115845 + "train/classification_loss": 0.19133809208869934, + "train/contrastive_loss": 1.7908800840377808, + "train/negative_loss": 1.5941218137741089, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.19675831496715546, + "train/total_loss": 0.54951411485672 + }, + { + "epoch": 7.577946768060836, + "step": 1000, + "train/classification_loss": 0.17724844813346863, + "train/contrastive_loss": 1.7639106512069702, + "train/negative_loss": 1.763830304145813, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 8.035029168240726e-05, + "train/total_loss": 0.5300306081771851 + }, + { + "epoch": 7.577946768060836, + "step": 1000, + "train/classification_loss": 0.23568691313266754, + "train/contrastive_loss": 1.5096843242645264, + "train/negative_loss": 0.5105292201042175, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 0.9991551637649536, + "train/total_loss": 0.5376237630844116 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.1818161904811859, - "train/contrastive_loss": 2.5703072547912598, - "train/negative_loss": 1.1342227458953857, + "train/classification_loss": 0.2120995670557022, + "train/contrastive_loss": 0.5877801775932312, + "train/negative_loss": 0.46068525314331055, "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.4360846281051636, - "train/total_loss": 0.6958776712417603 + "train/num_positives": 8, + "train/positive_loss": 0.12709492444992065, + "train/total_loss": 0.32965558767318726 + }, + { + "epoch": 7.577946768060836, + "step": 1000, + "train/classification_loss": 0.18819859623908997, + "train/contrastive_loss": 2.1361868381500244, + "train/negative_loss": 1.9836761951446533, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.1525106579065323, + "train/total_loss": 0.6154359579086304 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.1086404100060463, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.1086404100060463 - }, - { - "epoch": 9.526315789473685, - "eval_exact_match_accuracy": 0.3014354066985646, - "eval_hamming_loss": 0.06107514776245426, - "eval_loss": 0.6445569396018982, - "eval_macro_f1": 0.14631658687970386, - "eval_macro_precision": 0.2043892283614389, - "eval_macro_recall": 0.12156862745098038, - "eval_micro_f1": 0.4771084337349398, - "eval_micro_precision": 0.8319327731092437, - "eval_micro_recall": 0.3344594594594595, - "eval_runtime": 1.0471, - "eval_samples_per_second": 199.597, - "eval_steps_per_second": 25.785, + "train/classification_loss": 0.24615265429019928, + "train/contrastive_loss": 2.253807783126831, + "train/negative_loss": 2.216228485107422, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.03757930546998978, + "train/total_loss": 0.6969141960144043 + }, + { + "epoch": 7.577946768060836, + "step": 1000, + "train/classification_loss": 0.1565929502248764, + "train/contrastive_loss": 2.0112814903259277, + "train/negative_loss": 2.0091552734375, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.0021262106020003557, + "train/total_loss": 0.5588492751121521 + }, + { + "epoch": 7.577946768060836, + "eval_exact_match_accuracy": 0.21755725190839695, + "eval_hamming_loss": 0.07072294566681635, + "eval_loss": 0.6709954738616943, + "eval_macro_f1": 0.09932349989175998, + "eval_macro_precision": 0.1338772090389048, + "eval_macro_recall": 0.08645257067451657, + "eval_micro_f1": 0.4090056285178236, + "eval_micro_precision": 0.7730496453900709, + "eval_micro_recall": 0.2780612244897959, + "eval_runtime": 1.3218, + "eval_samples_per_second": 198.217, + "eval_steps_per_second": 24.966, "step": 1000 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.18626493215560913, - "train/contrastive_loss": 0.54546719789505, - "train/negative_loss": 0.5378743410110474, - "train/num_negatives": 38, + "train/classification_loss": 0.19921652972698212, + "train/contrastive_loss": 0.33700379729270935, + "train/negative_loss": 0.33375483751296997, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.007592828944325447, - "train/total_loss": 0.2953583598136902 + "train/positive_loss": 0.003248952329158783, + "train/total_loss": 0.2666172981262207 }, { - "epoch": 9.526315789473685, + "epoch": 7.577946768060836, "step": 1000, - "train/classification_loss": 0.1604236215353012, - "train/contrastive_loss": 0.6820098757743835, - "train/negative_loss": 0.6030079126358032, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.07900197803974152, - "train/total_loss": 0.2968255877494812 + "train/classification_loss": 0.21502281725406647, + "train/contrastive_loss": 0.5365382432937622, + "train/negative_loss": 0.4231454133987427, + "train/num_negatives": 24, + "train/num_positives": 22, + "train/positive_loss": 0.11339285224676132, + "train/total_loss": 0.3223304748535156 }, { - "epoch": 10.0, - "grad_norm": 34.500614166259766, - "learning_rate": 1.9638549618320613e-05, - "loss": 0.6812, + "epoch": 7.9581749049429655, + "grad_norm": 6.332681179046631, + "learning_rate": 1.971259484066768e-05, + "loss": 0.7731, "step": 1050 }, { - "epoch": 10.0, + "epoch": 7.9581749049429655, "step": 1050, - "train/classification_loss": 0.1448352038860321, - "train/contrastive_loss": 0.6293611526489258, - "train/negative_loss": 0.6017172336578369, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.027643907815217972, - "train/total_loss": 0.2707074284553528 + "train/classification_loss": 0.16990222036838531, + "train/contrastive_loss": 0.665253758430481, + "train/negative_loss": 0.6553784608840942, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.009875306859612465, + "train/total_loss": 0.30295297503471375 }, { - "epoch": 10.0, + "epoch": 7.9581749049429655, "step": 1050, - "train/classification_loss": 0.12870746850967407, - "train/contrastive_loss": 1.522359848022461, - "train/negative_loss": 1.5222282409667969, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.00013160442176740617, - "train/total_loss": 0.43317943811416626 + "train/classification_loss": 0.14874427020549774, + "train/contrastive_loss": 0.9709217548370361, + "train/negative_loss": 0.3836873471736908, + "train/num_negatives": 34, + "train/num_positives": 20, + "train/positive_loss": 0.5872344374656677, + "train/total_loss": 0.34292861819267273 }, { - "epoch": 10.47846889952153, - "grad_norm": 12.773706436157227, - "learning_rate": 1.9619465648854964e-05, - "loss": 0.6739, + "epoch": 8.334600760456274, + "grad_norm": 7.52630615234375, + "learning_rate": 1.969742033383915e-05, + "loss": 0.7086, "step": 1100 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, + "step": 1100, + "train/classification_loss": 0.12251343578100204, + "train/contrastive_loss": 0.22896680235862732, + "train/negative_loss": 0.13655462861061096, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.09241216629743576, + "train/total_loss": 0.16830679774284363 + }, + { + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.16197429597377777, - "train/contrastive_loss": 2.2197794914245605, - "train/negative_loss": 1.3641504049301147, + "train/classification_loss": 0.22633583843708038, + "train/contrastive_loss": 1.983313798904419, + "train/negative_loss": 1.4943736791610718, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.8556290864944458, - "train/total_loss": 0.6059302091598511 + "train/num_positives": 6, + "train/positive_loss": 0.4889400601387024, + "train/total_loss": 0.6229985952377319 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.14130955934524536, - "train/contrastive_loss": 1.9777063131332397, - "train/negative_loss": 1.519243597984314, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.4584626853466034, - "train/total_loss": 0.5368508100509644 + "train/classification_loss": 0.12605704367160797, + "train/contrastive_loss": 2.334038734436035, + "train/negative_loss": 2.2816648483276367, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.05237400531768799, + "train/total_loss": 0.5928648114204407 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.16651763021945953, - "train/contrastive_loss": 1.1608952283859253, - "train/negative_loss": 0.8254909515380859, - "train/num_negatives": 44, + "train/classification_loss": 0.22938711941242218, + "train/contrastive_loss": 1.634263515472412, + "train/negative_loss": 1.3560837507247925, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.27817970514297485, + "train/total_loss": 0.5562398433685303 + }, + { + "epoch": 8.334600760456274, + "step": 1100, + "train/classification_loss": 0.11625441163778305, + "train/contrastive_loss": 0.7092167139053345, + "train/negative_loss": 0.6367916464805603, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.07242506742477417, + "train/total_loss": 0.258097767829895 + }, + { + "epoch": 8.334600760456274, + "step": 1100, + "train/classification_loss": 0.22331994771957397, + "train/contrastive_loss": 2.228241443634033, + "train/negative_loss": 1.4063652753829956, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.33540424704551697, - "train/total_loss": 0.3986966609954834 + "train/positive_loss": 0.8218762278556824, + "train/total_loss": 0.6689682006835938 + }, + { + "epoch": 8.334600760456274, + "step": 1100, + "train/classification_loss": 0.2390061318874359, + "train/contrastive_loss": 3.6401987075805664, + "train/negative_loss": 1.783610224723816, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.85658860206604, + "train/total_loss": 0.9670459032058716 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.16691620647907257, - "train/contrastive_loss": 3.259784698486328, - "train/negative_loss": 1.6645179986953735, + "train/classification_loss": 0.2075033038854599, + "train/contrastive_loss": 1.389182686805725, + "train/negative_loss": 1.155967116355896, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.5952668190002441, - "train/total_loss": 0.8188731670379639 + "train/positive_loss": 0.23321561515331268, + "train/total_loss": 0.48533982038497925 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.12449623644351959, - "train/contrastive_loss": 2.8856053352355957, - "train/negative_loss": 0.8896602392196655, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.9959449768066406, - "train/total_loss": 0.7016173005104065 + "train/classification_loss": 0.13201560080051422, + "train/contrastive_loss": 0.21402695775032043, + "train/negative_loss": 0.16528062522411346, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.04874633997678757, + "train/total_loss": 0.17482098937034607 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.17125259339809418, - "train/contrastive_loss": 3.0673255920410156, - "train/negative_loss": 1.9484659433364868, - "train/num_negatives": 32, + "train/classification_loss": 0.2029905468225479, + "train/contrastive_loss": 2.673232316970825, + "train/negative_loss": 2.301500082015991, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.3717322051525116, + "train/total_loss": 0.7376369833946228 + }, + { + "epoch": 8.334600760456274, + "step": 1100, + "train/classification_loss": 0.18528428673744202, + "train/contrastive_loss": 4.296278476715088, + "train/negative_loss": 4.207477569580078, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.1188596487045288, - "train/total_loss": 0.7847177386283875 + "train/positive_loss": 0.08880069106817245, + "train/total_loss": 1.0445400476455688 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.18768465518951416, - "train/contrastive_loss": 1.5151920318603516, - "train/negative_loss": 1.3893792629241943, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.12581279873847961, - "train/total_loss": 0.4907230734825134 + "train/classification_loss": 0.19494333863258362, + "train/contrastive_loss": 3.2654266357421875, + "train/negative_loss": 1.9405418634414673, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 1.3248848915100098, + "train/total_loss": 0.8480286598205566 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.2540031969547272, - "train/contrastive_loss": 1.7753832340240479, - "train/negative_loss": 1.767555594444275, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.007827657274901867, - "train/total_loss": 0.6090798377990723 + "train/classification_loss": 0.23305070400238037, + "train/contrastive_loss": 2.0314908027648926, + "train/negative_loss": 1.229413390159607, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.8020774126052856, + "train/total_loss": 0.6393488645553589 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.1707945168018341, - "train/contrastive_loss": 1.135211706161499, - "train/negative_loss": 0.970426619052887, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.16478510200977325, - "train/total_loss": 0.3978368639945984 + "train/classification_loss": 0.22206592559814453, + "train/contrastive_loss": 2.730114221572876, + "train/negative_loss": 1.444948673248291, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.285165548324585, + "train/total_loss": 0.7680887579917908 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.2237662374973297, - "train/contrastive_loss": 2.4785070419311523, - "train/negative_loss": 1.0827609300613403, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 1.395746111869812, - "train/total_loss": 0.7194676399230957 + "train/classification_loss": 0.21598504483699799, + "train/contrastive_loss": 2.841259717941284, + "train/negative_loss": 2.100203514099121, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.7410562038421631, + "train/total_loss": 0.7842370271682739 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.2194347083568573, - "train/contrastive_loss": 3.620847463607788, - "train/negative_loss": 3.094550848007202, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.5262965559959412, - "train/total_loss": 0.9436042308807373 + "train/classification_loss": 0.18514758348464966, + "train/contrastive_loss": 1.032913088798523, + "train/negative_loss": 1.0315098762512207, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0014031573664397001, + "train/total_loss": 0.3917301893234253 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.16946591436862946, - "train/contrastive_loss": 1.433961272239685, - "train/negative_loss": 1.0363463163375854, + "train/classification_loss": 0.19285441935062408, + "train/contrastive_loss": 1.2532305717468262, + "train/negative_loss": 0.8396962285041809, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 0.3976149559020996, - "train/total_loss": 0.4562581777572632 + "train/positive_loss": 0.41353434324264526, + "train/total_loss": 0.4435005187988281 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.17939524352550507, - "train/contrastive_loss": 3.7352962493896484, - "train/negative_loss": 3.220642328262329, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.5146539211273193, - "train/total_loss": 0.9264545440673828 + "train/classification_loss": 0.1765381544828415, + "train/contrastive_loss": 2.9992740154266357, + "train/negative_loss": 2.6882519721984863, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.31102198362350464, + "train/total_loss": 0.7763929963111877 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.0994245707988739, - "train/contrastive_loss": 0.2713267505168915, - "train/negative_loss": 0.26938101649284363, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.001945739728398621, - "train/total_loss": 0.1536899209022522 + "train/classification_loss": 0.2415505051612854, + "train/contrastive_loss": 3.9982848167419434, + "train/negative_loss": 2.546706199645996, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.4515784978866577, + "train/total_loss": 1.0412075519561768 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.20828284323215485, - "train/contrastive_loss": 2.9073662757873535, - "train/negative_loss": 1.0185014009475708, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.8888648748397827, - "train/total_loss": 0.7897561192512512 + "train/classification_loss": 0.22855950891971588, + "train/contrastive_loss": 3.060380458831787, + "train/negative_loss": 1.891771674156189, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.1686089038848877, + "train/total_loss": 0.8406355977058411 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.14088210463523865, - "train/contrastive_loss": 1.1155530214309692, - "train/negative_loss": 1.0272313356399536, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.08832165598869324, - "train/total_loss": 0.36399269104003906 + "train/classification_loss": 0.26204755902290344, + "train/contrastive_loss": 0.7535474896430969, + "train/negative_loss": 0.7412403225898743, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.012307152152061462, + "train/total_loss": 0.4127570390701294 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.23819144070148468, - "train/contrastive_loss": 4.507602691650391, - "train/negative_loss": 2.1712241172790527, - "train/num_negatives": 46, + "train/classification_loss": 0.2552570104598999, + "train/contrastive_loss": 2.1103200912475586, + "train/negative_loss": 0.9599644541740417, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 2.336378574371338, - "train/total_loss": 1.1397119760513306 + "train/positive_loss": 1.150355577468872, + "train/total_loss": 0.6773210763931274 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.2316945493221283, - "train/contrastive_loss": 2.648344039916992, - "train/negative_loss": 1.9160257577896118, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.7323184013366699, - "train/total_loss": 0.7613633871078491 + "train/classification_loss": 0.2245423048734665, + "train/contrastive_loss": 1.8047780990600586, + "train/negative_loss": 1.0521111488342285, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.7526668906211853, + "train/total_loss": 0.5854979157447815 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.19705693423748016, - "train/contrastive_loss": 3.4346604347229004, - "train/negative_loss": 2.098921298980713, + "train/classification_loss": 0.21747775375843048, + "train/contrastive_loss": 1.1371403932571411, + "train/negative_loss": 1.137131929397583, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.3357391357421875, - "train/total_loss": 0.8839890360832214 + "train/num_positives": 8, + "train/positive_loss": 8.493709174217656e-06, + "train/total_loss": 0.4449058175086975 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.22427333891391754, - "train/contrastive_loss": 2.13596248626709, - "train/negative_loss": 1.9284706115722656, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.20749180018901825, - "train/total_loss": 0.6514658331871033 + "train/classification_loss": 0.15107844769954681, + "train/contrastive_loss": 4.67556619644165, + "train/negative_loss": 3.742248058319092, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.9333182573318481, + "train/total_loss": 1.0861916542053223 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.19471687078475952, - "train/contrastive_loss": 1.0363649129867554, - "train/negative_loss": 0.6711904406547546, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.36517447233200073, - "train/total_loss": 0.4019898772239685 + "train/classification_loss": 0.20439930260181427, + "train/contrastive_loss": 3.2523410320281982, + "train/negative_loss": 2.6617820262908936, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.5905590057373047, + "train/total_loss": 0.8548675179481506 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.18752555549144745, - "train/contrastive_loss": 5.93960428237915, - "train/negative_loss": 5.622648239135742, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.3169560432434082, - "train/total_loss": 1.3754464387893677 + "train/classification_loss": 0.20386584103107452, + "train/contrastive_loss": 2.7005093097686768, + "train/negative_loss": 1.329880714416504, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.3706285953521729, + "train/total_loss": 0.7439677119255066 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.18481187522411346, - "train/contrastive_loss": 2.496100425720215, - "train/negative_loss": 1.6814502477645874, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 0.8146501183509827, - "train/total_loss": 0.6840319633483887 + "train/classification_loss": 0.1687382310628891, + "train/contrastive_loss": 2.451442003250122, + "train/negative_loss": 2.45143461227417, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 7.391003236989491e-06, + "train/total_loss": 0.6590266227722168 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.11033163219690323, - "train/contrastive_loss": 1.5645242929458618, - "train/negative_loss": 1.4427884817123413, - "train/num_negatives": 42, + "train/classification_loss": 0.2263518124818802, + "train/contrastive_loss": 1.9750890731811523, + "train/negative_loss": 0.6379002332687378, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.12173578888177872, - "train/total_loss": 0.4232364892959595 + "train/positive_loss": 1.3371888399124146, + "train/total_loss": 0.6213696599006653 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.17155157029628754, - "train/contrastive_loss": 2.231532096862793, - "train/negative_loss": 1.399855613708496, - "train/num_negatives": 48, + "train/classification_loss": 0.2130688577890396, + "train/contrastive_loss": 0.8541478514671326, + "train/negative_loss": 0.776549220085144, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.8316763639450073, - "train/total_loss": 0.6178579926490784 + "train/positive_loss": 0.07759862393140793, + "train/total_loss": 0.38389843702316284 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.18335530161857605, - "train/contrastive_loss": 2.884373664855957, - "train/negative_loss": 1.1257481575012207, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.7586256265640259, - "train/total_loss": 0.7602300643920898 + "train/classification_loss": 0.19012342393398285, + "train/contrastive_loss": 3.2425622940063477, + "train/negative_loss": 2.0983047485351562, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.1442575454711914, + "train/total_loss": 0.8386359214782715 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.10455691814422607, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.10455691814422607 - }, - { - "epoch": 10.47846889952153, - "eval_exact_match_accuracy": 0.3253588516746411, - "eval_hamming_loss": 0.06051224317478188, - "eval_loss": 0.6664091944694519, - "eval_macro_f1": 0.17027789333983337, - "eval_macro_precision": 0.2546695434930729, - "eval_macro_recall": 0.1489777325071443, - "eval_micro_f1": 0.5011600928074246, - "eval_micro_precision": 0.8, - "eval_micro_recall": 0.36486486486486486, - "eval_runtime": 1.0641, - "eval_samples_per_second": 196.412, - "eval_steps_per_second": 25.374, - "step": 1100 + "train/classification_loss": 0.23661625385284424, + "train/contrastive_loss": 1.2800501585006714, + "train/negative_loss": 1.2601630687713623, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.019887108355760574, + "train/total_loss": 0.49262627959251404 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.17901572585105896, - "train/contrastive_loss": 0.9261891841888428, - "train/negative_loss": 0.6248205304145813, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.30136868357658386, - "train/total_loss": 0.36425358057022095 + "train/classification_loss": 0.14576472342014313, + "train/contrastive_loss": 1.8182165622711182, + "train/negative_loss": 1.8156219720840454, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.002594555728137493, + "train/total_loss": 0.5094080567359924 + }, + { + "epoch": 8.334600760456274, + "eval_exact_match_accuracy": 0.24427480916030533, + "eval_hamming_loss": 0.06668163448585541, + "eval_loss": 0.6410121917724609, + "eval_macro_f1": 0.11727486830181538, + "eval_macro_precision": 0.14232923938806294, + "eval_macro_recall": 0.1011559153428412, + "eval_micro_f1": 0.44692737430167595, + "eval_micro_precision": 0.8275862068965517, + "eval_micro_recall": 0.30612244897959184, + "eval_runtime": 1.3167, + "eval_samples_per_second": 198.975, + "eval_steps_per_second": 25.062, + "step": 1100 }, { - "epoch": 10.47846889952153, + "epoch": 8.334600760456274, "step": 1100, - "train/classification_loss": 0.17904946208000183, - "train/contrastive_loss": 0.287465900182724, - "train/negative_loss": 0.28102874755859375, - "train/num_negatives": 36, + "train/classification_loss": 0.21815134584903717, + "train/contrastive_loss": 1.0354244709014893, + "train/negative_loss": 0.6132058501243591, + "train/num_negatives": 34, "train/num_positives": 16, - "train/positive_loss": 0.006437139585614204, - "train/total_loss": 0.23654264211654663 + "train/positive_loss": 0.4222186207771301, + "train/total_loss": 0.42523622512817383 + }, + { + "epoch": 8.334600760456274, + "step": 1100, + "train/classification_loss": 0.15702518820762634, + "train/contrastive_loss": 0.859784722328186, + "train/negative_loss": 0.6582561731338501, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.20152854919433594, + "train/total_loss": 0.3289821147918701 }, { - "epoch": 10.956937799043063, - "grad_norm": 11.332853317260742, - "learning_rate": 1.9600381679389316e-05, - "loss": 0.618, + "epoch": 8.714828897338403, + "grad_norm": 8.422256469726562, + "learning_rate": 1.9682245827010626e-05, + "loss": 0.6933, "step": 1150 }, { - "epoch": 10.956937799043063, + "epoch": 8.714828897338403, "step": 1150, - "train/classification_loss": 0.16004574298858643, - "train/contrastive_loss": 1.0512441396713257, - "train/negative_loss": 0.7880906462669373, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.2631535232067108, - "train/total_loss": 0.37029457092285156 + "train/classification_loss": 0.13878974318504333, + "train/contrastive_loss": 1.083322525024414, + "train/negative_loss": 0.6921277642250061, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.39119473099708557, + "train/total_loss": 0.3554542660713196 }, { - "epoch": 10.956937799043063, + "epoch": 8.714828897338403, "step": 1150, - "train/classification_loss": 0.10191157460212708, - "train/contrastive_loss": 0.20182311534881592, - "train/negative_loss": 0.19957080483436584, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.002252305392175913, - "train/total_loss": 0.14227619767189026 + "train/classification_loss": 0.2356548011302948, + "train/contrastive_loss": 0.9873168468475342, + "train/negative_loss": 0.808514416217804, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.1788024604320526, + "train/total_loss": 0.43311816453933716 }, { - "epoch": 11.430622009569378, - "grad_norm": 8.150411605834961, - "learning_rate": 1.9581297709923667e-05, - "loss": 0.642, + "epoch": 9.091254752851711, + "grad_norm": 9.643750190734863, + "learning_rate": 1.9667071320182097e-05, + "loss": 0.6842, "step": 1200 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.16001516580581665, - "train/contrastive_loss": 2.294701337814331, - "train/negative_loss": 1.46543288230896, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.8292684555053711, - "train/total_loss": 0.6189554333686829 + "train/classification_loss": 0.11522156000137329, + "train/contrastive_loss": 0.6304599642753601, + "train/negative_loss": 0.5779423713684082, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.052517566829919815, + "train/total_loss": 0.24131356179714203 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.14391948282718658, - "train/contrastive_loss": 1.6825934648513794, - "train/negative_loss": 0.9074845314025879, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.7751089334487915, - "train/total_loss": 0.4804381728172302 + "train/classification_loss": 0.2299971729516983, + "train/contrastive_loss": 2.0912272930145264, + "train/negative_loss": 1.8536708354949951, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.2375563681125641, + "train/total_loss": 0.6482426524162292 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.15415945649147034, - "train/contrastive_loss": 1.1038308143615723, - "train/negative_loss": 0.5656692385673523, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.53816157579422, - "train/total_loss": 0.3749256134033203 + "train/classification_loss": 0.12348054349422455, + "train/contrastive_loss": 2.2564446926116943, + "train/negative_loss": 2.091764211654663, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.16468052566051483, + "train/total_loss": 0.5747694969177246 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.1754656285047531, - "train/contrastive_loss": 2.805241107940674, - "train/negative_loss": 1.4571949243545532, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.3480461835861206, - "train/total_loss": 0.7365138530731201 + "train/classification_loss": 0.23564372956752777, + "train/contrastive_loss": 1.7974443435668945, + "train/negative_loss": 1.1422911882400513, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.655153214931488, + "train/total_loss": 0.59513258934021 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.12056183815002441, - "train/contrastive_loss": 2.6797897815704346, - "train/negative_loss": 0.9693650603294373, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.7104246616363525, - "train/total_loss": 0.6565198302268982 + "train/classification_loss": 0.10654003918170929, + "train/contrastive_loss": 0.9370830059051514, + "train/negative_loss": 0.9335377812385559, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.003545238170772791, + "train/total_loss": 0.2939566373825073 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.16333021223545074, - "train/contrastive_loss": 2.764946222305298, - "train/negative_loss": 1.2136905193328857, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.551255702972412, - "train/total_loss": 0.7163194417953491 + "train/classification_loss": 0.22713603079319, + "train/contrastive_loss": 1.9015052318572998, + "train/negative_loss": 1.543819785118103, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.35768550634384155, + "train/total_loss": 0.6074370741844177 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.18110817670822144, - "train/contrastive_loss": 1.7087739706039429, - "train/negative_loss": 1.0957139730453491, + "train/classification_loss": 0.23175375163555145, + "train/contrastive_loss": 2.7911629676818848, + "train/negative_loss": 2.0473334789276123, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.6130599975585938, - "train/total_loss": 0.52286297082901 + "train/num_positives": 8, + "train/positive_loss": 0.7438293695449829, + "train/total_loss": 0.7899863719940186 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.24998508393764496, - "train/contrastive_loss": 1.0922963619232178, - "train/negative_loss": 1.0913116931915283, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.0009847183246165514, - "train/total_loss": 0.4684443473815918 + "train/classification_loss": 0.2095751315355301, + "train/contrastive_loss": 1.158883810043335, + "train/negative_loss": 0.9564789533615112, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.20240481197834015, + "train/total_loss": 0.44135189056396484 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.16859528422355652, - "train/contrastive_loss": 1.3366870880126953, - "train/negative_loss": 0.825068473815918, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.5116186141967773, - "train/total_loss": 0.4359326958656311 + "train/classification_loss": 0.1376168429851532, + "train/contrastive_loss": 0.510254442691803, + "train/negative_loss": 0.3015771210193634, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.20867732167243958, + "train/total_loss": 0.23966774344444275 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.20506130158901215, - "train/contrastive_loss": 3.530691623687744, - "train/negative_loss": 1.2335131168365479, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.2971785068511963, - "train/total_loss": 0.9111996293067932 + "train/classification_loss": 0.1969045102596283, + "train/contrastive_loss": 2.356926679611206, + "train/negative_loss": 2.3466153144836426, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.010311255231499672, + "train/total_loss": 0.668289840221405 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.2083723396062851, - "train/contrastive_loss": 1.9419358968734741, - "train/negative_loss": 1.8281525373458862, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.11378338932991028, - "train/total_loss": 0.5967594981193542 + "train/classification_loss": 0.1824135184288025, + "train/contrastive_loss": 4.847239017486572, + "train/negative_loss": 4.841297149658203, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.005942013114690781, + "train/total_loss": 1.151861310005188 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.16814740002155304, - "train/contrastive_loss": 1.1285849809646606, - "train/negative_loss": 0.8726739883422852, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.2559109926223755, - "train/total_loss": 0.39386439323425293 + "train/classification_loss": 0.1934165358543396, + "train/contrastive_loss": 3.668423891067505, + "train/negative_loss": 3.460658550262451, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.20776523649692535, + "train/total_loss": 0.9271013140678406 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.1799236238002777, - "train/contrastive_loss": 3.4011142253875732, - "train/negative_loss": 2.571828842163086, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.8292854428291321, - "train/total_loss": 0.8601465225219727 + "train/classification_loss": 0.23485150933265686, + "train/contrastive_loss": 2.848271608352661, + "train/negative_loss": 1.646693229675293, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.2015783786773682, + "train/total_loss": 0.8045058250427246 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.09177321195602417, - "train/contrastive_loss": 0.2999281585216522, - "train/negative_loss": 0.29682159423828125, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.0031065649818629026, - "train/total_loss": 0.1517588496208191 + "train/classification_loss": 0.22515466809272766, + "train/contrastive_loss": 2.569376230239868, + "train/negative_loss": 1.815672516822815, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 0.7537037134170532, + "train/total_loss": 0.7390298843383789 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.20968855917453766, - "train/contrastive_loss": 1.2754069566726685, - "train/negative_loss": 0.9297276735305786, + "train/classification_loss": 0.20222492516040802, + "train/contrastive_loss": 3.0450212955474854, + "train/negative_loss": 3.023350238800049, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.34567925333976746, - "train/total_loss": 0.46476995944976807 + "train/positive_loss": 0.02167099341750145, + "train/total_loss": 0.8112291693687439 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.13485682010650635, - "train/contrastive_loss": 1.1733124256134033, - "train/negative_loss": 0.8116272687911987, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.3616851568222046, - "train/total_loss": 0.36951929330825806 + "train/classification_loss": 0.1695021241903305, + "train/contrastive_loss": 0.9501579403877258, + "train/negative_loss": 0.950141966342926, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.5974186680978164e-05, + "train/total_loss": 0.35953372716903687 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.23820088803768158, - "train/contrastive_loss": 4.065998077392578, - "train/negative_loss": 2.3594019412994385, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.7065963745117188, - "train/total_loss": 1.0514005422592163 + "train/classification_loss": 0.1979752480983734, + "train/contrastive_loss": 0.9535108208656311, + "train/negative_loss": 0.8570928573608398, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.09641797840595245, + "train/total_loss": 0.3886774182319641 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.23777294158935547, - "train/contrastive_loss": 3.052964448928833, - "train/negative_loss": 1.7835121154785156, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.2694523334503174, - "train/total_loss": 0.848365843296051 + "train/classification_loss": 0.1797790825366974, + "train/contrastive_loss": 3.6323130130767822, + "train/negative_loss": 2.3202016353607178, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.3121113777160645, + "train/total_loss": 0.9062416553497314 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.2076777070760727, - "train/contrastive_loss": 3.575380325317383, - "train/negative_loss": 2.139606237411499, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.4357742071151733, - "train/total_loss": 0.9227538108825684 - }, + "train/classification_loss": 0.24152186512947083, + "train/contrastive_loss": 3.505556583404541, + "train/negative_loss": 2.6159703731536865, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.8895860910415649, + "train/total_loss": 0.9426331520080566 + }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.2257893830537796, - "train/contrastive_loss": 2.2169249057769775, - "train/negative_loss": 1.3995481729507446, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.8173767924308777, - "train/total_loss": 0.6691743731498718 + "train/classification_loss": 0.23048178851604462, + "train/contrastive_loss": 2.120089054107666, + "train/negative_loss": 1.587506651878357, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.5325824022293091, + "train/total_loss": 0.6544995903968811 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.1999325156211853, - "train/contrastive_loss": 1.4261304140090942, - "train/negative_loss": 1.0465446710586548, - "train/num_negatives": 44, + "train/classification_loss": 0.24754279851913452, + "train/contrastive_loss": 1.247743844985962, + "train/negative_loss": 0.7806310057640076, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.46711283922195435, + "train/total_loss": 0.4970915913581848 + }, + { + "epoch": 9.091254752851711, + "step": 1200, + "train/classification_loss": 0.24241851270198822, + "train/contrastive_loss": 2.8201231956481934, + "train/negative_loss": 0.6359479427337646, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.37958574295043945, - "train/total_loss": 0.4851585924625397 + "train/positive_loss": 2.1841752529144287, + "train/total_loss": 0.8064431548118591 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.18529735505580902, - "train/contrastive_loss": 5.070247173309326, - "train/negative_loss": 4.098145484924316, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.9721016883850098, - "train/total_loss": 1.1993467807769775 + "train/classification_loss": 0.2268640547990799, + "train/contrastive_loss": 1.390615701675415, + "train/negative_loss": 1.3187971115112305, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.07181864231824875, + "train/total_loss": 0.5049871802330017 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.18481363356113434, - "train/contrastive_loss": 3.3183343410491943, - "train/negative_loss": 1.4643878936767578, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.8539464473724365, - "train/total_loss": 0.8484805226325989 + "train/classification_loss": 0.21813519299030304, + "train/contrastive_loss": 0.8970542550086975, + "train/negative_loss": 0.8970496654510498, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 4.589572654367657e-06, + "train/total_loss": 0.39754605293273926 + }, + { + "epoch": 9.091254752851711, + "step": 1200, + "train/classification_loss": 0.15399520099163055, + "train/contrastive_loss": 4.8973307609558105, + "train/negative_loss": 4.280398368835449, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.6169325113296509, + "train/total_loss": 1.133461356163025 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.0917481854557991, - "train/contrastive_loss": 0.5597933530807495, - "train/negative_loss": 0.36864009499549866, + "train/classification_loss": 0.20128655433654785, + "train/contrastive_loss": 3.0032362937927246, + "train/negative_loss": 2.670208692550659, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.19115322828292847, - "train/total_loss": 0.20370686054229736 + "train/positive_loss": 0.33302757143974304, + "train/total_loss": 0.8019338250160217 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.16546712815761566, - "train/contrastive_loss": 1.7752612829208374, - "train/negative_loss": 0.9623466730117798, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.8129146099090576, - "train/total_loss": 0.5205193758010864 + "train/classification_loss": 0.20780350267887115, + "train/contrastive_loss": 2.902285575866699, + "train/negative_loss": 1.8058192729949951, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.0964664220809937, + "train/total_loss": 0.7882606387138367 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.18767699599266052, - "train/contrastive_loss": 2.5296854972839355, - "train/negative_loss": 1.0200064182281494, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.5096790790557861, - "train/total_loss": 0.69361412525177 + "train/classification_loss": 0.17104637622833252, + "train/contrastive_loss": 2.270787477493286, + "train/negative_loss": 2.2690377235412598, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.0017496744403615594, + "train/total_loss": 0.6252038478851318 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.09688097983598709, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.09688097983598709 - }, - { - "epoch": 11.430622009569378, - "eval_exact_match_accuracy": 0.3444976076555024, - "eval_hamming_loss": 0.05966788629327329, - "eval_loss": 0.6206148266792297, - "eval_macro_f1": 0.20351581928317886, - "eval_macro_precision": 0.2981490635469874, - "eval_macro_recall": 0.17397051749992928, - "eval_micro_f1": 0.5203619909502263, - "eval_micro_precision": 0.7876712328767124, - "eval_micro_recall": 0.3885135135135135, - "eval_runtime": 1.0626, - "eval_samples_per_second": 196.694, - "eval_steps_per_second": 25.41, - "step": 1200 + "train/classification_loss": 0.2270047515630722, + "train/contrastive_loss": 2.1957216262817383, + "train/negative_loss": 0.6205394864082336, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 1.5751821994781494, + "train/total_loss": 0.6661490797996521 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.1724703460931778, - "train/contrastive_loss": 0.6243595480918884, - "train/negative_loss": 0.599016547203064, + "train/classification_loss": 0.20535476505756378, + "train/contrastive_loss": 1.4335236549377441, + "train/negative_loss": 1.0636996030807495, "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.025342995300889015, - "train/total_loss": 0.2973422408103943 + "train/num_positives": 8, + "train/positive_loss": 0.369824081659317, + "train/total_loss": 0.49205952882766724 }, { - "epoch": 11.430622009569378, + "epoch": 9.091254752851711, "step": 1200, - "train/classification_loss": 0.13278736174106598, - "train/contrastive_loss": 0.39753156900405884, - "train/negative_loss": 0.39150819182395935, - "train/num_negatives": 38, + "train/classification_loss": 0.1863492876291275, + "train/contrastive_loss": 2.1735870838165283, + "train/negative_loss": 1.7748841047286987, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.3987028896808624, + "train/total_loss": 0.621066689491272 + }, + { + "epoch": 9.091254752851711, + "step": 1200, + "train/classification_loss": 0.22861790657043457, + "train/contrastive_loss": 1.8551541566848755, + "train/negative_loss": 1.8496192693710327, + "train/num_negatives": 42, "train/num_positives": 12, - "train/positive_loss": 0.006023370660841465, - "train/total_loss": 0.21229368448257446 + "train/positive_loss": 0.005534886848181486, + "train/total_loss": 0.5996487140655518 + }, + { + "epoch": 9.091254752851711, + "step": 1200, + "train/classification_loss": 0.13485419750213623, + "train/contrastive_loss": 1.4621001482009888, + "train/negative_loss": 1.461337924003601, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.000762276235036552, + "train/total_loss": 0.427274227142334 + }, + { + "epoch": 9.091254752851711, + "eval_exact_match_accuracy": 0.2480916030534351, + "eval_hamming_loss": 0.0673551863493489, + "eval_loss": 0.642435610294342, + "eval_macro_f1": 0.141847603748241, + "eval_macro_precision": 0.19350226649799704, + "eval_macro_recall": 0.1193832580486897, + "eval_micro_f1": 0.45652173913043476, + "eval_micro_precision": 0.7875, + "eval_micro_recall": 0.32142857142857145, + "eval_runtime": 1.3178, + "eval_samples_per_second": 198.821, + "eval_steps_per_second": 25.042, + "step": 1200 + }, + { + "epoch": 9.091254752851711, + "step": 1200, + "train/classification_loss": 0.12735812366008759, + "train/contrastive_loss": 0.598491370677948, + "train/negative_loss": 0.5459671020507812, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.05252426862716675, + "train/total_loss": 0.24705639481544495 + }, + { + "epoch": 9.091254752851711, + "step": 1200, + "train/classification_loss": 0.17101776599884033, + "train/contrastive_loss": 0.5124104619026184, + "train/negative_loss": 0.512086033821106, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 0.0003244396939408034, + "train/total_loss": 0.27349984645843506 }, { - "epoch": 11.909090909090908, - "grad_norm": 9.242064476013184, - "learning_rate": 1.9562213740458014e-05, - "loss": 0.6029, + "epoch": 9.47148288973384, + "grad_norm": 16.603618621826172, + "learning_rate": 1.965189681335357e-05, + "loss": 0.6554, "step": 1250 }, { - "epoch": 11.909090909090908, + "epoch": 9.47148288973384, "step": 1250, - "train/classification_loss": 0.10649878531694412, - "train/contrastive_loss": 0.36039918661117554, - "train/negative_loss": 0.2182859629392624, - "train/num_negatives": 32, - "train/num_positives": 24, - "train/positive_loss": 0.14211323857307434, - "train/total_loss": 0.17857861518859863 + "train/classification_loss": 0.16121473908424377, + "train/contrastive_loss": 0.45338547229766846, + "train/negative_loss": 0.40742233395576477, + "train/num_negatives": 38, + "train/num_positives": 18, + "train/positive_loss": 0.045963142067193985, + "train/total_loss": 0.2518918514251709 }, { - "epoch": 11.909090909090908, + "epoch": 9.47148288973384, "step": 1250, - "train/classification_loss": 0.12458126991987228, - "train/contrastive_loss": 0.2008017897605896, - "train/negative_loss": 0.15648171305656433, - "train/num_negatives": 38, - "train/num_positives": 16, - "train/positive_loss": 0.04432007297873497, - "train/total_loss": 0.1647416353225708 + "train/classification_loss": 0.1708284467458725, + "train/contrastive_loss": 0.6910529136657715, + "train/negative_loss": 0.6908326745033264, + "train/num_negatives": 50, + "train/num_positives": 4, + "train/positive_loss": 0.00022025511134415865, + "train/total_loss": 0.30903902649879456 }, { - "epoch": 12.382775119617225, - "grad_norm": 6.404537677764893, - "learning_rate": 1.954312977099237e-05, - "loss": 0.5809, + "epoch": 9.85171102661597, + "grad_norm": 13.335227966308594, + "learning_rate": 1.963672230652504e-05, + "loss": 0.644, "step": 1300 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, + "step": 1300, + "train/classification_loss": 0.10248986631631851, + "train/contrastive_loss": 0.3606616258621216, + "train/negative_loss": 0.2979530394077301, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.06270860135555267, + "train/total_loss": 0.17462219297885895 + }, + { + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.14872217178344727, - "train/contrastive_loss": 1.4446089267730713, - "train/negative_loss": 1.109423279762268, + "train/classification_loss": 0.22349895536899567, + "train/contrastive_loss": 1.997036099433899, + "train/negative_loss": 1.9433534145355225, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.3351856470108032, - "train/total_loss": 0.43764397501945496 + "train/num_positives": 6, + "train/positive_loss": 0.053682684898376465, + "train/total_loss": 0.6229062080383301 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.15272225439548492, - "train/contrastive_loss": 2.0208535194396973, - "train/negative_loss": 1.2218598127365112, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.7989935874938965, - "train/total_loss": 0.556892991065979 + "train/classification_loss": 0.10866637527942657, + "train/contrastive_loss": 2.1698367595672607, + "train/negative_loss": 2.1241445541381836, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.04569212347269058, + "train/total_loss": 0.5426337122917175 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.14189593493938446, - "train/contrastive_loss": 0.9813214540481567, - "train/negative_loss": 0.6098936200141907, - "train/num_negatives": 44, + "train/classification_loss": 0.1988143026828766, + "train/contrastive_loss": 1.5188987255096436, + "train/negative_loss": 1.3940938711166382, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.12480480968952179, + "train/total_loss": 0.5025940537452698 + }, + { + "epoch": 9.85171102661597, + "step": 1300, + "train/classification_loss": 0.10656008124351501, + "train/contrastive_loss": 0.9869464039802551, + "train/negative_loss": 0.9159125089645386, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.07103390991687775, + "train/total_loss": 0.30394935607910156 + }, + { + "epoch": 9.85171102661597, + "step": 1300, + "train/classification_loss": 0.23176509141921997, + "train/contrastive_loss": 1.9021978378295898, + "train/negative_loss": 1.4188199043273926, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.37142786383628845, - "train/total_loss": 0.3381602168083191 + "train/positive_loss": 0.48337799310684204, + "train/total_loss": 0.6122046709060669 + }, + { + "epoch": 9.85171102661597, + "step": 1300, + "train/classification_loss": 0.23929500579833984, + "train/contrastive_loss": 3.192241907119751, + "train/negative_loss": 2.2589659690856934, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.9332759380340576, + "train/total_loss": 0.8777434229850769 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.16442817449569702, - "train/contrastive_loss": 2.311100721359253, - "train/negative_loss": 1.310004711151123, + "train/classification_loss": 0.19203722476959229, + "train/contrastive_loss": 1.7736153602600098, + "train/negative_loss": 1.7328057289123535, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.0010960102081299, - "train/total_loss": 0.6266483068466187 + "train/positive_loss": 0.040809597820043564, + "train/total_loss": 0.5467603206634521 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.13230237364768982, - "train/contrastive_loss": 3.4720630645751953, - "train/negative_loss": 1.2752255201339722, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 2.1968376636505127, - "train/total_loss": 0.8267149925231934 + "train/classification_loss": 0.1387682855129242, + "train/contrastive_loss": 1.7775815725326538, + "train/negative_loss": 1.3157317638397217, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.46184980869293213, + "train/total_loss": 0.49428460001945496 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.1547008752822876, - "train/contrastive_loss": 2.670513391494751, - "train/negative_loss": 1.2977054119110107, - "train/num_negatives": 32, + "train/classification_loss": 0.1939309984445572, + "train/contrastive_loss": 2.8317439556121826, + "train/negative_loss": 2.194833993911743, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.6369099020957947, + "train/total_loss": 0.7602797746658325 + }, + { + "epoch": 9.85171102661597, + "step": 1300, + "train/classification_loss": 0.1738797426223755, + "train/contrastive_loss": 3.995946168899536, + "train/negative_loss": 3.915626049041748, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.3728079795837402, - "train/total_loss": 0.6888035535812378 + "train/positive_loss": 0.08032016456127167, + "train/total_loss": 0.9730690121650696 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.185238316655159, - "train/contrastive_loss": 2.0007500648498535, - "train/negative_loss": 1.1072773933410645, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.8934725522994995, - "train/total_loss": 0.5853883624076843 + "train/classification_loss": 0.1789366602897644, + "train/contrastive_loss": 3.2153639793395996, + "train/negative_loss": 3.046599864959717, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.1687641441822052, + "train/total_loss": 0.8220094442367554 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.256093293428421, - "train/contrastive_loss": 1.4009066820144653, - "train/negative_loss": 1.4000543355941772, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.0008523456053808331, - "train/total_loss": 0.5362746715545654 + "train/classification_loss": 0.21628570556640625, + "train/contrastive_loss": 2.1340293884277344, + "train/negative_loss": 1.2564226388931274, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.8776067495346069, + "train/total_loss": 0.6430915594100952 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.16700118780136108, - "train/contrastive_loss": 1.552871823310852, - "train/negative_loss": 0.9695398211479187, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.5833320021629333, - "train/total_loss": 0.4775755703449249 + "train/classification_loss": 0.20472365617752075, + "train/contrastive_loss": 2.2056353092193604, + "train/negative_loss": 1.0115885734558105, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.1940467357635498, + "train/total_loss": 0.6458507180213928 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.22046969830989838, - "train/contrastive_loss": 3.677873134613037, - "train/negative_loss": 1.4385451078414917, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.239327907562256, - "train/total_loss": 0.9560443758964539 + "train/classification_loss": 0.21196205914020538, + "train/contrastive_loss": 3.5819244384765625, + "train/negative_loss": 2.582134962081909, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.9997893571853638, + "train/total_loss": 0.9283469319343567 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.20636071264743805, - "train/contrastive_loss": 2.4596316814422607, - "train/negative_loss": 2.000361680984497, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.45927003026008606, - "train/total_loss": 0.6982870697975159 + "train/classification_loss": 0.17283444106578827, + "train/contrastive_loss": 1.1642478704452515, + "train/negative_loss": 1.1639028787612915, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.00034495373256504536, + "train/total_loss": 0.4056840240955353 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.15758508443832397, - "train/contrastive_loss": 1.5578044652938843, - "train/negative_loss": 1.1285693645477295, + "train/classification_loss": 0.19082209467887878, + "train/contrastive_loss": 1.245375633239746, + "train/negative_loss": 0.5975884199142456, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 0.4292351007461548, - "train/total_loss": 0.4691459834575653 + "train/positive_loss": 0.6477872729301453, + "train/total_loss": 0.43989723920822144 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.1809007078409195, - "train/contrastive_loss": 3.3025527000427246, - "train/negative_loss": 2.2386293411254883, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.0639233589172363, - "train/total_loss": 0.8414112329483032 + "train/classification_loss": 0.18263691663742065, + "train/contrastive_loss": 4.5198655128479, + "train/negative_loss": 2.8929924964904785, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.6268730163574219, + "train/total_loss": 1.0866100788116455 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.08700770139694214, - "train/contrastive_loss": 0.11996760219335556, - "train/negative_loss": 0.10864966362714767, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.011317937634885311, - "train/total_loss": 0.11100122332572937 + "train/classification_loss": 0.24095070362091064, + "train/contrastive_loss": 2.609123468399048, + "train/negative_loss": 2.0414252281188965, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.5676982402801514, + "train/total_loss": 0.7627754211425781 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.21315573155879974, - "train/contrastive_loss": 1.0115289688110352, - "train/negative_loss": 0.8475302457809448, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.1639987826347351, - "train/total_loss": 0.41546154022216797 + "train/classification_loss": 0.22925518453121185, + "train/contrastive_loss": 2.8529460430145264, + "train/negative_loss": 1.9892568588256836, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.8636891841888428, + "train/total_loss": 0.7998444437980652 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.12810149788856506, - "train/contrastive_loss": 1.5930066108703613, - "train/negative_loss": 0.6938771605491638, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.8991295099258423, - "train/total_loss": 0.44670283794403076 + "train/classification_loss": 0.26022523641586304, + "train/contrastive_loss": 1.3975938558578491, + "train/negative_loss": 1.1295288801193237, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.268064945936203, + "train/total_loss": 0.5397440195083618 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.22506822645664215, - "train/contrastive_loss": 4.663357734680176, - "train/negative_loss": 2.081484317779541, - "train/num_negatives": 46, + "train/classification_loss": 0.2487563043832779, + "train/contrastive_loss": 2.696424722671509, + "train/negative_loss": 0.7985318303108215, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 2.581873655319214, - "train/total_loss": 1.1577397584915161 + "train/positive_loss": 1.8978928327560425, + "train/total_loss": 0.7880412340164185 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.23891623318195343, - "train/contrastive_loss": 2.695413112640381, - "train/negative_loss": 1.5147284269332886, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.1806848049163818, - "train/total_loss": 0.7779988646507263 + "train/classification_loss": 0.2222227305173874, + "train/contrastive_loss": 1.2469182014465332, + "train/negative_loss": 1.0540047883987427, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.19291339814662933, + "train/total_loss": 0.47160637378692627 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.20941907167434692, - "train/contrastive_loss": 4.0196533203125, - "train/negative_loss": 2.0027313232421875, + "train/classification_loss": 0.22079302370548248, + "train/contrastive_loss": 1.892359972000122, + "train/negative_loss": 1.8923375606536865, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 2.0169219970703125, - "train/total_loss": 1.0133497714996338 + "train/num_positives": 8, + "train/positive_loss": 2.244148163299542e-05, + "train/total_loss": 0.5992650389671326 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.20294345915317535, - "train/contrastive_loss": 1.4998143911361694, - "train/negative_loss": 1.4039705991744995, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.09584376215934753, - "train/total_loss": 0.502906322479248 + "train/classification_loss": 0.1421300172805786, + "train/contrastive_loss": 4.875722885131836, + "train/negative_loss": 3.885855197906494, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.9898679256439209, + "train/total_loss": 1.117274522781372 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.19612102210521698, - "train/contrastive_loss": 1.6261684894561768, - "train/negative_loss": 0.8813042640686035, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.7448642253875732, - "train/total_loss": 0.5213547348976135 + "train/classification_loss": 0.20374208688735962, + "train/contrastive_loss": 3.1703922748565674, + "train/negative_loss": 2.5209426879882812, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.6494496464729309, + "train/total_loss": 0.8378205299377441 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.18507997691631317, - "train/contrastive_loss": 5.114315986633301, - "train/negative_loss": 4.594264030456543, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.520051896572113, - "train/total_loss": 1.2079432010650635 + "train/classification_loss": 0.21422965824604034, + "train/contrastive_loss": 2.7759714126586914, + "train/negative_loss": 1.6269564628601074, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.1490148305892944, + "train/total_loss": 0.7694239616394043 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.21782156825065613, - "train/contrastive_loss": 3.668837070465088, - "train/negative_loss": 1.7875946760177612, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.8812423944473267, - "train/total_loss": 0.9515889883041382 + "train/classification_loss": 0.1679249405860901, + "train/contrastive_loss": 2.7760956287384033, + "train/negative_loss": 2.7760908603668213, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 4.768382950715022e-06, + "train/total_loss": 0.7231440544128418 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.08931977301836014, - "train/contrastive_loss": 1.2202603816986084, - "train/negative_loss": 0.5320523977279663, - "train/num_negatives": 42, + "train/classification_loss": 0.22332635521888733, + "train/contrastive_loss": 2.060458183288574, + "train/negative_loss": 0.8563706278800964, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.6882080435752869, - "train/total_loss": 0.3333718478679657 + "train/positive_loss": 1.2040876150131226, + "train/total_loss": 0.6354179978370667 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.1523749977350235, - "train/contrastive_loss": 1.5381741523742676, - "train/negative_loss": 1.1566801071166992, - "train/num_negatives": 48, + "train/classification_loss": 0.20338626205921173, + "train/contrastive_loss": 0.4669026732444763, + "train/negative_loss": 0.35422590374946594, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.38149404525756836, - "train/total_loss": 0.4600098133087158 + "train/positive_loss": 0.11267676204442978, + "train/total_loss": 0.2967667877674103 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.18387828767299652, - "train/contrastive_loss": 2.548283338546753, - "train/negative_loss": 1.0425994396209717, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.5056838989257812, - "train/total_loss": 0.6935349702835083 + "train/classification_loss": 0.18401606380939484, + "train/contrastive_loss": 2.2188756465911865, + "train/negative_loss": 1.7771085500717163, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.44176703691482544, + "train/total_loss": 0.6277912259101868 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.10120268166065216, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.10120268166065216 - }, - { - "epoch": 12.382775119617225, - "eval_exact_match_accuracy": 0.3253588516746411, - "eval_hamming_loss": 0.06191950464396285, - "eval_loss": 0.6371140480041504, - "eval_macro_f1": 0.19376619379793045, - "eval_macro_precision": 0.2886134381699154, - "eval_macro_recall": 0.16283620518914635, - "eval_micro_f1": 0.4883720930232558, - "eval_micro_precision": 0.7835820895522388, - "eval_micro_recall": 0.3547297297297297, - "eval_runtime": 1.0636, - "eval_samples_per_second": 196.5, - "eval_steps_per_second": 25.385, + "train/classification_loss": 0.23315033316612244, + "train/contrastive_loss": 2.0753672122955322, + "train/negative_loss": 1.682298183441162, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.39306896924972534, + "train/total_loss": 0.6482237577438354 + }, + { + "epoch": 9.85171102661597, + "step": 1300, + "train/classification_loss": 0.1376163810491562, + "train/contrastive_loss": 1.523234486579895, + "train/negative_loss": 1.4811549186706543, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.042079534381628036, + "train/total_loss": 0.44226330518722534 + }, + { + "epoch": 9.85171102661597, + "eval_exact_match_accuracy": 0.2786259541984733, + "eval_hamming_loss": 0.06443646160754378, + "eval_loss": 0.6513397097587585, + "eval_macro_f1": 0.18118974261743379, + "eval_macro_precision": 0.29845390353872087, + "eval_macro_recall": 0.15053049875792604, + "eval_micro_f1": 0.49023090586145646, + "eval_micro_precision": 0.8070175438596491, + "eval_micro_recall": 0.3520408163265306, + "eval_runtime": 1.3057, + "eval_samples_per_second": 200.663, + "eval_steps_per_second": 25.274, "step": 1300 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.15810663998126984, - "train/contrastive_loss": 1.3554954528808594, - "train/negative_loss": 1.3048949241638184, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.050600528717041016, - "train/total_loss": 0.4292057156562805 + "train/classification_loss": 0.1186390221118927, + "train/contrastive_loss": 0.6909368634223938, + "train/negative_loss": 0.6895835995674133, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.0013532646698877215, + "train/total_loss": 0.25682640075683594 }, { - "epoch": 12.382775119617225, + "epoch": 9.85171102661597, "step": 1300, - "train/classification_loss": 0.1533185988664627, - "train/contrastive_loss": 0.1955500692129135, - "train/negative_loss": 0.19327017664909363, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 0.002279887208715081, - "train/total_loss": 0.1924286186695099 + "train/classification_loss": 0.13712094724178314, + "train/contrastive_loss": 0.4549577832221985, + "train/negative_loss": 0.4548714756965637, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 8.629389776615426e-05, + "train/total_loss": 0.22811250388622284 }, { - "epoch": 12.861244019138756, - "grad_norm": 10.22400951385498, - "learning_rate": 1.952404580152672e-05, - "loss": 0.549, + "epoch": 10.228136882129277, + "grad_norm": 14.350746154785156, + "learning_rate": 1.962154779969651e-05, + "loss": 0.6115, "step": 1350 }, { - "epoch": 12.861244019138756, + "epoch": 10.228136882129277, "step": 1350, - "train/classification_loss": 0.17971304059028625, - "train/contrastive_loss": 0.3504979908466339, - "train/negative_loss": 0.29014289379119873, - "train/num_negatives": 22, - "train/num_positives": 30, - "train/positive_loss": 0.06035509333014488, - "train/total_loss": 0.24981263279914856 + "train/classification_loss": 0.16284169256687164, + "train/contrastive_loss": 0.9056494235992432, + "train/negative_loss": 0.8341563940048218, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.07149302214384079, + "train/total_loss": 0.3439715802669525 }, { - "epoch": 12.861244019138756, + "epoch": 10.228136882129277, "step": 1350, - "train/classification_loss": 0.17026890814304352, - "train/contrastive_loss": 0.5720983743667603, - "train/negative_loss": 0.48068395256996155, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.0914144217967987, - "train/total_loss": 0.2846885919570923 + "train/classification_loss": 0.22291845083236694, + "train/contrastive_loss": 1.6815646886825562, + "train/negative_loss": 1.534811019897461, + "train/num_negatives": 46, + "train/num_positives": 4, + "train/positive_loss": 0.1467536836862564, + "train/total_loss": 0.5592314004898071 }, { - "epoch": 13.334928229665072, - "grad_norm": 11.235635757446289, - "learning_rate": 1.950496183206107e-05, - "loss": 0.5713, + "epoch": 10.608365019011407, + "grad_norm": 8.793706893920898, + "learning_rate": 1.9606373292867983e-05, + "loss": 0.6537, "step": 1400 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.15401560068130493, - "train/contrastive_loss": 2.398616313934326, - "train/negative_loss": 1.8603107929229736, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.5383055210113525, - "train/total_loss": 0.6337388753890991 + "train/classification_loss": 0.10600224882364273, + "train/contrastive_loss": 0.5431270003318787, + "train/negative_loss": 0.37425336241722107, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.1688736230134964, + "train/total_loss": 0.21462765336036682 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.15573574602603912, - "train/contrastive_loss": 1.5185545682907104, - "train/negative_loss": 0.37142547965049744, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.1471290588378906, - "train/total_loss": 0.45944666862487793 + "train/classification_loss": 0.21522219479084015, + "train/contrastive_loss": 1.6017496585845947, + "train/negative_loss": 1.0313148498535156, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.5704348087310791, + "train/total_loss": 0.5355721116065979 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.12959395349025726, - "train/contrastive_loss": 0.8592695593833923, - "train/negative_loss": 0.7700201869010925, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.0892493948340416, - "train/total_loss": 0.30144786834716797 + "train/classification_loss": 0.12544231116771698, + "train/contrastive_loss": 1.5225756168365479, + "train/negative_loss": 1.3148655891418457, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.20770996809005737, + "train/total_loss": 0.42995744943618774 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.17045821249485016, - "train/contrastive_loss": 3.2570104598999023, - "train/negative_loss": 1.3780624866485596, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.8789478540420532, - "train/total_loss": 0.8218603134155273 + "train/classification_loss": 0.20776265859603882, + "train/contrastive_loss": 1.3803324699401855, + "train/negative_loss": 1.2978060245513916, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.08252646028995514, + "train/total_loss": 0.48382917046546936 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.12776541709899902, - "train/contrastive_loss": 2.890228271484375, - "train/negative_loss": 1.2270727157592773, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.6631556749343872, - "train/total_loss": 0.705811083316803 + "train/classification_loss": 0.09268221259117126, + "train/contrastive_loss": 0.6137967109680176, + "train/negative_loss": 0.5668766498565674, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.046920038759708405, + "train/total_loss": 0.21544155478477478 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.1574447900056839, - "train/contrastive_loss": 3.5375261306762695, - "train/negative_loss": 1.869105339050293, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.668420672416687, - "train/total_loss": 0.8649500012397766 + "train/classification_loss": 0.22156493365764618, + "train/contrastive_loss": 1.7914345264434814, + "train/negative_loss": 1.6250545978546143, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.16637998819351196, + "train/total_loss": 0.5798518657684326 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.17266687750816345, - "train/contrastive_loss": 2.4955010414123535, - "train/negative_loss": 1.8012871742248535, + "train/classification_loss": 0.22928763926029205, + "train/contrastive_loss": 2.4524893760681152, + "train/negative_loss": 2.1494996547698975, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.6942137479782104, - "train/total_loss": 0.6717671155929565 + "train/num_positives": 8, + "train/positive_loss": 0.3029897212982178, + "train/total_loss": 0.7197855114936829 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.23075547814369202, - "train/contrastive_loss": 1.0541185140609741, - "train/negative_loss": 1.0540552139282227, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 6.334296631393954e-05, - "train/total_loss": 0.4415791630744934 + "train/classification_loss": 0.20166607201099396, + "train/contrastive_loss": 1.1722712516784668, + "train/negative_loss": 0.9921107888221741, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.18016044795513153, + "train/total_loss": 0.43612033128738403 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.14953632652759552, - "train/contrastive_loss": 0.9801151752471924, - "train/negative_loss": 0.8475226163864136, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.1325925886631012, - "train/total_loss": 0.34555935859680176 + "train/classification_loss": 0.1202576756477356, + "train/contrastive_loss": 0.3321623206138611, + "train/negative_loss": 0.31576234102249146, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.016399983316659927, + "train/total_loss": 0.18669015169143677 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.21339090168476105, - "train/contrastive_loss": 4.446648597717285, - "train/negative_loss": 1.842110514640808, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.6045379638671875, - "train/total_loss": 1.1027206182479858 + "train/classification_loss": 0.1943310648202896, + "train/contrastive_loss": 2.8339457511901855, + "train/negative_loss": 2.5553059577941895, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.27863985300064087, + "train/total_loss": 0.7611202001571655 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.2212502807378769, - "train/contrastive_loss": 2.3149685859680176, - "train/negative_loss": 1.9169089794158936, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.39805957674980164, - "train/total_loss": 0.6842439770698547 + "train/classification_loss": 0.17438973486423492, + "train/contrastive_loss": 4.325918197631836, + "train/negative_loss": 4.276722431182861, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.04919584468007088, + "train/total_loss": 1.0395734310150146 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.15477657318115234, - "train/contrastive_loss": 1.031294584274292, - "train/negative_loss": 0.6318318843841553, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.39946264028549194, - "train/total_loss": 0.3610354959964752 + "train/classification_loss": 0.18994003534317017, + "train/contrastive_loss": 3.0804343223571777, + "train/negative_loss": 1.8689204454421997, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 1.2115137577056885, + "train/total_loss": 0.8060269355773926 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.1913275271654129, - "train/contrastive_loss": 4.400562286376953, - "train/negative_loss": 3.317342758178711, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.0832197666168213, - "train/total_loss": 1.0714399814605713 + "train/classification_loss": 0.21049679815769196, + "train/contrastive_loss": 2.3709146976470947, + "train/negative_loss": 1.336519479751587, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.0343952178955078, + "train/total_loss": 0.6846797466278076 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.08461175113916397, - "train/contrastive_loss": 0.657322108745575, - "train/negative_loss": 0.44992002844810486, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.2074020653963089, - "train/total_loss": 0.21607616543769836 + "train/classification_loss": 0.19970807433128357, + "train/contrastive_loss": 2.2732653617858887, + "train/negative_loss": 1.6612293720245361, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 0.6120361089706421, + "train/total_loss": 0.6543611288070679 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.19520671665668488, - "train/contrastive_loss": 0.7475321292877197, - "train/negative_loss": 0.6566838622093201, + "train/classification_loss": 0.2038533091545105, + "train/contrastive_loss": 4.3009748458862305, + "train/negative_loss": 3.572389841079712, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.09084828197956085, - "train/total_loss": 0.34471315145492554 + "train/positive_loss": 0.7285849452018738, + "train/total_loss": 1.0640482902526855 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.12829196453094482, - "train/contrastive_loss": 2.2412290573120117, - "train/negative_loss": 1.4020726680755615, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.8391563296318054, - "train/total_loss": 0.5765377879142761 + "train/classification_loss": 0.1559639871120453, + "train/contrastive_loss": 0.9240552186965942, + "train/negative_loss": 0.9230062365531921, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.001048991922289133, + "train/total_loss": 0.3407750129699707 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.22272948920726776, - "train/contrastive_loss": 3.613645553588867, - "train/negative_loss": 2.2423923015594482, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.3712533712387085, - "train/total_loss": 0.9454586505889893 + "train/classification_loss": 0.17557525634765625, + "train/contrastive_loss": 1.4314385652542114, + "train/negative_loss": 0.8077163696289062, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.6237221956253052, + "train/total_loss": 0.4618629813194275 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.24463903903961182, - "train/contrastive_loss": 2.9006166458129883, - "train/negative_loss": 1.7863892316818237, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.114227294921875, - "train/total_loss": 0.8247624039649963 + "train/classification_loss": 0.18281519412994385, + "train/contrastive_loss": 3.7076287269592285, + "train/negative_loss": 2.7153849601745605, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.992243766784668, + "train/total_loss": 0.9243409633636475 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.2055075615644455, - "train/contrastive_loss": 3.433931589126587, - "train/negative_loss": 2.20204758644104, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.2318840026855469, - "train/total_loss": 0.8922939300537109 + "train/classification_loss": 0.2404552400112152, + "train/contrastive_loss": 3.473752737045288, + "train/negative_loss": 2.2733471393585205, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.2004055976867676, + "train/total_loss": 0.9352058172225952 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.21895363926887512, - "train/contrastive_loss": 2.054499864578247, - "train/negative_loss": 1.5983244180679321, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.4561755359172821, - "train/total_loss": 0.6298536062240601 + "train/classification_loss": 0.22150243818759918, + "train/contrastive_loss": 2.210860013961792, + "train/negative_loss": 1.3921306133270264, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.8187294006347656, + "train/total_loss": 0.6636744737625122 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.18600723147392273, - "train/contrastive_loss": 1.1082426309585571, - "train/negative_loss": 1.086058259010315, - "train/num_negatives": 44, + "train/classification_loss": 0.235976442694664, + "train/contrastive_loss": 1.2563800811767578, + "train/negative_loss": 0.7188432812690735, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.5375368595123291, + "train/total_loss": 0.48725247383117676 + }, + { + "epoch": 10.608365019011407, + "step": 1400, + "train/classification_loss": 0.254368394613266, + "train/contrastive_loss": 2.386457920074463, + "train/negative_loss": 0.7480190992355347, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.022184399887919426, - "train/total_loss": 0.4076557755470276 + "train/positive_loss": 1.6384388208389282, + "train/total_loss": 0.731660008430481 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.1875167191028595, - "train/contrastive_loss": 4.873600006103516, - "train/negative_loss": 3.768026828765869, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.1055729389190674, - "train/total_loss": 1.1622366905212402 + "train/classification_loss": 0.2137077897787094, + "train/contrastive_loss": 1.5000706911087036, + "train/negative_loss": 1.1118580102920532, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.388212651014328, + "train/total_loss": 0.5137219429016113 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.20904788374900818, - "train/contrastive_loss": 4.045825004577637, - "train/negative_loss": 2.157090902328491, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.888734221458435, - "train/total_loss": 1.018212914466858 + "train/classification_loss": 0.22045668959617615, + "train/contrastive_loss": 1.0816925764083862, + "train/negative_loss": 1.0816847085952759, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 7.867854947107844e-06, + "train/total_loss": 0.4367952048778534 + }, + { + "epoch": 10.608365019011407, + "step": 1400, + "train/classification_loss": 0.15887607634067535, + "train/contrastive_loss": 6.121879577636719, + "train/negative_loss": 5.428781509399414, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.6930980682373047, + "train/total_loss": 1.3832520246505737 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.09990839660167694, - "train/contrastive_loss": 2.3212673664093018, - "train/negative_loss": 0.8012881875038147, + "train/classification_loss": 0.21047914028167725, + "train/contrastive_loss": 3.857605457305908, + "train/negative_loss": 2.8552820682525635, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 1.5199792385101318, - "train/total_loss": 0.5641618967056274 + "train/positive_loss": 1.0023232698440552, + "train/total_loss": 0.9820002317428589 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.14654961228370667, - "train/contrastive_loss": 1.959504246711731, - "train/negative_loss": 1.0908310413360596, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.8686732053756714, - "train/total_loss": 0.5384504795074463 + "train/classification_loss": 0.20117418467998505, + "train/contrastive_loss": 3.1132454872131348, + "train/negative_loss": 1.70205557346344, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.4111900329589844, + "train/total_loss": 0.8238233327865601 + }, + { + "epoch": 10.608365019011407, + "step": 1400, + "train/classification_loss": 0.16946175694465637, + "train/contrastive_loss": 2.992823600769043, + "train/negative_loss": 2.9928174018859863, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 6.07969241173123e-06, + "train/total_loss": 0.7680264711380005 + }, + { + "epoch": 10.608365019011407, + "step": 1400, + "train/classification_loss": 0.2259078174829483, + "train/contrastive_loss": 1.663398265838623, + "train/negative_loss": 0.45091524720191956, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 1.2124830484390259, + "train/total_loss": 0.5585874915122986 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.18375958502292633, - "train/contrastive_loss": 1.8042848110198975, - "train/negative_loss": 1.0479329824447632, + "train/classification_loss": 0.19829413294792175, + "train/contrastive_loss": 0.9380596876144409, + "train/negative_loss": 0.9150434136390686, "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.7563517689704895, - "train/total_loss": 0.5446165800094604 + "train/num_positives": 8, + "train/positive_loss": 0.023016277700662613, + "train/total_loss": 0.38590607047080994 + }, + { + "epoch": 10.608365019011407, + "step": 1400, + "train/classification_loss": 0.19864819943904877, + "train/contrastive_loss": 2.107515335083008, + "train/negative_loss": 1.4555439949035645, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.6519712805747986, + "train/total_loss": 0.6201512813568115 + }, + { + "epoch": 10.608365019011407, + "step": 1400, + "train/classification_loss": 0.2148272544145584, + "train/contrastive_loss": 1.1791287660598755, + "train/negative_loss": 0.9530178308486938, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.22611096501350403, + "train/total_loss": 0.4506530165672302 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.07102499902248383, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.07102499902248383 - }, - { - "epoch": 13.334928229665072, - "eval_exact_match_accuracy": 0.3397129186602871, - "eval_hamming_loss": 0.06051224317478188, - "eval_loss": 0.6560577154159546, - "eval_macro_f1": 0.21424010147125802, - "eval_macro_precision": 0.2854020165941139, - "eval_macro_recall": 0.18528478057889822, - "eval_micro_f1": 0.5232815964523282, - "eval_micro_precision": 0.7612903225806451, - "eval_micro_recall": 0.39864864864864863, - "eval_runtime": 1.0601, - "eval_samples_per_second": 197.157, - "eval_steps_per_second": 25.47, + "train/classification_loss": 0.12215259671211243, + "train/contrastive_loss": 1.1831159591674805, + "train/negative_loss": 1.1779812574386597, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.005134749226272106, + "train/total_loss": 0.358775794506073 + }, + { + "epoch": 10.608365019011407, + "eval_exact_match_accuracy": 0.2633587786259542, + "eval_hamming_loss": 0.06533453075886843, + "eval_loss": 0.6274337768554688, + "eval_macro_f1": 0.19171207566794313, + "eval_macro_precision": 0.3078056840713814, + "eval_macro_recall": 0.15468769671175395, + "eval_micro_f1": 0.4756756756756757, + "eval_micro_precision": 0.8098159509202454, + "eval_micro_recall": 0.336734693877551, + "eval_runtime": 1.3106, + "eval_samples_per_second": 199.91, + "eval_steps_per_second": 25.179, "step": 1400 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.16362348198890686, - "train/contrastive_loss": 1.6566376686096191, - "train/negative_loss": 0.6526287794113159, + "train/classification_loss": 0.16673992574214935, + "train/contrastive_loss": 1.1085630655288696, + "train/negative_loss": 0.8438073992729187, "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.0040088891983032, - "train/total_loss": 0.4949510097503662 + "train/num_positives": 18, + "train/positive_loss": 0.26475563645362854, + "train/total_loss": 0.38845252990722656 }, { - "epoch": 13.334928229665072, + "epoch": 10.608365019011407, "step": 1400, - "train/classification_loss": 0.1398027092218399, - "train/contrastive_loss": 0.8765386343002319, - "train/negative_loss": 0.861359179019928, - "train/num_negatives": 42, - "train/num_positives": 6, - "train/positive_loss": 0.01517943199723959, - "train/total_loss": 0.315110445022583 + "train/classification_loss": 0.1913316398859024, + "train/contrastive_loss": 1.1688536405563354, + "train/negative_loss": 1.1393665075302124, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 0.029487131163477898, + "train/total_loss": 0.4251023530960083 }, { - "epoch": 13.813397129186603, - "grad_norm": 9.308649063110352, - "learning_rate": 1.9485877862595422e-05, - "loss": 0.5319, + "epoch": 10.988593155893536, + "grad_norm": 9.907177925109863, + "learning_rate": 1.9591198786039458e-05, + "loss": 0.6066, "step": 1450 }, { - "epoch": 13.813397129186603, + "epoch": 10.988593155893536, "step": 1450, - "train/classification_loss": 0.16096529364585876, - "train/contrastive_loss": 1.160418152809143, - "train/negative_loss": 1.1592220067977905, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.001196112367324531, - "train/total_loss": 0.3930489420890808 + "train/classification_loss": 0.17901188135147095, + "train/contrastive_loss": 1.6280736923217773, + "train/negative_loss": 1.035062313079834, + "train/num_negatives": 34, + "train/num_positives": 18, + "train/positive_loss": 0.5930113196372986, + "train/total_loss": 0.5046266317367554 }, { - "epoch": 13.813397129186603, + "epoch": 10.988593155893536, "step": 1450, - "train/classification_loss": 0.1259065568447113, - "train/contrastive_loss": 0.2798709273338318, - "train/negative_loss": 0.25372952222824097, - "train/num_negatives": 40, - "train/num_positives": 14, - "train/positive_loss": 0.026141412556171417, - "train/total_loss": 0.18188074231147766 + "train/classification_loss": 0.08104603737592697, + "train/contrastive_loss": 0.26719337701797485, + "train/negative_loss": 0.24767369031906128, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.01951969973742962, + "train/total_loss": 0.13448470830917358 }, { - "epoch": 14.287081339712918, - "grad_norm": 8.477154731750488, - "learning_rate": 1.9466793893129773e-05, - "loss": 0.5107, + "epoch": 11.365019011406844, + "grad_norm": 13.740861892700195, + "learning_rate": 1.957602427921093e-05, + "loss": 0.5838, "step": 1500 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, + "step": 1500, + "train/classification_loss": 0.09861591458320618, + "train/contrastive_loss": 0.5697097182273865, + "train/negative_loss": 0.4217391312122345, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.14797058701515198, + "train/total_loss": 0.212557852268219 + }, + { + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.15041065216064453, - "train/contrastive_loss": 2.0114243030548096, - "train/negative_loss": 1.5327328443527222, + "train/classification_loss": 0.2256750464439392, + "train/contrastive_loss": 2.1529288291931152, + "train/negative_loss": 1.620274543762207, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.4786915183067322, - "train/total_loss": 0.5526955127716064 + "train/num_positives": 6, + "train/positive_loss": 0.5326541662216187, + "train/total_loss": 0.6562608480453491 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.1484643965959549, - "train/contrastive_loss": 0.9840456247329712, - "train/negative_loss": 0.6074565052986145, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.3765891194343567, - "train/total_loss": 0.34527352452278137 + "train/classification_loss": 0.10564269870519638, + "train/contrastive_loss": 2.4086146354675293, + "train/negative_loss": 1.7858558893203735, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.6227588057518005, + "train/total_loss": 0.5873656272888184 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.12409380078315735, - "train/contrastive_loss": 0.820763111114502, - "train/negative_loss": 0.5240992903709412, - "train/num_negatives": 44, + "train/classification_loss": 0.2446737289428711, + "train/contrastive_loss": 1.7636499404907227, + "train/negative_loss": 1.0223649740219116, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.7412850260734558, + "train/total_loss": 0.5974037647247314 + }, + { + "epoch": 11.365019011406844, + "step": 1500, + "train/classification_loss": 0.10936953127384186, + "train/contrastive_loss": 1.6717777252197266, + "train/negative_loss": 1.092523217201233, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.5792545080184937, + "train/total_loss": 0.4437251091003418 + }, + { + "epoch": 11.365019011406844, + "step": 1500, + "train/classification_loss": 0.21978646516799927, + "train/contrastive_loss": 1.7573573589324951, + "train/negative_loss": 1.38262939453125, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.2966638207435608, - "train/total_loss": 0.28824642300605774 + "train/positive_loss": 0.3747280240058899, + "train/total_loss": 0.5712579488754272 + }, + { + "epoch": 11.365019011406844, + "step": 1500, + "train/classification_loss": 0.21850526332855225, + "train/contrastive_loss": 2.573582172393799, + "train/negative_loss": 1.5654726028442383, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.0081095695495605, + "train/total_loss": 0.733221709728241 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.17011766135692596, - "train/contrastive_loss": 3.168403148651123, - "train/negative_loss": 1.8615789413452148, + "train/classification_loss": 0.18763099610805511, + "train/contrastive_loss": 1.3034762144088745, + "train/negative_loss": 1.0619337558746338, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.3068242073059082, - "train/total_loss": 0.8037983179092407 + "train/positive_loss": 0.24154244363307953, + "train/total_loss": 0.4483262300491333 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.12072156369686127, - "train/contrastive_loss": 2.427509307861328, - "train/negative_loss": 1.1673847436904907, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.2601245641708374, - "train/total_loss": 0.6062234044075012 + "train/classification_loss": 0.11989963799715042, + "train/contrastive_loss": 0.9044489860534668, + "train/negative_loss": 0.41513317823410034, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.48931580781936646, + "train/total_loss": 0.3007894456386566 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.14975042641162872, - "train/contrastive_loss": 3.8349971771240234, - "train/negative_loss": 2.274657726287842, - "train/num_negatives": 32, + "train/classification_loss": 0.1795686036348343, + "train/contrastive_loss": 2.5499327182769775, + "train/negative_loss": 2.0652108192443848, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.484721839427948, + "train/total_loss": 0.6895551681518555 + }, + { + "epoch": 11.365019011406844, + "step": 1500, + "train/classification_loss": 0.17391210794448853, + "train/contrastive_loss": 3.9912476539611816, + "train/negative_loss": 3.827375650405884, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.5603395700454712, - "train/total_loss": 0.9167498350143433 + "train/positive_loss": 0.16387192904949188, + "train/total_loss": 0.9721616506576538 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.1680203378200531, - "train/contrastive_loss": 1.4227616786956787, - "train/negative_loss": 0.9078612327575684, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.5149003863334656, - "train/total_loss": 0.45257267355918884 + "train/classification_loss": 0.17105048894882202, + "train/contrastive_loss": 2.4079978466033936, + "train/negative_loss": 1.5418744087219238, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.8661234378814697, + "train/total_loss": 0.6526500582695007 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.23357421159744263, - "train/contrastive_loss": 1.2457858324050903, - "train/negative_loss": 1.2442820072174072, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.0015037957346066833, - "train/total_loss": 0.4827314019203186 + "train/classification_loss": 0.1999681144952774, + "train/contrastive_loss": 1.9556995630264282, + "train/negative_loss": 1.0867838859558105, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.8689156770706177, + "train/total_loss": 0.5911080241203308 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.15182389318943024, - "train/contrastive_loss": 0.9530190825462341, - "train/negative_loss": 0.8639364242553711, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.08908264338970184, - "train/total_loss": 0.34242773056030273 + "train/classification_loss": 0.18541331589221954, + "train/contrastive_loss": 1.9243738651275635, + "train/negative_loss": 1.0939668416976929, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 0.8304070234298706, + "train/total_loss": 0.5702881217002869 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.23716884851455688, - "train/contrastive_loss": 4.45583963394165, - "train/negative_loss": 1.9793386459350586, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.476500988006592, - "train/total_loss": 1.128336787223816 + "train/classification_loss": 0.21109513938426971, + "train/contrastive_loss": 3.1149120330810547, + "train/negative_loss": 1.9363877773284912, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 1.1785242557525635, + "train/total_loss": 0.8340775966644287 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.21465329825878143, - "train/contrastive_loss": 2.666534185409546, - "train/negative_loss": 1.8841923475265503, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.7823418974876404, - "train/total_loss": 0.7479601502418518 + "train/classification_loss": 0.15814556181430817, + "train/contrastive_loss": 0.7412458658218384, + "train/negative_loss": 0.7392727732658386, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0019731041975319386, + "train/total_loss": 0.3063947558403015 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.15987999737262726, - "train/contrastive_loss": 3.0645322799682617, - "train/negative_loss": 1.147494912147522, + "train/classification_loss": 0.17618605494499207, + "train/contrastive_loss": 1.6600050926208496, + "train/negative_loss": 0.6872461438179016, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 1.9170374870300293, - "train/total_loss": 0.7727864384651184 + "train/positive_loss": 0.972758948802948, + "train/total_loss": 0.5081870555877686 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.193498894572258, - "train/contrastive_loss": 4.333597183227539, - "train/negative_loss": 3.243507146835327, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.0900901556015015, - "train/total_loss": 1.060218334197998 + "train/classification_loss": 0.18346929550170898, + "train/contrastive_loss": 4.167718410491943, + "train/negative_loss": 2.4209201335906982, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.7467982769012451, + "train/total_loss": 1.0170130729675293 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.08087383955717087, - "train/contrastive_loss": 0.2484678477048874, - "train/negative_loss": 0.1985999047756195, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.04986794665455818, - "train/total_loss": 0.13056740164756775 + "train/classification_loss": 0.24648116528987885, + "train/contrastive_loss": 4.084001064300537, + "train/negative_loss": 2.50249981880188, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.5815011262893677, + "train/total_loss": 1.0632814168930054 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.19818207621574402, - "train/contrastive_loss": 1.1603983640670776, - "train/negative_loss": 0.8341845273971558, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.3262138366699219, - "train/total_loss": 0.4302617311477661 + "train/classification_loss": 0.22530514001846313, + "train/contrastive_loss": 2.4877090454101562, + "train/negative_loss": 1.4027713537216187, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.0849378108978271, + "train/total_loss": 0.7228469848632812 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.11618407815694809, - "train/contrastive_loss": 1.6980681419372559, - "train/negative_loss": 0.9510011076927185, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.7470670938491821, - "train/total_loss": 0.4557977318763733 + "train/classification_loss": 0.22520890831947327, + "train/contrastive_loss": 1.4399610757827759, + "train/negative_loss": 0.7399224638938904, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.7000386118888855, + "train/total_loss": 0.513201117515564 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.24536380171775818, - "train/contrastive_loss": 3.8345460891723633, - "train/negative_loss": 2.325089693069458, - "train/num_negatives": 46, + "train/classification_loss": 0.2470000833272934, + "train/contrastive_loss": 2.9370217323303223, + "train/negative_loss": 0.7357836961746216, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.5094562768936157, - "train/total_loss": 1.0122730731964111 + "train/positive_loss": 2.201237916946411, + "train/total_loss": 0.834404468536377 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.22948160767555237, - "train/contrastive_loss": 2.3956096172332764, - "train/negative_loss": 1.838942527770996, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.5566670894622803, - "train/total_loss": 0.7086035013198853 + "train/classification_loss": 0.21994034945964813, + "train/contrastive_loss": 1.4201785326004028, + "train/negative_loss": 1.038430094718933, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.38174840807914734, + "train/total_loss": 0.503976047039032 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.20356106758117676, - "train/contrastive_loss": 3.6371278762817383, - "train/negative_loss": 2.0163304805755615, + "train/classification_loss": 0.20702700316905975, + "train/contrastive_loss": 0.3645065724849701, + "train/negative_loss": 0.36450013518333435, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.6207975149154663, - "train/total_loss": 0.9309866428375244 + "train/num_positives": 8, + "train/positive_loss": 6.437332103814697e-06, + "train/total_loss": 0.2799283266067505 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.20002706348896027, - "train/contrastive_loss": 1.8054094314575195, - "train/negative_loss": 1.4137924909591675, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.39161691069602966, - "train/total_loss": 0.5611089468002319 + "train/classification_loss": 0.1449035406112671, + "train/contrastive_loss": 4.5464186668396, + "train/negative_loss": 3.724026679992676, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.8223918676376343, + "train/total_loss": 1.054187297821045 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.1909698098897934, - "train/contrastive_loss": 1.5531387329101562, - "train/negative_loss": 1.0342121124267578, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.5189266204833984, - "train/total_loss": 0.5015975832939148 + "train/classification_loss": 0.21166524291038513, + "train/contrastive_loss": 3.6249351501464844, + "train/negative_loss": 2.532609701156616, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 1.0923255681991577, + "train/total_loss": 0.9366523027420044 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.19898301362991333, - "train/contrastive_loss": 5.836091041564941, - "train/negative_loss": 5.118866920471191, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.7172243595123291, - "train/total_loss": 1.3662011623382568 + "train/classification_loss": 0.20557758212089539, + "train/contrastive_loss": 3.573507785797119, + "train/negative_loss": 1.1342341899871826, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 2.4392735958099365, + "train/total_loss": 0.9202791452407837 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.20964300632476807, - "train/contrastive_loss": 3.815145969390869, - "train/negative_loss": 1.9445394277572632, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.870606541633606, - "train/total_loss": 0.9726722240447998 + "train/classification_loss": 0.1620628982782364, + "train/contrastive_loss": 2.4350478649139404, + "train/negative_loss": 2.4347612857818604, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.0002865605929400772, + "train/total_loss": 0.6490724682807922 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.08370140194892883, - "train/contrastive_loss": 1.808200716972351, - "train/negative_loss": 0.9184941053390503, - "train/num_negatives": 42, + "train/classification_loss": 0.22110071778297424, + "train/contrastive_loss": 2.5741734504699707, + "train/negative_loss": 0.7382262945175171, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.8897066116333008, - "train/total_loss": 0.445341557264328 + "train/positive_loss": 1.8359471559524536, + "train/total_loss": 0.7359354496002197 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.1512787789106369, - "train/contrastive_loss": 1.834839105606079, - "train/negative_loss": 1.1589303016662598, - "train/num_negatives": 48, + "train/classification_loss": 0.19608284533023834, + "train/contrastive_loss": 0.5419977307319641, + "train/negative_loss": 0.518947184085846, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.6759087443351746, - "train/total_loss": 0.518246591091156 + "train/positive_loss": 0.023050544783473015, + "train/total_loss": 0.3044824004173279 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.18929502367973328, - "train/contrastive_loss": 2.058727741241455, - "train/negative_loss": 0.97370445728302, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.0850234031677246, - "train/total_loss": 0.6010406017303467 + "train/classification_loss": 0.1826229691505432, + "train/contrastive_loss": 2.2586710453033447, + "train/negative_loss": 1.3235772848129272, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.9350938200950623, + "train/total_loss": 0.634357213973999 + }, + { + "epoch": 11.365019011406844, + "step": 1500, + "train/classification_loss": 0.2100907564163208, + "train/contrastive_loss": 1.3962537050247192, + "train/negative_loss": 1.3048955202102661, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.09135816991329193, + "train/total_loss": 0.48934149742126465 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.06077631935477257, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.06077631935477257 - }, - { - "epoch": 14.287081339712918, - "eval_exact_match_accuracy": 0.3492822966507177, - "eval_hamming_loss": 0.061356600056290456, - "eval_loss": 0.6561652421951294, - "eval_macro_f1": 0.21534015547797633, - "eval_macro_precision": 0.27719368290302554, - "eval_macro_recall": 0.1905191975780211, - "eval_micro_f1": 0.5301724137931034, - "eval_micro_precision": 0.7321428571428571, - "eval_micro_recall": 0.4155405405405405, - "eval_runtime": 1.0669, - "eval_samples_per_second": 195.893, - "eval_steps_per_second": 25.307, + "train/classification_loss": 0.12245427072048187, + "train/contrastive_loss": 1.3281551599502563, + "train/negative_loss": 1.262970209121704, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.06518500298261642, + "train/total_loss": 0.3880853056907654 + }, + { + "epoch": 11.365019011406844, + "eval_exact_match_accuracy": 0.2862595419847328, + "eval_hamming_loss": 0.06196677144140099, + "eval_loss": 0.6297818422317505, + "eval_macro_f1": 0.20938993987756407, + "eval_macro_precision": 0.29611470301003284, + "eval_macro_recall": 0.1695964455277758, + "eval_micro_f1": 0.5157894736842106, + "eval_micro_precision": 0.8258426966292135, + "eval_micro_recall": 0.375, + "eval_runtime": 1.3205, + "eval_samples_per_second": 198.413, + "eval_steps_per_second": 24.991, "step": 1500 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.12984482944011688, - "train/contrastive_loss": 0.9209384322166443, - "train/negative_loss": 0.9209294319152832, - "train/num_negatives": 52, - "train/num_positives": 4, - "train/positive_loss": 9.000341378850862e-06, - "train/total_loss": 0.31403252482414246 + "train/classification_loss": 0.11807133257389069, + "train/contrastive_loss": 0.9185880422592163, + "train/negative_loss": 0.49032899737358093, + "train/num_negatives": 36, + "train/num_positives": 18, + "train/positive_loss": 0.4282590448856354, + "train/total_loss": 0.30178892612457275 }, { - "epoch": 14.287081339712918, + "epoch": 11.365019011406844, "step": 1500, - "train/classification_loss": 0.08383780717849731, - "train/contrastive_loss": 0.2104165405035019, - "train/negative_loss": 0.20067375898361206, - "train/num_negatives": 38, + "train/classification_loss": 0.1774991750717163, + "train/contrastive_loss": 0.3154314160346985, + "train/negative_loss": 0.1840473860502243, + "train/num_negatives": 30, "train/num_positives": 18, - "train/positive_loss": 0.009742782451212406, - "train/total_loss": 0.1259211152791977 + "train/positive_loss": 0.131384015083313, + "train/total_loss": 0.24058546125888824 }, { - "epoch": 14.76555023923445, - "grad_norm": 3.672731399536133, - "learning_rate": 1.9447709923664124e-05, - "loss": 0.4716, + "epoch": 11.745247148288973, + "grad_norm": 9.16875171661377, + "learning_rate": 1.95608497723824e-05, + "loss": 0.557, "step": 1550 }, { - "epoch": 14.76555023923445, + "epoch": 11.745247148288973, "step": 1550, - "train/classification_loss": 0.08121229708194733, - "train/contrastive_loss": 0.4330196678638458, - "train/negative_loss": 0.4329761266708374, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 4.355326746008359e-05, - "train/total_loss": 0.16781622171401978 + "train/classification_loss": 0.17551130056381226, + "train/contrastive_loss": 1.2519142627716064, + "train/negative_loss": 1.2440593242645264, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.007854917086660862, + "train/total_loss": 0.425894170999527 }, { - "epoch": 14.76555023923445, + "epoch": 11.745247148288973, "step": 1550, - "train/classification_loss": 0.07607679069042206, - "train/contrastive_loss": 0.0918704941868782, - "train/negative_loss": 0.09184063971042633, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 2.985690844070632e-05, - "train/total_loss": 0.09445089101791382 + "train/classification_loss": 0.12335247546434402, + "train/contrastive_loss": 0.5997835993766785, + "train/negative_loss": 0.5997729897499084, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 1.0609683158691041e-05, + "train/total_loss": 0.24330919981002808 }, { - "epoch": 15.239234449760765, - "grad_norm": 5.174161911010742, - "learning_rate": 1.9428625954198476e-05, - "loss": 0.4788, + "epoch": 12.12167300380228, + "grad_norm": 6.07202672958374, + "learning_rate": 1.9545675265553872e-05, + "loss": 0.5526, "step": 1600 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.15752071142196655, - "train/contrastive_loss": 3.1958930492401123, - "train/negative_loss": 2.215540885925293, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.9803522229194641, - "train/total_loss": 0.7966993451118469 + "train/classification_loss": 0.08894173055887222, + "train/contrastive_loss": 0.2236195057630539, + "train/negative_loss": 0.19990557432174683, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.023713933303952217, + "train/total_loss": 0.13366563618183136 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.1409478783607483, - "train/contrastive_loss": 0.9171524047851562, - "train/negative_loss": 0.30135682225227356, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.6157955527305603, - "train/total_loss": 0.3243783712387085 + "train/classification_loss": 0.21339821815490723, + "train/contrastive_loss": 1.3455613851547241, + "train/negative_loss": 1.3189923763275146, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.02656901627779007, + "train/total_loss": 0.482510507106781 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.13494743406772614, - "train/contrastive_loss": 1.3400497436523438, - "train/negative_loss": 0.8955064415931702, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.44454336166381836, - "train/total_loss": 0.40295737981796265 + "train/classification_loss": 0.11260194331407547, + "train/contrastive_loss": 1.6041719913482666, + "train/negative_loss": 1.4210484027862549, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.18312358856201172, + "train/total_loss": 0.4334363341331482 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.1492130607366562, - "train/contrastive_loss": 2.7980263233184814, - "train/negative_loss": 1.6946392059326172, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.1033871173858643, - "train/total_loss": 0.7088183760643005 + "train/classification_loss": 0.21479199826717377, + "train/contrastive_loss": 1.4357836246490479, + "train/negative_loss": 1.2801661491394043, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.15561750531196594, + "train/total_loss": 0.5019487142562866 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.11373357474803925, - "train/contrastive_loss": 2.873638153076172, - "train/negative_loss": 1.2761237621307373, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.5975145101547241, - "train/total_loss": 0.6884612441062927 + "train/classification_loss": 0.11261561512947083, + "train/contrastive_loss": 1.3402072191238403, + "train/negative_loss": 1.2441537380218506, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.09605353325605392, + "train/total_loss": 0.3806570768356323 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.13899821043014526, - "train/contrastive_loss": 3.6482725143432617, - "train/negative_loss": 1.9483106136322021, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.6999620199203491, - "train/total_loss": 0.8686527013778687 + "train/classification_loss": 0.22225457429885864, + "train/contrastive_loss": 1.717200517654419, + "train/negative_loss": 1.4871206283569336, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.23007993400096893, + "train/total_loss": 0.5656946897506714 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.1708022803068161, - "train/contrastive_loss": 1.1557947397232056, - "train/negative_loss": 0.5793697237968445, + "train/classification_loss": 0.22670303285121918, + "train/contrastive_loss": 2.4150938987731934, + "train/negative_loss": 1.8397369384765625, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.5764250159263611, - "train/total_loss": 0.40196123719215393 + "train/num_positives": 8, + "train/positive_loss": 0.5753570795059204, + "train/total_loss": 0.7097218036651611 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.23990294337272644, - "train/contrastive_loss": 1.999047040939331, - "train/negative_loss": 1.999038577079773, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 8.503644494339824e-06, - "train/total_loss": 0.6397123336791992 + "train/classification_loss": 0.18728338181972504, + "train/contrastive_loss": 1.1736338138580322, + "train/negative_loss": 1.1439838409423828, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.02964996173977852, + "train/total_loss": 0.4220101535320282 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.1549234837293625, - "train/contrastive_loss": 1.2002921104431152, - "train/negative_loss": 0.7258384823799133, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.47445356845855713, - "train/total_loss": 0.39498192071914673 + "train/classification_loss": 0.11804933845996857, + "train/contrastive_loss": 0.50077223777771, + "train/negative_loss": 0.22932720184326172, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.27144506573677063, + "train/total_loss": 0.21820378303527832 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.21038368344306946, - "train/contrastive_loss": 3.3087167739868164, - "train/negative_loss": 1.1217433214187622, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.1869735717773438, - "train/total_loss": 0.8721270561218262 + "train/classification_loss": 0.17583192884922028, + "train/contrastive_loss": 2.3222224712371826, + "train/negative_loss": 2.3032164573669434, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.01900612935423851, + "train/total_loss": 0.6402764320373535 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.21623048186302185, - "train/contrastive_loss": 3.3868465423583984, - "train/negative_loss": 2.481123685836792, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.9057227373123169, - "train/total_loss": 0.8935997486114502 + "train/classification_loss": 0.1774006187915802, + "train/contrastive_loss": 6.202363967895508, + "train/negative_loss": 6.202311038970947, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 5.303501893649809e-05, + "train/total_loss": 1.4178733825683594 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.14718136191368103, - "train/contrastive_loss": 2.3015453815460205, - "train/negative_loss": 1.3150310516357422, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.9865142703056335, - "train/total_loss": 0.6074904203414917 + "train/classification_loss": 0.1684599667787552, + "train/contrastive_loss": 1.9690700769424438, + "train/negative_loss": 1.0491468906402588, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.9199231863021851, + "train/total_loss": 0.5622739791870117 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.1903207153081894, - "train/contrastive_loss": 4.034844875335693, - "train/negative_loss": 3.31681227684021, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.7180324196815491, - "train/total_loss": 0.9972897171974182 + "train/classification_loss": 0.20292598009109497, + "train/contrastive_loss": 2.0782015323638916, + "train/negative_loss": 1.2045344114303589, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.8736671209335327, + "train/total_loss": 0.6185662746429443 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.07176332175731659, - "train/contrastive_loss": 0.3468479812145233, - "train/negative_loss": 0.34566056728363037, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.001187402056530118, - "train/total_loss": 0.1411329209804535 + "train/classification_loss": 0.1835746020078659, + "train/contrastive_loss": 2.703171968460083, + "train/negative_loss": 1.3359017372131348, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.3672702312469482, + "train/total_loss": 0.7242090106010437 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.1978733390569687, - "train/contrastive_loss": 1.111863374710083, - "train/negative_loss": 0.8425741195678711, + "train/classification_loss": 0.2059800773859024, + "train/contrastive_loss": 3.025761842727661, + "train/negative_loss": 2.4487171173095703, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.2692892849445343, - "train/total_loss": 0.4202460050582886 + "train/positive_loss": 0.577044665813446, + "train/total_loss": 0.8111324310302734 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.1136418804526329, - "train/contrastive_loss": 1.8500757217407227, - "train/negative_loss": 1.3314672708511353, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.5186085104942322, - "train/total_loss": 0.48365703225135803 + "train/classification_loss": 0.15136948227882385, + "train/contrastive_loss": 0.8036243319511414, + "train/negative_loss": 0.8034980297088623, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.00012627772230189294, + "train/total_loss": 0.3120943307876587 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.2817772626876831, - "train/contrastive_loss": 4.158234596252441, - "train/negative_loss": 2.3905513286590576, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.7676830291748047, - "train/total_loss": 1.1134241819381714 + "train/classification_loss": 0.18112443387508392, + "train/contrastive_loss": 1.2086983919143677, + "train/negative_loss": 0.7162982225418091, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.4924001395702362, + "train/total_loss": 0.4228641092777252 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.2338811159133911, - "train/contrastive_loss": 3.13173246383667, - "train/negative_loss": 2.4842865467071533, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.6474459767341614, - "train/total_loss": 0.860227644443512 + "train/classification_loss": 0.18197710812091827, + "train/contrastive_loss": 3.688178062438965, + "train/negative_loss": 2.2303266525268555, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.4578514099121094, + "train/total_loss": 0.91961270570755 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.21189749240875244, - "train/contrastive_loss": 4.1148905754089355, - "train/negative_loss": 1.9970492124557495, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 2.1178412437438965, - "train/total_loss": 1.0348756313323975 + "train/classification_loss": 0.2508490979671478, + "train/contrastive_loss": 2.8669023513793945, + "train/negative_loss": 2.078296661376953, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.7886056900024414, + "train/total_loss": 0.8242295980453491 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.19474296271800995, - "train/contrastive_loss": 2.483468770980835, - "train/negative_loss": 2.481419086456299, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.0020497289951890707, - "train/total_loss": 0.6914367079734802 + "train/classification_loss": 0.2320588082075119, + "train/contrastive_loss": 2.7234549522399902, + "train/negative_loss": 1.5700616836547852, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.1533933877944946, + "train/total_loss": 0.776749849319458 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.1763027012348175, - "train/contrastive_loss": 1.288729190826416, - "train/negative_loss": 0.7800418138504028, - "train/num_negatives": 44, + "train/classification_loss": 0.2182902991771698, + "train/contrastive_loss": 0.5984843969345093, + "train/negative_loss": 0.5748553276062012, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.023629089817404747, + "train/total_loss": 0.33798718452453613 + }, + { + "epoch": 12.12167300380228, + "step": 1600, + "train/classification_loss": 0.24809111654758453, + "train/contrastive_loss": 3.1582770347595215, + "train/negative_loss": 0.6579371094703674, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.508687436580658, - "train/total_loss": 0.43404853343963623 + "train/positive_loss": 2.500339984893799, + "train/total_loss": 0.8797464966773987 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.21076472103595734, - "train/contrastive_loss": 6.931009292602539, - "train/negative_loss": 6.5306396484375, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.4003695845603943, - "train/total_loss": 1.5969666242599487 + "train/classification_loss": 0.2034873068332672, + "train/contrastive_loss": 1.7792260646820068, + "train/negative_loss": 0.9678674936294556, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.811358630657196, + "train/total_loss": 0.5593324899673462 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.2068892866373062, - "train/contrastive_loss": 3.5128965377807617, - "train/negative_loss": 1.8621296882629395, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.6507669687271118, - "train/total_loss": 0.9094685912132263 + "train/classification_loss": 0.20935265719890594, + "train/contrastive_loss": 0.5558663606643677, + "train/negative_loss": 0.5558643937110901, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 1.9669557786983205e-06, + "train/total_loss": 0.3205259442329407 + }, + { + "epoch": 12.12167300380228, + "step": 1600, + "train/classification_loss": 0.14814211428165436, + "train/contrastive_loss": 5.133417129516602, + "train/negative_loss": 4.3707661628723145, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.7626511454582214, + "train/total_loss": 1.1748255491256714 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.07736042886972427, - "train/contrastive_loss": 1.6321200132369995, - "train/negative_loss": 1.5932475328445435, + "train/classification_loss": 0.21286523342132568, + "train/contrastive_loss": 3.177407741546631, + "train/negative_loss": 2.7748420238494873, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.038872458040714264, - "train/total_loss": 0.4037844240665436 + "train/positive_loss": 0.4025656282901764, + "train/total_loss": 0.8483467698097229 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.1488143354654312, - "train/contrastive_loss": 2.000516891479492, - "train/negative_loss": 1.2192745208740234, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.7812424302101135, - "train/total_loss": 0.5489177107810974 + "train/classification_loss": 0.20272955298423767, + "train/contrastive_loss": 2.2746124267578125, + "train/negative_loss": 1.037997841835022, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.236614465713501, + "train/total_loss": 0.6576520204544067 + }, + { + "epoch": 12.12167300380228, + "step": 1600, + "train/classification_loss": 0.1649957299232483, + "train/contrastive_loss": 2.676151990890503, + "train/negative_loss": 2.6761300563812256, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 2.1934749383945018e-05, + "train/total_loss": 0.7002261281013489 + }, + { + "epoch": 12.12167300380228, + "step": 1600, + "train/classification_loss": 0.2287149280309677, + "train/contrastive_loss": 2.351236581802368, + "train/negative_loss": 0.6586307287216187, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 1.6926058530807495, + "train/total_loss": 0.6989622712135315 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.19744686782360077, - "train/contrastive_loss": 2.8055343627929688, - "train/negative_loss": 1.6830716133117676, + "train/classification_loss": 0.20071808993816376, + "train/contrastive_loss": 0.7104887366294861, + "train/negative_loss": 0.5824899077415466, "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.1224627494812012, - "train/total_loss": 0.7585537433624268 + "train/num_positives": 8, + "train/positive_loss": 0.12799881398677826, + "train/total_loss": 0.3428158462047577 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.056874822825193405, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.056874822825193405 - }, - { - "epoch": 15.239234449760765, - "eval_exact_match_accuracy": 0.36363636363636365, - "eval_hamming_loss": 0.059949338587109484, - "eval_loss": 0.6890327334403992, - "eval_macro_f1": 0.2245930031841075, - "eval_macro_precision": 0.330176376770804, - "eval_macro_recall": 0.2028880999815256, - "eval_micro_f1": 0.5534591194968553, - "eval_micro_precision": 0.7292817679558011, - "eval_micro_recall": 0.44594594594594594, - "eval_runtime": 1.055, - "eval_samples_per_second": 198.109, - "eval_steps_per_second": 25.593, - "step": 1600 + "train/classification_loss": 0.18567177653312683, + "train/contrastive_loss": 2.0488533973693848, + "train/negative_loss": 1.471875786781311, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.5769777297973633, + "train/total_loss": 0.5954424738883972 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.13198119401931763, - "train/contrastive_loss": 0.9257301092147827, - "train/negative_loss": 0.2407333254814148, - "train/num_negatives": 44, + "train/classification_loss": 0.2031201422214508, + "train/contrastive_loss": 1.5375924110412598, + "train/negative_loss": 1.5013926029205322, + "train/num_negatives": 42, "train/num_positives": 12, - "train/positive_loss": 0.6849967837333679, - "train/total_loss": 0.3171272277832031 + "train/positive_loss": 0.03619980439543724, + "train/total_loss": 0.5106385946273804 + }, + { + "epoch": 12.12167300380228, + "step": 1600, + "train/classification_loss": 0.11557672917842865, + "train/contrastive_loss": 1.0100412368774414, + "train/negative_loss": 0.9667158722877502, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.043325312435626984, + "train/total_loss": 0.3175849914550781 + }, + { + "epoch": 12.12167300380228, + "eval_exact_match_accuracy": 0.29389312977099236, + "eval_hamming_loss": 0.06286484059272564, + "eval_loss": 0.6028220653533936, + "eval_macro_f1": 0.2045373155804306, + "eval_macro_precision": 0.30171582225661925, + "eval_macro_recall": 0.16570599374740877, + "eval_micro_f1": 0.5087719298245614, + "eval_micro_precision": 0.8146067415730337, + "eval_micro_recall": 0.36989795918367346, + "eval_runtime": 1.3113, + "eval_samples_per_second": 199.808, + "eval_steps_per_second": 25.167, + "step": 1600 }, { - "epoch": 15.239234449760765, + "epoch": 12.12167300380228, "step": 1600, - "train/classification_loss": 0.11313965171575546, - "train/contrastive_loss": 1.5864315032958984, - "train/negative_loss": 1.3034659624099731, + "train/classification_loss": 0.08570905774831772, + "train/contrastive_loss": 0.2990018129348755, + "train/negative_loss": 0.2971798777580261, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.28296560049057007, - "train/total_loss": 0.4304259717464447 + "train/positive_loss": 0.0018219402991235256, + "train/total_loss": 0.14550942182540894 }, { - "epoch": 15.717703349282296, - "grad_norm": 10.801045417785645, - "learning_rate": 1.9409541984732827e-05, - "loss": 0.5203, + "epoch": 12.12167300380228, + "step": 1600, + "train/classification_loss": 0.13388659060001373, + "train/contrastive_loss": 0.6260982751846313, + "train/negative_loss": 0.6112869381904602, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.01481135655194521, + "train/total_loss": 0.25910624861717224 + }, + { + "epoch": 12.50190114068441, + "grad_norm": 7.73107385635376, + "learning_rate": 1.9530500758725343e-05, + "loss": 0.5407, "step": 1650 }, { - "epoch": 15.717703349282296, + "epoch": 12.50190114068441, "step": 1650, - "train/classification_loss": 0.16223503649234772, - "train/contrastive_loss": 0.6221617460250854, - "train/negative_loss": 0.412134051322937, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.21002767980098724, - "train/total_loss": 0.2866673767566681 + "train/classification_loss": 0.11878790706396103, + "train/contrastive_loss": 0.5832277536392212, + "train/negative_loss": 0.5756916999816895, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.007536050397902727, + "train/total_loss": 0.2354334592819214 }, { - "epoch": 15.717703349282296, + "epoch": 12.50190114068441, "step": 1650, - "train/classification_loss": 0.13385009765625, - "train/contrastive_loss": 0.4339475929737091, - "train/negative_loss": 0.3740254044532776, - "train/num_negatives": 34, - "train/num_positives": 20, - "train/positive_loss": 0.059922195971012115, - "train/total_loss": 0.22063961625099182 + "train/classification_loss": 0.11889126896858215, + "train/contrastive_loss": 0.5771357417106628, + "train/negative_loss": 0.5771155953407288, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 2.0116911400691606e-05, + "train/total_loss": 0.23431842029094696 }, { - "epoch": 16.19138755980861, - "grad_norm": 5.265356540679932, - "learning_rate": 1.9390458015267178e-05, - "loss": 0.4637, + "epoch": 12.88212927756654, + "grad_norm": 10.54318904876709, + "learning_rate": 1.9515326251896815e-05, + "loss": 0.5283, "step": 1700 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, + "step": 1700, + "train/classification_loss": 0.09606307744979858, + "train/contrastive_loss": 0.44607722759246826, + "train/negative_loss": 0.2081630975008011, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.23791413009166718, + "train/total_loss": 0.1852785348892212 + }, + { + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.1460825502872467, - "train/contrastive_loss": 1.8468589782714844, - "train/negative_loss": 1.503981351852417, + "train/classification_loss": 0.22282098233699799, + "train/contrastive_loss": 1.528801679611206, + "train/negative_loss": 1.2383153438568115, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.34287765622138977, - "train/total_loss": 0.5154543519020081 + "train/num_positives": 6, + "train/positive_loss": 0.29048627614974976, + "train/total_loss": 0.5285813212394714 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.16673989593982697, - "train/contrastive_loss": 2.528071641921997, - "train/negative_loss": 1.41066312789917, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.1174085140228271, - "train/total_loss": 0.6723542213439941 + "train/classification_loss": 0.13247670233249664, + "train/contrastive_loss": 1.917529582977295, + "train/negative_loss": 1.5841270685195923, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.3334025740623474, + "train/total_loss": 0.5159826278686523 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.1403646618127823, - "train/contrastive_loss": 0.8411633968353271, - "train/negative_loss": 0.5411331057548523, - "train/num_negatives": 44, + "train/classification_loss": 0.25282517075538635, + "train/contrastive_loss": 1.626455545425415, + "train/negative_loss": 1.3423837423324585, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.2840718626976013, + "train/total_loss": 0.5781162977218628 + }, + { + "epoch": 12.88212927756654, + "step": 1700, + "train/classification_loss": 0.07956813275814056, + "train/contrastive_loss": 0.6915061473846436, + "train/negative_loss": 0.6914976239204407, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 8.547396646463312e-06, + "train/total_loss": 0.217869371175766 + }, + { + "epoch": 12.88212927756654, + "step": 1700, + "train/classification_loss": 0.22155509889125824, + "train/contrastive_loss": 1.7423001527786255, + "train/negative_loss": 1.661501169204712, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.30003029108047485, - "train/total_loss": 0.3085973262786865 + "train/positive_loss": 0.08079902827739716, + "train/total_loss": 0.5700151324272156 + }, + { + "epoch": 12.88212927756654, + "step": 1700, + "train/classification_loss": 0.22545653581619263, + "train/contrastive_loss": 2.551382541656494, + "train/negative_loss": 1.6049325466156006, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.9464499354362488, + "train/total_loss": 0.7357330322265625 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.1665392518043518, - "train/contrastive_loss": 3.0270309448242188, - "train/negative_loss": 1.7476743459701538, + "train/classification_loss": 0.19748640060424805, + "train/contrastive_loss": 1.2935124635696411, + "train/negative_loss": 1.2756943702697754, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.2793567180633545, - "train/total_loss": 0.7719454765319824 + "train/positive_loss": 0.01781807281076908, + "train/total_loss": 0.4561888873577118 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.13227206468582153, - "train/contrastive_loss": 2.963927745819092, - "train/negative_loss": 1.2840406894683838, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.6798871755599976, - "train/total_loss": 0.7250576019287109 + "train/classification_loss": 0.10938296467065811, + "train/contrastive_loss": 0.14993321895599365, + "train/negative_loss": 0.14991207420825958, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 2.1151698092580773e-05, + "train/total_loss": 0.13936960697174072 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.15310640633106232, - "train/contrastive_loss": 2.882631778717041, - "train/negative_loss": 1.4283075332641602, - "train/num_negatives": 32, + "train/classification_loss": 0.19253912568092346, + "train/contrastive_loss": 2.4842779636383057, + "train/negative_loss": 2.461676836013794, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.022601084783673286, + "train/total_loss": 0.6893947124481201 + }, + { + "epoch": 12.88212927756654, + "step": 1700, + "train/classification_loss": 0.18020649254322052, + "train/contrastive_loss": 6.5282673835754395, + "train/negative_loss": 6.5282511711120605, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.4543243646621704, - "train/total_loss": 0.7296327352523804 + "train/positive_loss": 1.6104877431644127e-05, + "train/total_loss": 1.485859990119934 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.16978925466537476, - "train/contrastive_loss": 1.626012921333313, - "train/negative_loss": 1.351974368095398, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.27403855323791504, - "train/total_loss": 0.49499183893203735 + "train/classification_loss": 0.1793949007987976, + "train/contrastive_loss": 2.6745691299438477, + "train/negative_loss": 1.7313265800476074, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.9432426691055298, + "train/total_loss": 0.7143087387084961 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.2524758279323578, - "train/contrastive_loss": 1.426585078239441, - "train/negative_loss": 1.3498964309692383, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.07668867707252502, - "train/total_loss": 0.5377928614616394 + "train/classification_loss": 0.21810562908649445, + "train/contrastive_loss": 2.1919286251068115, + "train/negative_loss": 1.8644332885742188, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.327495276927948, + "train/total_loss": 0.6564913392066956 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.14746323227882385, - "train/contrastive_loss": 1.8775863647460938, - "train/negative_loss": 0.9509301781654358, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.9266561269760132, - "train/total_loss": 0.5229805111885071 + "train/classification_loss": 0.1950128823518753, + "train/contrastive_loss": 2.2918832302093506, + "train/negative_loss": 1.7090336084365845, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 0.5828496813774109, + "train/total_loss": 0.6533895134925842 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.2260521948337555, - "train/contrastive_loss": 2.904604434967041, - "train/negative_loss": 1.2982759475708008, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 1.6063283681869507, - "train/total_loss": 0.8069730997085571 + "train/classification_loss": 0.2087969332933426, + "train/contrastive_loss": 2.6403746604919434, + "train/negative_loss": 2.2302703857421875, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.41010424494743347, + "train/total_loss": 0.7368718385696411 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.19383873045444489, - "train/contrastive_loss": 1.9140510559082031, - "train/negative_loss": 1.56093168258667, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.3531193137168884, - "train/total_loss": 0.5766489505767822 + "train/classification_loss": 0.13864614069461823, + "train/contrastive_loss": 0.7644107937812805, + "train/negative_loss": 0.764297604560852, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.00011316063319100067, + "train/total_loss": 0.29152828454971313 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.1246759444475174, - "train/contrastive_loss": 0.6045861840248108, - "train/negative_loss": 0.43279314041137695, + "train/classification_loss": 0.16396041214466095, + "train/contrastive_loss": 0.8738212585449219, + "train/negative_loss": 0.7734999060630798, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 0.17179302871227264, - "train/total_loss": 0.24559319019317627 + "train/positive_loss": 0.10032138228416443, + "train/total_loss": 0.33872467279434204 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.1917956918478012, - "train/contrastive_loss": 3.6683974266052246, - "train/negative_loss": 2.776566982269287, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.8918304443359375, - "train/total_loss": 0.9254751801490784 + "train/classification_loss": 0.19523188471794128, + "train/contrastive_loss": 3.546950101852417, + "train/negative_loss": 2.522944211959839, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.0240058898925781, + "train/total_loss": 0.904621958732605 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.06658864766359329, - "train/contrastive_loss": 0.31333044171333313, - "train/negative_loss": 0.3005511164665222, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.012779326178133488, - "train/total_loss": 0.12925472855567932 + "train/classification_loss": 0.24949976801872253, + "train/contrastive_loss": 3.202735424041748, + "train/negative_loss": 2.0317044258117676, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.17103111743927, + "train/total_loss": 0.8900468349456787 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.21705736219882965, - "train/contrastive_loss": 2.644564390182495, - "train/negative_loss": 0.8175681829452515, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.8269962072372437, - "train/total_loss": 0.7459702491760254 + "train/classification_loss": 0.24016907811164856, + "train/contrastive_loss": 3.4043455123901367, + "train/negative_loss": 2.8826074600219727, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.5217381119728088, + "train/total_loss": 0.9210381507873535 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.10301677137613297, - "train/contrastive_loss": 1.457463026046753, - "train/negative_loss": 0.8628287315368652, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.5946343541145325, - "train/total_loss": 0.39450937509536743 + "train/classification_loss": 0.22149565815925598, + "train/contrastive_loss": 1.2266966104507446, + "train/negative_loss": 0.46073174476623535, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.7659648656845093, + "train/total_loss": 0.4668349623680115 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.2538357377052307, - "train/contrastive_loss": 4.588127613067627, - "train/negative_loss": 2.273132085800171, - "train/num_negatives": 46, + "train/classification_loss": 0.26768800616264343, + "train/contrastive_loss": 2.0071263313293457, + "train/negative_loss": 0.8446502685546875, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 2.314995527267456, - "train/total_loss": 1.1714613437652588 + "train/positive_loss": 1.1624759435653687, + "train/total_loss": 0.669113278388977 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.24315491318702698, - "train/contrastive_loss": 2.160088300704956, - "train/negative_loss": 1.1869938373565674, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.9730944633483887, - "train/total_loss": 0.6751725673675537 + "train/classification_loss": 0.2098824381828308, + "train/contrastive_loss": 1.408147931098938, + "train/negative_loss": 1.4068970680236816, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.001250898465514183, + "train/total_loss": 0.4915120303630829 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.21998827159404755, - "train/contrastive_loss": 4.244344711303711, - "train/negative_loss": 1.6891050338745117, + "train/classification_loss": 0.1932513564825058, + "train/contrastive_loss": 0.30781248211860657, + "train/negative_loss": 0.3078111708164215, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 2.555239677429199, - "train/total_loss": 1.068857192993164 + "train/num_positives": 8, + "train/positive_loss": 1.3113030945532955e-06, + "train/total_loss": 0.2548138499259949 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.19766205549240112, - "train/contrastive_loss": 1.9258180856704712, - "train/negative_loss": 1.6455812454223633, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.2802368104457855, - "train/total_loss": 0.5828256607055664 + "train/classification_loss": 0.18092581629753113, + "train/contrastive_loss": 6.833359241485596, + "train/negative_loss": 6.3279242515563965, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.5054348111152649, + "train/total_loss": 1.5475977659225464 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.18755750358104706, - "train/contrastive_loss": 1.1898468732833862, - "train/negative_loss": 0.6151241660118103, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.5747227072715759, - "train/total_loss": 0.425526887178421 + "train/classification_loss": 0.21519052982330322, + "train/contrastive_loss": 3.8773326873779297, + "train/negative_loss": 2.484271764755249, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 1.3930610418319702, + "train/total_loss": 0.9906570911407471 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.19355224072933197, - "train/contrastive_loss": 5.258101463317871, - "train/negative_loss": 4.625899791717529, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.6322014331817627, - "train/total_loss": 1.2451726198196411 + "train/classification_loss": 0.20568615198135376, + "train/contrastive_loss": 3.015080690383911, + "train/negative_loss": 1.1093021631240845, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.9057785272598267, + "train/total_loss": 0.808702290058136 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.2236928641796112, - "train/contrastive_loss": 3.4001102447509766, - "train/negative_loss": 1.8576185703277588, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.5424917936325073, - "train/total_loss": 0.9037148952484131 + "train/classification_loss": 0.1743355691432953, + "train/contrastive_loss": 2.344377279281616, + "train/negative_loss": 2.34047269821167, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.0039045645389705896, + "train/total_loss": 0.6432110071182251 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.08759450912475586, - "train/contrastive_loss": 1.7441842555999756, - "train/negative_loss": 1.024582862854004, - "train/num_negatives": 42, + "train/classification_loss": 0.21885588765144348, + "train/contrastive_loss": 1.8821320533752441, + "train/negative_loss": 0.48158735036849976, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.7196013927459717, - "train/total_loss": 0.4364313781261444 + "train/positive_loss": 1.4005446434020996, + "train/total_loss": 0.5952823162078857 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.1263348013162613, - "train/contrastive_loss": 1.7909915447235107, - "train/negative_loss": 1.7692804336547852, - "train/num_negatives": 48, + "train/classification_loss": 0.19661563634872437, + "train/contrastive_loss": 0.8967550992965698, + "train/negative_loss": 0.8499459028244019, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.021711155772209167, - "train/total_loss": 0.4845331311225891 + "train/positive_loss": 0.046809203922748566, + "train/total_loss": 0.3759666681289673 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.1869618445634842, - "train/contrastive_loss": 2.03935170173645, - "train/negative_loss": 1.1856013536453247, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.8537504076957703, - "train/total_loss": 0.594832181930542 + "train/classification_loss": 0.21133507788181305, + "train/contrastive_loss": 1.6143351793289185, + "train/negative_loss": 1.2117116451263428, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.4026235342025757, + "train/total_loss": 0.5342020988464355 + }, + { + "epoch": 12.88212927756654, + "step": 1700, + "train/classification_loss": 0.2109963744878769, + "train/contrastive_loss": 1.9284276962280273, + "train/negative_loss": 1.7287086248397827, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.1997191309928894, + "train/total_loss": 0.5966818928718567 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.06219205632805824, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.06219205632805824 - }, - { - "epoch": 16.19138755980861, - "eval_exact_match_accuracy": 0.33014354066985646, - "eval_hamming_loss": 0.058823529411764705, - "eval_loss": 0.6392166018486023, - "eval_macro_f1": 0.2148674043221563, - "eval_macro_precision": 0.3527557075386074, - "eval_macro_recall": 0.17871850819947704, - "eval_micro_f1": 0.5173210161662818, - "eval_micro_precision": 0.8175182481751825, - "eval_micro_recall": 0.3783783783783784, - "eval_runtime": 1.0608, - "eval_samples_per_second": 197.028, - "eval_steps_per_second": 25.453, + "train/classification_loss": 0.11695797741413116, + "train/contrastive_loss": 1.8156991004943848, + "train/negative_loss": 1.7989892959594727, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.016709819436073303, + "train/total_loss": 0.48009783029556274 + }, + { + "epoch": 12.88212927756654, + "eval_exact_match_accuracy": 0.27099236641221375, + "eval_hamming_loss": 0.06645711719802425, + "eval_loss": 0.6272995471954346, + "eval_macro_f1": 0.20644859753856293, + "eval_macro_precision": 0.31226664854818936, + "eval_macro_recall": 0.1678703970410817, + "eval_micro_f1": 0.4861111111111111, + "eval_micro_precision": 0.7608695652173914, + "eval_micro_recall": 0.35714285714285715, + "eval_runtime": 1.3209, + "eval_samples_per_second": 198.346, + "eval_steps_per_second": 24.982, "step": 1700 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.14450135827064514, - "train/contrastive_loss": 0.2882414758205414, - "train/negative_loss": 0.16855844855308533, - "train/num_negatives": 40, + "train/classification_loss": 0.09757482260465622, + "train/contrastive_loss": 0.32933735847473145, + "train/negative_loss": 0.32916712760925293, + "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.11968302726745605, - "train/total_loss": 0.2021496593952179 + "train/positive_loss": 0.00017023234977386892, + "train/total_loss": 0.16344229876995087 }, { - "epoch": 16.19138755980861, + "epoch": 12.88212927756654, "step": 1700, - "train/classification_loss": 0.14777609705924988, - "train/contrastive_loss": 0.3725033402442932, - "train/negative_loss": 0.29374825954437256, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.07875506579875946, - "train/total_loss": 0.22227677702903748 + "train/classification_loss": 0.09597595036029816, + "train/contrastive_loss": 1.0215736627578735, + "train/negative_loss": 1.0086164474487305, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.012957226485013962, + "train/total_loss": 0.30029070377349854 }, { - "epoch": 16.669856459330145, - "grad_norm": 14.991618156433105, - "learning_rate": 1.937137404580153e-05, - "loss": 0.4526, + "epoch": 13.258555133079849, + "grad_norm": 6.615599155426025, + "learning_rate": 1.9500151745068286e-05, + "loss": 0.5197, "step": 1750 }, { - "epoch": 16.669856459330145, + "epoch": 13.258555133079849, "step": 1750, - "train/classification_loss": 0.11514256149530411, - "train/contrastive_loss": 0.6555768251419067, - "train/negative_loss": 0.31947025656700134, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.3361065983772278, - "train/total_loss": 0.2462579309940338 + "train/classification_loss": 0.12247705459594727, + "train/contrastive_loss": 0.4755876958370209, + "train/negative_loss": 0.47536373138427734, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.00022397536667995155, + "train/total_loss": 0.21759459376335144 }, { - "epoch": 16.669856459330145, + "epoch": 13.258555133079849, "step": 1750, - "train/classification_loss": 0.10076004266738892, - "train/contrastive_loss": 0.5821503400802612, - "train/negative_loss": 0.5819196701049805, + "train/classification_loss": 0.10521535575389862, + "train/contrastive_loss": 0.17052458226680756, + "train/negative_loss": 0.17050223052501678, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.0002306577516719699, - "train/total_loss": 0.21719011664390564 + "train/positive_loss": 2.23520582949277e-05, + "train/total_loss": 0.1393202692270279 }, { - "epoch": 17.14354066985646, - "grad_norm": 13.831987380981445, - "learning_rate": 1.935229007633588e-05, - "loss": 0.4466, + "epoch": 13.638783269961976, + "grad_norm": 7.799471378326416, + "learning_rate": 1.948497723823976e-05, + "loss": 0.5101, "step": 1800 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.14061635732650757, - "train/contrastive_loss": 1.7993816137313843, - "train/negative_loss": 1.303756833076477, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.4956247806549072, - "train/total_loss": 0.5004926919937134 + "train/classification_loss": 0.0735933780670166, + "train/contrastive_loss": 0.3083081543445587, + "train/negative_loss": 0.3009565472602844, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.007351607549935579, + "train/total_loss": 0.13525500893592834 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.17739702761173248, - "train/contrastive_loss": 2.434443235397339, - "train/negative_loss": 1.614537000656128, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.8199062347412109, - "train/total_loss": 0.6642856597900391 + "train/classification_loss": 0.20378831028938293, + "train/contrastive_loss": 1.1491307020187378, + "train/negative_loss": 1.1467385292053223, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.002392140682786703, + "train/total_loss": 0.43361443281173706 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.12042595446109772, - "train/contrastive_loss": 0.7969041466712952, - "train/negative_loss": 0.7360256314277649, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.06087850034236908, - "train/total_loss": 0.27980679273605347 + "train/classification_loss": 0.10379533469676971, + "train/contrastive_loss": 1.7083336114883423, + "train/negative_loss": 1.6632187366485596, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.04511485621333122, + "train/total_loss": 0.44546204805374146 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.1683027148246765, - "train/contrastive_loss": 3.5032272338867188, - "train/negative_loss": 1.669071078300476, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.8341560363769531, - "train/total_loss": 0.8689481616020203 + "train/classification_loss": 0.2229107767343521, + "train/contrastive_loss": 1.3486955165863037, + "train/negative_loss": 1.016568899154663, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.3321266174316406, + "train/total_loss": 0.4926499128341675 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.13928598165512085, - "train/contrastive_loss": 3.3332743644714355, - "train/negative_loss": 1.4015512466430664, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.9317232370376587, - "train/total_loss": 0.8059408664703369 + "train/classification_loss": 0.08722729235887527, + "train/contrastive_loss": 1.101468801498413, + "train/negative_loss": 1.1005933284759521, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.0008755002054385841, + "train/total_loss": 0.3075210452079773 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.14172042906284332, - "train/contrastive_loss": 2.6390676498413086, - "train/negative_loss": 1.6251447200775146, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.0139228105545044, - "train/total_loss": 0.6695339679718018 + "train/classification_loss": 0.2185172736644745, + "train/contrastive_loss": 1.8610458374023438, + "train/negative_loss": 1.443930983543396, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.417114794254303, + "train/total_loss": 0.5907264351844788 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.14169815182685852, - "train/contrastive_loss": 1.3769944906234741, - "train/negative_loss": 1.1471415758132935, + "train/classification_loss": 0.22605638206005096, + "train/contrastive_loss": 2.362730026245117, + "train/negative_loss": 1.4725955724716187, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.22985294461250305, - "train/total_loss": 0.4170970618724823 + "train/num_positives": 8, + "train/positive_loss": 0.8901345729827881, + "train/total_loss": 0.6986023783683777 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.2502813935279846, - "train/contrastive_loss": 0.9636480212211609, - "train/negative_loss": 0.876903235912323, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.0867447704076767, - "train/total_loss": 0.44301098585128784 + "train/classification_loss": 0.18716482818126678, + "train/contrastive_loss": 1.1390804052352905, + "train/negative_loss": 0.7269158959388733, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.4121645390987396, + "train/total_loss": 0.4149809181690216 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.1346329003572464, - "train/contrastive_loss": 0.9466884136199951, - "train/negative_loss": 0.7966229915618896, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.15006545186042786, - "train/total_loss": 0.32397058606147766 + "train/classification_loss": 0.11455750465393066, + "train/contrastive_loss": 0.1048976257443428, + "train/negative_loss": 0.10159383714199066, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.003303785575553775, + "train/total_loss": 0.1355370283126831 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.23915411531925201, - "train/contrastive_loss": 3.9910662174224854, - "train/negative_loss": 1.6632277965545654, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.32783842086792, - "train/total_loss": 1.037367343902588 + "train/classification_loss": 0.18756996095180511, + "train/contrastive_loss": 2.6788322925567627, + "train/negative_loss": 2.5148730278015137, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.1639593094587326, + "train/total_loss": 0.7233364582061768 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.21238869428634644, - "train/contrastive_loss": 2.1614491939544678, - "train/negative_loss": 1.7848687171936035, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.37658050656318665, - "train/total_loss": 0.64467853307724 + "train/classification_loss": 0.17400898039340973, + "train/contrastive_loss": 5.34821891784668, + "train/negative_loss": 5.347267150878906, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.0009516936843283474, + "train/total_loss": 1.2436528205871582 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.12946940958499908, - "train/contrastive_loss": 1.664920449256897, - "train/negative_loss": 1.0716230869293213, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.5932973623275757, - "train/total_loss": 0.4624534845352173 + "train/classification_loss": 0.15718485414981842, + "train/contrastive_loss": 2.9785475730895996, + "train/negative_loss": 2.1047356128692627, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.8738118410110474, + "train/total_loss": 0.7528943419456482 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.206929013133049, - "train/contrastive_loss": 4.448216438293457, - "train/negative_loss": 2.931344747543335, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.5168718099594116, - "train/total_loss": 1.0965722799301147 + "train/classification_loss": 0.20287369191646576, + "train/contrastive_loss": 2.464920997619629, + "train/negative_loss": 1.3021577596664429, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.1627633571624756, + "train/total_loss": 0.6958578824996948 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.06715616583824158, - "train/contrastive_loss": 0.3084225654602051, - "train/negative_loss": 0.19461597502231598, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.11380660533905029, - "train/total_loss": 0.12884068489074707 + "train/classification_loss": 0.17473071813583374, + "train/contrastive_loss": 2.7827088832855225, + "train/negative_loss": 1.8537956476211548, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 0.9289132356643677, + "train/total_loss": 0.7312725186347961 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.17946024239063263, - "train/contrastive_loss": 1.039685845375061, - "train/negative_loss": 0.6710777878761292, + "train/classification_loss": 0.18917325139045715, + "train/contrastive_loss": 2.334197759628296, + "train/negative_loss": 2.211106777191162, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.3686080574989319, - "train/total_loss": 0.3873974084854126 + "train/positive_loss": 0.12309092283248901, + "train/total_loss": 0.656012773513794 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.10463927686214447, - "train/contrastive_loss": 1.9791920185089111, - "train/negative_loss": 1.0661441087722778, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.9130479693412781, - "train/total_loss": 0.50047767162323 + "train/classification_loss": 0.1425389051437378, + "train/contrastive_loss": 0.7888224720954895, + "train/negative_loss": 0.7830746173858643, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.005747850518673658, + "train/total_loss": 0.3003033995628357 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.26085853576660156, - "train/contrastive_loss": 3.4794609546661377, - "train/negative_loss": 2.1488842964172363, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.3305766582489014, - "train/total_loss": 0.956750750541687 + "train/classification_loss": 0.16108070313930511, + "train/contrastive_loss": 0.7935317754745483, + "train/negative_loss": 0.6890529990196228, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.10447880625724792, + "train/total_loss": 0.31978705525398254 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.24982166290283203, - "train/contrastive_loss": 2.4127278327941895, - "train/negative_loss": 1.8199793100357056, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.5927486419677734, - "train/total_loss": 0.7323672771453857 + "train/classification_loss": 0.18144001066684723, + "train/contrastive_loss": 4.074300765991211, + "train/negative_loss": 2.6255767345428467, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.4487237930297852, + "train/total_loss": 0.9963001608848572 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.18774455785751343, - "train/contrastive_loss": 2.819167137145996, - "train/negative_loss": 1.7669833898544312, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.0521836280822754, - "train/total_loss": 0.7515779733657837 + "train/classification_loss": 0.23920726776123047, + "train/contrastive_loss": 4.001072406768799, + "train/negative_loss": 2.600663900375366, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.4004086256027222, + "train/total_loss": 1.039421796798706 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.1862604320049286, - "train/contrastive_loss": 1.5428426265716553, - "train/negative_loss": 1.2280055284500122, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.3148370683193207, - "train/total_loss": 0.4948289692401886 + "train/classification_loss": 0.23295101523399353, + "train/contrastive_loss": 2.693228244781494, + "train/negative_loss": 1.6854667663574219, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.0077615976333618, + "train/total_loss": 0.7715966701507568 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.18744990229606628, - "train/contrastive_loss": 1.498857021331787, - "train/negative_loss": 0.9413409233093262, - "train/num_negatives": 44, + "train/classification_loss": 0.21859842538833618, + "train/contrastive_loss": 0.7781258225440979, + "train/negative_loss": 0.6700326800346375, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.10809312760829926, + "train/total_loss": 0.37422358989715576 + }, + { + "epoch": 13.638783269961976, + "step": 1800, + "train/classification_loss": 0.2328457534313202, + "train/contrastive_loss": 2.078765392303467, + "train/negative_loss": 0.46748557686805725, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.5575160384178162, - "train/total_loss": 0.48722130060195923 + "train/positive_loss": 1.611279845237732, + "train/total_loss": 0.648598849773407 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.18996690213680267, - "train/contrastive_loss": 4.838502883911133, - "train/negative_loss": 4.023083209991455, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.815419614315033, - "train/total_loss": 1.1576675176620483 + "train/classification_loss": 0.2054598480463028, + "train/contrastive_loss": 2.190652847290039, + "train/negative_loss": 1.2297484874725342, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.9609042406082153, + "train/total_loss": 0.6435904502868652 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.22032149136066437, - "train/contrastive_loss": 3.890558958053589, - "train/negative_loss": 2.072878360748291, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.8176805973052979, - "train/total_loss": 0.9984332919120789 + "train/classification_loss": 0.1836058646440506, + "train/contrastive_loss": 0.724223792552948, + "train/negative_loss": 0.7242207527160645, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 3.010040018125437e-06, + "train/total_loss": 0.32845062017440796 + }, + { + "epoch": 13.638783269961976, + "step": 1800, + "train/classification_loss": 0.16242405772209167, + "train/contrastive_loss": 5.85664701461792, + "train/negative_loss": 4.683841705322266, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 1.1728054285049438, + "train/total_loss": 1.3337534666061401 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.06860590726137161, - "train/contrastive_loss": 2.0582070350646973, - "train/negative_loss": 0.8874230980873108, + "train/classification_loss": 0.2233675867319107, + "train/contrastive_loss": 3.7644975185394287, + "train/negative_loss": 2.020059823989868, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 1.1707838773727417, - "train/total_loss": 0.4802473187446594 + "train/positive_loss": 1.7444376945495605, + "train/total_loss": 0.9762670993804932 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.11691358685493469, - "train/contrastive_loss": 1.6546388864517212, - "train/negative_loss": 1.6167664527893066, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.037872470915317535, - "train/total_loss": 0.4478413760662079 + "train/classification_loss": 0.1994556039571762, + "train/contrastive_loss": 2.2390120029449463, + "train/negative_loss": 1.1627429723739624, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.0762690305709839, + "train/total_loss": 0.6472579836845398 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.20305833220481873, - "train/contrastive_loss": 2.0574569702148438, - "train/negative_loss": 1.4394279718399048, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.6180289387702942, - "train/total_loss": 0.6145497560501099 + "train/classification_loss": 0.16719815135002136, + "train/contrastive_loss": 2.831435441970825, + "train/negative_loss": 2.8314037322998047, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 3.171017306158319e-05, + "train/total_loss": 0.733485221862793 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.03922242671251297, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.03922242671251297 - }, - { - "epoch": 17.14354066985646, - "eval_exact_match_accuracy": 0.36363636363636365, - "eval_hamming_loss": 0.05854207711792851, - "eval_loss": 0.6261152625083923, - "eval_macro_f1": 0.2577961043805446, - "eval_macro_precision": 0.3946746845075309, - "eval_macro_recall": 0.2238741010367308, - "eval_micro_f1": 0.5555555555555556, - "eval_micro_precision": 0.7558139534883721, - "eval_micro_recall": 0.4391891891891892, - "eval_runtime": 1.0592, - "eval_samples_per_second": 197.319, - "eval_steps_per_second": 25.491, - "step": 1800 + "train/classification_loss": 0.22703701257705688, + "train/contrastive_loss": 1.7075855731964111, + "train/negative_loss": 0.7542730569839478, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 0.9533125758171082, + "train/total_loss": 0.568554162979126 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.08715925365686417, - "train/contrastive_loss": 0.34230443835258484, - "train/negative_loss": 0.3402731418609619, + "train/classification_loss": 0.1994522511959076, + "train/contrastive_loss": 0.9040136337280273, + "train/negative_loss": 0.6801341772079468, "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.0020313048735260963, - "train/total_loss": 0.15562014281749725 + "train/positive_loss": 0.22387947142124176, + "train/total_loss": 0.38025498390197754 }, { - "epoch": 17.14354066985646, + "epoch": 13.638783269961976, "step": 1800, - "train/classification_loss": 0.10722032934427261, - "train/contrastive_loss": 0.40641629695892334, - "train/negative_loss": 0.32449793815612793, + "train/classification_loss": 0.18274252116680145, + "train/contrastive_loss": 2.0476491451263428, + "train/negative_loss": 1.236582636833191, "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.8110665082931519, + "train/total_loss": 0.5922723412513733 + }, + { + "epoch": 13.638783269961976, + "step": 1800, + "train/classification_loss": 0.19177789986133575, + "train/contrastive_loss": 0.8378923535346985, + "train/negative_loss": 0.8364318013191223, + "train/num_negatives": 42, "train/num_positives": 12, - "train/positive_loss": 0.08191834390163422, - "train/total_loss": 0.18850359320640564 + "train/positive_loss": 0.0014605596661567688, + "train/total_loss": 0.3593563735485077 + }, + { + "epoch": 13.638783269961976, + "step": 1800, + "train/classification_loss": 0.11341644078493118, + "train/contrastive_loss": 1.2300479412078857, + "train/negative_loss": 1.2292712926864624, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.0007765923510305583, + "train/total_loss": 0.35942602157592773 + }, + { + "epoch": 13.638783269961976, + "eval_exact_match_accuracy": 0.31297709923664124, + "eval_hamming_loss": 0.06061966771441401, + "eval_loss": 0.602760910987854, + "eval_macro_f1": 0.22219844944598655, + "eval_macro_precision": 0.31402127284480225, + "eval_macro_recall": 0.18616688158255434, + "eval_micro_f1": 0.5454545454545454, + "eval_micro_precision": 0.801980198019802, + "eval_micro_recall": 0.413265306122449, + "eval_runtime": 1.3254, + "eval_samples_per_second": 197.677, + "eval_steps_per_second": 24.898, + "step": 1800 + }, + { + "epoch": 13.638783269961976, + "step": 1800, + "train/classification_loss": 0.08451979607343674, + "train/contrastive_loss": 0.18734052777290344, + "train/negative_loss": 0.1682884842157364, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.0190520491451025, + "train/total_loss": 0.12198790162801743 + }, + { + "epoch": 13.638783269961976, + "step": 1800, + "train/classification_loss": 0.13382044434547424, + "train/contrastive_loss": 2.2609713077545166, + "train/negative_loss": 1.7917320728302002, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.4692392647266388, + "train/total_loss": 0.5860147476196289 }, { - "epoch": 17.62200956937799, - "grad_norm": 7.960711479187012, - "learning_rate": 1.933320610687023e-05, - "loss": 0.4687, + "epoch": 14.015209125475286, + "grad_norm": 12.249835968017578, + "learning_rate": 1.9469802731411232e-05, + "loss": 0.5195, "step": 1850 }, { - "epoch": 17.62200956937799, + "epoch": 14.015209125475286, "step": 1850, - "train/classification_loss": 0.12341137230396271, - "train/contrastive_loss": 0.6320751905441284, - "train/negative_loss": 0.28440263867378235, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.34767258167266846, - "train/total_loss": 0.24982641637325287 + "train/classification_loss": 0.09767338633537292, + "train/contrastive_loss": 0.3340766727924347, + "train/negative_loss": 0.17878377437591553, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.15529289841651917, + "train/total_loss": 0.16448873281478882 }, { - "epoch": 17.62200956937799, + "epoch": 14.015209125475286, "step": 1850, - "train/classification_loss": 0.06309905648231506, - "train/contrastive_loss": 0.04067939519882202, - "train/negative_loss": 0.04055425524711609, - "train/num_negatives": 30, - "train/num_positives": 26, - "train/positive_loss": 0.000125140868476592, - "train/total_loss": 0.07123493403196335 + "train/classification_loss": 0.09878262132406235, + "train/contrastive_loss": 0.34147679805755615, + "train/negative_loss": 0.3376036584377289, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.003873141948133707, + "train/total_loss": 0.16707798838615417 }, { - "epoch": 18.095693779904305, - "grad_norm": 11.400513648986816, - "learning_rate": 1.931412213740458e-05, - "loss": 0.4072, + "epoch": 14.395437262357415, + "grad_norm": 7.7965617179870605, + "learning_rate": 1.9454628224582704e-05, + "loss": 0.517, "step": 1900 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, + "step": 1900, + "train/classification_loss": 0.07511281967163086, + "train/contrastive_loss": 0.27653616666793823, + "train/negative_loss": 0.22097273170948029, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.05556342378258705, + "train/total_loss": 0.13042005896568298 + }, + { + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.1300882250070572, - "train/contrastive_loss": 1.950577735900879, - "train/negative_loss": 1.3486342430114746, + "train/classification_loss": 0.22325165569782257, + "train/contrastive_loss": 1.2811579704284668, + "train/negative_loss": 1.2744929790496826, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.6019435524940491, - "train/total_loss": 0.5202037692070007 + "train/num_positives": 6, + "train/positive_loss": 0.006664990913122892, + "train/total_loss": 0.4794832468032837 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.180917888879776, - "train/contrastive_loss": 2.421323776245117, - "train/negative_loss": 1.0704470872879028, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.3508766889572144, - "train/total_loss": 0.6651826500892639 + "train/classification_loss": 0.11254758387804031, + "train/contrastive_loss": 2.3951406478881836, + "train/negative_loss": 1.9852935075759888, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.4098471999168396, + "train/total_loss": 0.5915757417678833 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.13968943059444427, - "train/contrastive_loss": 1.244077205657959, - "train/negative_loss": 0.5094669461250305, - "train/num_negatives": 44, + "train/classification_loss": 0.20526641607284546, + "train/contrastive_loss": 1.7252683639526367, + "train/negative_loss": 1.0884147882461548, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.6368535757064819, + "train/total_loss": 0.5503200888633728 + }, + { + "epoch": 14.395437262357415, + "step": 1900, + "train/classification_loss": 0.09438164532184601, + "train/contrastive_loss": 0.8368391394615173, + "train/negative_loss": 0.7901600003242493, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.04667915031313896, + "train/total_loss": 0.2617494761943817 + }, + { + "epoch": 14.395437262357415, + "step": 1900, + "train/classification_loss": 0.22050265967845917, + "train/contrastive_loss": 1.581840991973877, + "train/negative_loss": 1.1081961393356323, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.7346101999282837, - "train/total_loss": 0.38850486278533936 + "train/positive_loss": 0.47364485263824463, + "train/total_loss": 0.5368708372116089 + }, + { + "epoch": 14.395437262357415, + "step": 1900, + "train/classification_loss": 0.2175942361354828, + "train/contrastive_loss": 1.8274791240692139, + "train/negative_loss": 1.3424395322799683, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.4850395619869232, + "train/total_loss": 0.58309006690979 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.15488077700138092, - "train/contrastive_loss": 3.7024433612823486, - "train/negative_loss": 1.8312009572982788, + "train/classification_loss": 0.16620810329914093, + "train/contrastive_loss": 2.598825454711914, + "train/negative_loss": 1.1574183702468872, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.8712424039840698, - "train/total_loss": 0.8953694701194763 + "train/positive_loss": 1.4414070844650269, + "train/total_loss": 0.6859731674194336 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.13165463507175446, - "train/contrastive_loss": 3.3967180252075195, - "train/negative_loss": 1.1302891969680786, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 2.2664289474487305, - "train/total_loss": 0.810998260974884 + "train/classification_loss": 0.12259801477193832, + "train/contrastive_loss": 0.7445568442344666, + "train/negative_loss": 0.3526812195777893, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.39187562465667725, + "train/total_loss": 0.2715093791484833 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.1512242704629898, - "train/contrastive_loss": 3.2102413177490234, - "train/negative_loss": 1.5527098178863525, - "train/num_negatives": 32, + "train/classification_loss": 0.1537664383649826, + "train/contrastive_loss": 1.1059975624084473, + "train/negative_loss": 1.0545599460601807, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.051437605172395706, + "train/total_loss": 0.37496596574783325 + }, + { + "epoch": 14.395437262357415, + "step": 1900, + "train/classification_loss": 0.18275803327560425, + "train/contrastive_loss": 4.051697254180908, + "train/negative_loss": 3.911844491958618, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.657531499862671, - "train/total_loss": 0.7932725548744202 + "train/positive_loss": 0.13985270261764526, + "train/total_loss": 0.9930974841117859 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.15925730764865875, - "train/contrastive_loss": 1.7629473209381104, - "train/negative_loss": 1.343664526939392, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.4192827343940735, - "train/total_loss": 0.5118467807769775 + "train/classification_loss": 0.1755467802286148, + "train/contrastive_loss": 2.0721240043640137, + "train/negative_loss": 1.6101200580596924, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.4620038568973541, + "train/total_loss": 0.5899716019630432 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.22161756455898285, - "train/contrastive_loss": 1.070312261581421, - "train/negative_loss": 1.0537030696868896, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.016609206795692444, - "train/total_loss": 0.4356800317764282 + "train/classification_loss": 0.16711030900478363, + "train/contrastive_loss": 1.6092967987060547, + "train/negative_loss": 0.963678777217865, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.6456179618835449, + "train/total_loss": 0.48896968364715576 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.13964523375034332, - "train/contrastive_loss": 1.7134759426116943, - "train/negative_loss": 0.7751818895339966, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.9382941126823425, - "train/total_loss": 0.4823404550552368 + "train/classification_loss": 0.19343402981758118, + "train/contrastive_loss": 1.8655927181243896, + "train/negative_loss": 1.4780042171478271, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 0.3875885009765625, + "train/total_loss": 0.5665525794029236 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.21902143955230713, - "train/contrastive_loss": 3.448808431625366, - "train/negative_loss": 1.1691901683807373, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.279618263244629, - "train/total_loss": 0.9087831377983093 + "train/classification_loss": 0.1965893656015396, + "train/contrastive_loss": 3.4597222805023193, + "train/negative_loss": 2.415802478790283, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 1.0439198017120361, + "train/total_loss": 0.8885338306427002 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.2331944853067398, - "train/contrastive_loss": 2.388911247253418, - "train/negative_loss": 1.9757474660873413, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.4131636619567871, - "train/total_loss": 0.7109767198562622 + "train/classification_loss": 0.1421116441488266, + "train/contrastive_loss": 1.6673777103424072, + "train/negative_loss": 0.9821317195892334, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.6852459907531738, + "train/total_loss": 0.4755871891975403 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.12599752843379974, - "train/contrastive_loss": 1.548302173614502, - "train/negative_loss": 0.6962335109710693, + "train/classification_loss": 0.16773027181625366, + "train/contrastive_loss": 1.1012933254241943, + "train/negative_loss": 0.6711786985397339, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 0.8520686030387878, - "train/total_loss": 0.43565797805786133 + "train/positive_loss": 0.43011459708213806, + "train/total_loss": 0.3879889249801636 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.1920221596956253, - "train/contrastive_loss": 3.912412643432617, - "train/negative_loss": 2.660000801086426, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.2524118423461914, - "train/total_loss": 0.9745047092437744 + "train/classification_loss": 0.17112725973129272, + "train/contrastive_loss": 4.739749908447266, + "train/negative_loss": 2.259692430496216, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 2.4800572395324707, + "train/total_loss": 1.119077205657959 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.06035440042614937, - "train/contrastive_loss": 0.247762992978096, - "train/negative_loss": 0.2470719963312149, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.0006909968215040863, - "train/total_loss": 0.10990700125694275 + "train/classification_loss": 0.2452780157327652, + "train/contrastive_loss": 3.9520883560180664, + "train/negative_loss": 2.2219271659851074, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.7301610708236694, + "train/total_loss": 1.0356956720352173 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.19224901497364044, - "train/contrastive_loss": 1.0481619834899902, - "train/negative_loss": 0.782869815826416, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.265292227268219, - "train/total_loss": 0.4018813967704773 + "train/classification_loss": 0.22113758325576782, + "train/contrastive_loss": 2.6628007888793945, + "train/negative_loss": 1.508427619934082, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.1543731689453125, + "train/total_loss": 0.7536977529525757 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.11466037482023239, - "train/contrastive_loss": 2.4035134315490723, - "train/negative_loss": 1.2248822450637817, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 1.178631067276001, - "train/total_loss": 0.5953630805015564 + "train/classification_loss": 0.23403191566467285, + "train/contrastive_loss": 1.9398785829544067, + "train/negative_loss": 0.5723521709442139, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 1.3675264120101929, + "train/total_loss": 0.6220076084136963 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.2612779140472412, - "train/contrastive_loss": 3.0463156700134277, - "train/negative_loss": 1.7878063917160034, - "train/num_negatives": 46, + "train/classification_loss": 0.2529774308204651, + "train/contrastive_loss": 3.260683536529541, + "train/negative_loss": 0.9763698577880859, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.2585092782974243, - "train/total_loss": 0.8705410361289978 + "train/positive_loss": 2.284313678741455, + "train/total_loss": 0.9051141738891602 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.275688111782074, - "train/contrastive_loss": 3.0787177085876465, - "train/negative_loss": 1.3709449768066406, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.7077727317810059, - "train/total_loss": 0.8914316892623901 + "train/classification_loss": 0.22186987102031708, + "train/contrastive_loss": 0.8460968136787415, + "train/negative_loss": 0.8445189595222473, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.001577861374244094, + "train/total_loss": 0.39108923077583313 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.18282058835029602, - "train/contrastive_loss": 2.2708425521850586, - "train/negative_loss": 1.8263403177261353, + "train/classification_loss": 0.1900271326303482, + "train/contrastive_loss": 0.42415663599967957, + "train/negative_loss": 0.42415350675582886, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.4445022940635681, - "train/total_loss": 0.6369891166687012 + "train/num_positives": 8, + "train/positive_loss": 3.129250671918271e-06, + "train/total_loss": 0.2748584747314453 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.19021157920360565, - "train/contrastive_loss": 1.9337000846862793, - "train/negative_loss": 1.3600244522094727, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.5736756324768066, - "train/total_loss": 0.5769516229629517 + "train/classification_loss": 0.13769856095314026, + "train/contrastive_loss": 5.143195629119873, + "train/negative_loss": 3.7602338790893555, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 1.382961630821228, + "train/total_loss": 1.1663377285003662 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.16940529644489288, - "train/contrastive_loss": 1.0859380960464478, - "train/negative_loss": 0.6929154396057129, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.3930226266384125, - "train/total_loss": 0.38659292459487915 + "train/classification_loss": 0.2199779897928238, + "train/contrastive_loss": 4.182306289672852, + "train/negative_loss": 2.45135760307312, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 1.730948805809021, + "train/total_loss": 1.0564392805099487 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.21280179917812347, - "train/contrastive_loss": 6.917343616485596, - "train/negative_loss": 6.091588973999023, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.8257545232772827, - "train/total_loss": 1.5962705612182617 + "train/classification_loss": 0.22804193198680878, + "train/contrastive_loss": 2.949676036834717, + "train/negative_loss": 1.135170340538025, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.8145058155059814, + "train/total_loss": 0.8179771900177002 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.2624678313732147, - "train/contrastive_loss": 5.12508487701416, - "train/negative_loss": 2.3681819438934326, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.7569031715393066, - "train/total_loss": 1.287484884262085 + "train/classification_loss": 0.16327829658985138, + "train/contrastive_loss": 2.7236990928649902, + "train/negative_loss": 2.204535484313965, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.5191634893417358, + "train/total_loss": 0.7080181241035461 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.08168107271194458, - "train/contrastive_loss": 1.378607988357544, - "train/negative_loss": 0.9890571236610413, - "train/num_negatives": 42, + "train/classification_loss": 0.2405744194984436, + "train/contrastive_loss": 2.792715072631836, + "train/negative_loss": 0.8956336975097656, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.3895508944988251, - "train/total_loss": 0.3574026823043823 + "train/positive_loss": 1.8970814943313599, + "train/total_loss": 0.7991174459457397 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.11688434332609177, - "train/contrastive_loss": 1.6404467821121216, - "train/negative_loss": 1.6221144199371338, - "train/num_negatives": 48, + "train/classification_loss": 0.19685977697372437, + "train/contrastive_loss": 0.5284190773963928, + "train/negative_loss": 0.5243779420852661, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.018332362174987793, - "train/total_loss": 0.4449737071990967 + "train/positive_loss": 0.0040411632508039474, + "train/total_loss": 0.302543580532074 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.1763734221458435, - "train/contrastive_loss": 2.32942533493042, - "train/negative_loss": 1.3264862298965454, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.002939224243164, - "train/total_loss": 0.6422585248947144 + "train/classification_loss": 0.2039702981710434, + "train/contrastive_loss": 3.3690242767333984, + "train/negative_loss": 2.011583089828491, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.3574413061141968, + "train/total_loss": 0.8777751922607422 + }, + { + "epoch": 14.395437262357415, + "step": 1900, + "train/classification_loss": 0.22212478518486023, + "train/contrastive_loss": 1.4453644752502441, + "train/negative_loss": 1.2268857955932617, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.21847864985466003, + "train/total_loss": 0.5111976861953735 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.06805432587862015, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.06805432587862015 - }, - { - "epoch": 18.095693779904305, - "eval_exact_match_accuracy": 0.3588516746411483, - "eval_hamming_loss": 0.05966788629327329, - "eval_loss": 0.6637274026870728, - "eval_macro_f1": 0.2800415500912601, - "eval_macro_precision": 0.4523007493595729, - "eval_macro_recall": 0.2323639925370029, - "eval_micro_f1": 0.537117903930131, - "eval_micro_precision": 0.7592592592592593, - "eval_micro_recall": 0.4155405405405405, - "eval_runtime": 1.0474, - "eval_samples_per_second": 199.548, - "eval_steps_per_second": 25.779, + "train/classification_loss": 0.10083791613578796, + "train/contrastive_loss": 1.671883225440979, + "train/negative_loss": 1.6491361856460571, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.022747086361050606, + "train/total_loss": 0.4352145791053772 + }, + { + "epoch": 14.395437262357415, + "eval_exact_match_accuracy": 0.2748091603053435, + "eval_hamming_loss": 0.06264032330489448, + "eval_loss": 0.6266875267028809, + "eval_macro_f1": 0.21183228135919496, + "eval_macro_precision": 0.3498425322942244, + "eval_macro_recall": 0.17067757973584496, + "eval_micro_f1": 0.5026737967914439, + "eval_micro_precision": 0.834319526627219, + "eval_micro_recall": 0.3596938775510204, + "eval_runtime": 1.3236, + "eval_samples_per_second": 197.943, + "eval_steps_per_second": 24.932, "step": 1900 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.06403530389070511, - "train/contrastive_loss": 0.2694249451160431, - "train/negative_loss": 0.14247694611549377, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.12694799900054932, - "train/total_loss": 0.11792029440402985 + "train/classification_loss": 0.12609916925430298, + "train/contrastive_loss": 1.6805546283721924, + "train/negative_loss": 1.4588472843170166, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.22170737385749817, + "train/total_loss": 0.462210088968277 }, { - "epoch": 18.095693779904305, + "epoch": 14.395437262357415, "step": 1900, - "train/classification_loss": 0.1095554307103157, - "train/contrastive_loss": 0.2449251413345337, - "train/negative_loss": 0.1431448757648468, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 0.10178027302026749, - "train/total_loss": 0.15854045748710632 + "train/classification_loss": 0.12419541925191879, + "train/contrastive_loss": 0.3176477551460266, + "train/negative_loss": 0.19100196659564972, + "train/num_negatives": 28, + "train/num_positives": 26, + "train/positive_loss": 0.1266457736492157, + "train/total_loss": 0.1877249777317047 }, { - "epoch": 18.574162679425836, - "grad_norm": 5.415369510650635, - "learning_rate": 1.9295038167938933e-05, - "loss": 0.4354, + "epoch": 14.775665399239543, + "grad_norm": 7.7437920570373535, + "learning_rate": 1.9439453717754175e-05, + "loss": 0.4681, "step": 1950 }, { - "epoch": 18.574162679425836, + "epoch": 14.775665399239543, "step": 1950, - "train/classification_loss": 0.09987133741378784, - "train/contrastive_loss": 0.15129117667675018, - "train/negative_loss": 0.15126165747642517, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 2.9516788345063105e-05, - "train/total_loss": 0.13012957572937012 + "train/classification_loss": 0.1337612122297287, + "train/contrastive_loss": 1.0767085552215576, + "train/negative_loss": 0.1463315784931183, + "train/num_negatives": 36, + "train/num_positives": 18, + "train/positive_loss": 0.9303770065307617, + "train/total_loss": 0.3491029143333435 }, { - "epoch": 18.574162679425836, + "epoch": 14.775665399239543, "step": 1950, - "train/classification_loss": 0.0940314456820488, - "train/contrastive_loss": 0.1791023164987564, - "train/negative_loss": 0.17909574508666992, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 6.576401119673392e-06, - "train/total_loss": 0.12985190749168396 + "train/classification_loss": 0.16035369038581848, + "train/contrastive_loss": 1.0763345956802368, + "train/negative_loss": 0.3993370234966278, + "train/num_negatives": 24, + "train/num_positives": 24, + "train/positive_loss": 0.6769976019859314, + "train/total_loss": 0.37562060356140137 }, { - "epoch": 19.047846889952154, - "grad_norm": 3.289793014526367, - "learning_rate": 1.9275954198473285e-05, - "loss": 0.399, + "epoch": 15.152091254752852, + "grad_norm": 15.70482063293457, + "learning_rate": 1.9424279210925646e-05, + "loss": 0.4855, "step": 2000 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.1255495399236679, - "train/contrastive_loss": 1.2353180646896362, - "train/negative_loss": 0.9131842851638794, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.3221338093280792, - "train/total_loss": 0.37261316180229187 + "train/classification_loss": 0.0701872855424881, + "train/contrastive_loss": 0.3416714668273926, + "train/negative_loss": 0.27912577986717224, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.06254567950963974, + "train/total_loss": 0.13852158188819885 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.1792978197336197, - "train/contrastive_loss": 2.5038623809814453, - "train/negative_loss": 1.1844357252120972, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.3194267749786377, - "train/total_loss": 0.6800702810287476 + "train/classification_loss": 0.2204667180776596, + "train/contrastive_loss": 1.417641043663025, + "train/negative_loss": 1.4015660285949707, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.01607504114508629, + "train/total_loss": 0.5039949417114258 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.14922794699668884, - "train/contrastive_loss": 1.62973153591156, - "train/negative_loss": 0.7028886675834656, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.9268428683280945, - "train/total_loss": 0.4751742482185364 + "train/classification_loss": 0.1275085061788559, + "train/contrastive_loss": 1.756780982017517, + "train/negative_loss": 1.579646348953247, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.1771346628665924, + "train/total_loss": 0.47886472940444946 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.1579681634902954, - "train/contrastive_loss": 3.410350799560547, - "train/negative_loss": 1.2774404287338257, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 2.1329104900360107, - "train/total_loss": 0.8400383591651917 + "train/classification_loss": 0.24467767775058746, + "train/contrastive_loss": 2.1059181690216064, + "train/negative_loss": 1.5017712116241455, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.6041468977928162, + "train/total_loss": 0.6658613085746765 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.13095320761203766, - "train/contrastive_loss": 2.8803296089172363, - "train/negative_loss": 0.9476401209831238, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.9326894283294678, - "train/total_loss": 0.7070191502571106 + "train/classification_loss": 0.09588751941919327, + "train/contrastive_loss": 0.6157394051551819, + "train/negative_loss": 0.615626871585846, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.00011255087883910164, + "train/total_loss": 0.21903540194034576 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.14211405813694, - "train/contrastive_loss": 1.7911438941955566, - "train/negative_loss": 0.4722535312175751, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.3188903331756592, - "train/total_loss": 0.500342845916748 + "train/classification_loss": 0.2212655395269394, + "train/contrastive_loss": 1.4973254203796387, + "train/negative_loss": 1.0142096281051636, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.4831157624721527, + "train/total_loss": 0.5207306146621704 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.151325061917305, - "train/contrastive_loss": 2.14174222946167, - "train/negative_loss": 0.8474525809288025, + "train/classification_loss": 0.21601499617099762, + "train/contrastive_loss": 2.2753102779388428, + "train/negative_loss": 1.086225152015686, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 1.2942897081375122, - "train/total_loss": 0.5796735286712646 + "train/num_positives": 8, + "train/positive_loss": 1.1890851259231567, + "train/total_loss": 0.6710770726203918 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.22604325413703918, - "train/contrastive_loss": 0.9500680565834045, - "train/negative_loss": 0.9446334838867188, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.005434592254459858, - "train/total_loss": 0.41605687141418457 + "train/classification_loss": 0.17806853353977203, + "train/contrastive_loss": 2.291654586791992, + "train/negative_loss": 0.98368239402771, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 1.3079723119735718, + "train/total_loss": 0.6363994479179382 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.13013683259487152, - "train/contrastive_loss": 0.9880148768424988, - "train/negative_loss": 0.709029495716095, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.2789853811264038, - "train/total_loss": 0.32773980498313904 + "train/classification_loss": 0.12032925337553024, + "train/contrastive_loss": 0.7517608404159546, + "train/negative_loss": 0.5136995911598206, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.23806126415729523, + "train/total_loss": 0.2706814110279083 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.2185635268688202, - "train/contrastive_loss": 2.581798553466797, - "train/negative_loss": 0.987783670425415, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 1.5940148830413818, - "train/total_loss": 0.734923243522644 + "train/classification_loss": 0.149476557970047, + "train/contrastive_loss": 1.3941333293914795, + "train/negative_loss": 1.3860453367233276, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.008087958209216595, + "train/total_loss": 0.42830324172973633 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.19959506392478943, - "train/contrastive_loss": 2.9170122146606445, - "train/negative_loss": 1.6903659105300903, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 1.2266463041305542, - "train/total_loss": 0.7829974889755249 + "train/classification_loss": 0.16592974960803986, + "train/contrastive_loss": 4.552746295928955, + "train/negative_loss": 4.53888463973999, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.013861631043255329, + "train/total_loss": 1.0764790773391724 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.13289913535118103, - "train/contrastive_loss": 1.538935661315918, - "train/negative_loss": 0.7338822484016418, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.8050534129142761, - "train/total_loss": 0.44068628549575806 + "train/classification_loss": 0.1618216186761856, + "train/contrastive_loss": 2.7983341217041016, + "train/negative_loss": 2.1871378421783447, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.6111963987350464, + "train/total_loss": 0.7214884161949158 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.19674789905548096, - "train/contrastive_loss": 4.154658794403076, - "train/negative_loss": 2.8674323558807373, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.2872265577316284, - "train/total_loss": 1.027679681777954 + "train/classification_loss": 0.18450935184955597, + "train/contrastive_loss": 2.4198522567749023, + "train/negative_loss": 1.3456653356552124, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.0741868019104004, + "train/total_loss": 0.6684798002243042 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.059077080339193344, - "train/contrastive_loss": 0.16576889157295227, - "train/negative_loss": 0.14539194107055664, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.020376943051815033, - "train/total_loss": 0.09223085641860962 + "train/classification_loss": 0.178411602973938, + "train/contrastive_loss": 2.6308062076568604, + "train/negative_loss": 1.4339452981948853, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.196860909461975, + "train/total_loss": 0.704572856426239 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.2080760896205902, - "train/contrastive_loss": 1.2858028411865234, - "train/negative_loss": 0.7147717475891113, + "train/classification_loss": 0.19731207191944122, + "train/contrastive_loss": 3.9429972171783447, + "train/negative_loss": 2.8154184818267822, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.5710310339927673, - "train/total_loss": 0.4652366638183594 + "train/positive_loss": 1.1275787353515625, + "train/total_loss": 0.98591148853302 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.1266353726387024, - "train/contrastive_loss": 2.319145917892456, - "train/negative_loss": 0.9754383563995361, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 1.34370756149292, - "train/total_loss": 0.5904645919799805 + "train/classification_loss": 0.1317308098077774, + "train/contrastive_loss": 0.5705164074897766, + "train/negative_loss": 0.570435643196106, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 8.078908285824582e-05, + "train/total_loss": 0.245834082365036 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.22007127106189728, - "train/contrastive_loss": 3.932006359100342, - "train/negative_loss": 1.88003671169281, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 2.051969528198242, - "train/total_loss": 1.0064725875854492 + "train/classification_loss": 0.16273875534534454, + "train/contrastive_loss": 1.830619215965271, + "train/negative_loss": 1.0599654912948608, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.7706537246704102, + "train/total_loss": 0.5288625955581665 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.2293570637702942, - "train/contrastive_loss": 1.7917921543121338, - "train/negative_loss": 0.6668974161148071, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.1248947381973267, - "train/total_loss": 0.5877155065536499 + "train/classification_loss": 0.17538650333881378, + "train/contrastive_loss": 3.3942928314208984, + "train/negative_loss": 2.1493260860443115, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.244966745376587, + "train/total_loss": 0.8542450666427612 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.198373481631279, - "train/contrastive_loss": 2.9558897018432617, - "train/negative_loss": 1.8758512735366821, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.08003830909729, - "train/total_loss": 0.7895514369010925 + "train/classification_loss": 0.2501019239425659, + "train/contrastive_loss": 3.7514586448669434, + "train/negative_loss": 3.004185199737549, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.747273325920105, + "train/total_loss": 1.0003936290740967 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.1815413385629654, - "train/contrastive_loss": 1.6114490032196045, - "train/negative_loss": 0.9843201637268066, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.6271288990974426, - "train/total_loss": 0.503831148147583 + "train/classification_loss": 0.2371411919593811, + "train/contrastive_loss": 3.2503271102905273, + "train/negative_loss": 1.5636756420135498, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.686651587486267, + "train/total_loss": 0.8872066140174866 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.18509411811828613, - "train/contrastive_loss": 1.3037470579147339, - "train/negative_loss": 0.860169529914856, - "train/num_negatives": 44, + "train/classification_loss": 0.21357402205467224, + "train/contrastive_loss": 1.1567103862762451, + "train/negative_loss": 1.0273030996322632, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.12940728664398193, + "train/total_loss": 0.44491609930992126 + }, + { + "epoch": 15.152091254752852, + "step": 2000, + "train/classification_loss": 0.2665392756462097, + "train/contrastive_loss": 2.3554248809814453, + "train/negative_loss": 0.6516381502151489, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.44357749819755554, - "train/total_loss": 0.44584354758262634 + "train/positive_loss": 1.703786849975586, + "train/total_loss": 0.7376242876052856 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.19148804247379303, - "train/contrastive_loss": 5.882230281829834, - "train/negative_loss": 4.948007106781006, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.9342233538627625, - "train/total_loss": 1.3679341077804565 + "train/classification_loss": 0.21713358163833618, + "train/contrastive_loss": 1.5772277116775513, + "train/negative_loss": 1.3599326610565186, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.21729503571987152, + "train/total_loss": 0.5325791239738464 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.20950788259506226, - "train/contrastive_loss": 3.4437789916992188, - "train/negative_loss": 1.4027955532073975, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.0409834384918213, - "train/total_loss": 0.898263692855835 + "train/classification_loss": 0.2059563547372818, + "train/contrastive_loss": 0.6350311040878296, + "train/negative_loss": 0.6350284218788147, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 2.6822137897397624e-06, + "train/total_loss": 0.3329625725746155 + }, + { + "epoch": 15.152091254752852, + "step": 2000, + "train/classification_loss": 0.1646677851676941, + "train/contrastive_loss": 6.983513832092285, + "train/negative_loss": 5.618288516998291, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 1.3652251958847046, + "train/total_loss": 1.561370611190796 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.07107966393232346, - "train/contrastive_loss": 1.565699815750122, - "train/negative_loss": 1.2137773036956787, + "train/classification_loss": 0.24211613833904266, + "train/contrastive_loss": 3.4684481620788574, + "train/negative_loss": 2.157989978790283, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.35192248225212097, - "train/total_loss": 0.3842196464538574 + "train/positive_loss": 1.3104580640792847, + "train/total_loss": 0.9358057975769043 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.12522931396961212, - "train/contrastive_loss": 1.3406131267547607, - "train/negative_loss": 1.18857741355896, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.15203575789928436, - "train/total_loss": 0.3933519721031189 + "train/classification_loss": 0.18394196033477783, + "train/contrastive_loss": 2.5449228286743164, + "train/negative_loss": 1.1443510055541992, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.4005719423294067, + "train/total_loss": 0.6929265260696411 + }, + { + "epoch": 15.152091254752852, + "step": 2000, + "train/classification_loss": 0.16763746738433838, + "train/contrastive_loss": 2.7025187015533447, + "train/negative_loss": 2.136430501937866, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.5660881996154785, + "train/total_loss": 0.7081412076950073 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.16623488068580627, - "train/contrastive_loss": 2.130695104598999, - "train/negative_loss": 1.2392009496688843, + "train/classification_loss": 0.2382870316505432, + "train/contrastive_loss": 2.0244216918945312, + "train/negative_loss": 0.6599618196487427, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 1.3644598722457886, + "train/total_loss": 0.6431713700294495 + }, + { + "epoch": 15.152091254752852, + "step": 2000, + "train/classification_loss": 0.20026318728923798, + "train/contrastive_loss": 0.7749852538108826, + "train/negative_loss": 0.7744566798210144, "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.89149409532547, - "train/total_loss": 0.5923739075660706 + "train/num_positives": 8, + "train/positive_loss": 0.0005285913357511163, + "train/total_loss": 0.3552602529525757 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.030721938237547874, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.030721938237547874 - }, - { - "epoch": 19.047846889952154, - "eval_exact_match_accuracy": 0.37320574162679426, - "eval_hamming_loss": 0.05826062482409232, - "eval_loss": 0.612683117389679, - "eval_macro_f1": 0.28855635083072373, - "eval_macro_precision": 0.4495679816345608, - "eval_macro_recall": 0.23682501402224587, - "eval_micro_f1": 0.5470459518599562, - "eval_micro_precision": 0.7763975155279503, - "eval_micro_recall": 0.4222972972972973, - "eval_runtime": 1.061, - "eval_samples_per_second": 196.991, - "eval_steps_per_second": 25.449, - "step": 2000 + "train/classification_loss": 0.2075064778327942, + "train/contrastive_loss": 1.5055947303771973, + "train/negative_loss": 0.9475614428520203, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.5580332279205322, + "train/total_loss": 0.5086253881454468 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.10624876618385315, - "train/contrastive_loss": 0.5208284854888916, - "train/negative_loss": 0.5168348550796509, - "train/num_negatives": 52, - "train/num_positives": 4, - "train/positive_loss": 0.003993641119450331, - "train/total_loss": 0.21041446924209595 + "train/classification_loss": 0.19756203889846802, + "train/contrastive_loss": 1.0618505477905273, + "train/negative_loss": 1.055565595626831, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.006285009440034628, + "train/total_loss": 0.40993213653564453 + }, + { + "epoch": 15.152091254752852, + "step": 2000, + "train/classification_loss": 0.1058526337146759, + "train/contrastive_loss": 1.4905683994293213, + "train/negative_loss": 1.4896771907806396, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.0008911883924156427, + "train/total_loss": 0.4039663076400757 + }, + { + "epoch": 15.152091254752852, + "eval_exact_match_accuracy": 0.30916030534351147, + "eval_hamming_loss": 0.061517736865738665, + "eval_loss": 0.6220834255218506, + "eval_macro_f1": 0.2529952669767114, + "eval_macro_precision": 0.36244905027972046, + "eval_macro_recall": 0.2121604656713523, + "eval_micro_f1": 0.535593220338983, + "eval_micro_precision": 0.797979797979798, + "eval_micro_recall": 0.4030612244897959, + "eval_runtime": 1.3227, + "eval_samples_per_second": 198.074, + "eval_steps_per_second": 24.948, + "step": 2000 }, { - "epoch": 19.047846889952154, + "epoch": 15.152091254752852, "step": 2000, - "train/classification_loss": 0.07537344843149185, - "train/contrastive_loss": 0.10696140676736832, - "train/negative_loss": 0.10384932160377502, + "train/classification_loss": 0.10543461889028549, + "train/contrastive_loss": 0.5911973714828491, + "train/negative_loss": 0.36928582191467285, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.0031120816711336374, - "train/total_loss": 0.09676572680473328 + "train/positive_loss": 0.22191157937049866, + "train/total_loss": 0.22367408871650696 + }, + { + "epoch": 15.152091254752852, + "step": 2000, + "train/classification_loss": 0.0895552784204483, + "train/contrastive_loss": 0.43293771147727966, + "train/negative_loss": 0.43284404277801514, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 9.366181620862335e-05, + "train/total_loss": 0.17614281177520752 }, { - "epoch": 19.526315789473685, - "grad_norm": 4.325128078460693, - "learning_rate": 1.9256870229007636e-05, - "loss": 0.3909, + "epoch": 15.532319391634982, + "grad_norm": 6.463346481323242, + "learning_rate": 1.9409104704097118e-05, + "loss": 0.4389, "step": 2050 }, { - "epoch": 19.526315789473685, + "epoch": 15.532319391634982, "step": 2050, - "train/classification_loss": 0.11700613051652908, - "train/contrastive_loss": 0.209782212972641, - "train/negative_loss": 0.14757585525512695, - "train/num_negatives": 44, - "train/num_positives": 8, - "train/positive_loss": 0.06220636144280434, - "train/total_loss": 0.15896257758140564 + "train/classification_loss": 0.11535491049289703, + "train/contrastive_loss": 0.3327261805534363, + "train/negative_loss": 0.14566968381404877, + "train/num_negatives": 28, + "train/num_positives": 24, + "train/positive_loss": 0.1870565116405487, + "train/total_loss": 0.18190014362335205 }, { - "epoch": 19.526315789473685, + "epoch": 15.532319391634982, "step": 2050, - "train/classification_loss": 0.07519859820604324, - "train/contrastive_loss": 0.2604178488254547, - "train/negative_loss": 0.2604111433029175, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 6.69561404720298e-06, - "train/total_loss": 0.12728217244148254 + "train/classification_loss": 0.0955490842461586, + "train/contrastive_loss": 0.19425924122333527, + "train/negative_loss": 0.1686285436153412, + "train/num_negatives": 32, + "train/num_positives": 22, + "train/positive_loss": 0.025630703195929527, + "train/total_loss": 0.13440093398094177 }, { - "epoch": 20.0, - "grad_norm": 8.813825607299805, - "learning_rate": 1.9237786259541987e-05, - "loss": 0.3888, + "epoch": 15.912547528517111, + "grad_norm": 9.845856666564941, + "learning_rate": 1.9393930197268593e-05, + "loss": 0.4724, "step": 2100 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, + "step": 2100, + "train/classification_loss": 0.05955767631530762, + "train/contrastive_loss": 0.38289347290992737, + "train/negative_loss": 0.36766818165779114, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.015225290320813656, + "train/total_loss": 0.13613638281822205 + }, + { + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.12741154432296753, - "train/contrastive_loss": 1.368431806564331, - "train/negative_loss": 0.8387002348899841, + "train/classification_loss": 0.21058617532253265, + "train/contrastive_loss": 1.4797617197036743, + "train/negative_loss": 1.4778345823287964, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.5297315120697021, - "train/total_loss": 0.4010979235172272 + "train/num_positives": 6, + "train/positive_loss": 0.001927086035721004, + "train/total_loss": 0.5065385103225708 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.16757048666477203, - "train/contrastive_loss": 1.9360377788543701, - "train/negative_loss": 1.0179245471954346, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.9181132912635803, - "train/total_loss": 0.5547780394554138 + "train/classification_loss": 0.11189567297697067, + "train/contrastive_loss": 3.0799148082733154, + "train/negative_loss": 2.753957748413086, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.3259570300579071, + "train/total_loss": 0.7278786301612854 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.1167624220252037, - "train/contrastive_loss": 0.6678293347358704, - "train/negative_loss": 0.5209057927131653, - "train/num_negatives": 44, + "train/classification_loss": 0.24435198307037354, + "train/contrastive_loss": 1.8478825092315674, + "train/negative_loss": 1.110702633857727, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.7371799349784851, + "train/total_loss": 0.613928496837616 + }, + { + "epoch": 15.912547528517111, + "step": 2100, + "train/classification_loss": 0.12918618321418762, + "train/contrastive_loss": 1.5327796936035156, + "train/negative_loss": 1.4445537328720093, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.08822593837976456, + "train/total_loss": 0.4357421398162842 + }, + { + "epoch": 15.912547528517111, + "step": 2100, + "train/classification_loss": 0.2131003439426422, + "train/contrastive_loss": 1.4814600944519043, + "train/negative_loss": 0.9198795557022095, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.14692354202270508, - "train/total_loss": 0.25032830238342285 + "train/positive_loss": 0.5615805387496948, + "train/total_loss": 0.5093923807144165 + }, + { + "epoch": 15.912547528517111, + "step": 2100, + "train/classification_loss": 0.2220860719680786, + "train/contrastive_loss": 3.0596532821655273, + "train/negative_loss": 1.7154020071029663, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.344251275062561, + "train/total_loss": 0.834016740322113 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.17067407071590424, - "train/contrastive_loss": 2.861807346343994, - "train/negative_loss": 1.382408857345581, + "train/classification_loss": 0.1607942283153534, + "train/contrastive_loss": 1.0233904123306274, + "train/negative_loss": 0.9503950476646423, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.4793986082077026, - "train/total_loss": 0.7430355548858643 + "train/positive_loss": 0.07299536466598511, + "train/total_loss": 0.36547231674194336 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.1375819444656372, - "train/contrastive_loss": 3.569960594177246, - "train/negative_loss": 1.4642633199691772, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 2.1056973934173584, - "train/total_loss": 0.8515740633010864 + "train/classification_loss": 0.13154079020023346, + "train/contrastive_loss": 1.463498592376709, + "train/negative_loss": 0.8615715503692627, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.6019271016120911, + "train/total_loss": 0.4242405295372009 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.13350629806518555, - "train/contrastive_loss": 1.6272976398468018, - "train/negative_loss": 0.5489711761474609, - "train/num_negatives": 32, + "train/classification_loss": 0.17926768958568573, + "train/contrastive_loss": 2.587658166885376, + "train/negative_loss": 2.1226646900177, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.46499353647232056, + "train/total_loss": 0.6967993378639221 + }, + { + "epoch": 15.912547528517111, + "step": 2100, + "train/classification_loss": 0.17781037092208862, + "train/contrastive_loss": 3.995781421661377, + "train/negative_loss": 3.938577175140381, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.0783264636993408, - "train/total_loss": 0.45896583795547485 + "train/positive_loss": 0.057204265147447586, + "train/total_loss": 0.9769666790962219 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.13811790943145752, - "train/contrastive_loss": 1.5580633878707886, - "train/negative_loss": 0.8085144758224487, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.7495489120483398, - "train/total_loss": 0.44973060488700867 + "train/classification_loss": 0.14956338703632355, + "train/contrastive_loss": 2.877876043319702, + "train/negative_loss": 2.4293456077575684, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.44853049516677856, + "train/total_loss": 0.7251386046409607 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.24079735577106476, - "train/contrastive_loss": 0.7057647109031677, - "train/negative_loss": 0.7009286284446716, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.004836088977754116, - "train/total_loss": 0.381950318813324 + "train/classification_loss": 0.217518612742424, + "train/contrastive_loss": 2.2334628105163574, + "train/negative_loss": 1.1864458322525024, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.0470170974731445, + "train/total_loss": 0.6642111539840698 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.13373251259326935, - "train/contrastive_loss": 1.3215610980987549, - "train/negative_loss": 0.7234039306640625, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.5981571674346924, - "train/total_loss": 0.39804476499557495 + "train/classification_loss": 0.17286691069602966, + "train/contrastive_loss": 3.322479248046875, + "train/negative_loss": 1.0818465948104858, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.2406327724456787, + "train/total_loss": 0.8373627662658691 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.2448260635137558, - "train/contrastive_loss": 3.737464427947998, - "train/negative_loss": 0.9323051571846008, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.805159330368042, - "train/total_loss": 0.9923189878463745 + "train/classification_loss": 0.1977982223033905, + "train/contrastive_loss": 2.9000954627990723, + "train/negative_loss": 1.9652061462402344, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.9348894357681274, + "train/total_loss": 0.7778173685073853 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.22938589751720428, - "train/contrastive_loss": 2.599924325942993, - "train/negative_loss": 1.7884129285812378, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.8115113973617554, - "train/total_loss": 0.749370813369751 + "train/classification_loss": 0.13611455261707306, + "train/contrastive_loss": 0.4448603093624115, + "train/negative_loss": 0.4448261857032776, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 3.4134613088099286e-05, + "train/total_loss": 0.22508661448955536 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.13538114726543427, - "train/contrastive_loss": 1.6560474634170532, - "train/negative_loss": 0.5799732208251953, + "train/classification_loss": 0.1490538865327835, + "train/contrastive_loss": 1.359757900238037, + "train/negative_loss": 1.3388371467590332, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 1.076074242591858, - "train/total_loss": 0.46659064292907715 + "train/positive_loss": 0.020920734852552414, + "train/total_loss": 0.4210054874420166 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.20184487104415894, - "train/contrastive_loss": 3.4578235149383545, - "train/negative_loss": 2.3472676277160645, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.11055588722229, - "train/total_loss": 0.8934096097946167 + "train/classification_loss": 0.16416874527931213, + "train/contrastive_loss": 3.706799268722534, + "train/negative_loss": 1.9022960662841797, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.8045032024383545, + "train/total_loss": 0.9055286645889282 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.06134726107120514, - "train/contrastive_loss": 0.1660657525062561, - "train/negative_loss": 0.14378894865512848, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.022276800125837326, - "train/total_loss": 0.0945604145526886 + "train/classification_loss": 0.2685341536998749, + "train/contrastive_loss": 3.4862382411956787, + "train/negative_loss": 2.7996833324432373, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.6865549683570862, + "train/total_loss": 0.9657818078994751 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.19925084710121155, - "train/contrastive_loss": 1.1920158863067627, - "train/negative_loss": 0.6894676685333252, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.5025482177734375, - "train/total_loss": 0.4376540184020996 + "train/classification_loss": 0.23766851425170898, + "train/contrastive_loss": 2.083160161972046, + "train/negative_loss": 1.3453514575958252, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.7378087043762207, + "train/total_loss": 0.6543005704879761 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.12260229140520096, - "train/contrastive_loss": 2.3667943477630615, - "train/negative_loss": 0.9732067584991455, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 1.393587589263916, - "train/total_loss": 0.5959611535072327 + "train/classification_loss": 0.20962558686733246, + "train/contrastive_loss": 1.429269552230835, + "train/negative_loss": 0.9721832275390625, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.45708638429641724, + "train/total_loss": 0.4954795241355896 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.26682543754577637, - "train/contrastive_loss": 3.5168070793151855, - "train/negative_loss": 1.92378830909729, - "train/num_negatives": 46, + "train/classification_loss": 0.2354494035243988, + "train/contrastive_loss": 1.632279396057129, + "train/negative_loss": 0.3893807828426361, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.593018889427185, - "train/total_loss": 0.9701868891716003 + "train/positive_loss": 1.2428985834121704, + "train/total_loss": 0.5619052648544312 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.2701972424983978, - "train/contrastive_loss": 1.7366790771484375, - "train/negative_loss": 0.9631988406181335, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.7734801769256592, - "train/total_loss": 0.6175330877304077 + "train/classification_loss": 0.23342491686344147, + "train/contrastive_loss": 2.0714824199676514, + "train/negative_loss": 1.0788813829421997, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.9926010966300964, + "train/total_loss": 0.6477214097976685 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.2000785768032074, - "train/contrastive_loss": 3.0332813262939453, - "train/negative_loss": 1.7841318845748901, + "train/classification_loss": 0.21403159201145172, + "train/contrastive_loss": 1.4382604360580444, + "train/negative_loss": 1.4382487535476685, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.2491495609283447, - "train/total_loss": 0.8067348003387451 + "train/num_positives": 8, + "train/positive_loss": 1.165282264992129e-05, + "train/total_loss": 0.5016837120056152 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.16905641555786133, - "train/contrastive_loss": 1.3817839622497559, - "train/negative_loss": 1.0096195936203003, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.37216439843177795, - "train/total_loss": 0.445413202047348 + "train/classification_loss": 0.14449208974838257, + "train/contrastive_loss": 5.596737861633301, + "train/negative_loss": 4.631580352783203, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.9651573896408081, + "train/total_loss": 1.2638397216796875 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.1840933859348297, - "train/contrastive_loss": 1.2156600952148438, - "train/negative_loss": 0.8469054698944092, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.3687545955181122, - "train/total_loss": 0.42722541093826294 + "train/classification_loss": 0.24022048711776733, + "train/contrastive_loss": 3.789987564086914, + "train/negative_loss": 2.4678375720977783, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 1.3221501111984253, + "train/total_loss": 0.9982179999351501 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.1968679279088974, - "train/contrastive_loss": 5.169436931610107, - "train/negative_loss": 4.083937168121338, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.085499882698059, - "train/total_loss": 1.23075532913208 + "train/classification_loss": 0.19758810102939606, + "train/contrastive_loss": 1.9545950889587402, + "train/negative_loss": 1.0959234237670898, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.8586716055870056, + "train/total_loss": 0.5885071158409119 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.2574560344219208, - "train/contrastive_loss": 5.069339752197266, - "train/negative_loss": 2.4615728855133057, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.60776686668396, - "train/total_loss": 1.2713240385055542 + "train/classification_loss": 0.16642902791500092, + "train/contrastive_loss": 1.6492385864257812, + "train/negative_loss": 1.3912630081176758, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.25797563791275024, + "train/total_loss": 0.49627673625946045 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.06536085903644562, - "train/contrastive_loss": 1.4914284944534302, - "train/negative_loss": 0.49872663617134094, - "train/num_negatives": 42, + "train/classification_loss": 0.25256168842315674, + "train/contrastive_loss": 2.732414484024048, + "train/negative_loss": 0.8212307095527649, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.9927018284797668, - "train/total_loss": 0.36364656686782837 + "train/positive_loss": 1.9111838340759277, + "train/total_loss": 0.7990446090698242 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.11234592646360397, - "train/contrastive_loss": 1.7103300094604492, - "train/negative_loss": 1.6838805675506592, - "train/num_negatives": 48, + "train/classification_loss": 0.19866162538528442, + "train/contrastive_loss": 0.3874627351760864, + "train/negative_loss": 0.3871746063232422, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.026449397206306458, - "train/total_loss": 0.45441195368766785 + "train/positive_loss": 0.00028813129756599665, + "train/total_loss": 0.27615416049957275 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.18249927461147308, - "train/contrastive_loss": 1.6949949264526367, - "train/negative_loss": 0.9129961729049683, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.7819988131523132, - "train/total_loss": 0.5214982628822327 + "train/classification_loss": 0.20212675631046295, + "train/contrastive_loss": 3.4701485633850098, + "train/negative_loss": 2.7435567378997803, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.7265918850898743, + "train/total_loss": 0.8961564898490906 + }, + { + "epoch": 15.912547528517111, + "step": 2100, + "train/classification_loss": 0.1843351125717163, + "train/contrastive_loss": 0.7982025146484375, + "train/negative_loss": 0.7975988984107971, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.0006036045379005373, + "train/total_loss": 0.34397560358047485 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.0403926782310009, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.0403926782310009 - }, - { - "epoch": 20.0, - "eval_exact_match_accuracy": 0.3397129186602871, - "eval_hamming_loss": 0.06079369546861807, - "eval_loss": 0.6060535907745361, - "eval_macro_f1": 0.2681587020663775, - "eval_macro_precision": 0.3523009368654857, - "eval_macro_recall": 0.23161436410571357, - "eval_micro_f1": 0.5344827586206896, - "eval_micro_precision": 0.7380952380952381, - "eval_micro_recall": 0.4189189189189189, - "eval_runtime": 1.0689, - "eval_samples_per_second": 195.524, - "eval_steps_per_second": 25.259, + "train/classification_loss": 0.1035948395729065, + "train/contrastive_loss": 1.2339072227478027, + "train/negative_loss": 1.222368836402893, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.011538418009877205, + "train/total_loss": 0.35037630796432495 + }, + { + "epoch": 15.912547528517111, + "eval_exact_match_accuracy": 0.29770992366412213, + "eval_hamming_loss": 0.061068702290076333, + "eval_loss": 0.6270256638526917, + "eval_macro_f1": 0.24698850252472024, + "eval_macro_precision": 0.34211122176174386, + "eval_macro_recall": 0.2142507597857037, + "eval_micro_f1": 0.5526315789473685, + "eval_micro_precision": 0.7777777777777778, + "eval_micro_recall": 0.42857142857142855, + "eval_runtime": 1.3187, + "eval_samples_per_second": 198.678, + "eval_steps_per_second": 25.024, "step": 2100 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.07597373425960541, - "train/contrastive_loss": 0.4649398922920227, - "train/negative_loss": 0.4648740291595459, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 6.585595838259906e-05, - "train/total_loss": 0.16896170377731323 + "train/classification_loss": 0.09331023693084717, + "train/contrastive_loss": 0.6021571755409241, + "train/negative_loss": 0.5833436846733093, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.018813518807291985, + "train/total_loss": 0.21374167501926422 }, { - "epoch": 20.0, + "epoch": 15.912547528517111, "step": 2100, - "train/classification_loss": 0.0714992880821228, - "train/contrastive_loss": 0.12948665022850037, - "train/negative_loss": 0.11443259567022324, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.01505405642092228, - "train/total_loss": 0.097396619617939 + "train/classification_loss": 0.1134042739868164, + "train/contrastive_loss": 0.35022836923599243, + "train/negative_loss": 0.3412408232688904, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.008987538516521454, + "train/total_loss": 0.18344995379447937 }, { - "epoch": 20.47846889952153, - "grad_norm": 5.691018104553223, - "learning_rate": 1.9218702290076338e-05, - "loss": 0.4242, + "epoch": 16.288973384030417, + "grad_norm": 2.0515518188476562, + "learning_rate": 1.9378755690440064e-05, + "loss": 0.4677, "step": 2150 }, { - "epoch": 20.47846889952153, + "epoch": 16.288973384030417, "step": 2150, - "train/classification_loss": 0.04841278865933418, - "train/contrastive_loss": 0.14306855201721191, - "train/negative_loss": 0.14305877685546875, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 9.775225407793187e-06, - "train/total_loss": 0.07702650129795074 + "train/classification_loss": 0.15665249526500702, + "train/contrastive_loss": 0.7897652983665466, + "train/negative_loss": 0.6750941276550293, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.11467118561267853, + "train/total_loss": 0.31460556387901306 }, { - "epoch": 20.47846889952153, + "epoch": 16.288973384030417, "step": 2150, - "train/classification_loss": 0.08810149133205414, - "train/contrastive_loss": 0.3284183442592621, - "train/negative_loss": 0.327949583530426, + "train/classification_loss": 0.08651706576347351, + "train/contrastive_loss": 0.2877170741558075, + "train/negative_loss": 0.12036720663309097, "train/num_negatives": 36, - "train/num_positives": 16, - "train/positive_loss": 0.0004687635228037834, - "train/total_loss": 0.15378516912460327 + "train/num_positives": 20, + "train/positive_loss": 0.16734986007213593, + "train/total_loss": 0.14406047761440277 }, { - "epoch": 20.95693779904306, - "grad_norm": 3.2255311012268066, - "learning_rate": 1.919961832061069e-05, - "loss": 0.3246, + "epoch": 16.669201520912548, + "grad_norm": 7.582057476043701, + "learning_rate": 1.9363884673748105e-05, + "loss": 0.4243, "step": 2200 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.13607454299926758, - "train/contrastive_loss": 1.53801429271698, - "train/negative_loss": 1.0994112491607666, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.4386030435562134, - "train/total_loss": 0.4436773955821991 + "train/classification_loss": 0.06304914504289627, + "train/contrastive_loss": 0.4401205778121948, + "train/negative_loss": 0.30185261368751526, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.13826796412467957, + "train/total_loss": 0.15107326209545135 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.16833680868148804, - "train/contrastive_loss": 2.5009822845458984, - "train/negative_loss": 1.0988625288009644, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.4021196365356445, - "train/total_loss": 0.6685332655906677 + "train/classification_loss": 0.21518191695213318, + "train/contrastive_loss": 1.450197696685791, + "train/negative_loss": 1.4500049352645874, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.00019271441851742566, + "train/total_loss": 0.5052214860916138 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.11516279727220535, - "train/contrastive_loss": 0.8062045574188232, - "train/negative_loss": 0.6328814625740051, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.17332309484481812, - "train/total_loss": 0.2764037251472473 + "train/classification_loss": 0.09896998107433319, + "train/contrastive_loss": 1.1896674633026123, + "train/negative_loss": 1.0190216302871704, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.17064586281776428, + "train/total_loss": 0.33690348267555237 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.17051047086715698, - "train/contrastive_loss": 4.238770008087158, - "train/negative_loss": 2.0653131008148193, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 2.173456907272339, - "train/total_loss": 1.0182645320892334 + "train/classification_loss": 0.24951279163360596, + "train/contrastive_loss": 2.230431079864502, + "train/negative_loss": 1.424292802810669, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.8061383366584778, + "train/total_loss": 0.6955990195274353 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.11551028490066528, - "train/contrastive_loss": 2.902541399002075, - "train/negative_loss": 1.1680169105529785, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.7345244884490967, - "train/total_loss": 0.6960185766220093 + "train/classification_loss": 0.11221382021903992, + "train/contrastive_loss": 1.2096120119094849, + "train/negative_loss": 1.1709370613098145, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.038674935698509216, + "train/total_loss": 0.35413622856140137 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.12542724609375, - "train/contrastive_loss": 1.7968759536743164, - "train/negative_loss": 0.4631419777870178, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.3337340354919434, - "train/total_loss": 0.4848024547100067 + "train/classification_loss": 0.21357405185699463, + "train/contrastive_loss": 1.59758460521698, + "train/negative_loss": 1.2045930624008179, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.3929915428161621, + "train/total_loss": 0.5330909490585327 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.13145045936107635, - "train/contrastive_loss": 0.8915443420410156, - "train/negative_loss": 0.7334885597229004, + "train/classification_loss": 0.21710538864135742, + "train/contrastive_loss": 2.3491365909576416, + "train/negative_loss": 1.2755091190338135, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.15805581212043762, - "train/total_loss": 0.30975931882858276 + "train/num_positives": 8, + "train/positive_loss": 1.0736274719238281, + "train/total_loss": 0.6869326829910278 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.22389067709445953, - "train/contrastive_loss": 1.0008851289749146, - "train/negative_loss": 1.0007643699645996, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.00012077945575583726, - "train/total_loss": 0.4240677058696747 + "train/classification_loss": 0.17245590686798096, + "train/contrastive_loss": 0.8999598622322083, + "train/negative_loss": 0.88002610206604, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.019933752715587616, + "train/total_loss": 0.35244786739349365 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.13976667821407318, - "train/contrastive_loss": 1.3182792663574219, - "train/negative_loss": 0.7280686497688293, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.5902106761932373, - "train/total_loss": 0.4034225344657898 + "train/classification_loss": 0.1374269425868988, + "train/contrastive_loss": 1.1133636236190796, + "train/negative_loss": 0.3809535801410675, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.7324100732803345, + "train/total_loss": 0.3600996732711792 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.23203957080841064, - "train/contrastive_loss": 3.3997745513916016, - "train/negative_loss": 1.0621532201766968, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.3376212120056152, - "train/total_loss": 0.9119945168495178 + "train/classification_loss": 0.17397893965244293, + "train/contrastive_loss": 2.587181568145752, + "train/negative_loss": 2.3492043018341064, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.23797717690467834, + "train/total_loss": 0.6914152503013611 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.21034255623817444, - "train/contrastive_loss": 2.9265847206115723, - "train/negative_loss": 1.7979613542556763, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 1.128623366355896, - "train/total_loss": 0.7956595420837402 + "train/classification_loss": 0.17513199150562286, + "train/contrastive_loss": 4.163281440734863, + "train/negative_loss": 4.094315052032471, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.06896619498729706, + "train/total_loss": 1.0077883005142212 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.1353205293416977, - "train/contrastive_loss": 1.9040862321853638, - "train/negative_loss": 1.0244134664535522, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.8796727657318115, - "train/total_loss": 0.5161377787590027 + "train/classification_loss": 0.14343780279159546, + "train/contrastive_loss": 2.078705072402954, + "train/negative_loss": 1.284332036972046, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.7943730354309082, + "train/total_loss": 0.5591788291931152 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.1965874582529068, - "train/contrastive_loss": 3.504218101501465, - "train/negative_loss": 2.585940361022949, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.9182776212692261, - "train/total_loss": 0.8974310755729675 + "train/classification_loss": 0.20816941559314728, + "train/contrastive_loss": 2.2095859050750732, + "train/negative_loss": 1.2686837911605835, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.9409021139144897, + "train/total_loss": 0.6500865817070007 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.053256306797266006, - "train/contrastive_loss": 0.4460504353046417, - "train/negative_loss": 0.4415407180786133, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.004509711172431707, - "train/total_loss": 0.14246639609336853 + "train/classification_loss": 0.18374207615852356, + "train/contrastive_loss": 4.006612777709961, + "train/negative_loss": 1.8609925508499146, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.145620107650757, + "train/total_loss": 0.9850646257400513 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.21066394448280334, - "train/contrastive_loss": 1.5733082294464111, - "train/negative_loss": 0.8320735692977905, + "train/classification_loss": 0.17177166044712067, + "train/contrastive_loss": 2.3939993381500244, + "train/negative_loss": 2.3815205097198486, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.7412347197532654, - "train/total_loss": 0.52532559633255 + "train/positive_loss": 0.01247886847704649, + "train/total_loss": 0.6505715250968933 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.10031905770301819, - "train/contrastive_loss": 2.042243719100952, - "train/negative_loss": 1.239901065826416, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.8023425936698914, - "train/total_loss": 0.50876784324646 + "train/classification_loss": 0.12874162197113037, + "train/contrastive_loss": 0.7990466356277466, + "train/negative_loss": 0.7987974882125854, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0002491234918124974, + "train/total_loss": 0.2885509729385376 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.2529829442501068, - "train/contrastive_loss": 4.592385768890381, - "train/negative_loss": 2.291137456893921, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 2.30124831199646, - "train/total_loss": 1.1714601516723633 + "train/classification_loss": 0.15942957997322083, + "train/contrastive_loss": 1.2781269550323486, + "train/negative_loss": 0.9842009544372559, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.29392603039741516, + "train/total_loss": 0.41505497694015503 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.24250759184360504, - "train/contrastive_loss": 2.8234407901763916, - "train/negative_loss": 1.4827384948730469, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.3407022953033447, - "train/total_loss": 0.8071957230567932 + "train/classification_loss": 0.1687476933002472, + "train/contrastive_loss": 3.4054675102233887, + "train/negative_loss": 2.1807448863983154, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.2247226238250732, + "train/total_loss": 0.8498412370681763 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.20271144807338715, - "train/contrastive_loss": 3.5783257484436035, - "train/negative_loss": 1.7796132564544678, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.7987124919891357, - "train/total_loss": 0.918376624584198 + "train/classification_loss": 0.24414440989494324, + "train/contrastive_loss": 3.551377534866333, + "train/negative_loss": 2.230901002883911, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.3204765319824219, + "train/total_loss": 0.9544199705123901 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.16951385140419006, - "train/contrastive_loss": 1.5057713985443115, - "train/negative_loss": 0.9644584655761719, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.5413128733634949, - "train/total_loss": 0.47066813707351685 + "train/classification_loss": 0.23190227150917053, + "train/contrastive_loss": 2.5864784717559814, + "train/negative_loss": 1.3677351474761963, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.2187433242797852, + "train/total_loss": 0.7491979598999023 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.1583126038312912, - "train/contrastive_loss": 1.1766419410705566, - "train/negative_loss": 0.635985791683197, - "train/num_negatives": 44, + "train/classification_loss": 0.2266465425491333, + "train/contrastive_loss": 1.7696895599365234, + "train/negative_loss": 0.6427833437919617, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 1.1269062757492065, + "train/total_loss": 0.5805844664573669 + }, + { + "epoch": 16.669201520912548, + "step": 2200, + "train/classification_loss": 0.23799028992652893, + "train/contrastive_loss": 2.699742317199707, + "train/negative_loss": 0.8483810424804688, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.5406560897827148, - "train/total_loss": 0.39364099502563477 + "train/positive_loss": 1.8513611555099487, + "train/total_loss": 0.777938723564148 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.20440343022346497, - "train/contrastive_loss": 6.158827304840088, - "train/negative_loss": 4.91609525680542, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.2427319288253784, - "train/total_loss": 1.436168909072876 + "train/classification_loss": 0.21879571676254272, + "train/contrastive_loss": 1.6723886728286743, + "train/negative_loss": 1.370291829109192, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.30209681391716003, + "train/total_loss": 0.5532734394073486 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.22479590773582458, - "train/contrastive_loss": 4.118882179260254, - "train/negative_loss": 1.725520133972168, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.393362283706665, - "train/total_loss": 1.048572301864624 + "train/classification_loss": 0.1973317265510559, + "train/contrastive_loss": 0.718858003616333, + "train/negative_loss": 0.7188538908958435, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 4.11273413192248e-06, + "train/total_loss": 0.34110331535339355 + }, + { + "epoch": 16.669201520912548, + "step": 2200, + "train/classification_loss": 0.15299901366233826, + "train/contrastive_loss": 5.899017333984375, + "train/negative_loss": 4.804858684539795, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 1.0941585302352905, + "train/total_loss": 1.3328025341033936 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.0530916191637516, - "train/contrastive_loss": 0.6326392292976379, - "train/negative_loss": 0.24216434359550476, + "train/classification_loss": 0.2322429120540619, + "train/contrastive_loss": 3.759376049041748, + "train/negative_loss": 3.223592519760132, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.3904748857021332, - "train/total_loss": 0.1796194612979889 + "train/positive_loss": 0.5357835292816162, + "train/total_loss": 0.9841181039810181 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.1141776442527771, - "train/contrastive_loss": 1.523531198501587, - "train/negative_loss": 1.4819433689117432, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.04158787429332733, - "train/total_loss": 0.41888388991355896 + "train/classification_loss": 0.21721595525741577, + "train/contrastive_loss": 3.3090107440948486, + "train/negative_loss": 1.550881266593933, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.7581294775009155, + "train/total_loss": 0.8790181279182434 + }, + { + "epoch": 16.669201520912548, + "step": 2200, + "train/classification_loss": 0.16223053634166718, + "train/contrastive_loss": 3.0543041229248047, + "train/negative_loss": 2.0447354316711426, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 1.0095685720443726, + "train/total_loss": 0.7730913758277893 + }, + { + "epoch": 16.669201520912548, + "step": 2200, + "train/classification_loss": 0.2528938055038452, + "train/contrastive_loss": 2.297661542892456, + "train/negative_loss": 0.7968876957893372, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 1.5007739067077637, + "train/total_loss": 0.7124261260032654 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.1930762380361557, - "train/contrastive_loss": 2.658128261566162, - "train/negative_loss": 1.5363292694091797, + "train/classification_loss": 0.20382282137870789, + "train/contrastive_loss": 1.0683226585388184, + "train/negative_loss": 1.0008041858673096, "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.121799111366272, - "train/total_loss": 0.7247019410133362 + "train/num_positives": 8, + "train/positive_loss": 0.06751848757266998, + "train/total_loss": 0.41748735308647156 + }, + { + "epoch": 16.669201520912548, + "step": 2200, + "train/classification_loss": 0.20214764773845673, + "train/contrastive_loss": 1.3152387142181396, + "train/negative_loss": 0.9214023947715759, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.39383625984191895, + "train/total_loss": 0.4651954174041748 + }, + { + "epoch": 16.669201520912548, + "step": 2200, + "train/classification_loss": 0.20532679557800293, + "train/contrastive_loss": 0.9626813530921936, + "train/negative_loss": 0.959811270236969, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.002870062133297324, + "train/total_loss": 0.39786309003829956 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.046590063720941544, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.046590063720941544 - }, - { - "epoch": 20.95693779904306, - "eval_exact_match_accuracy": 0.3875598086124402, - "eval_hamming_loss": 0.05797917253025612, - "eval_loss": 0.6353241205215454, - "eval_macro_f1": 0.27719528703067176, - "eval_macro_precision": 0.39167250233426704, - "eval_macro_recall": 0.24042888932162285, - "eval_micro_f1": 0.5708333333333333, - "eval_micro_precision": 0.7445652173913043, - "eval_micro_recall": 0.46283783783783783, - "eval_runtime": 1.0611, - "eval_samples_per_second": 196.969, - "eval_steps_per_second": 25.446, + "train/classification_loss": 0.09461088478565216, + "train/contrastive_loss": 1.6569318771362305, + "train/negative_loss": 1.6546976566314697, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.002234211191534996, + "train/total_loss": 0.425997257232666 + }, + { + "epoch": 16.669201520912548, + "eval_exact_match_accuracy": 0.31297709923664124, + "eval_hamming_loss": 0.061068702290076333, + "eval_loss": 0.6201854944229126, + "eval_macro_f1": 0.2536582639144722, + "eval_macro_precision": 0.3543259772148423, + "eval_macro_recall": 0.20795193727868955, + "eval_micro_f1": 0.5436241610738255, + "eval_micro_precision": 0.7941176470588235, + "eval_micro_recall": 0.413265306122449, + "eval_runtime": 1.3164, + "eval_samples_per_second": 199.031, + "eval_steps_per_second": 25.069, "step": 2200 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.05690139904618263, - "train/contrastive_loss": 0.22760194540023804, - "train/negative_loss": 0.2275824248790741, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 1.9520948626450263e-05, - "train/total_loss": 0.10242179036140442 + "train/classification_loss": 0.1145583912730217, + "train/contrastive_loss": 0.6275790929794312, + "train/negative_loss": 0.6275706887245178, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 8.384442480746657e-06, + "train/total_loss": 0.24007421731948853 }, { - "epoch": 20.95693779904306, + "epoch": 16.669201520912548, "step": 2200, - "train/classification_loss": 0.07079331576824188, - "train/contrastive_loss": 0.10769735276699066, - "train/negative_loss": 0.10769196599721909, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 5.388275440054713e-06, - "train/total_loss": 0.09233278781175613 + "train/classification_loss": 0.12628410756587982, + "train/contrastive_loss": 0.2735500633716583, + "train/negative_loss": 0.26789674162864685, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.005653317552059889, + "train/total_loss": 0.18099412322044373 }, { - "epoch": 21.43062200956938, - "grad_norm": 17.026758193969727, - "learning_rate": 1.918053435114504e-05, - "loss": 0.3534, + "epoch": 17.045627376425855, + "grad_norm": 1.881194829940796, + "learning_rate": 1.9348710166919576e-05, + "loss": 0.3991, "step": 2250 }, { - "epoch": 21.43062200956938, + "epoch": 17.045627376425855, "step": 2250, - "train/classification_loss": 0.08988539129495621, - "train/contrastive_loss": 1.0063384771347046, - "train/negative_loss": 1.0060614347457886, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.00027706247055903077, - "train/total_loss": 0.29115310311317444 + "train/classification_loss": 0.11281248927116394, + "train/contrastive_loss": 0.3633287847042084, + "train/negative_loss": 0.24530592560768127, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.1180228590965271, + "train/total_loss": 0.18547824025154114 }, { - "epoch": 21.43062200956938, + "epoch": 17.045627376425855, "step": 2250, - "train/classification_loss": 0.09080985188484192, - "train/contrastive_loss": 0.6869128346443176, - "train/negative_loss": 0.16513274610042572, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.5217800736427307, - "train/total_loss": 0.22819241881370544 + "train/classification_loss": 0.06783324480056763, + "train/contrastive_loss": 0.3524426817893982, + "train/negative_loss": 0.13985775411128998, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.21258491277694702, + "train/total_loss": 0.13832178711891174 }, { - "epoch": 21.90909090909091, - "grad_norm": 7.447392463684082, - "learning_rate": 1.916145038167939e-05, - "loss": 0.3715, + "epoch": 17.425855513307983, + "grad_norm": 7.014838218688965, + "learning_rate": 1.933353566009105e-05, + "loss": 0.4161, "step": 2300 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, + "step": 2300, + "train/classification_loss": 0.058833811432123184, + "train/contrastive_loss": 0.35479283332824707, + "train/negative_loss": 0.31344324350357056, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.04134959727525711, + "train/total_loss": 0.12979237735271454 + }, + { + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.1259078085422516, - "train/contrastive_loss": 1.2593520879745483, - "train/negative_loss": 0.9297453761100769, + "train/classification_loss": 0.22031082212924957, + "train/contrastive_loss": 1.2736599445343018, + "train/negative_loss": 1.2710506916046143, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.32960668206214905, - "train/total_loss": 0.37777823209762573 + "train/num_positives": 6, + "train/positive_loss": 0.002609258983284235, + "train/total_loss": 0.47504281997680664 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.19811026751995087, - "train/contrastive_loss": 3.4957571029663086, - "train/negative_loss": 1.8964003324508667, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.5993568897247314, - "train/total_loss": 0.8972617387771606 + "train/classification_loss": 0.12942002713680267, + "train/contrastive_loss": 2.251772403717041, + "train/negative_loss": 1.6370941400527954, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.6146782636642456, + "train/total_loss": 0.5797744989395142 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.10413289070129395, - "train/contrastive_loss": 0.609454333782196, - "train/negative_loss": 0.5788906812667847, - "train/num_negatives": 44, + "train/classification_loss": 0.24288839101791382, + "train/contrastive_loss": 2.1154351234436035, + "train/negative_loss": 1.4812310934066772, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.6342039704322815, + "train/total_loss": 0.6659754514694214 + }, + { + "epoch": 17.425855513307983, + "step": 2300, + "train/classification_loss": 0.10862905532121658, + "train/contrastive_loss": 1.0897237062454224, + "train/negative_loss": 0.9631260633468628, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.12659762799739838, + "train/total_loss": 0.32657378911972046 + }, + { + "epoch": 17.425855513307983, + "step": 2300, + "train/classification_loss": 0.21865755319595337, + "train/contrastive_loss": 1.5750186443328857, + "train/negative_loss": 1.0912379026412964, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.030563656240701675, - "train/total_loss": 0.22602376341819763 + "train/positive_loss": 0.48378074169158936, + "train/total_loss": 0.5336612462997437 + }, + { + "epoch": 17.425855513307983, + "step": 2300, + "train/classification_loss": 0.2301435023546219, + "train/contrastive_loss": 3.7126407623291016, + "train/negative_loss": 1.4760569334030151, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 2.236583948135376, + "train/total_loss": 0.972671627998352 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.16424623131752014, - "train/contrastive_loss": 2.9641623497009277, - "train/negative_loss": 1.3978941440582275, + "train/classification_loss": 0.17100591957569122, + "train/contrastive_loss": 2.874896764755249, + "train/negative_loss": 1.3093689680099487, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.5662680864334106, - "train/total_loss": 0.7570787668228149 + "train/positive_loss": 1.5655277967453003, + "train/total_loss": 0.7459852695465088 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.14039196074008942, - "train/contrastive_loss": 3.2275421619415283, - "train/negative_loss": 1.289525032043457, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.9380171298980713, - "train/total_loss": 0.7859004139900208 + "train/classification_loss": 0.15754921734333038, + "train/contrastive_loss": 1.7975473403930664, + "train/negative_loss": 0.8368979096412659, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.9606493711471558, + "train/total_loss": 0.5170586705207825 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.15231680870056152, - "train/contrastive_loss": 2.342830181121826, - "train/negative_loss": 0.8237559795379639, - "train/num_negatives": 32, + "train/classification_loss": 0.15340128540992737, + "train/contrastive_loss": 2.169710874557495, + "train/negative_loss": 2.1455435752868652, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.024167383089661598, + "train/total_loss": 0.5873434543609619 + }, + { + "epoch": 17.425855513307983, + "step": 2300, + "train/classification_loss": 0.16682182252407074, + "train/contrastive_loss": 4.208253860473633, + "train/negative_loss": 4.014573574066162, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.5190742015838623, - "train/total_loss": 0.6208828687667847 + "train/positive_loss": 0.19368022680282593, + "train/total_loss": 1.0084725618362427 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.13399244844913483, - "train/contrastive_loss": 1.2283793687820435, - "train/negative_loss": 0.9163599014282227, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.3120194375514984, - "train/total_loss": 0.37966832518577576 + "train/classification_loss": 0.1728552132844925, + "train/contrastive_loss": 3.176663398742676, + "train/negative_loss": 2.0008904933929443, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 1.1757729053497314, + "train/total_loss": 0.8081879019737244 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.21481600403785706, - "train/contrastive_loss": 0.7955127954483032, - "train/negative_loss": 0.7661020755767822, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.029410740360617638, - "train/total_loss": 0.3739185631275177 + "train/classification_loss": 0.20872148871421814, + "train/contrastive_loss": 2.3771066665649414, + "train/negative_loss": 1.1719424724578857, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.2051640748977661, + "train/total_loss": 0.6841428279876709 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.138074591755867, - "train/contrastive_loss": 2.6850132942199707, - "train/negative_loss": 1.288662075996399, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 1.3963510990142822, - "train/total_loss": 0.6750772595405579 + "train/classification_loss": 0.17612940073013306, + "train/contrastive_loss": 2.838968276977539, + "train/negative_loss": 1.6468650102615356, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.1921032667160034, + "train/total_loss": 0.7439230680465698 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.2502129375934601, - "train/contrastive_loss": 3.759060859680176, - "train/negative_loss": 1.6247795820236206, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.1342811584472656, - "train/total_loss": 1.0020251274108887 + "train/classification_loss": 0.19879724085330963, + "train/contrastive_loss": 4.102169990539551, + "train/negative_loss": 2.757352828979492, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 1.3448169231414795, + "train/total_loss": 1.0192313194274902 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.23006457090377808, - "train/contrastive_loss": 2.614903450012207, - "train/negative_loss": 1.7405953407287598, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.8743082284927368, - "train/total_loss": 0.7530452609062195 + "train/classification_loss": 0.12039431184530258, + "train/contrastive_loss": 0.6283483505249023, + "train/negative_loss": 0.6142286658287048, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.01411969494074583, + "train/total_loss": 0.2460639774799347 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.12079191207885742, - "train/contrastive_loss": 1.8550293445587158, - "train/negative_loss": 0.6831962466239929, + "train/classification_loss": 0.14738988876342773, + "train/contrastive_loss": 1.525869369506836, + "train/negative_loss": 0.7305834293365479, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 1.1718331575393677, - "train/total_loss": 0.4917977750301361 + "train/positive_loss": 0.7952859401702881, + "train/total_loss": 0.4525637626647949 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.21073220670223236, - "train/contrastive_loss": 4.06158447265625, - "train/negative_loss": 2.8635880947113037, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.1979961395263672, - "train/total_loss": 1.0230491161346436 + "train/classification_loss": 0.17028363049030304, + "train/contrastive_loss": 2.679755210876465, + "train/negative_loss": 1.7128340005874634, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.9669210910797119, + "train/total_loss": 0.7062346935272217 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.05460936576128006, - "train/contrastive_loss": 0.1769816279411316, - "train/negative_loss": 0.13560926914215088, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.041372355073690414, - "train/total_loss": 0.09000569581985474 + "train/classification_loss": 0.26073572039604187, + "train/contrastive_loss": 2.96062970161438, + "train/negative_loss": 1.9417033195495605, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.0189263820648193, + "train/total_loss": 0.8528616428375244 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.1955001950263977, - "train/contrastive_loss": 1.4461673498153687, - "train/negative_loss": 0.8938241004943848, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.5523432493209839, - "train/total_loss": 0.4847336709499359 + "train/classification_loss": 0.2226603627204895, + "train/contrastive_loss": 2.8980512619018555, + "train/negative_loss": 1.4363616704940796, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.4616897106170654, + "train/total_loss": 0.8022706508636475 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.11947701871395111, - "train/contrastive_loss": 2.2215824127197266, - "train/negative_loss": 0.8260148763656616, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 1.395567536354065, - "train/total_loss": 0.5637934803962708 + "train/classification_loss": 0.24691444635391235, + "train/contrastive_loss": 2.095137119293213, + "train/negative_loss": 1.2179800271987915, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.8771571516990662, + "train/total_loss": 0.6659418344497681 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.22717680037021637, - "train/contrastive_loss": 3.6050209999084473, - "train/negative_loss": 1.9912391901016235, - "train/num_negatives": 46, + "train/classification_loss": 0.2773834764957428, + "train/contrastive_loss": 2.8165156841278076, + "train/negative_loss": 0.9828323721885681, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.6137818098068237, - "train/total_loss": 0.9481809735298157 + "train/positive_loss": 1.8336833715438843, + "train/total_loss": 0.8406866788864136 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.26704367995262146, - "train/contrastive_loss": 1.9984209537506104, - "train/negative_loss": 0.9037294387817383, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.094691514968872, - "train/total_loss": 0.6667279005050659 + "train/classification_loss": 0.23445159196853638, + "train/contrastive_loss": 1.826267957687378, + "train/negative_loss": 1.2999050617218018, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.5263628363609314, + "train/total_loss": 0.5997052192687988 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.19606326520442963, - "train/contrastive_loss": 3.531022310256958, - "train/negative_loss": 1.667274832725525, + "train/classification_loss": 0.17643703520298004, + "train/contrastive_loss": 0.4981665015220642, + "train/negative_loss": 0.4980768859386444, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.863747477531433, - "train/total_loss": 0.9022677540779114 + "train/num_positives": 8, + "train/positive_loss": 8.960890409070998e-05, + "train/total_loss": 0.27607032656669617 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.2159067690372467, - "train/contrastive_loss": 3.1789402961730957, - "train/negative_loss": 1.5687106847763062, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 1.610229730606079, - "train/total_loss": 0.8516948223114014 + "train/classification_loss": 0.14553014934062958, + "train/contrastive_loss": 4.521634578704834, + "train/negative_loss": 3.239170551300049, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 1.2824640274047852, + "train/total_loss": 1.0498570203781128 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.17137636244297028, - "train/contrastive_loss": 1.7944128513336182, - "train/negative_loss": 1.0179449319839478, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.7764679789543152, - "train/total_loss": 0.5302589535713196 + "train/classification_loss": 0.23319123685359955, + "train/contrastive_loss": 4.04248571395874, + "train/negative_loss": 2.460202932357788, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 1.5822827816009521, + "train/total_loss": 1.041688323020935 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.19492340087890625, - "train/contrastive_loss": 5.545324325561523, - "train/negative_loss": 4.146565914154053, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.3987586498260498, - "train/total_loss": 1.3039883375167847 + "train/classification_loss": 0.2138170599937439, + "train/contrastive_loss": 2.5913572311401367, + "train/negative_loss": 1.036714792251587, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.5546423196792603, + "train/total_loss": 0.7320885062217712 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.25617724657058716, - "train/contrastive_loss": 5.208540916442871, - "train/negative_loss": 2.3466145992279053, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.861926317214966, - "train/total_loss": 1.2978854179382324 + "train/classification_loss": 0.1550111472606659, + "train/contrastive_loss": 2.210012912750244, + "train/negative_loss": 1.9524074792861938, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.2576054632663727, + "train/total_loss": 0.5970137119293213 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.06164241582155228, - "train/contrastive_loss": 1.6357975006103516, - "train/negative_loss": 0.3919445276260376, - "train/num_negatives": 42, + "train/classification_loss": 0.25706061720848083, + "train/contrastive_loss": 2.23604154586792, + "train/negative_loss": 0.8411686420440674, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 1.243852972984314, - "train/total_loss": 0.3888019025325775 + "train/positive_loss": 1.3948729038238525, + "train/total_loss": 0.7042689323425293 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.11135431379079819, - "train/contrastive_loss": 1.8291467428207397, - "train/negative_loss": 1.6375898122787476, - "train/num_negatives": 48, + "train/classification_loss": 0.2035166472196579, + "train/contrastive_loss": 0.8024842739105225, + "train/negative_loss": 0.7611761093139648, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.19155696034431458, - "train/total_loss": 0.47718366980552673 + "train/positive_loss": 0.041308194398880005, + "train/total_loss": 0.3640134930610657 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.18736690282821655, - "train/contrastive_loss": 2.0889763832092285, - "train/negative_loss": 1.301356315612793, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.7876200079917908, - "train/total_loss": 0.6051621437072754 + "train/classification_loss": 0.21028324961662292, + "train/contrastive_loss": 1.4958069324493408, + "train/negative_loss": 0.8295918703079224, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.6662150621414185, + "train/total_loss": 0.5094446539878845 + }, + { + "epoch": 17.425855513307983, + "step": 2300, + "train/classification_loss": 0.1973634958267212, + "train/contrastive_loss": 1.0015912055969238, + "train/negative_loss": 0.9768807888031006, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.024710362777113914, + "train/total_loss": 0.39768174290657043 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.026981400325894356, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.026981400325894356 - }, - { - "epoch": 21.90909090909091, - "eval_exact_match_accuracy": 0.3588516746411483, - "eval_hamming_loss": 0.059949338587109484, - "eval_loss": 0.6689977049827576, - "eval_macro_f1": 0.2874034761697888, - "eval_macro_precision": 0.36760600792040954, - "eval_macro_recall": 0.24484199383853358, - "eval_micro_f1": 0.5458422174840085, - "eval_micro_precision": 0.7398843930635838, - "eval_micro_recall": 0.43243243243243246, - "eval_runtime": 1.0619, - "eval_samples_per_second": 196.813, - "eval_steps_per_second": 25.426, + "train/classification_loss": 0.09828422218561172, + "train/contrastive_loss": 1.9746288061141968, + "train/negative_loss": 1.9612061977386475, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.013422584161162376, + "train/total_loss": 0.49320998787879944 + }, + { + "epoch": 17.425855513307983, + "eval_exact_match_accuracy": 0.30916030534351147, + "eval_hamming_loss": 0.06331387516838796, + "eval_loss": 0.6414106488227844, + "eval_macro_f1": 0.2740934119735401, + "eval_macro_precision": 0.35602021704962883, + "eval_macro_recall": 0.22691618258410118, + "eval_micro_f1": 0.5284280936454849, + "eval_micro_precision": 0.7669902912621359, + "eval_micro_recall": 0.4030612244897959, + "eval_runtime": 1.3053, + "eval_samples_per_second": 200.723, + "eval_steps_per_second": 25.282, "step": 2300 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.07241903990507126, - "train/contrastive_loss": 0.26751935482025146, - "train/negative_loss": 0.267509788274765, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 9.566608241584618e-06, - "train/total_loss": 0.12592291831970215 + "train/classification_loss": 0.0912293791770935, + "train/contrastive_loss": 0.10950058698654175, + "train/negative_loss": 0.10848486423492432, + "train/num_negatives": 34, + "train/num_positives": 14, + "train/positive_loss": 0.0010157203068956733, + "train/total_loss": 0.11312949657440186 }, { - "epoch": 21.90909090909091, + "epoch": 17.425855513307983, "step": 2300, - "train/classification_loss": 0.09625313431024551, - "train/contrastive_loss": 0.4620448648929596, - "train/negative_loss": 0.42757901549339294, - "train/num_negatives": 38, - "train/num_positives": 16, - "train/positive_loss": 0.03446585685014725, - "train/total_loss": 0.1886621117591858 + "train/classification_loss": 0.11572025716304779, + "train/contrastive_loss": 0.176627516746521, + "train/negative_loss": 0.17555075883865356, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.0010767554631456733, + "train/total_loss": 0.1510457694530487 }, { - "epoch": 22.382775119617225, - "grad_norm": 4.871077537536621, - "learning_rate": 1.9142366412213742e-05, - "loss": 0.3556, + "epoch": 17.806083650190114, + "grad_norm": 4.2402520179748535, + "learning_rate": 1.9318361153262522e-05, + "loss": 0.4436, "step": 2350 }, { - "epoch": 22.382775119617225, + "epoch": 17.806083650190114, "step": 2350, - "train/classification_loss": 0.06534972786903381, - "train/contrastive_loss": 0.12165739387273788, - "train/negative_loss": 0.12085486203432083, - "train/num_negatives": 26, - "train/num_positives": 28, - "train/positive_loss": 0.0008025291608646512, - "train/total_loss": 0.08968120813369751 + "train/classification_loss": 0.055575139820575714, + "train/contrastive_loss": 0.1426246464252472, + "train/negative_loss": 0.13868117332458496, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.003943477291613817, + "train/total_loss": 0.08410006761550903 }, { - "epoch": 22.382775119617225, + "epoch": 17.806083650190114, "step": 2350, - "train/classification_loss": 0.10662441700696945, - "train/contrastive_loss": 0.15221214294433594, - "train/negative_loss": 0.15216578543186188, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 4.6363984438357875e-05, - "train/total_loss": 0.13706684112548828 + "train/classification_loss": 0.10090979188680649, + "train/contrastive_loss": 0.9170433282852173, + "train/negative_loss": 0.2914513945579529, + "train/num_negatives": 38, + "train/num_positives": 18, + "train/positive_loss": 0.6255919337272644, + "train/total_loss": 0.2843184471130371 }, { - "epoch": 22.861244019138756, - "grad_norm": 7.3902764320373535, - "learning_rate": 1.9123282442748094e-05, - "loss": 0.3334, + "epoch": 18.182509505703422, + "grad_norm": 5.00245475769043, + "learning_rate": 1.9303186646433994e-05, + "loss": 0.396, "step": 2400 }, { - "epoch": 22.861244019138756, - "step": 2400, - "train/classification_loss": 0.13269813358783722, - "train/contrastive_loss": 1.4186406135559082, - "train/negative_loss": 1.1982452869415283, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.22039535641670227, - "train/total_loss": 0.41642624139785767 - }, - { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.1946112960577011, - "train/contrastive_loss": 2.2636001110076904, - "train/negative_loss": 1.3154717683792114, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.9481282830238342, - "train/total_loss": 0.6473312973976135 + "train/classification_loss": 0.060825858265161514, + "train/contrastive_loss": 0.3736673891544342, + "train/negative_loss": 0.15713104605674744, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.21653634309768677, + "train/total_loss": 0.1355593353509903 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.11884357035160065, - "train/contrastive_loss": 0.9946784377098083, - "train/negative_loss": 0.8872374296188354, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.1074410006403923, - "train/total_loss": 0.3177792429924011 + "train/classification_loss": 0.23395930230617523, + "train/contrastive_loss": 1.630251407623291, + "train/negative_loss": 1.6137356758117676, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.01651577837765217, + "train/total_loss": 0.5600095987319946 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.1530592292547226, - "train/contrastive_loss": 2.608339309692383, - "train/negative_loss": 1.3266665935516357, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.2816728353500366, - "train/total_loss": 0.6747271418571472 + "train/classification_loss": 0.1048639640212059, + "train/contrastive_loss": 2.448622226715088, + "train/negative_loss": 2.201669216156006, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.2469530552625656, + "train/total_loss": 0.5945883989334106 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.14304761588573456, - "train/contrastive_loss": 2.655427932739258, - "train/negative_loss": 1.4142804145812988, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.241147518157959, - "train/total_loss": 0.6741332411766052 + "train/classification_loss": 0.21451503038406372, + "train/contrastive_loss": 1.9686616659164429, + "train/negative_loss": 1.2893317937850952, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.6793298721313477, + "train/total_loss": 0.6082473993301392 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.13373611867427826, - "train/contrastive_loss": 2.0157852172851562, - "train/negative_loss": 0.617508053779602, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.3982771635055542, - "train/total_loss": 0.5368931889533997 + "train/classification_loss": 0.10143005102872849, + "train/contrastive_loss": 0.8170071244239807, + "train/negative_loss": 0.8146344423294067, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.0023726734798401594, + "train/total_loss": 0.2648314833641052 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.11863918602466583, - "train/contrastive_loss": 1.7504560947418213, - "train/negative_loss": 0.9922500848770142, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.7582060098648071, - "train/total_loss": 0.4687303900718689 + "train/classification_loss": 0.20287169516086578, + "train/contrastive_loss": 1.2878674268722534, + "train/negative_loss": 1.2287638187408447, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.059103574603796005, + "train/total_loss": 0.4604451656341553 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.23499971628189087, - "train/contrastive_loss": 0.6423831582069397, - "train/negative_loss": 0.576582133769989, + "train/classification_loss": 0.2384653091430664, + "train/contrastive_loss": 3.849626064300537, + "train/negative_loss": 1.8963642120361328, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.06580103933811188, - "train/total_loss": 0.36347633600234985 + "train/num_positives": 8, + "train/positive_loss": 1.9532617330551147, + "train/total_loss": 1.0083905458450317 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.1148604303598404, - "train/contrastive_loss": 0.6528195738792419, - "train/negative_loss": 0.5945950150489807, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.05822458118200302, - "train/total_loss": 0.24542434513568878 + "train/classification_loss": 0.16559313237667084, + "train/contrastive_loss": 1.4069759845733643, + "train/negative_loss": 1.2354000806808472, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.1715758591890335, + "train/total_loss": 0.4469883441925049 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.2494233101606369, - "train/contrastive_loss": 3.361999034881592, - "train/negative_loss": 1.1518245935440063, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.210174322128296, - "train/total_loss": 0.9218231439590454 + "train/classification_loss": 0.10818913578987122, + "train/contrastive_loss": 0.45971450209617615, + "train/negative_loss": 0.3063635230064392, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.15335097908973694, + "train/total_loss": 0.20013204216957092 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.24271897971630096, - "train/contrastive_loss": 2.8653616905212402, - "train/negative_loss": 1.8338308334350586, - "train/num_negatives": 46, + "train/classification_loss": 0.16306866705417633, + "train/contrastive_loss": 2.023575782775879, + "train/negative_loss": 1.9771751165390015, + "train/num_negatives": 48, "train/num_positives": 8, - "train/positive_loss": 1.0315308570861816, - "train/total_loss": 0.8157913684844971 + "train/positive_loss": 0.046400681138038635, + "train/total_loss": 0.5677838325500488 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.12171508371829987, - "train/contrastive_loss": 1.6888409852981567, - "train/negative_loss": 0.3681544363498688, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 1.3206865787506104, - "train/total_loss": 0.45948326587677 + "train/classification_loss": 0.17327429354190826, + "train/contrastive_loss": 5.389758586883545, + "train/negative_loss": 5.3882293701171875, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.0015291100135073066, + "train/total_loss": 1.2512260675430298 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.21203282475471497, - "train/contrastive_loss": 4.2433762550354, - "train/negative_loss": 3.046238660812378, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.197137475013733, - "train/total_loss": 1.0607080459594727 + "train/classification_loss": 0.16968287527561188, + "train/contrastive_loss": 3.609344244003296, + "train/negative_loss": 2.9121713638305664, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.6971728801727295, + "train/total_loss": 0.8915517330169678 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.05384403094649315, - "train/contrastive_loss": 0.5872567892074585, - "train/negative_loss": 0.5460790991783142, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.04117768257856369, - "train/total_loss": 0.1712953895330429 + "train/classification_loss": 0.20022986829280853, + "train/contrastive_loss": 2.2896223068237305, + "train/negative_loss": 1.3466452360153198, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.9429770112037659, + "train/total_loss": 0.658154308795929 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.19349339604377747, - "train/contrastive_loss": 1.984846830368042, - "train/negative_loss": 0.850542426109314, + "train/classification_loss": 0.17223259806632996, + "train/contrastive_loss": 3.0502676963806152, + "train/negative_loss": 1.2675079107284546, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.7827599048614502, + "train/total_loss": 0.7822861671447754 + }, + { + "epoch": 18.182509505703422, + "step": 2400, + "train/classification_loss": 0.19881576299667358, + "train/contrastive_loss": 2.7032222747802734, + "train/negative_loss": 2.2782795429229736, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 1.134304404258728, - "train/total_loss": 0.5904628038406372 + "train/positive_loss": 0.42494261264801025, + "train/total_loss": 0.7394602298736572 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.10817869752645493, - "train/contrastive_loss": 2.225862741470337, - "train/negative_loss": 1.2672168016433716, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.9586458802223206, - "train/total_loss": 0.5533512234687805 + "train/classification_loss": 0.11366742104291916, + "train/contrastive_loss": 0.5158471465110779, + "train/negative_loss": 0.5158442854881287, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 2.8610279514396098e-06, + "train/total_loss": 0.2168368548154831 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.26903751492500305, - "train/contrastive_loss": 5.193462371826172, - "train/negative_loss": 3.10563063621521, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 2.087831497192383, - "train/total_loss": 1.307729959487915 + "train/classification_loss": 0.14389677345752716, + "train/contrastive_loss": 1.1191126108169556, + "train/negative_loss": 0.868039608001709, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.2510729730129242, + "train/total_loss": 0.36771929264068604 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.269782155752182, - "train/contrastive_loss": 2.1090164184570312, - "train/negative_loss": 1.0870609283447266, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.0219554901123047, - "train/total_loss": 0.6915854215621948 + "train/classification_loss": 0.180295929312706, + "train/contrastive_loss": 3.973062515258789, + "train/negative_loss": 2.0384559631347656, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.9346064329147339, + "train/total_loss": 0.9749084711074829 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.18097630143165588, - "train/contrastive_loss": 3.0168991088867188, - "train/negative_loss": 1.7169110774993896, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.2999879121780396, - "train/total_loss": 0.7843561172485352 + "train/classification_loss": 0.26392507553100586, + "train/contrastive_loss": 3.6391055583953857, + "train/negative_loss": 2.401292324066162, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.2378132343292236, + "train/total_loss": 0.991746187210083 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.1807979941368103, - "train/contrastive_loss": 1.9735524654388428, - "train/negative_loss": 1.2691739797592163, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.7043784856796265, - "train/total_loss": 0.5755084753036499 + "train/classification_loss": 0.23418201506137848, + "train/contrastive_loss": 2.9386892318725586, + "train/negative_loss": 1.4227432012557983, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.5159461498260498, + "train/total_loss": 0.821919858455658 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.1834668070077896, - "train/contrastive_loss": 1.603402018547058, - "train/negative_loss": 0.9308879375457764, - "train/num_negatives": 44, + "train/classification_loss": 0.2140362411737442, + "train/contrastive_loss": 0.4556373357772827, + "train/negative_loss": 0.4393577575683594, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.016279567033052444, + "train/total_loss": 0.305163711309433 + }, + { + "epoch": 18.182509505703422, + "step": 2400, + "train/classification_loss": 0.2477482706308365, + "train/contrastive_loss": 2.0471482276916504, + "train/negative_loss": 0.34427598118782043, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.6725140810012817, - "train/total_loss": 0.5041472315788269 + "train/positive_loss": 1.7028721570968628, + "train/total_loss": 0.6571779251098633 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.1832117736339569, - "train/contrastive_loss": 4.777559280395508, - "train/negative_loss": 3.5519089698791504, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.2256500720977783, - "train/total_loss": 1.138723611831665 + "train/classification_loss": 0.22450657188892365, + "train/contrastive_loss": 1.927932620048523, + "train/negative_loss": 1.0931626558303833, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.8347699642181396, + "train/total_loss": 0.6100931167602539 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.22090663015842438, - "train/contrastive_loss": 5.395474910736084, - "train/negative_loss": 2.564561605453491, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.8309133052825928, - "train/total_loss": 1.300001621246338 + "train/classification_loss": 0.21489639580249786, + "train/contrastive_loss": 1.453914999961853, + "train/negative_loss": 1.4538739919662476, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 4.09505155403167e-05, + "train/total_loss": 0.5056794285774231 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.05589883401989937, - "train/contrastive_loss": 1.2581605911254883, - "train/negative_loss": 0.25482356548309326, + "train/classification_loss": 0.16431249678134918, + "train/contrastive_loss": 6.768932342529297, + "train/negative_loss": 5.7731451988220215, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.9957873821258545, + "train/total_loss": 1.5180989503860474 + }, + { + "epoch": 18.182509505703422, + "step": 2400, + "train/classification_loss": 0.2283271700143814, + "train/contrastive_loss": 3.207368850708008, + "train/negative_loss": 2.614288091659546, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 1.003337025642395, - "train/total_loss": 0.3075309693813324 + "train/positive_loss": 0.5930808186531067, + "train/total_loss": 0.8698009252548218 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.10560169070959091, - "train/contrastive_loss": 1.8003445863723755, - "train/negative_loss": 1.378133773803711, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.42221084237098694, - "train/total_loss": 0.4656706154346466 + "train/classification_loss": 0.20788678526878357, + "train/contrastive_loss": 1.822659969329834, + "train/negative_loss": 1.009602427482605, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.813057541847229, + "train/total_loss": 0.5724188089370728 + }, + { + "epoch": 18.182509505703422, + "step": 2400, + "train/classification_loss": 0.13966910541057587, + "train/contrastive_loss": 2.2813544273376465, + "train/negative_loss": 2.260056734085083, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.021297700703144073, + "train/total_loss": 0.5959399938583374 + }, + { + "epoch": 18.182509505703422, + "step": 2400, + "train/classification_loss": 0.2549617886543274, + "train/contrastive_loss": 2.151254653930664, + "train/negative_loss": 0.8255473375320435, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 1.325707197189331, + "train/total_loss": 0.6852127313613892 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.2095080316066742, - "train/contrastive_loss": 1.6817681789398193, - "train/negative_loss": 0.9502595067024231, + "train/classification_loss": 0.2031390219926834, + "train/contrastive_loss": 0.579549252986908, + "train/negative_loss": 0.5611429810523987, "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.7315086722373962, - "train/total_loss": 0.5458616614341736 + "train/num_positives": 8, + "train/positive_loss": 0.018406296148896217, + "train/total_loss": 0.3190488815307617 + }, + { + "epoch": 18.182509505703422, + "step": 2400, + "train/classification_loss": 0.2036266028881073, + "train/contrastive_loss": 1.668156623840332, + "train/negative_loss": 1.3023062944412231, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.36585038900375366, + "train/total_loss": 0.5372579097747803 + }, + { + "epoch": 18.182509505703422, + "step": 2400, + "train/classification_loss": 0.20243412256240845, + "train/contrastive_loss": 2.3171918392181396, + "train/negative_loss": 2.2643423080444336, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.05284957215189934, + "train/total_loss": 0.6658724546432495 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.02165534906089306, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.02165534906089306 - }, - { - "epoch": 22.861244019138756, - "eval_exact_match_accuracy": 0.3827751196172249, - "eval_hamming_loss": 0.061638052350126656, - "eval_loss": 0.6331735253334045, - "eval_macro_f1": 0.2998069235671563, - "eval_macro_precision": 0.37110809183055926, - "eval_macro_recall": 0.2705308148214723, - "eval_micro_f1": 0.5575757575757576, - "eval_micro_precision": 0.6934673366834171, - "eval_micro_recall": 0.46621621621621623, - "eval_runtime": 1.0637, - "eval_samples_per_second": 196.478, - "eval_steps_per_second": 25.382, + "train/classification_loss": 0.09846001118421555, + "train/contrastive_loss": 1.7588176727294922, + "train/negative_loss": 1.753343939781189, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.005473725497722626, + "train/total_loss": 0.45022356510162354 + }, + { + "epoch": 18.182509505703422, + "eval_exact_match_accuracy": 0.33587786259541985, + "eval_hamming_loss": 0.06039515042658285, + "eval_loss": 0.6327394843101501, + "eval_macro_f1": 0.29405504892138623, + "eval_macro_precision": 0.40126718199484984, + "eval_macro_recall": 0.24414882485853334, + "eval_micro_f1": 0.5696, + "eval_micro_precision": 0.7639484978540773, + "eval_micro_recall": 0.45408163265306123, + "eval_runtime": 1.3085, + "eval_samples_per_second": 200.232, + "eval_steps_per_second": 25.22, "step": 2400 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.0761108547449112, - "train/contrastive_loss": 0.16759860515594482, - "train/negative_loss": 0.13036693632602692, + "train/classification_loss": 0.10018738359212875, + "train/contrastive_loss": 0.2741122245788574, + "train/negative_loss": 0.1645783931016922, "train/num_negatives": 36, "train/num_positives": 20, - "train/positive_loss": 0.03723166510462761, - "train/total_loss": 0.10963057726621628 + "train/positive_loss": 0.10953383892774582, + "train/total_loss": 0.15500983595848083 }, { - "epoch": 22.861244019138756, + "epoch": 18.182509505703422, "step": 2400, - "train/classification_loss": 0.028262780979275703, - "train/contrastive_loss": 0.2484734207391739, - "train/negative_loss": 0.2111663967370987, - "train/num_negatives": 36, - "train/num_positives": 20, - "train/positive_loss": 0.037307024002075195, - "train/total_loss": 0.07795746624469757 + "train/classification_loss": 0.07538611441850662, + "train/contrastive_loss": 0.1463598608970642, + "train/negative_loss": 0.14635634422302246, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 3.516681317705661e-06, + "train/total_loss": 0.1046580895781517 }, { - "epoch": 23.33492822966507, - "grad_norm": 2.706557512283325, - "learning_rate": 1.9104198473282445e-05, - "loss": 0.2923, + "epoch": 18.562737642585553, + "grad_norm": 9.8873929977417, + "learning_rate": 1.9288012139605465e-05, + "loss": 0.3781, "step": 2450 }, { - "epoch": 23.33492822966507, + "epoch": 18.562737642585553, "step": 2450, - "train/classification_loss": 0.13168157637119293, - "train/contrastive_loss": 0.926624059677124, - "train/negative_loss": 0.2523733377456665, - "train/num_negatives": 48, + "train/classification_loss": 0.12244893610477448, + "train/contrastive_loss": 0.23092828691005707, + "train/negative_loss": 0.23089471459388733, + "train/num_negatives": 46, "train/num_positives": 6, - "train/positive_loss": 0.6742507219314575, - "train/total_loss": 0.3170064091682434 + "train/positive_loss": 3.357830428285524e-05, + "train/total_loss": 0.1686345934867859 }, { - "epoch": 23.33492822966507, + "epoch": 18.562737642585553, "step": 2450, - "train/classification_loss": 0.1377706229686737, - "train/contrastive_loss": 0.41647517681121826, - "train/negative_loss": 0.4158691167831421, - "train/num_negatives": 42, - "train/num_positives": 8, - "train/positive_loss": 0.0006060732412151992, - "train/total_loss": 0.2210656702518463 + "train/classification_loss": 0.09521232545375824, + "train/contrastive_loss": 0.8519479036331177, + "train/negative_loss": 0.7950224280357361, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.05692550539970398, + "train/total_loss": 0.26560190320014954 }, { - "epoch": 23.813397129186605, - "grad_norm": 4.433479309082031, - "learning_rate": 1.9085114503816796e-05, - "loss": 0.3133, + "epoch": 18.94296577946768, + "grad_norm": 5.7090630531311035, + "learning_rate": 1.9272837632776936e-05, + "loss": 0.3831, "step": 2500 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, + "step": 2500, + "train/classification_loss": 0.04731317237019539, + "train/contrastive_loss": 0.145561084151268, + "train/negative_loss": 0.11568094044923782, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.029880139976739883, + "train/total_loss": 0.07642538845539093 + }, + { + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.13394975662231445, - "train/contrastive_loss": 1.229429006576538, - "train/negative_loss": 1.0219714641571045, + "train/classification_loss": 0.22679442167282104, + "train/contrastive_loss": 1.5200704336166382, + "train/negative_loss": 1.5198935270309448, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.2074575126171112, - "train/total_loss": 0.3798355460166931 + "train/num_positives": 6, + "train/positive_loss": 0.00017694869893603027, + "train/total_loss": 0.5308085083961487 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.17271657288074493, - "train/contrastive_loss": 2.993584394454956, - "train/negative_loss": 1.9566650390625, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.036919355392456, - "train/total_loss": 0.7714334726333618 + "train/classification_loss": 0.11324331164360046, + "train/contrastive_loss": 3.3938276767730713, + "train/negative_loss": 2.6165101528167725, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.7773175239562988, + "train/total_loss": 0.7920088768005371 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.11925823241472244, - "train/contrastive_loss": 0.8469833135604858, - "train/negative_loss": 0.7033913731575012, - "train/num_negatives": 44, + "train/classification_loss": 0.2364424616098404, + "train/contrastive_loss": 1.947588562965393, + "train/negative_loss": 1.075680136680603, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.87190842628479, + "train/total_loss": 0.6259601712226868 + }, + { + "epoch": 18.94296577946768, + "step": 2500, + "train/classification_loss": 0.11556568741798401, + "train/contrastive_loss": 1.3833638429641724, + "train/negative_loss": 1.338081955909729, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.04528184235095978, + "train/total_loss": 0.39223846793174744 + }, + { + "epoch": 18.94296577946768, + "step": 2500, + "train/classification_loss": 0.20521940290927887, + "train/contrastive_loss": 1.2720438241958618, + "train/negative_loss": 0.9632143378257751, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.14359194040298462, - "train/total_loss": 0.2886548936367035 + "train/positive_loss": 0.30882948637008667, + "train/total_loss": 0.459628164768219 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.15368276834487915, - "train/contrastive_loss": 2.7024285793304443, - "train/negative_loss": 1.1553196907043457, + "train/classification_loss": 0.2440956085920334, + "train/contrastive_loss": 3.7085585594177246, + "train/negative_loss": 1.3067049980163574, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 2.401853561401367, + "train/total_loss": 0.9858073592185974 + }, + { + "epoch": 18.94296577946768, + "step": 2500, + "train/classification_loss": 0.17272672057151794, + "train/contrastive_loss": 0.9356433153152466, + "train/negative_loss": 0.6259652972221375, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.5471088886260986, - "train/total_loss": 0.6941685080528259 + "train/positive_loss": 0.30967798829078674, + "train/total_loss": 0.35985538363456726 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.12575550377368927, - "train/contrastive_loss": 2.927257537841797, - "train/negative_loss": 1.502415418624878, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.4248420000076294, - "train/total_loss": 0.7112070322036743 + "train/classification_loss": 0.12624533474445343, + "train/contrastive_loss": 1.0344828367233276, + "train/negative_loss": 0.5056060552597046, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.528876781463623, + "train/total_loss": 0.33314192295074463 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.16140732169151306, - "train/contrastive_loss": 3.2411441802978516, - "train/negative_loss": 1.880916953086853, - "train/num_negatives": 32, + "train/classification_loss": 0.1666499376296997, + "train/contrastive_loss": 3.013045072555542, + "train/negative_loss": 2.0333402156829834, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.9797049164772034, + "train/total_loss": 0.769258975982666 + }, + { + "epoch": 18.94296577946768, + "step": 2500, + "train/classification_loss": 0.1870373785495758, + "train/contrastive_loss": 6.566862106323242, + "train/negative_loss": 6.566845893859863, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.360227108001709, - "train/total_loss": 0.809636116027832 + "train/positive_loss": 1.6440693798358552e-05, + "train/total_loss": 1.5004098415374756 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.13192395865917206, - "train/contrastive_loss": 1.02787446975708, - "train/negative_loss": 0.5810546875, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.44681984186172485, - "train/total_loss": 0.33749884366989136 + "train/classification_loss": 0.14235498011112213, + "train/contrastive_loss": 2.239654541015625, + "train/negative_loss": 1.4641391038894653, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.7755153179168701, + "train/total_loss": 0.5902858972549438 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.22113412618637085, - "train/contrastive_loss": 0.9258909821510315, - "train/negative_loss": 0.9258767366409302, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 1.4225845916371327e-05, - "train/total_loss": 0.40631234645843506 + "train/classification_loss": 0.20826289057731628, + "train/contrastive_loss": 2.0289041996002197, + "train/negative_loss": 1.25517737865448, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.773726761341095, + "train/total_loss": 0.6140437126159668 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.1395042985677719, - "train/contrastive_loss": 1.4150408506393433, - "train/negative_loss": 0.732489287853241, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.6825515627861023, - "train/total_loss": 0.4225124716758728 + "train/classification_loss": 0.174913689494133, + "train/contrastive_loss": 3.5629372596740723, + "train/negative_loss": 1.2925275564193726, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.27040958404541, + "train/total_loss": 0.8875011801719666 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.24286136031150818, - "train/contrastive_loss": 3.136279821395874, - "train/negative_loss": 1.2720892429351807, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 1.8641905784606934, - "train/total_loss": 0.8701173067092896 + "train/classification_loss": 0.21699240803718567, + "train/contrastive_loss": 2.563103675842285, + "train/negative_loss": 1.9586668014526367, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.6044368147850037, + "train/total_loss": 0.729613184928894 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.21856100857257843, - "train/contrastive_loss": 2.197334051132202, - "train/negative_loss": 1.7858563661575317, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.4114777445793152, - "train/total_loss": 0.6580278277397156 + "train/classification_loss": 0.12402063608169556, + "train/contrastive_loss": 0.6105921268463135, + "train/negative_loss": 0.6105103492736816, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 8.17828913568519e-05, + "train/total_loss": 0.2461390644311905 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.12145136296749115, - "train/contrastive_loss": 1.8010122776031494, - "train/negative_loss": 1.0063754320144653, + "train/classification_loss": 0.15560723841190338, + "train/contrastive_loss": 1.013145089149475, + "train/negative_loss": 0.7235718369483948, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 0.7946369051933289, - "train/total_loss": 0.4816538095474243 + "train/positive_loss": 0.2895732820034027, + "train/total_loss": 0.35823625326156616 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.2238074690103531, - "train/contrastive_loss": 4.315067291259766, - "train/negative_loss": 2.8351962566375732, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.4798709154129028, - "train/total_loss": 1.0868209600448608 + "train/classification_loss": 0.18332161009311676, + "train/contrastive_loss": 4.035504341125488, + "train/negative_loss": 2.0447118282318115, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.9907922744750977, + "train/total_loss": 0.9904224872589111 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.05181095749139786, - "train/contrastive_loss": 0.5136381983757019, - "train/negative_loss": 0.5085988640785217, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.0050393324345350266, - "train/total_loss": 0.1545386016368866 + "train/classification_loss": 0.25147131085395813, + "train/contrastive_loss": 4.420267105102539, + "train/negative_loss": 2.7550837993621826, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.665183424949646, + "train/total_loss": 1.1355247497558594 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.190106600522995, - "train/contrastive_loss": 1.2489343881607056, - "train/negative_loss": 1.0331065654754639, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.2158278524875641, - "train/total_loss": 0.4398934841156006 + "train/classification_loss": 0.24191178381443024, + "train/contrastive_loss": 2.4743587970733643, + "train/negative_loss": 1.4745750427246094, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.9997837543487549, + "train/total_loss": 0.7367835640907288 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.08917197585105896, - "train/contrastive_loss": 2.2841601371765137, - "train/negative_loss": 1.1362003087997437, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 1.1479599475860596, - "train/total_loss": 0.5460039973258972 + "train/classification_loss": 0.20542606711387634, + "train/contrastive_loss": 0.21042241156101227, + "train/negative_loss": 0.2088732272386551, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.0015491892118006945, + "train/total_loss": 0.24751055240631104 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.28470858931541443, - "train/contrastive_loss": 4.177870273590088, - "train/negative_loss": 2.409210681915283, - "train/num_negatives": 46, + "train/classification_loss": 0.24380874633789062, + "train/contrastive_loss": 1.479932188987732, + "train/negative_loss": 0.44405654072761536, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.7686595916748047, - "train/total_loss": 1.1202826499938965 + "train/positive_loss": 1.035875678062439, + "train/total_loss": 0.5397951602935791 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.2675986588001251, - "train/contrastive_loss": 3.336737632751465, - "train/negative_loss": 2.3252086639404297, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.0115290880203247, - "train/total_loss": 0.9349461793899536 + "train/classification_loss": 0.2345714122056961, + "train/contrastive_loss": 2.847850799560547, + "train/negative_loss": 1.2612992525100708, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.5865516662597656, + "train/total_loss": 0.8041415810585022 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.20072387158870697, - "train/contrastive_loss": 3.4655632972717285, - "train/negative_loss": 1.7673330307006836, + "train/classification_loss": 0.22197021543979645, + "train/contrastive_loss": 1.585049033164978, + "train/negative_loss": 1.5841337442398071, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.6982301473617554, - "train/total_loss": 0.8938365578651428 + "train/num_positives": 8, + "train/positive_loss": 0.0009152587736025453, + "train/total_loss": 0.5389800071716309 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.18538600206375122, - "train/contrastive_loss": 1.3248718976974487, - "train/negative_loss": 1.0491575002670288, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.27571436762809753, - "train/total_loss": 0.45036038756370544 + "train/classification_loss": 0.150671124458313, + "train/contrastive_loss": 5.797352313995361, + "train/negative_loss": 5.068827152252197, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.7285252809524536, + "train/total_loss": 1.3101415634155273 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.20289970934391022, - "train/contrastive_loss": 2.151657819747925, - "train/negative_loss": 1.2247767448425293, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.9268811345100403, - "train/total_loss": 0.6332312822341919 + "train/classification_loss": 0.24683138728141785, + "train/contrastive_loss": 3.389397144317627, + "train/negative_loss": 2.314236640930176, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 1.0751605033874512, + "train/total_loss": 0.9247108697891235 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.20821522176265717, - "train/contrastive_loss": 6.184883117675781, - "train/negative_loss": 5.128342628479004, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.0565402507781982, - "train/total_loss": 1.4451918601989746 + "train/classification_loss": 0.18139326572418213, + "train/contrastive_loss": 1.622804880142212, + "train/negative_loss": 1.0358093976974487, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.5869954228401184, + "train/total_loss": 0.5059542655944824 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.25261572003364563, - "train/contrastive_loss": 4.639723300933838, - "train/negative_loss": 2.1566243171691895, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.4830989837646484, - "train/total_loss": 1.1805603504180908 + "train/classification_loss": 0.14523743093013763, + "train/contrastive_loss": 2.2011449337005615, + "train/negative_loss": 2.0776803493499756, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.12346456199884415, + "train/total_loss": 0.5854664444923401 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.05637431889772415, - "train/contrastive_loss": 1.2931867837905884, - "train/negative_loss": 0.6301845908164978, - "train/num_negatives": 42, + "train/classification_loss": 0.2727917432785034, + "train/contrastive_loss": 2.2381370067596436, + "train/negative_loss": 0.8912444710731506, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.6630021929740906, - "train/total_loss": 0.3150116801261902 + "train/positive_loss": 1.3468925952911377, + "train/total_loss": 0.72041916847229 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.10713080316781998, - "train/contrastive_loss": 1.611513614654541, - "train/negative_loss": 1.3453491926193237, - "train/num_negatives": 48, + "train/classification_loss": 0.1986035257577896, + "train/contrastive_loss": 0.3167102038860321, + "train/negative_loss": 0.3102244436740875, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.2661644518375397, - "train/total_loss": 0.42943352460861206 + "train/positive_loss": 0.006485761608928442, + "train/total_loss": 0.26194557547569275 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.20963171124458313, - "train/contrastive_loss": 1.4862478971481323, - "train/negative_loss": 0.75822913646698, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.7280187606811523, - "train/total_loss": 0.5068812966346741 + "train/classification_loss": 0.2012515664100647, + "train/contrastive_loss": 2.0012366771698, + "train/negative_loss": 1.3981997966766357, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.6030368208885193, + "train/total_loss": 0.6014989018440247 + }, + { + "epoch": 18.94296577946768, + "step": 2500, + "train/classification_loss": 0.17119084298610687, + "train/contrastive_loss": 0.791343629360199, + "train/negative_loss": 0.7890623211860657, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.002281289082020521, + "train/total_loss": 0.3294595777988434 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.036070890724658966, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.036070890724658966 - }, - { - "epoch": 23.813397129186605, - "eval_exact_match_accuracy": 0.41148325358851673, - "eval_hamming_loss": 0.058823529411764705, - "eval_loss": 0.6496673822402954, - "eval_macro_f1": 0.33738364164793627, - "eval_macro_precision": 0.40807837502817224, - "eval_macro_recall": 0.29976545817376266, - "eval_micro_f1": 0.584493041749503, - "eval_micro_precision": 0.7101449275362319, - "eval_micro_recall": 0.4966216216216216, - "eval_runtime": 1.0621, - "eval_samples_per_second": 196.772, - "eval_steps_per_second": 25.42, + "train/classification_loss": 0.10481034219264984, + "train/contrastive_loss": 1.8564534187316895, + "train/negative_loss": 1.7962459325790405, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.060207486152648926, + "train/total_loss": 0.4761010408401489 + }, + { + "epoch": 18.94296577946768, + "eval_exact_match_accuracy": 0.3320610687022901, + "eval_hamming_loss": 0.05949708127525819, + "eval_loss": 0.6363722681999207, + "eval_macro_f1": 0.2924164236878677, + "eval_macro_precision": 0.45678184527318294, + "eval_macro_recall": 0.24763616810128475, + "eval_micro_f1": 0.5786963434022258, + "eval_micro_precision": 0.7679324894514767, + "eval_micro_recall": 0.4642857142857143, + "eval_runtime": 1.3173, + "eval_samples_per_second": 198.887, + "eval_steps_per_second": 25.051, "step": 2500 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.08467818051576614, - "train/contrastive_loss": 0.0988035500049591, - "train/negative_loss": 0.09503602981567383, - "train/num_negatives": 30, - "train/num_positives": 22, - "train/positive_loss": 0.003767516929656267, - "train/total_loss": 0.1044388934969902 + "train/classification_loss": 0.09101570397615433, + "train/contrastive_loss": 0.5434196591377258, + "train/negative_loss": 0.5434113144874573, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 8.344684829353355e-06, + "train/total_loss": 0.19969964027404785 }, { - "epoch": 23.813397129186605, + "epoch": 18.94296577946768, "step": 2500, - "train/classification_loss": 0.06276270747184753, - "train/contrastive_loss": 0.07960370182991028, - "train/negative_loss": 0.07950983196496964, - "train/num_negatives": 44, - "train/num_positives": 8, - "train/positive_loss": 9.387116006109864e-05, - "train/total_loss": 0.07868345081806183 + "train/classification_loss": 0.0617167092859745, + "train/contrastive_loss": 0.1312294602394104, + "train/negative_loss": 0.13122566044330597, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 3.797675844907644e-06, + "train/total_loss": 0.08796259760856628 }, { - "epoch": 24.28708133971292, - "grad_norm": 6.330770015716553, - "learning_rate": 1.9066030534351147e-05, - "loss": 0.2955, + "epoch": 19.319391634980988, + "grad_norm": 5.150401592254639, + "learning_rate": 1.9257663125948408e-05, + "loss": 0.3403, "step": 2550 }, { - "epoch": 24.28708133971292, + "epoch": 19.319391634980988, "step": 2550, - "train/classification_loss": 0.042592961341142654, - "train/contrastive_loss": 0.11330509185791016, - "train/negative_loss": 0.11330121755599976, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 3.874310095852707e-06, - "train/total_loss": 0.06525398045778275 + "train/classification_loss": 0.09046201407909393, + "train/contrastive_loss": 0.05033661797642708, + "train/negative_loss": 0.04955242574214935, + "train/num_negatives": 38, + "train/num_positives": 16, + "train/positive_loss": 0.0007841916522011161, + "train/total_loss": 0.10052933543920517 }, { - "epoch": 24.28708133971292, + "epoch": 19.319391634980988, "step": 2550, - "train/classification_loss": 0.07914221286773682, - "train/contrastive_loss": 0.25798478722572327, - "train/negative_loss": 0.15705133974552155, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.10093344002962112, - "train/total_loss": 0.13073916733264923 + "train/classification_loss": 0.10332977026700974, + "train/contrastive_loss": 0.189910426735878, + "train/negative_loss": 0.18194347620010376, + "train/num_negatives": 36, + "train/num_positives": 18, + "train/positive_loss": 0.007966955192387104, + "train/total_loss": 0.1413118541240692 }, { - "epoch": 24.76555023923445, - "grad_norm": 3.0690431594848633, - "learning_rate": 1.9046946564885498e-05, - "loss": 0.3031, + "epoch": 19.69961977186312, + "grad_norm": 5.378705024719238, + "learning_rate": 1.9242488619119883e-05, + "loss": 0.386, "step": 2600 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.13498210906982422, - "train/contrastive_loss": 1.2645633220672607, - "train/negative_loss": 0.8803446292877197, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.38421863317489624, - "train/total_loss": 0.38789477944374084 + "train/classification_loss": 0.05025104805827141, + "train/contrastive_loss": 0.20192904770374298, + "train/negative_loss": 0.17020410299301147, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.031724948436021805, + "train/total_loss": 0.09063685685396194 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.1889919936656952, - "train/contrastive_loss": 2.943314552307129, - "train/negative_loss": 1.8116073608398438, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.1317071914672852, - "train/total_loss": 0.7776548862457275 + "train/classification_loss": 0.2243623435497284, + "train/contrastive_loss": 1.3388066291809082, + "train/negative_loss": 1.3370131254196167, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.001793482108041644, + "train/total_loss": 0.49212366342544556 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.13153378665447235, - "train/contrastive_loss": 0.994266152381897, - "train/negative_loss": 0.7133355140686035, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.28093063831329346, - "train/total_loss": 0.33038702607154846 + "train/classification_loss": 0.12883396446704865, + "train/contrastive_loss": 3.6052026748657227, + "train/negative_loss": 2.7592148780822754, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.8459876775741577, + "train/total_loss": 0.8498744964599609 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.16689729690551758, - "train/contrastive_loss": 3.2972848415374756, - "train/negative_loss": 1.6167206764221191, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.6805641651153564, - "train/total_loss": 0.8263542652130127 + "train/classification_loss": 0.24950255453586578, + "train/contrastive_loss": 1.8088736534118652, + "train/negative_loss": 0.8170151710510254, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.9918584823608398, + "train/total_loss": 0.6112772822380066 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.14607354998588562, - "train/contrastive_loss": 2.569248914718628, - "train/negative_loss": 1.3629138469696045, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.2063350677490234, - "train/total_loss": 0.6599233150482178 + "train/classification_loss": 0.10373694449663162, + "train/contrastive_loss": 1.0448983907699585, + "train/negative_loss": 1.0431584119796753, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.0017400048673152924, + "train/total_loss": 0.31271663308143616 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.17026017606258392, - "train/contrastive_loss": 3.421651840209961, - "train/negative_loss": 1.9346544742584229, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.4869974851608276, - "train/total_loss": 0.8545905947685242 + "train/classification_loss": 0.19370251893997192, + "train/contrastive_loss": 1.4251106977462769, + "train/negative_loss": 1.281157374382019, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.1439533233642578, + "train/total_loss": 0.4787246584892273 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.12829940021038055, - "train/contrastive_loss": 1.3115735054016113, - "train/negative_loss": 1.2041373252868652, + "train/classification_loss": 0.24770870804786682, + "train/contrastive_loss": 2.443418502807617, + "train/negative_loss": 1.4409583806991577, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.10743623226881027, - "train/total_loss": 0.3906140923500061 + "train/num_positives": 8, + "train/positive_loss": 1.00246000289917, + "train/total_loss": 0.7363923788070679 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.22340339422225952, - "train/contrastive_loss": 0.7629526853561401, - "train/negative_loss": 0.762153148651123, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.0007995376363396645, - "train/total_loss": 0.375993937253952 + "train/classification_loss": 0.16828328371047974, + "train/contrastive_loss": 1.9470064640045166, + "train/negative_loss": 1.1108851432800293, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.8361213803291321, + "train/total_loss": 0.5576845407485962 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.10873979330062866, - "train/contrastive_loss": 0.9713344573974609, - "train/negative_loss": 0.6380152702331543, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.33331918716430664, - "train/total_loss": 0.30300670862197876 + "train/classification_loss": 0.11295194178819656, + "train/contrastive_loss": 0.6084390878677368, + "train/negative_loss": 0.36807313561439514, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.24036595225334167, + "train/total_loss": 0.23463976383209229 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.25384631752967834, - "train/contrastive_loss": 2.804058074951172, - "train/negative_loss": 1.367546796798706, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 1.4365112781524658, - "train/total_loss": 0.8146579265594482 + "train/classification_loss": 0.16157159209251404, + "train/contrastive_loss": 3.2931785583496094, + "train/negative_loss": 2.3628172874450684, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.930361270904541, + "train/total_loss": 0.8202073574066162 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.22703032195568085, - "train/contrastive_loss": 2.6415464878082275, - "train/negative_loss": 1.8330538272857666, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.8084927201271057, - "train/total_loss": 0.7553396224975586 + "train/classification_loss": 0.17458026111125946, + "train/contrastive_loss": 5.838879585266113, + "train/negative_loss": 5.838565826416016, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.0003138380707241595, + "train/total_loss": 1.3423562049865723 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.12385080754756927, - "train/contrastive_loss": 1.7815660238265991, - "train/negative_loss": 0.35580775141716003, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 1.4257582426071167, - "train/total_loss": 0.4801639914512634 + "train/classification_loss": 0.15171803534030914, + "train/contrastive_loss": 4.067558765411377, + "train/negative_loss": 3.363480567932129, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.704078197479248, + "train/total_loss": 0.9652298092842102 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.22661571204662323, - "train/contrastive_loss": 3.612557888031006, - "train/negative_loss": 2.7073440551757812, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.9052138924598694, - "train/total_loss": 0.9491273164749146 + "train/classification_loss": 0.2167922407388687, + "train/contrastive_loss": 1.8707177639007568, + "train/negative_loss": 1.3368258476257324, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.5338919162750244, + "train/total_loss": 0.5909358263015747 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.04783937707543373, - "train/contrastive_loss": 0.7350449562072754, - "train/negative_loss": 0.7315009236335754, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.0035440572537481785, - "train/total_loss": 0.19484837353229523 + "train/classification_loss": 0.17223258316516876, + "train/contrastive_loss": 3.3995862007141113, + "train/negative_loss": 1.2540441751480103, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.1455419063568115, + "train/total_loss": 0.8521498441696167 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.21017321944236755, - "train/contrastive_loss": 1.4857150316238403, - "train/negative_loss": 0.9414215087890625, + "train/classification_loss": 0.21649539470672607, + "train/contrastive_loss": 2.692234992980957, + "train/negative_loss": 1.6579599380493164, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.5442935228347778, - "train/total_loss": 0.5073162317276001 + "train/positive_loss": 1.034274935722351, + "train/total_loss": 0.7549424171447754 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.09407968074083328, - "train/contrastive_loss": 1.9122037887573242, - "train/negative_loss": 0.9934880137443542, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 0.9187157154083252, - "train/total_loss": 0.47652044892311096 + "train/classification_loss": 0.10601767152547836, + "train/contrastive_loss": 0.5552632808685303, + "train/negative_loss": 0.5528170466423035, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0024462472647428513, + "train/total_loss": 0.2170703262090683 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.2998703718185425, - "train/contrastive_loss": 4.978605270385742, - "train/negative_loss": 3.1790287494659424, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.799576759338379, - "train/total_loss": 1.2955913543701172 + "train/classification_loss": 0.13864736258983612, + "train/contrastive_loss": 0.9307587742805481, + "train/negative_loss": 0.8569555282592773, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.07380324602127075, + "train/total_loss": 0.324799120426178 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.26762840151786804, - "train/contrastive_loss": 3.0022289752960205, - "train/negative_loss": 1.4940359592437744, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.508193016052246, - "train/total_loss": 0.8680741786956787 + "train/classification_loss": 0.17891521751880646, + "train/contrastive_loss": 4.334905624389648, + "train/negative_loss": 2.221419095993042, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 2.1134867668151855, + "train/total_loss": 1.0458964109420776 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.2136717140674591, - "train/contrastive_loss": 4.045363426208496, - "train/negative_loss": 2.2001047134399414, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.8452585935592651, - "train/total_loss": 1.0227444171905518 + "train/classification_loss": 0.25715795159339905, + "train/contrastive_loss": 3.7047462463378906, + "train/negative_loss": 2.731210708618164, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.9735355973243713, + "train/total_loss": 0.9981071949005127 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.18114164471626282, - "train/contrastive_loss": 1.4979159832000732, - "train/negative_loss": 1.1005526781082153, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.3973632752895355, - "train/total_loss": 0.48072484135627747 + "train/classification_loss": 0.23772229254245758, + "train/contrastive_loss": 2.696626663208008, + "train/negative_loss": 1.357496738433838, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.3391300439834595, + "train/total_loss": 0.7770476341247559 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.18002620339393616, - "train/contrastive_loss": 1.5564134120941162, - "train/negative_loss": 1.0303367376327515, - "train/num_negatives": 44, + "train/classification_loss": 0.20269355177879333, + "train/contrastive_loss": 1.0740340948104858, + "train/negative_loss": 1.0704396963119507, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.003594358218833804, + "train/total_loss": 0.417500376701355 + }, + { + "epoch": 19.69961977186312, + "step": 2600, + "train/classification_loss": 0.2487742155790329, + "train/contrastive_loss": 1.949742317199707, + "train/negative_loss": 0.4492781460285187, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.5260767340660095, - "train/total_loss": 0.49130889773368835 + "train/positive_loss": 1.5004642009735107, + "train/total_loss": 0.6387226581573486 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.21966244280338287, - "train/contrastive_loss": 5.881505966186523, - "train/negative_loss": 4.859813213348389, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.0216927528381348, - "train/total_loss": 1.3959636688232422 + "train/classification_loss": 0.23145778477191925, + "train/contrastive_loss": 2.765547275543213, + "train/negative_loss": 1.4009863138198853, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.3645610809326172, + "train/total_loss": 0.7845672369003296 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.2887398302555084, - "train/contrastive_loss": 3.6235830783843994, - "train/negative_loss": 1.778088092803955, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 1.8454949855804443, - "train/total_loss": 1.0134564638137817 + "train/classification_loss": 0.2037326842546463, + "train/contrastive_loss": 0.8369959592819214, + "train/negative_loss": 0.8364999294281006, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 0.0004960251390002668, + "train/total_loss": 0.37113189697265625 + }, + { + "epoch": 19.69961977186312, + "step": 2600, + "train/classification_loss": 0.1828833818435669, + "train/contrastive_loss": 6.823410987854004, + "train/negative_loss": 5.654133319854736, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 1.1692777872085571, + "train/total_loss": 1.5475655794143677 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.07824526727199554, - "train/contrastive_loss": 1.484796404838562, - "train/negative_loss": 1.3122512102127075, + "train/classification_loss": 0.2502679228782654, + "train/contrastive_loss": 3.6883108615875244, + "train/negative_loss": 1.8328686952590942, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.1725451499223709, - "train/total_loss": 0.37520456314086914 + "train/positive_loss": 1.8554421663284302, + "train/total_loss": 0.9879301190376282 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.09136409312486649, - "train/contrastive_loss": 1.9319281578063965, - "train/negative_loss": 1.5730180740356445, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.35891008377075195, - "train/total_loss": 0.4777497351169586 + "train/classification_loss": 0.22621731460094452, + "train/contrastive_loss": 1.9434621334075928, + "train/negative_loss": 1.1530812978744507, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.7903807759284973, + "train/total_loss": 0.6149097681045532 + }, + { + "epoch": 19.69961977186312, + "step": 2600, + "train/classification_loss": 0.14251872897148132, + "train/contrastive_loss": 2.615741729736328, + "train/negative_loss": 2.615391969680786, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.00034982123179361224, + "train/total_loss": 0.6656670570373535 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.18156307935714722, - "train/contrastive_loss": 2.508430004119873, - "train/negative_loss": 1.2794535160064697, + "train/classification_loss": 0.2560277283191681, + "train/contrastive_loss": 2.444533348083496, + "train/negative_loss": 0.8075316548347473, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 1.637001633644104, + "train/total_loss": 0.7449344396591187 + }, + { + "epoch": 19.69961977186312, + "step": 2600, + "train/classification_loss": 0.19624795019626617, + "train/contrastive_loss": 0.5038001537322998, + "train/negative_loss": 0.5029491186141968, "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.2289766073226929, - "train/total_loss": 0.6832491159439087 + "train/num_positives": 8, + "train/positive_loss": 0.0008510468760505319, + "train/total_loss": 0.2970079779624939 + }, + { + "epoch": 19.69961977186312, + "step": 2600, + "train/classification_loss": 0.20240476727485657, + "train/contrastive_loss": 2.002354621887207, + "train/negative_loss": 1.0796339511871338, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.9227206707000732, + "train/total_loss": 0.6028757095336914 + }, + { + "epoch": 19.69961977186312, + "step": 2600, + "train/classification_loss": 0.17470766603946686, + "train/contrastive_loss": 1.0763269662857056, + "train/negative_loss": 1.0749403238296509, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.0013866957742720842, + "train/total_loss": 0.3899730443954468 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.026815257966518402, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.026815257966518402 - }, - { - "epoch": 24.76555023923445, - "eval_exact_match_accuracy": 0.41148325358851673, - "eval_hamming_loss": 0.05741626794258373, - "eval_loss": 0.65805983543396, - "eval_macro_f1": 0.38484327945112257, - "eval_macro_precision": 0.5044376138304033, - "eval_macro_recall": 0.3475562512586734, - "eval_micro_f1": 0.5984251968503937, - "eval_micro_precision": 0.7169811320754716, - "eval_micro_recall": 0.5135135135135135, - "eval_runtime": 1.059, - "eval_samples_per_second": 197.365, - "eval_steps_per_second": 25.497, + "train/classification_loss": 0.10378376394510269, + "train/contrastive_loss": 1.4150664806365967, + "train/negative_loss": 1.2416012287139893, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.17346520721912384, + "train/total_loss": 0.38679707050323486 + }, + { + "epoch": 19.69961977186312, + "eval_exact_match_accuracy": 0.3435114503816794, + "eval_hamming_loss": 0.05927256398742703, + "eval_loss": 0.653609037399292, + "eval_macro_f1": 0.3125675805307233, + "eval_macro_precision": 0.40963976988644923, + "eval_macro_recall": 0.26882793592042875, + "eval_micro_f1": 0.5849056603773585, + "eval_micro_precision": 0.7622950819672131, + "eval_micro_recall": 0.4744897959183674, + "eval_runtime": 1.313, + "eval_samples_per_second": 199.541, + "eval_steps_per_second": 25.133, "step": 2600 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.0650646984577179, - "train/contrastive_loss": 0.13743744790554047, - "train/negative_loss": 0.13561201095581055, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 0.0018254347378388047, - "train/total_loss": 0.09255218505859375 + "train/classification_loss": 0.09976828098297119, + "train/contrastive_loss": 0.4085346460342407, + "train/negative_loss": 0.19176478683948517, + "train/num_negatives": 40, + "train/num_positives": 14, + "train/positive_loss": 0.21676987409591675, + "train/total_loss": 0.1814752221107483 }, { - "epoch": 24.76555023923445, + "epoch": 19.69961977186312, "step": 2600, - "train/classification_loss": 0.05799359455704689, - "train/contrastive_loss": 0.40042057633399963, - "train/negative_loss": 0.34136006236076355, - "train/num_negatives": 34, - "train/num_positives": 20, - "train/positive_loss": 0.059060510247945786, - "train/total_loss": 0.13807770609855652 + "train/classification_loss": 0.12180746346712112, + "train/contrastive_loss": 0.8859732151031494, + "train/negative_loss": 0.3271149694919586, + "train/num_negatives": 44, + "train/num_positives": 6, + "train/positive_loss": 0.5588582754135132, + "train/total_loss": 0.29900211095809937 }, { - "epoch": 25.239234449760765, - "grad_norm": 3.049635648727417, - "learning_rate": 1.902786259541985e-05, - "loss": 0.274, + "epoch": 20.076045627376427, + "grad_norm": 2.8606159687042236, + "learning_rate": 1.9227314112291354e-05, + "loss": 0.3645, "step": 2650 }, { - "epoch": 25.239234449760765, + "epoch": 20.076045627376427, "step": 2650, - "train/classification_loss": 0.08034232258796692, - "train/contrastive_loss": 0.37516534328460693, - "train/negative_loss": 0.3723876178264618, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.0027777180075645447, - "train/total_loss": 0.1553753912448883 + "train/classification_loss": 0.11522707343101501, + "train/contrastive_loss": 0.25256243348121643, + "train/negative_loss": 0.24772606790065765, + "train/num_negatives": 40, + "train/num_positives": 14, + "train/positive_loss": 0.004836364649236202, + "train/total_loss": 0.16573956608772278 }, { - "epoch": 25.239234449760765, + "epoch": 20.076045627376427, "step": 2650, - "train/classification_loss": 0.06996670365333557, - "train/contrastive_loss": 0.41410166025161743, - "train/negative_loss": 0.1551651507616043, + "train/classification_loss": 0.10388641804456711, + "train/contrastive_loss": 0.05149182677268982, + "train/negative_loss": 0.0514877550303936, "train/num_negatives": 42, - "train/num_positives": 12, - "train/positive_loss": 0.25893649458885193, - "train/total_loss": 0.15278702974319458 + "train/num_positives": 14, + "train/positive_loss": 4.070157501701033e-06, + "train/total_loss": 0.11418478190898895 }, { - "epoch": 25.717703349282296, - "grad_norm": 2.1142029762268066, - "learning_rate": 1.90087786259542e-05, - "loss": 0.3042, + "epoch": 20.456273764258555, + "grad_norm": 4.0417866706848145, + "learning_rate": 1.9212139605462825e-05, + "loss": 0.3456, "step": 2700 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.13080711662769318, - "train/contrastive_loss": 1.4347331523895264, - "train/negative_loss": 0.8294667601585388, + "train/classification_loss": 0.05859171971678734, + "train/contrastive_loss": 1.0164811611175537, + "train/negative_loss": 0.6765485405921936, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.3399326503276825, + "train/total_loss": 0.26188793778419495 + }, + { + "epoch": 20.456273764258555, + "step": 2700, + "train/classification_loss": 0.23217910528182983, + "train/contrastive_loss": 1.1300222873687744, + "train/negative_loss": 1.1280847787857056, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.6052663326263428, - "train/total_loss": 0.41775375604629517 + "train/num_positives": 6, + "train/positive_loss": 0.0019374985713511705, + "train/total_loss": 0.4581835865974426 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.1867365539073944, - "train/contrastive_loss": 3.050706386566162, - "train/negative_loss": 1.5208978652954102, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.5298084020614624, - "train/total_loss": 0.7968778610229492 + "train/classification_loss": 0.10137128829956055, + "train/contrastive_loss": 3.12966251373291, + "train/negative_loss": 2.4063920974731445, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.7232702970504761, + "train/total_loss": 0.7273038029670715 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.1527557224035263, - "train/contrastive_loss": 1.5301684141159058, - "train/negative_loss": 0.5699876546859741, - "train/num_negatives": 44, + "train/classification_loss": 0.19987106323242188, + "train/contrastive_loss": 1.5268120765686035, + "train/negative_loss": 1.082521677017212, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.4442903399467468, + "train/total_loss": 0.5052335262298584 + }, + { + "epoch": 20.456273764258555, + "step": 2700, + "train/classification_loss": 0.09737250208854675, + "train/contrastive_loss": 1.645639181137085, + "train/negative_loss": 1.211020827293396, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.43461838364601135, + "train/total_loss": 0.4265003502368927 + }, + { + "epoch": 20.456273764258555, + "step": 2700, + "train/classification_loss": 0.2113673835992813, + "train/contrastive_loss": 1.602341890335083, + "train/negative_loss": 1.2001748085021973, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.9601807594299316, - "train/total_loss": 0.4587894082069397 + "train/positive_loss": 0.40216702222824097, + "train/total_loss": 0.5318357944488525 + }, + { + "epoch": 20.456273764258555, + "step": 2700, + "train/classification_loss": 0.2585236728191376, + "train/contrastive_loss": 3.022456169128418, + "train/negative_loss": 1.2945339679718018, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.7279222011566162, + "train/total_loss": 0.8630149364471436 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.15141598880290985, - "train/contrastive_loss": 3.7045159339904785, - "train/negative_loss": 1.6485867500305176, + "train/classification_loss": 0.15164175629615784, + "train/contrastive_loss": 2.444613218307495, + "train/negative_loss": 1.01923406124115, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 2.055929183959961, - "train/total_loss": 0.8923192024230957 + "train/positive_loss": 1.4253791570663452, + "train/total_loss": 0.6405644416809082 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.16160018742084503, - "train/contrastive_loss": 3.129535675048828, - "train/negative_loss": 1.2519632577896118, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.8775724172592163, - "train/total_loss": 0.7875072956085205 + "train/classification_loss": 0.1428641676902771, + "train/contrastive_loss": 0.746272087097168, + "train/negative_loss": 0.42117074131965637, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.3251013457775116, + "train/total_loss": 0.2921186089515686 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.17014020681381226, - "train/contrastive_loss": 2.744396209716797, - "train/negative_loss": 1.4884288311004639, - "train/num_negatives": 32, + "train/classification_loss": 0.13964125514030457, + "train/contrastive_loss": 1.8053290843963623, + "train/negative_loss": 0.6603575348854065, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.1449716091156006, + "train/total_loss": 0.5007070899009705 + }, + { + "epoch": 20.456273764258555, + "step": 2700, + "train/classification_loss": 0.19590479135513306, + "train/contrastive_loss": 4.215064525604248, + "train/negative_loss": 3.7378671169281006, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.2559672594070435, - "train/total_loss": 0.7190194725990295 + "train/positive_loss": 0.47719743847846985, + "train/total_loss": 1.0389177799224854 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.13504447042942047, - "train/contrastive_loss": 1.067433476448059, - "train/negative_loss": 0.5776059627532959, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.4898274838924408, - "train/total_loss": 0.34853118658065796 + "train/classification_loss": 0.1608041375875473, + "train/contrastive_loss": 2.997180461883545, + "train/negative_loss": 2.38226318359375, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.6149172782897949, + "train/total_loss": 0.7602402567863464 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.21083272993564606, - "train/contrastive_loss": 0.7917493581771851, - "train/negative_loss": 0.7917453646659851, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 3.97365738535882e-06, - "train/total_loss": 0.3691825866699219 + "train/classification_loss": 0.16649900376796722, + "train/contrastive_loss": 1.387982726097107, + "train/negative_loss": 0.8604939579963684, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.5274887681007385, + "train/total_loss": 0.44409555196762085 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.11570761352777481, - "train/contrastive_loss": 1.0960111618041992, - "train/negative_loss": 0.5694347023963928, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.5265765190124512, - "train/total_loss": 0.3349098563194275 + "train/classification_loss": 0.19241948425769806, + "train/contrastive_loss": 3.6471638679504395, + "train/negative_loss": 1.3898497819900513, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.2573139667510986, + "train/total_loss": 0.9218522310256958 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.25608962774276733, - "train/contrastive_loss": 2.988679885864258, - "train/negative_loss": 1.7704013586044312, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 1.218278408050537, - "train/total_loss": 0.8538256287574768 + "train/classification_loss": 0.22408746182918549, + "train/contrastive_loss": 3.2080368995666504, + "train/negative_loss": 1.9905940294265747, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 1.2174429893493652, + "train/total_loss": 0.8656948804855347 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.2270900160074234, - "train/contrastive_loss": 2.7067649364471436, - "train/negative_loss": 2.0596225261688232, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.6471423506736755, - "train/total_loss": 0.7684429883956909 + "train/classification_loss": 0.11685330420732498, + "train/contrastive_loss": 1.1976516246795654, + "train/negative_loss": 0.6973995566368103, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.5002520084381104, + "train/total_loss": 0.35638362169265747 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.11609337478876114, - "train/contrastive_loss": 2.1363720893859863, - "train/negative_loss": 0.4150735139846802, + "train/classification_loss": 0.1448003202676773, + "train/contrastive_loss": 0.9657196402549744, + "train/negative_loss": 0.7926670908927917, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 1.7212986946105957, - "train/total_loss": 0.5433678030967712 + "train/positive_loss": 0.17305253446102142, + "train/total_loss": 0.33794426918029785 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.22340574860572815, - "train/contrastive_loss": 3.6932291984558105, - "train/negative_loss": 2.0816473960876465, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.611581802368164, - "train/total_loss": 0.9620516300201416 + "train/classification_loss": 0.21254409849643707, + "train/contrastive_loss": 4.7097296714782715, + "train/negative_loss": 3.2952499389648438, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.4144796133041382, + "train/total_loss": 1.1544901132583618 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.04803686961531639, - "train/contrastive_loss": 0.21485953032970428, - "train/negative_loss": 0.2030380815267563, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.011821446940302849, - "train/total_loss": 0.09100877493619919 + "train/classification_loss": 0.25035378336906433, + "train/contrastive_loss": 3.1871848106384277, + "train/negative_loss": 1.8960590362548828, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.291125774383545, + "train/total_loss": 0.8877907991409302 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.1931370049715042, - "train/contrastive_loss": 1.3653607368469238, - "train/negative_loss": 1.0153295993804932, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.3500310778617859, - "train/total_loss": 0.46620917320251465 + "train/classification_loss": 0.21842944622039795, + "train/contrastive_loss": 1.9859956502914429, + "train/negative_loss": 1.5422465801239014, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.4437490403652191, + "train/total_loss": 0.6156286001205444 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.120521180331707, - "train/contrastive_loss": 2.4757754802703857, - "train/negative_loss": 0.9247766733169556, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 1.5509988069534302, - "train/total_loss": 0.6156762838363647 + "train/classification_loss": 0.22519202530384064, + "train/contrastive_loss": 1.065685510635376, + "train/negative_loss": 1.055258870124817, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.010426650755107403, + "train/total_loss": 0.43832913041114807 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.2796458601951599, - "train/contrastive_loss": 3.2161407470703125, - "train/negative_loss": 2.0098705291748047, - "train/num_negatives": 46, + "train/classification_loss": 0.25382131338119507, + "train/contrastive_loss": 2.718515634536743, + "train/negative_loss": 0.49194470047950745, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.2062700986862183, - "train/total_loss": 0.9228740334510803 + "train/positive_loss": 2.2265708446502686, + "train/total_loss": 0.7975244522094727 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.31322067975997925, - "train/contrastive_loss": 2.924895763397217, - "train/negative_loss": 1.2326667308807373, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.692229151725769, - "train/total_loss": 0.8981998562812805 + "train/classification_loss": 0.23242183029651642, + "train/contrastive_loss": 1.6241562366485596, + "train/negative_loss": 1.0119949579238892, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.6121612787246704, + "train/total_loss": 0.5572530627250671 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.20866705477237701, - "train/contrastive_loss": 3.9898767471313477, - "train/negative_loss": 2.0444576740264893, + "train/classification_loss": 0.2155867964029312, + "train/contrastive_loss": 0.9061272144317627, + "train/negative_loss": 0.903988778591156, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.9454190731048584, - "train/total_loss": 1.006642460823059 + "train/num_positives": 8, + "train/positive_loss": 0.002138406503945589, + "train/total_loss": 0.3968122601509094 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.18712829053401947, - "train/contrastive_loss": 1.6004730463027954, - "train/negative_loss": 1.1439001560211182, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.45657292008399963, - "train/total_loss": 0.5072228908538818 + "train/classification_loss": 0.1471550464630127, + "train/contrastive_loss": 4.736666679382324, + "train/negative_loss": 3.9489643573760986, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.7877020835876465, + "train/total_loss": 1.0944883823394775 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.19616861641407013, - "train/contrastive_loss": 1.408005952835083, - "train/negative_loss": 0.9543436765670776, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.4536622166633606, - "train/total_loss": 0.47776979207992554 + "train/classification_loss": 0.23852106928825378, + "train/contrastive_loss": 3.313732862472534, + "train/negative_loss": 1.8370627164840698, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 1.4766701459884644, + "train/total_loss": 0.9012676477432251 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.24998553097248077, - "train/contrastive_loss": 7.527674674987793, - "train/negative_loss": 5.546813488006592, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.9808611869812012, - "train/total_loss": 1.7555205821990967 + "train/classification_loss": 0.2683336138725281, + "train/contrastive_loss": 2.813425064086914, + "train/negative_loss": 1.1091681718826294, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.7042570114135742, + "train/total_loss": 0.8310186266899109 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.28469353914260864, - "train/contrastive_loss": 4.908812046051025, - "train/negative_loss": 2.256782293319702, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.6520297527313232, - "train/total_loss": 1.266455888748169 + "train/classification_loss": 0.16792596876621246, + "train/contrastive_loss": 2.7828493118286133, + "train/negative_loss": 2.7828369140625, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 1.2278632311790716e-05, + "train/total_loss": 0.7244958281517029 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.07549601793289185, - "train/contrastive_loss": 1.6986604928970337, - "train/negative_loss": 1.168606162071228, - "train/num_negatives": 42, + "train/classification_loss": 0.26594796776771545, + "train/contrastive_loss": 2.319901943206787, + "train/negative_loss": 0.8332549333572388, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.5300543308258057, - "train/total_loss": 0.41522812843322754 + "train/positive_loss": 1.486647129058838, + "train/total_loss": 0.7299283742904663 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.10540276765823364, - "train/contrastive_loss": 2.099207639694214, - "train/negative_loss": 1.2550535202026367, - "train/num_negatives": 48, + "train/classification_loss": 0.19985376298427582, + "train/contrastive_loss": 0.29728275537490845, + "train/negative_loss": 0.27876633405685425, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.8441541790962219, - "train/total_loss": 0.5252442955970764 + "train/positive_loss": 0.01851641573011875, + "train/total_loss": 0.2593103051185608 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.16453523933887482, - "train/contrastive_loss": 2.126314163208008, - "train/negative_loss": 0.9598888158798218, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.1664252281188965, - "train/total_loss": 0.589798092842102 + "train/classification_loss": 0.2265719473361969, + "train/contrastive_loss": 3.665708065032959, + "train/negative_loss": 2.2164881229400635, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.4492199420928955, + "train/total_loss": 0.9597135782241821 + }, + { + "epoch": 20.456273764258555, + "step": 2700, + "train/classification_loss": 0.1977507323026657, + "train/contrastive_loss": 1.4005084037780762, + "train/negative_loss": 1.3900343179702759, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.010474035516381264, + "train/total_loss": 0.47785240411758423 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.021147510036826134, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.021147510036826134 - }, - { - "epoch": 25.717703349282296, - "eval_exact_match_accuracy": 0.39712918660287083, - "eval_hamming_loss": 0.058823529411764705, - "eval_loss": 0.6810745000839233, - "eval_macro_f1": 0.367737387624819, - "eval_macro_precision": 0.5016054325104558, - "eval_macro_recall": 0.3269569812476386, - "eval_micro_f1": 0.5743380855397149, - "eval_micro_precision": 0.7230769230769231, - "eval_micro_recall": 0.47635135135135137, - "eval_runtime": 1.0606, - "eval_samples_per_second": 197.052, - "eval_steps_per_second": 25.456, + "train/classification_loss": 0.07871989160776138, + "train/contrastive_loss": 1.2423882484436035, + "train/negative_loss": 1.2298789024353027, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.0125093599781394, + "train/total_loss": 0.3271975517272949 + }, + { + "epoch": 20.456273764258555, + "eval_exact_match_accuracy": 0.3435114503816794, + "eval_hamming_loss": 0.06084418500224517, + "eval_loss": 0.6413367986679077, + "eval_macro_f1": 0.3211021745582854, + "eval_macro_precision": 0.4018734241334861, + "eval_macro_recall": 0.28636799152325687, + "eval_micro_f1": 0.573228346456693, + "eval_micro_precision": 0.7489711934156379, + "eval_micro_recall": 0.4642857142857143, + "eval_runtime": 1.3082, + "eval_samples_per_second": 200.277, + "eval_steps_per_second": 25.226, "step": 2700 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.024969661608338356, - "train/contrastive_loss": 0.03595958277583122, - "train/negative_loss": 0.03420748934149742, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.0017520916881039739, - "train/total_loss": 0.03216157853603363 + "train/classification_loss": 0.12659543752670288, + "train/contrastive_loss": 0.9059181213378906, + "train/negative_loss": 0.2168639451265335, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.6890541911125183, + "train/total_loss": 0.30777907371520996 }, { - "epoch": 25.717703349282296, + "epoch": 20.456273764258555, "step": 2700, - "train/classification_loss": 0.07281262427568436, - "train/contrastive_loss": 0.11026860773563385, - "train/negative_loss": 0.10937732458114624, + "train/classification_loss": 0.081440769135952, + "train/contrastive_loss": 0.7406312227249146, + "train/negative_loss": 0.7396841645240784, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 0.0008912809425964952, - "train/total_loss": 0.09486635029315948 + "train/positive_loss": 0.0009470360237173736, + "train/total_loss": 0.2295670211315155 }, { - "epoch": 26.19138755980861, - "grad_norm": 2.8260304927825928, - "learning_rate": 1.898969465648855e-05, - "loss": 0.2847, + "epoch": 20.836501901140686, + "grad_norm": 7.412676811218262, + "learning_rate": 1.9196965098634297e-05, + "loss": 0.3568, "step": 2750 }, { - "epoch": 26.19138755980861, + "epoch": 20.836501901140686, "step": 2750, - "train/classification_loss": 0.09658903628587723, - "train/contrastive_loss": 0.1094699427485466, - "train/negative_loss": 0.10695277154445648, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.0025171716697514057, - "train/total_loss": 0.11848302185535431 + "train/classification_loss": 0.10550791770219803, + "train/contrastive_loss": 1.8021348714828491, + "train/negative_loss": 0.7133928537368774, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 1.0887420177459717, + "train/total_loss": 0.4659349024295807 }, { - "epoch": 26.19138755980861, + "epoch": 20.836501901140686, "step": 2750, - "train/classification_loss": 0.038197144865989685, - "train/contrastive_loss": 0.6622310876846313, - "train/negative_loss": 0.6610731482505798, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 0.0011579312849789858, - "train/total_loss": 0.17064335942268372 + "train/classification_loss": 0.07917799800634384, + "train/contrastive_loss": 0.2591489851474762, + "train/negative_loss": 0.2556912302970886, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.0034577439073473215, + "train/total_loss": 0.13100779056549072 }, { - "epoch": 26.669856459330145, - "grad_norm": 2.138977289199829, - "learning_rate": 1.8970610687022903e-05, - "loss": 0.289, + "epoch": 21.212927756653993, + "grad_norm": 12.66663932800293, + "learning_rate": 1.9181790591805768e-05, + "loss": 0.331, "step": 2800 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.15551839768886566, - "train/contrastive_loss": 3.2606165409088135, - "train/negative_loss": 2.661928653717041, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.5986879467964172, - "train/total_loss": 0.8076417446136475 + "train/classification_loss": 0.05719170719385147, + "train/contrastive_loss": 0.5844509601593018, + "train/negative_loss": 0.46072012186050415, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.12373083829879761, + "train/total_loss": 0.17408189177513123 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.17370271682739258, - "train/contrastive_loss": 2.5138134956359863, - "train/negative_loss": 1.5766963958740234, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.9371170997619629, - "train/total_loss": 0.6764654517173767 + "train/classification_loss": 0.22535613179206848, + "train/contrastive_loss": 1.2418160438537598, + "train/negative_loss": 1.1874825954437256, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.054333414882421494, + "train/total_loss": 0.47371935844421387 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.1335412859916687, - "train/contrastive_loss": 1.101709246635437, - "train/negative_loss": 0.27200984954833984, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.8296993970870972, - "train/total_loss": 0.35388314723968506 + "train/classification_loss": 0.11702582985162735, + "train/contrastive_loss": 4.285498142242432, + "train/negative_loss": 3.6322317123413086, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.6532663702964783, + "train/total_loss": 0.9741255044937134 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.14799922704696655, - "train/contrastive_loss": 3.3204941749572754, - "train/negative_loss": 2.0660295486450195, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.2544645071029663, - "train/total_loss": 0.8120980858802795 + "train/classification_loss": 0.2045333832502365, + "train/contrastive_loss": 1.691235899925232, + "train/negative_loss": 1.0561021566390991, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.6351337432861328, + "train/total_loss": 0.5427805781364441 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.136515274643898, - "train/contrastive_loss": 3.6886982917785645, - "train/negative_loss": 1.9920003414154053, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.6966979503631592, - "train/total_loss": 0.8742549419403076 + "train/classification_loss": 0.082425057888031, + "train/contrastive_loss": 0.9789115786552429, + "train/negative_loss": 0.9717419743537903, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.007169579155743122, + "train/total_loss": 0.27820736169815063 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.1092243567109108, - "train/contrastive_loss": 2.342686414718628, - "train/negative_loss": 1.2416682243347168, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.1010181903839111, - "train/total_loss": 0.5777616500854492 + "train/classification_loss": 0.19986069202423096, + "train/contrastive_loss": 1.5217393636703491, + "train/negative_loss": 1.3838598728179932, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.13787949085235596, + "train/total_loss": 0.5042085647583008 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.12427968531847, - "train/contrastive_loss": 1.2397325038909912, - "train/negative_loss": 0.6874176263809204, + "train/classification_loss": 0.23955386877059937, + "train/contrastive_loss": 3.2377071380615234, + "train/negative_loss": 1.304320216178894, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.5523148775100708, - "train/total_loss": 0.37222617864608765 + "train/num_positives": 8, + "train/positive_loss": 1.9333869218826294, + "train/total_loss": 0.8870953321456909 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.22248636186122894, - "train/contrastive_loss": 1.7367368936538696, - "train/negative_loss": 1.7366600036621094, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 7.691838254686445e-05, - "train/total_loss": 0.5698337554931641 + "train/classification_loss": 0.18278683722019196, + "train/contrastive_loss": 1.1992104053497314, + "train/negative_loss": 1.0023913383483887, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.19681905210018158, + "train/total_loss": 0.4226289391517639 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.14542455971240997, - "train/contrastive_loss": 0.9438638687133789, - "train/negative_loss": 0.6258833408355713, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.3179805278778076, - "train/total_loss": 0.33419734239578247 + "train/classification_loss": 0.1220925822854042, + "train/contrastive_loss": 0.5647119283676147, + "train/negative_loss": 0.31393754482269287, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.2507743537425995, + "train/total_loss": 0.2350349724292755 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.24655768275260925, - "train/contrastive_loss": 3.6754279136657715, - "train/negative_loss": 1.0723466873168945, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.603081226348877, - "train/total_loss": 0.9816433191299438 + "train/classification_loss": 0.1614987701177597, + "train/contrastive_loss": 2.064613103866577, + "train/negative_loss": 1.9367752075195312, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.12783782184123993, + "train/total_loss": 0.5744214057922363 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.2547297775745392, - "train/contrastive_loss": 3.634153127670288, - "train/negative_loss": 2.732689619064331, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.9014635682106018, - "train/total_loss": 0.981560468673706 + "train/classification_loss": 0.1698465645313263, + "train/contrastive_loss": 5.36143684387207, + "train/negative_loss": 5.360783100128174, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.0006538458983413875, + "train/total_loss": 1.2421339750289917 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.11301751434803009, - "train/contrastive_loss": 1.3884385824203491, - "train/negative_loss": 0.5440245866775513, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.8444139957427979, - "train/total_loss": 0.3907052278518677 + "train/classification_loss": 0.1707490086555481, + "train/contrastive_loss": 4.149271011352539, + "train/negative_loss": 3.783967971801758, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.3653032183647156, + "train/total_loss": 1.000603199005127 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.23935219645500183, - "train/contrastive_loss": 4.0582966804504395, - "train/negative_loss": 2.93163800239563, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.1266587972640991, - "train/total_loss": 1.051011562347412 + "train/classification_loss": 0.17349164187908173, + "train/contrastive_loss": 2.1297483444213867, + "train/negative_loss": 1.2499650716781616, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.8797832131385803, + "train/total_loss": 0.5994412899017334 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.04545772820711136, - "train/contrastive_loss": 0.1281871497631073, - "train/negative_loss": 0.1277613490819931, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.00042579800356179476, - "train/total_loss": 0.07109516113996506 + "train/classification_loss": 0.19602665305137634, + "train/contrastive_loss": 3.258887767791748, + "train/negative_loss": 1.3412584066390991, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.9176294803619385, + "train/total_loss": 0.8478041887283325 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.18430829048156738, - "train/contrastive_loss": 1.7082706689834595, - "train/negative_loss": 0.7639150023460388, + "train/classification_loss": 0.21568870544433594, + "train/contrastive_loss": 3.2895805835723877, + "train/negative_loss": 2.0955073833465576, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.9443556666374207, - "train/total_loss": 0.5259624719619751 + "train/positive_loss": 1.19407320022583, + "train/total_loss": 0.8736048340797424 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.09643382579088211, - "train/contrastive_loss": 2.600656747817993, - "train/negative_loss": 1.5425114631652832, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 1.05814528465271, - "train/total_loss": 0.6165651679039001 + "train/classification_loss": 0.10292685031890869, + "train/contrastive_loss": 0.8443986177444458, + "train/negative_loss": 0.8442986607551575, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 9.994461288442835e-05, + "train/total_loss": 0.27180659770965576 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.3238472640514374, - "train/contrastive_loss": 4.1576690673828125, - "train/negative_loss": 2.112665891647339, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 2.0450031757354736, - "train/total_loss": 1.1553810834884644 + "train/classification_loss": 0.14201106131076813, + "train/contrastive_loss": 1.0048683881759644, + "train/negative_loss": 0.8609304428100586, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.14393794536590576, + "train/total_loss": 0.34298473596572876 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.2968192994594574, - "train/contrastive_loss": 3.408097505569458, - "train/negative_loss": 2.6538779735565186, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.7542195320129395, - "train/total_loss": 0.9784388542175293 + "train/classification_loss": 0.20021870732307434, + "train/contrastive_loss": 5.5714521408081055, + "train/negative_loss": 2.9886422157287598, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 2.5828099250793457, + "train/total_loss": 1.3145091533660889 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.2341563105583191, - "train/contrastive_loss": 3.9285879135131836, - "train/negative_loss": 2.1788201332092285, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.7497676610946655, - "train/total_loss": 1.019873857498169 + "train/classification_loss": 0.2678006589412689, + "train/contrastive_loss": 4.003244876861572, + "train/negative_loss": 2.50020432472229, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.5030405521392822, + "train/total_loss": 1.06844961643219 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.1752365678548813, - "train/contrastive_loss": 1.9965174198150635, - "train/negative_loss": 1.9205322265625, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.07598515599966049, - "train/total_loss": 0.5745400786399841 + "train/classification_loss": 0.2160794883966446, + "train/contrastive_loss": 2.7545619010925293, + "train/negative_loss": 1.4096320867538452, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.344929814338684, + "train/total_loss": 0.7669918537139893 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.1858842521905899, - "train/contrastive_loss": 0.7194234728813171, - "train/negative_loss": 0.5113207101821899, - "train/num_negatives": 44, + "train/classification_loss": 0.21002978086471558, + "train/contrastive_loss": 0.555666983127594, + "train/negative_loss": 0.5516024231910706, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.004064539447426796, + "train/total_loss": 0.3211631774902344 + }, + { + "epoch": 21.212927756653993, + "step": 2800, + "train/classification_loss": 0.21293118596076965, + "train/contrastive_loss": 1.892226219177246, + "train/negative_loss": 0.33868739008903503, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.208102747797966, - "train/total_loss": 0.32976895570755005 + "train/positive_loss": 1.5535387992858887, + "train/total_loss": 0.5913764238357544 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.25368842482566833, - "train/contrastive_loss": 7.388097286224365, - "train/negative_loss": 6.072396278381348, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.3157011270523071, - "train/total_loss": 1.7313079833984375 + "train/classification_loss": 0.2307271510362625, + "train/contrastive_loss": 1.5698111057281494, + "train/negative_loss": 1.1828062534332275, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.38700491189956665, + "train/total_loss": 0.5446893572807312 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.28240782022476196, - "train/contrastive_loss": 4.564754486083984, - "train/negative_loss": 2.3067080974578857, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.2580461502075195, - "train/total_loss": 1.1953587532043457 + "train/classification_loss": 0.23744572699069977, + "train/contrastive_loss": 1.993817925453186, + "train/negative_loss": 1.9937162399291992, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 0.00010169903544010594, + "train/total_loss": 0.6362093091011047 + }, + { + "epoch": 21.212927756653993, + "step": 2800, + "train/classification_loss": 0.1619071364402771, + "train/contrastive_loss": 5.203372955322266, + "train/negative_loss": 4.4633684158325195, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.7400047183036804, + "train/total_loss": 1.2025816440582275 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.066621333360672, - "train/contrastive_loss": 2.08371639251709, - "train/negative_loss": 1.713018774986267, + "train/classification_loss": 0.25086191296577454, + "train/contrastive_loss": 3.743448257446289, + "train/negative_loss": 2.355262279510498, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.3706977069377899, - "train/total_loss": 0.48336461186408997 + "train/positive_loss": 1.3881858587265015, + "train/total_loss": 0.99955153465271 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.09921960532665253, - "train/contrastive_loss": 1.504751443862915, - "train/negative_loss": 1.4596712589263916, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.04508022218942642, - "train/total_loss": 0.40016990900039673 + "train/classification_loss": 0.2125963270664215, + "train/contrastive_loss": 1.669679880142212, + "train/negative_loss": 1.0572137832641602, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.612466037273407, + "train/total_loss": 0.5465322732925415 + }, + { + "epoch": 21.212927756653993, + "step": 2800, + "train/classification_loss": 0.1488211303949356, + "train/contrastive_loss": 2.6756420135498047, + "train/negative_loss": 2.6756255626678467, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 1.6451016563223675e-05, + "train/total_loss": 0.6839495301246643 + }, + { + "epoch": 21.212927756653993, + "step": 2800, + "train/classification_loss": 0.27302852272987366, + "train/contrastive_loss": 2.4143848419189453, + "train/negative_loss": 0.819159746170044, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 1.595225214958191, + "train/total_loss": 0.7559055089950562 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.2002348154783249, - "train/contrastive_loss": 2.3184826374053955, - "train/negative_loss": 1.2333152294158936, + "train/classification_loss": 0.19461506605148315, + "train/contrastive_loss": 0.5699907541275024, + "train/negative_loss": 0.5150886178016663, "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.085167407989502, - "train/total_loss": 0.6639313697814941 + "train/num_positives": 8, + "train/positive_loss": 0.054902154952287674, + "train/total_loss": 0.30861321091651917 + }, + { + "epoch": 21.212927756653993, + "step": 2800, + "train/classification_loss": 0.2079331874847412, + "train/contrastive_loss": 3.302799940109253, + "train/negative_loss": 2.262810707092285, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.0399892330169678, + "train/total_loss": 0.8684931993484497 + }, + { + "epoch": 21.212927756653993, + "step": 2800, + "train/classification_loss": 0.20371410250663757, + "train/contrastive_loss": 1.8009443283081055, + "train/negative_loss": 1.620246410369873, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.1806979477405548, + "train/total_loss": 0.5639029741287231 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.020845936611294746, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.020845936611294746 - }, - { - "epoch": 26.669856459330145, - "eval_exact_match_accuracy": 0.41626794258373206, - "eval_hamming_loss": 0.05741626794258373, - "eval_loss": 0.7093453407287598, - "eval_macro_f1": 0.41057754265571333, - "eval_macro_precision": 0.5696738465501143, - "eval_macro_recall": 0.3607206685407377, - "eval_micro_f1": 0.6076923076923076, - "eval_micro_precision": 0.7053571428571429, - "eval_micro_recall": 0.5337837837837838, - "eval_runtime": 1.0499, - "eval_samples_per_second": 199.063, - "eval_steps_per_second": 25.716, + "train/classification_loss": 0.08969129621982574, + "train/contrastive_loss": 1.8944575786590576, + "train/negative_loss": 1.4483088254928589, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.44614875316619873, + "train/total_loss": 0.46858280897140503 + }, + { + "epoch": 21.212927756653993, + "eval_exact_match_accuracy": 0.32061068702290074, + "eval_hamming_loss": 0.06286484059272564, + "eval_loss": 0.6647034883499146, + "eval_macro_f1": 0.3141031068511214, + "eval_macro_precision": 0.40793908982522226, + "eval_macro_recall": 0.27380092559113933, + "eval_micro_f1": 0.5679012345679012, + "eval_micro_precision": 0.71875, + "eval_micro_recall": 0.46938775510204084, + "eval_runtime": 1.3165, + "eval_samples_per_second": 199.012, + "eval_steps_per_second": 25.066, "step": 2800 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.07552523165941238, - "train/contrastive_loss": 0.3778338134288788, - "train/negative_loss": 0.29860928654670715, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.07922452688217163, - "train/total_loss": 0.15109199285507202 + "train/classification_loss": 0.046241775155067444, + "train/contrastive_loss": 0.16929829120635986, + "train/negative_loss": 0.16929320991039276, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 5.08628181705717e-06, + "train/total_loss": 0.08010143041610718 }, { - "epoch": 26.669856459330145, + "epoch": 21.212927756653993, "step": 2800, - "train/classification_loss": 0.05359485745429993, - "train/contrastive_loss": 0.18125492334365845, - "train/negative_loss": 0.17896035313606262, - "train/num_negatives": 30, - "train/num_positives": 26, - "train/positive_loss": 0.0022945657838135958, - "train/total_loss": 0.08984584361314774 + "train/classification_loss": 0.09662207216024399, + "train/contrastive_loss": 1.206395149230957, + "train/negative_loss": 0.3859325051307678, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.820462703704834, + "train/total_loss": 0.33790111541748047 }, { - "epoch": 27.14354066985646, - "grad_norm": 3.190964460372925, - "learning_rate": 1.8951526717557254e-05, - "loss": 0.2827, + "epoch": 21.59315589353612, + "grad_norm": 2.34725284576416, + "learning_rate": 1.916661608497724e-05, + "loss": 0.3346, "step": 2850 }, { - "epoch": 27.14354066985646, + "epoch": 21.59315589353612, "step": 2850, - "train/classification_loss": 0.04475223273038864, - "train/contrastive_loss": 0.07806459069252014, - "train/negative_loss": 0.07626266032457352, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.0018019293202087283, - "train/total_loss": 0.06036515161395073 + "train/classification_loss": 0.10153967887163162, + "train/contrastive_loss": 0.2588789165019989, + "train/negative_loss": 0.16841836273670197, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.09046054631471634, + "train/total_loss": 0.1533154547214508 }, { - "epoch": 27.14354066985646, + "epoch": 21.59315589353612, "step": 2850, - "train/classification_loss": 0.08392240852117538, - "train/contrastive_loss": 0.31907492876052856, - "train/negative_loss": 0.14141422510147095, - "train/num_negatives": 36, - "train/num_positives": 20, - "train/positive_loss": 0.1776607185602188, - "train/total_loss": 0.14773739874362946 + "train/classification_loss": 0.07756659388542175, + "train/contrastive_loss": 0.14228606224060059, + "train/negative_loss": 0.138946533203125, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.0033395341597497463, + "train/total_loss": 0.10602380335330963 }, { - "epoch": 27.62200956937799, - "grad_norm": 2.986572504043579, - "learning_rate": 1.8932442748091605e-05, - "loss": 0.265, + "epoch": 21.973384030418252, + "grad_norm": 6.23289155960083, + "learning_rate": 1.915144157814871e-05, + "loss": 0.3419, "step": 2900 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, + "step": 2900, + "train/classification_loss": 0.041735704988241196, + "train/contrastive_loss": 0.21933475136756897, + "train/negative_loss": 0.16358120739459991, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.055753543972969055, + "train/total_loss": 0.08560265600681305 + }, + { + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.1350754350423813, - "train/contrastive_loss": 1.7961560487747192, - "train/negative_loss": 1.2230117321014404, + "train/classification_loss": 0.24991591274738312, + "train/contrastive_loss": 1.3975976705551147, + "train/negative_loss": 1.3855326175689697, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.5731443166732788, - "train/total_loss": 0.49430662393569946 + "train/num_positives": 6, + "train/positive_loss": 0.012065049260854721, + "train/total_loss": 0.5294354557991028 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.19461657106876373, - "train/contrastive_loss": 2.593592643737793, - "train/negative_loss": 1.6091980934143066, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 0.9843944907188416, - "train/total_loss": 0.7133350968360901 + "train/classification_loss": 0.09108658134937286, + "train/contrastive_loss": 2.682494640350342, + "train/negative_loss": 2.0981104373931885, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.5843840837478638, + "train/total_loss": 0.6275855302810669 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.1589219719171524, - "train/contrastive_loss": 1.4979604482650757, - "train/negative_loss": 0.5614158511161804, - "train/num_negatives": 44, + "train/classification_loss": 0.2313794642686844, + "train/contrastive_loss": 2.2463698387145996, + "train/negative_loss": 1.4202908277511597, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.8260790109634399, + "train/total_loss": 0.68065345287323 + }, + { + "epoch": 21.973384030418252, + "step": 2900, + "train/classification_loss": 0.10731256753206253, + "train/contrastive_loss": 1.8004977703094482, + "train/negative_loss": 1.1448204517364502, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.655677318572998, + "train/total_loss": 0.4674121141433716 + }, + { + "epoch": 21.973384030418252, + "step": 2900, + "train/classification_loss": 0.20372197031974792, + "train/contrastive_loss": 1.3511736392974854, + "train/negative_loss": 1.0137407779693604, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.9365445971488953, - "train/total_loss": 0.45851409435272217 + "train/positive_loss": 0.3374328315258026, + "train/total_loss": 0.4739567041397095 + }, + { + "epoch": 21.973384030418252, + "step": 2900, + "train/classification_loss": 0.2330053448677063, + "train/contrastive_loss": 3.7396810054779053, + "train/negative_loss": 1.5073983669281006, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 2.2322826385498047, + "train/total_loss": 0.9809415340423584 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.14964258670806885, - "train/contrastive_loss": 3.294684886932373, - "train/negative_loss": 1.6775013208389282, + "train/classification_loss": 0.19800832867622375, + "train/contrastive_loss": 1.3377667665481567, + "train/negative_loss": 0.7210788726806641, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.6171835660934448, - "train/total_loss": 0.8085795640945435 + "train/positive_loss": 0.6166878938674927, + "train/total_loss": 0.4655616879463196 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.1440880447626114, - "train/contrastive_loss": 2.5090432167053223, - "train/negative_loss": 1.508595585823059, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.0004475116729736, - "train/total_loss": 0.6458966732025146 + "train/classification_loss": 0.1386839598417282, + "train/contrastive_loss": 0.49289727210998535, + "train/negative_loss": 0.16484612226486206, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.3280511498451233, + "train/total_loss": 0.23726341128349304 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.13252127170562744, - "train/contrastive_loss": 1.6155930757522583, - "train/negative_loss": 0.3971262276172638, - "train/num_negatives": 32, + "train/classification_loss": 0.16388054192066193, + "train/contrastive_loss": 1.9985064268112183, + "train/negative_loss": 1.9362869262695312, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.062219519168138504, + "train/total_loss": 0.5635818243026733 + }, + { + "epoch": 21.973384030418252, + "step": 2900, + "train/classification_loss": 0.1713174730539322, + "train/contrastive_loss": 4.6073737144470215, + "train/negative_loss": 4.5884623527526855, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.218466877937317, - "train/total_loss": 0.45563989877700806 + "train/positive_loss": 0.018911141902208328, + "train/total_loss": 1.092792272567749 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.12430979311466217, - "train/contrastive_loss": 1.1635749340057373, - "train/negative_loss": 0.4980207681655884, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.6655541658401489, - "train/total_loss": 0.35702478885650635 + "train/classification_loss": 0.16178852319717407, + "train/contrastive_loss": 3.9701623916625977, + "train/negative_loss": 3.4444117546081543, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.5257506370544434, + "train/total_loss": 0.9558210372924805 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.20326359570026398, - "train/contrastive_loss": 1.0254210233688354, - "train/negative_loss": 1.0251281261444092, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.0002928848844021559, - "train/total_loss": 0.4083477854728699 + "train/classification_loss": 0.1936434805393219, + "train/contrastive_loss": 2.104376792907715, + "train/negative_loss": 1.2255613803863525, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.8788154125213623, + "train/total_loss": 0.6145188808441162 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.10622522234916687, - "train/contrastive_loss": 0.7751947641372681, - "train/negative_loss": 0.5668174624443054, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.20837730169296265, - "train/total_loss": 0.2612641751766205 + "train/classification_loss": 0.1560455560684204, + "train/contrastive_loss": 3.2568979263305664, + "train/negative_loss": 1.173117995262146, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.08378005027771, + "train/total_loss": 0.8074251413345337 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.23456262052059174, - "train/contrastive_loss": 2.8382279872894287, - "train/negative_loss": 0.7066481113433838, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.131579875946045, - "train/total_loss": 0.8022082448005676 + "train/classification_loss": 0.2245616614818573, + "train/contrastive_loss": 3.139453887939453, + "train/negative_loss": 2.1164968013763428, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 1.0229570865631104, + "train/total_loss": 0.8524523973464966 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.2301967740058899, - "train/contrastive_loss": 2.850341320037842, - "train/negative_loss": 1.9484007358551025, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.9019404649734497, - "train/total_loss": 0.8002650737762451 + "train/classification_loss": 0.10528393089771271, + "train/contrastive_loss": 0.944165825843811, + "train/negative_loss": 0.9429017305374146, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.001264124526642263, + "train/total_loss": 0.2941170930862427 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.11944998055696487, - "train/contrastive_loss": 1.5868878364562988, - "train/negative_loss": 0.29893797636032104, + "train/classification_loss": 0.14484301209449768, + "train/contrastive_loss": 0.805377721786499, + "train/negative_loss": 0.6296384930610657, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 1.287949800491333, - "train/total_loss": 0.43682754039764404 + "train/positive_loss": 0.17573922872543335, + "train/total_loss": 0.3059185743331909 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.23720228672027588, - "train/contrastive_loss": 3.9003961086273193, - "train/negative_loss": 2.399158477783203, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.5012376308441162, - "train/total_loss": 1.0172815322875977 + "train/classification_loss": 0.16981954872608185, + "train/contrastive_loss": 2.6723828315734863, + "train/negative_loss": 1.5600494146347046, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.1123335361480713, + "train/total_loss": 0.7042961120605469 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.03820775821805, - "train/contrastive_loss": 0.5532017946243286, - "train/negative_loss": 0.5523915886878967, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.0008102335268631577, - "train/total_loss": 0.14884811639785767 + "train/classification_loss": 0.2690731883049011, + "train/contrastive_loss": 3.749177932739258, + "train/negative_loss": 2.8107669353485107, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.9384109973907471, + "train/total_loss": 1.0189087390899658 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.18541313707828522, - "train/contrastive_loss": 1.1238718032836914, - "train/negative_loss": 0.8570102453231812, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.2668614983558655, - "train/total_loss": 0.4101874828338623 + "train/classification_loss": 0.23705364763736725, + "train/contrastive_loss": 2.2228739261627197, + "train/negative_loss": 1.3422768115997314, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.8805971145629883, + "train/total_loss": 0.6816284656524658 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.11672545969486237, - "train/contrastive_loss": 2.335811138153076, - "train/negative_loss": 1.313269019126892, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 1.0225422382354736, - "train/total_loss": 0.5838876962661743 + "train/classification_loss": 0.23124369978904724, + "train/contrastive_loss": 0.7072563767433167, + "train/negative_loss": 0.5035107731819153, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.20374560356140137, + "train/total_loss": 0.3726949691772461 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.2792483866214752, - "train/contrastive_loss": 2.964496374130249, - "train/negative_loss": 1.6006083488464355, - "train/num_negatives": 46, + "train/classification_loss": 0.24296656250953674, + "train/contrastive_loss": 2.4146335124969482, + "train/negative_loss": 0.4423455595970154, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.3638880252838135, - "train/total_loss": 0.8721476793289185 + "train/positive_loss": 1.9722880125045776, + "train/total_loss": 0.7258932590484619 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.2744080126285553, - "train/contrastive_loss": 3.1693954467773438, - "train/negative_loss": 1.8714386224746704, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.2979568243026733, - "train/total_loss": 0.9082871675491333 + "train/classification_loss": 0.24033145606517792, + "train/contrastive_loss": 2.5493290424346924, + "train/negative_loss": 1.2036656141281128, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.3456634283065796, + "train/total_loss": 0.7501972913742065 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.19467777013778687, - "train/contrastive_loss": 3.148101329803467, - "train/negative_loss": 1.7097265720367432, + "train/classification_loss": 0.1959974616765976, + "train/contrastive_loss": 0.3057321310043335, + "train/negative_loss": 0.30572938919067383, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.4383748769760132, - "train/total_loss": 0.8242980241775513 + "train/num_positives": 8, + "train/positive_loss": 2.741818889262504e-06, + "train/total_loss": 0.25714388489723206 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.17163652181625366, - "train/contrastive_loss": 0.8299056887626648, - "train/negative_loss": 0.5914888978004456, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.23841679096221924, - "train/total_loss": 0.3376176655292511 + "train/classification_loss": 0.15147237479686737, + "train/contrastive_loss": 5.759326934814453, + "train/negative_loss": 5.109811305999756, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.6495158672332764, + "train/total_loss": 1.3033376932144165 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.1937088668346405, - "train/contrastive_loss": 1.2692232131958008, - "train/negative_loss": 0.8977059721946716, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.3715173006057739, - "train/total_loss": 0.44755351543426514 + "train/classification_loss": 0.2531725764274597, + "train/contrastive_loss": 3.989531993865967, + "train/negative_loss": 2.7101848125457764, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 1.2793470621109009, + "train/total_loss": 1.0510790348052979 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.21626080572605133, - "train/contrastive_loss": 6.468845367431641, - "train/negative_loss": 4.947849273681641, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.5209962129592896, - "train/total_loss": 1.510029911994934 + "train/classification_loss": 0.20137447118759155, + "train/contrastive_loss": 2.0978450775146484, + "train/negative_loss": 1.0433108806610107, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.0545343160629272, + "train/total_loss": 0.6209434866905212 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.25402191281318665, - "train/contrastive_loss": 4.00527811050415, - "train/negative_loss": 1.8291914463043213, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.176086664199829, - "train/total_loss": 1.0550775527954102 + "train/classification_loss": 0.1260501891374588, + "train/contrastive_loss": 2.165342092514038, + "train/negative_loss": 2.0606744289398193, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.10466764867305756, + "train/total_loss": 0.5591186285018921 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.05897409841418266, - "train/contrastive_loss": 0.8137833476066589, - "train/negative_loss": 0.6802762746810913, - "train/num_negatives": 42, + "train/classification_loss": 0.2801283597946167, + "train/contrastive_loss": 2.8385696411132812, + "train/negative_loss": 0.8390555381774902, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.13350708782672882, - "train/total_loss": 0.2217307686805725 + "train/positive_loss": 1.9995142221450806, + "train/total_loss": 0.847842276096344 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.10981236398220062, - "train/contrastive_loss": 2.404169797897339, - "train/negative_loss": 1.1630936861038208, - "train/num_negatives": 48, + "train/classification_loss": 0.19724705815315247, + "train/contrastive_loss": 0.7248622179031372, + "train/negative_loss": 0.48744356632232666, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 1.241076111793518, - "train/total_loss": 0.5906463265419006 + "train/positive_loss": 0.23741862177848816, + "train/total_loss": 0.3422195017337799 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.17037135362625122, - "train/contrastive_loss": 1.800424575805664, - "train/negative_loss": 0.9456534385681152, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.854771077632904, - "train/total_loss": 0.5304563045501709 + "train/classification_loss": 0.2035117894411087, + "train/contrastive_loss": 1.7434511184692383, + "train/negative_loss": 0.9907554984092712, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.7526956796646118, + "train/total_loss": 0.552202045917511 + }, + { + "epoch": 21.973384030418252, + "step": 2900, + "train/classification_loss": 0.2002890706062317, + "train/contrastive_loss": 1.7123980522155762, + "train/negative_loss": 1.7086163759231567, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.0037816851399838924, + "train/total_loss": 0.5427687168121338 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.015245171263813972, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.015245171263813972 - }, - { - "epoch": 27.62200956937799, - "eval_exact_match_accuracy": 0.44976076555023925, - "eval_hamming_loss": 0.05432029271038559, - "eval_loss": 0.6163508296012878, - "eval_macro_f1": 0.4533317319936565, - "eval_macro_precision": 0.5668586125876253, - "eval_macro_recall": 0.4124797155939024, - "eval_micro_f1": 0.6237816764132553, - "eval_micro_precision": 0.7373271889400922, - "eval_micro_recall": 0.5405405405405406, - "eval_runtime": 1.0546, - "eval_samples_per_second": 198.188, - "eval_steps_per_second": 25.603, + "train/classification_loss": 0.0939951166510582, + "train/contrastive_loss": 2.120361566543579, + "train/negative_loss": 2.105099678039551, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.015261889435350895, + "train/total_loss": 0.5180674195289612 + }, + { + "epoch": 21.973384030418252, + "eval_exact_match_accuracy": 0.3320610687022901, + "eval_hamming_loss": 0.06196677144140099, + "eval_loss": 0.6337057948112488, + "eval_macro_f1": 0.3056526842242511, + "eval_macro_precision": 0.3889522010143292, + "eval_macro_recall": 0.263708610689339, + "eval_micro_f1": 0.56875, + "eval_micro_precision": 0.7338709677419355, + "eval_micro_recall": 0.4642857142857143, + "eval_runtime": 1.3162, + "eval_samples_per_second": 199.055, + "eval_steps_per_second": 25.072, "step": 2900 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.08472827076911926, - "train/contrastive_loss": 0.4758649468421936, - "train/negative_loss": 0.44248348474502563, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.03338146582245827, - "train/total_loss": 0.17990127205848694 + "train/classification_loss": 0.06108995899558067, + "train/contrastive_loss": 0.10891393572092056, + "train/negative_loss": 0.10881856083869934, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 9.53757917159237e-05, + "train/total_loss": 0.08287274837493896 }, { - "epoch": 27.62200956937799, + "epoch": 21.973384030418252, "step": 2900, - "train/classification_loss": 0.09065502136945724, - "train/contrastive_loss": 1.440433382987976, - "train/negative_loss": 1.0095561742782593, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.4308772087097168, - "train/total_loss": 0.37874171137809753 + "train/classification_loss": 0.09315165132284164, + "train/contrastive_loss": 0.8508111834526062, + "train/negative_loss": 0.8484206795692444, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 0.002390496665611863, + "train/total_loss": 0.263313889503479 }, { - "epoch": 28.095693779904305, - "grad_norm": 1.8979859352111816, - "learning_rate": 1.8913358778625956e-05, - "loss": 0.2513, + "epoch": 22.34980988593156, + "grad_norm": 4.498758316040039, + "learning_rate": 1.9136267071320182e-05, + "loss": 0.3165, "step": 2950 }, { - "epoch": 28.095693779904305, + "epoch": 22.34980988593156, "step": 2950, - "train/classification_loss": 0.05062108486890793, - "train/contrastive_loss": 0.13499273359775543, - "train/negative_loss": 0.13235673308372498, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.0026360007468611, - "train/total_loss": 0.07761963456869125 + "train/classification_loss": 0.06855756044387817, + "train/contrastive_loss": 0.1269378811120987, + "train/negative_loss": 0.12687140703201294, + "train/num_negatives": 34, + "train/num_positives": 14, + "train/positive_loss": 6.647333793807775e-05, + "train/total_loss": 0.09394513815641403 }, { - "epoch": 28.095693779904305, + "epoch": 22.34980988593156, "step": 2950, - "train/classification_loss": 0.045265648514032364, - "train/contrastive_loss": 0.12915562093257904, - "train/negative_loss": 0.07938095927238464, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 0.0497746616601944, - "train/total_loss": 0.07109677046537399 + "train/classification_loss": 0.10494504123926163, + "train/contrastive_loss": 0.28756922483444214, + "train/negative_loss": 0.287564218044281, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 5.006804713048041e-06, + "train/total_loss": 0.1624588817358017 }, { - "epoch": 28.574162679425836, - "grad_norm": 1.283211588859558, - "learning_rate": 1.8894274809160307e-05, - "loss": 0.2171, + "epoch": 22.730038022813687, + "grad_norm": 3.5064268112182617, + "learning_rate": 1.9121092564491654e-05, + "loss": 0.3119, "step": 3000 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.13044509291648865, - "train/contrastive_loss": 1.3225327730178833, - "train/negative_loss": 0.7743901610374451, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.5481426119804382, - "train/total_loss": 0.39495164155960083 + "train/classification_loss": 0.05024823918938637, + "train/contrastive_loss": 0.23837293684482574, + "train/negative_loss": 0.0847877711057663, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.15358516573905945, + "train/total_loss": 0.09792283177375793 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.19967179000377655, - "train/contrastive_loss": 2.662342071533203, - "train/negative_loss": 1.1279144287109375, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.5344277620315552, - "train/total_loss": 0.7321402430534363 + "train/classification_loss": 0.23739080131053925, + "train/contrastive_loss": 1.2242213487625122, + "train/negative_loss": 1.2178022861480713, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.0064190952107310295, + "train/total_loss": 0.4822350740432739 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.11364533752202988, - "train/contrastive_loss": 0.6944937705993652, - "train/negative_loss": 0.47404810786247253, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.2204456329345703, - "train/total_loss": 0.252544105052948 + "train/classification_loss": 0.11826159060001373, + "train/contrastive_loss": 2.606766700744629, + "train/negative_loss": 1.971703290939331, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.6350635290145874, + "train/total_loss": 0.6396149396896362 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.16432122886180878, - "train/contrastive_loss": 4.30855655670166, - "train/negative_loss": 2.466432809829712, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.8421236276626587, - "train/total_loss": 1.026032567024231 + "train/classification_loss": 0.2481301873922348, + "train/contrastive_loss": 2.1053128242492676, + "train/negative_loss": 1.0449202060699463, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.0603927373886108, + "train/total_loss": 0.6691927313804626 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.14210495352745056, - "train/contrastive_loss": 3.4216113090515137, - "train/negative_loss": 1.490955114364624, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.9306561946868896, - "train/total_loss": 0.8264272212982178 + "train/classification_loss": 0.08890162408351898, + "train/contrastive_loss": 0.8281328678131104, + "train/negative_loss": 0.8257669806480408, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.0023659057915210724, + "train/total_loss": 0.2545281946659088 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.1276906281709671, - "train/contrastive_loss": 1.9666051864624023, - "train/negative_loss": 0.560606062412262, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.4059991836547852, - "train/total_loss": 0.5210116505622864 + "train/classification_loss": 0.1932091861963272, + "train/contrastive_loss": 1.2443757057189941, + "train/negative_loss": 0.9160003662109375, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.3283753991127014, + "train/total_loss": 0.44208431243896484 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.12824128568172455, - "train/contrastive_loss": 0.8989971876144409, - "train/negative_loss": 0.4885387420654297, + "train/classification_loss": 0.254597932100296, + "train/contrastive_loss": 3.5138492584228516, + "train/negative_loss": 1.5584208965301514, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.41045841574668884, - "train/total_loss": 0.3080407381057739 + "train/num_positives": 8, + "train/positive_loss": 1.9554283618927002, + "train/total_loss": 0.9573677778244019 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.1933402270078659, - "train/contrastive_loss": 0.7706634402275085, - "train/negative_loss": 0.7706462740898132, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 1.7166508769150823e-05, - "train/total_loss": 0.3474729061126709 + "train/classification_loss": 0.20340026915073395, + "train/contrastive_loss": 1.1237661838531494, + "train/negative_loss": 0.7336217761039734, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.3901444375514984, + "train/total_loss": 0.42815351486206055 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.12954136729240417, - "train/contrastive_loss": 1.2097007036209106, - "train/negative_loss": 0.6424413323402405, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.5672593712806702, - "train/total_loss": 0.3714815080165863 + "train/classification_loss": 0.09930409491062164, + "train/contrastive_loss": 0.20740973949432373, + "train/negative_loss": 0.20599615573883057, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.0014135872479528189, + "train/total_loss": 0.1407860517501831 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.2454867959022522, - "train/contrastive_loss": 2.8941876888275146, - "train/negative_loss": 0.770233690738678, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.1239540576934814, - "train/total_loss": 0.824324369430542 + "train/classification_loss": 0.1641291379928589, + "train/contrastive_loss": 2.2256696224212646, + "train/negative_loss": 2.088256359100342, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.13741333782672882, + "train/total_loss": 0.6092630624771118 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.25338301062583923, - "train/contrastive_loss": 2.985530138015747, - "train/negative_loss": 1.9204994440078735, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 1.0650306940078735, - "train/total_loss": 0.8504890203475952 + "train/classification_loss": 0.17835447192192078, + "train/contrastive_loss": 5.804915904998779, + "train/negative_loss": 5.8046488761901855, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.00026723367045633495, + "train/total_loss": 1.339337706565857 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.13389842212200165, - "train/contrastive_loss": 2.9735217094421387, - "train/negative_loss": 0.9405983686447144, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 2.0329232215881348, - "train/total_loss": 0.7286027669906616 + "train/classification_loss": 0.16066159307956696, + "train/contrastive_loss": 3.770228385925293, + "train/negative_loss": 3.118920087814331, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.6513082981109619, + "train/total_loss": 0.9147072434425354 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.25010085105895996, - "train/contrastive_loss": 4.694190502166748, - "train/negative_loss": 2.825225353240967, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.8689651489257812, - "train/total_loss": 1.1889389753341675 + "train/classification_loss": 0.181002676486969, + "train/contrastive_loss": 1.8734900951385498, + "train/negative_loss": 1.35397207736969, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.5195180177688599, + "train/total_loss": 0.5557006597518921 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.04358405992388725, - "train/contrastive_loss": 0.2643384337425232, - "train/negative_loss": 0.256952166557312, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.007386254612356424, - "train/total_loss": 0.09645174443721771 + "train/classification_loss": 0.18481767177581787, + "train/contrastive_loss": 3.04640531539917, + "train/negative_loss": 1.2302279472351074, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.816177248954773, + "train/total_loss": 0.7940987348556519 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.14453233778476715, - "train/contrastive_loss": 1.1862539052963257, - "train/negative_loss": 0.9576621651649475, + "train/classification_loss": 0.19779425859451294, + "train/contrastive_loss": 2.4390251636505127, + "train/negative_loss": 1.6624621152877808, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 0.2285917103290558, - "train/total_loss": 0.381783127784729 + "train/positive_loss": 0.7765630483627319, + "train/total_loss": 0.6855993270874023 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.11011961847543716, - "train/contrastive_loss": 2.5235369205474854, - "train/negative_loss": 0.9049144387245178, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 1.6186224222183228, - "train/total_loss": 0.614827036857605 + "train/classification_loss": 0.1103302463889122, + "train/contrastive_loss": 1.0377720594406128, + "train/negative_loss": 1.0377579927444458, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.406683622917626e-05, + "train/total_loss": 0.3178846538066864 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.29478928446769714, - "train/contrastive_loss": 3.7032952308654785, - "train/negative_loss": 2.0779495239257812, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.6253455877304077, - "train/total_loss": 1.0354483127593994 + "train/classification_loss": 0.15018680691719055, + "train/contrastive_loss": 1.1917169094085693, + "train/negative_loss": 0.7325969338417053, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.4591200053691864, + "train/total_loss": 0.3885301947593689 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.29838377237319946, - "train/contrastive_loss": 3.299133539199829, - "train/negative_loss": 1.7681015729904175, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.5310319662094116, - "train/total_loss": 0.9582104682922363 + "train/classification_loss": 0.18561242520809174, + "train/contrastive_loss": 3.5760393142700195, + "train/negative_loss": 2.3553178310394287, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 1.2207213640213013, + "train/total_loss": 0.9008203148841858 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.2255016416311264, - "train/contrastive_loss": 4.0118408203125, - "train/negative_loss": 2.1044745445251465, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.907366156578064, - "train/total_loss": 1.0278698205947876 + "train/classification_loss": 0.23808659613132477, + "train/contrastive_loss": 3.8615736961364746, + "train/negative_loss": 2.2455945014953613, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.6159791946411133, + "train/total_loss": 1.0104013681411743 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.18558986485004425, - "train/contrastive_loss": 1.7945789098739624, - "train/negative_loss": 1.3192301988601685, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.47534868121147156, - "train/total_loss": 0.5445056557655334 + "train/classification_loss": 0.22659024596214294, + "train/contrastive_loss": 2.658191680908203, + "train/negative_loss": 1.4774677753448486, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.180724024772644, + "train/total_loss": 0.7582285404205322 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.20715872943401337, - "train/contrastive_loss": 1.522726058959961, - "train/negative_loss": 0.8759582042694092, - "train/num_negatives": 44, + "train/classification_loss": 0.20052507519721985, + "train/contrastive_loss": 0.9355710744857788, + "train/negative_loss": 0.9269508719444275, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.008620200678706169, + "train/total_loss": 0.38763928413391113 + }, + { + "epoch": 22.730038022813687, + "step": 3000, + "train/classification_loss": 0.25221824645996094, + "train/contrastive_loss": 2.1961376667022705, + "train/negative_loss": 0.339395672082901, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 0.6467678546905518, - "train/total_loss": 0.5117039680480957 + "train/positive_loss": 1.8567419052124023, + "train/total_loss": 0.6914458274841309 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.21786928176879883, - "train/contrastive_loss": 5.7593793869018555, - "train/negative_loss": 4.017568111419678, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.7418110370635986, - "train/total_loss": 1.369745135307312 + "train/classification_loss": 0.24159327149391174, + "train/contrastive_loss": 1.7659056186676025, + "train/negative_loss": 1.2668263912200928, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.499079167842865, + "train/total_loss": 0.5947743654251099 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.26852166652679443, - "train/contrastive_loss": 4.360503673553467, - "train/negative_loss": 1.6296322345733643, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.7308714389801025, - "train/total_loss": 1.1406223773956299 + "train/classification_loss": 0.21647636592388153, + "train/contrastive_loss": 1.1484295129776, + "train/negative_loss": 1.148427963256836, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 1.5199198060145136e-06, + "train/total_loss": 0.44616228342056274 + }, + { + "epoch": 22.730038022813687, + "step": 3000, + "train/classification_loss": 0.17634446918964386, + "train/contrastive_loss": 6.783164024353027, + "train/negative_loss": 5.750717639923096, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 1.0324461460113525, + "train/total_loss": 1.5329773426055908 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.0493873655796051, - "train/contrastive_loss": 1.227917194366455, - "train/negative_loss": 0.7322075963020325, + "train/classification_loss": 0.27542853355407715, + "train/contrastive_loss": 3.8689792156219482, + "train/negative_loss": 2.756824016571045, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.49570953845977783, - "train/total_loss": 0.2949708104133606 + "train/positive_loss": 1.1121551990509033, + "train/total_loss": 1.0492243766784668 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.09585792571306229, - "train/contrastive_loss": 1.4614700078964233, - "train/negative_loss": 1.3141181468963623, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.14735184609889984, - "train/total_loss": 0.3881519138813019 + "train/classification_loss": 0.219009131193161, + "train/contrastive_loss": 2.3311758041381836, + "train/negative_loss": 1.0090065002441406, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.3221694231033325, + "train/total_loss": 0.6852443218231201 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.2295001894235611, - "train/contrastive_loss": 1.9096217155456543, - "train/negative_loss": 0.7837393283843994, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.1258823871612549, - "train/total_loss": 0.6114245653152466 + "train/classification_loss": 0.1396012008190155, + "train/contrastive_loss": 2.2591357231140137, + "train/negative_loss": 2.055718421936035, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.20341721177101135, + "train/total_loss": 0.5914283394813538 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.013830125331878662, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.013830125331878662 - }, - { - "epoch": 28.574162679425836, - "eval_exact_match_accuracy": 0.4354066985645933, - "eval_hamming_loss": 0.05657191106107515, - "eval_loss": 0.6641110777854919, - "eval_macro_f1": 0.44677413106362723, - "eval_macro_precision": 0.5524087302894152, - "eval_macro_recall": 0.4115477064265991, - "eval_micro_f1": 0.615678776290631, - "eval_micro_precision": 0.7092511013215859, - "eval_micro_recall": 0.543918918918919, - "eval_runtime": 1.0546, - "eval_samples_per_second": 198.172, - "eval_steps_per_second": 25.601, - "step": 3000 + "train/classification_loss": 0.2621252238750458, + "train/contrastive_loss": 2.595353603363037, + "train/negative_loss": 0.8300564885139465, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 1.7652971744537354, + "train/total_loss": 0.7811959981918335 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.07961928844451904, - "train/contrastive_loss": 0.26182952523231506, - "train/negative_loss": 0.2547517418861389, + "train/classification_loss": 0.22145959734916687, + "train/contrastive_loss": 0.5869923830032349, + "train/negative_loss": 0.5529500246047974, "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.007077791728079319, - "train/total_loss": 0.13198518753051758 + "train/positive_loss": 0.0340423509478569, + "train/total_loss": 0.33885806798934937 + }, + { + "epoch": 22.730038022813687, + "step": 3000, + "train/classification_loss": 0.22363658249378204, + "train/contrastive_loss": 1.387320637702942, + "train/negative_loss": 0.8971138000488281, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.49020683765411377, + "train/total_loss": 0.5011007189750671 }, { - "epoch": 28.574162679425836, + "epoch": 22.730038022813687, "step": 3000, - "train/classification_loss": 0.03360428661108017, - "train/contrastive_loss": 0.1529385894536972, - "train/negative_loss": 0.15203040838241577, + "train/classification_loss": 0.22825860977172852, + "train/contrastive_loss": 2.2215986251831055, + "train/negative_loss": 1.9383509159088135, "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.0009081767057068646, - "train/total_loss": 0.06419200450181961 + "train/num_positives": 12, + "train/positive_loss": 0.28324776887893677, + "train/total_loss": 0.6725783348083496 + }, + { + "epoch": 22.730038022813687, + "step": 3000, + "train/classification_loss": 0.09187773615121841, + "train/contrastive_loss": 1.4055594205856323, + "train/negative_loss": 1.374021291732788, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.031538113951683044, + "train/total_loss": 0.37298962473869324 + }, + { + "epoch": 22.730038022813687, + "eval_exact_match_accuracy": 0.30916030534351147, + "eval_hamming_loss": 0.06443646160754378, + "eval_loss": 0.6393535137176514, + "eval_macro_f1": 0.3160130086093044, + "eval_macro_precision": 0.3685990561251953, + "eval_macro_recall": 0.2826752037930375, + "eval_micro_f1": 0.5618320610687023, + "eval_micro_precision": 0.6996197718631179, + "eval_micro_recall": 0.46938775510204084, + "eval_runtime": 1.3157, + "eval_samples_per_second": 199.129, + "eval_steps_per_second": 25.081, + "step": 3000 + }, + { + "epoch": 22.730038022813687, + "step": 3000, + "train/classification_loss": 0.06666012108325958, + "train/contrastive_loss": 0.14042483270168304, + "train/negative_loss": 0.14041411876678467, + "train/num_negatives": 38, + "train/num_positives": 18, + "train/positive_loss": 1.0715687494666781e-05, + "train/total_loss": 0.09474508464336395 + }, + { + "epoch": 22.730038022813687, + "step": 3000, + "train/classification_loss": 0.048709750175476074, + "train/contrastive_loss": 0.37545520067214966, + "train/negative_loss": 0.3752516806125641, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0002035334473475814, + "train/total_loss": 0.12380079180002213 }, { - "epoch": 29.047846889952154, - "grad_norm": 2.623663902282715, - "learning_rate": 1.8875190839694658e-05, - "loss": 0.2618, + "epoch": 23.106463878326995, + "grad_norm": 9.799505233764648, + "learning_rate": 1.9105918057663125e-05, + "loss": 0.2706, "step": 3050 }, { - "epoch": 29.047846889952154, + "epoch": 23.106463878326995, "step": 3050, - "train/classification_loss": 0.04132596775889397, - "train/contrastive_loss": 0.2238161563873291, - "train/negative_loss": 0.18330925703048706, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 0.04050689935684204, - "train/total_loss": 0.08608920127153397 + "train/classification_loss": 0.04519231244921684, + "train/contrastive_loss": 0.12822826206684113, + "train/negative_loss": 0.12809649109840393, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.00013176986249163747, + "train/total_loss": 0.07083796709775925 }, { - "epoch": 29.047846889952154, + "epoch": 23.106463878326995, "step": 3050, - "train/classification_loss": 0.026460621505975723, - "train/contrastive_loss": 0.10660955309867859, - "train/negative_loss": 0.10550732165575027, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.0011022307444363832, - "train/total_loss": 0.0477825328707695 + "train/classification_loss": 0.04827345162630081, + "train/contrastive_loss": 0.3491438627243042, + "train/negative_loss": 0.315034419298172, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.034109439700841904, + "train/total_loss": 0.11810222268104553 }, { - "epoch": 29.526315789473685, - "grad_norm": 5.849625587463379, - "learning_rate": 1.885610687022901e-05, - "loss": 0.2337, + "epoch": 23.486692015209126, + "grad_norm": 3.6271841526031494, + "learning_rate": 1.90907435508346e-05, + "loss": 0.301, "step": 3100 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, + "step": 3100, + "train/classification_loss": 0.04460832104086876, + "train/contrastive_loss": 0.1377183496952057, + "train/negative_loss": 0.11040845513343811, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.02730989269912243, + "train/total_loss": 0.07215198874473572 + }, + { + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.128273144364357, - "train/contrastive_loss": 0.9626711010932922, - "train/negative_loss": 0.7343614101409912, + "train/classification_loss": 0.23689772188663483, + "train/contrastive_loss": 1.3933662176132202, + "train/negative_loss": 1.3926423788070679, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.22830969095230103, - "train/total_loss": 0.3208073675632477 + "train/num_positives": 6, + "train/positive_loss": 0.000723804987501353, + "train/total_loss": 0.5155709981918335 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.20729929208755493, - "train/contrastive_loss": 2.5886752605438232, - "train/negative_loss": 1.3106322288513184, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.2780430316925049, - "train/total_loss": 0.7250343561172485 + "train/classification_loss": 0.10107841342687607, + "train/contrastive_loss": 2.6117289066314697, + "train/negative_loss": 2.4704384803771973, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.14129039645195007, + "train/total_loss": 0.6234241724014282 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.13024286925792694, - "train/contrastive_loss": 0.8691848516464233, - "train/negative_loss": 0.6231476068496704, - "train/num_negatives": 44, + "train/classification_loss": 0.22328615188598633, + "train/contrastive_loss": 1.9988946914672852, + "train/negative_loss": 1.1539627313613892, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.844931960105896, + "train/total_loss": 0.6230651140213013 + }, + { + "epoch": 23.486692015209126, + "step": 3100, + "train/classification_loss": 0.1086631491780281, + "train/contrastive_loss": 1.1936149597167969, + "train/negative_loss": 1.164294719696045, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.029320213943719864, + "train/total_loss": 0.3473861515522003 + }, + { + "epoch": 23.486692015209126, + "step": 3100, + "train/classification_loss": 0.1771281510591507, + "train/contrastive_loss": 1.2458733320236206, + "train/negative_loss": 0.894384503364563, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 0.24603721499443054, - "train/total_loss": 0.3040798306465149 + "train/positive_loss": 0.35148879885673523, + "train/total_loss": 0.42630282044410706 + }, + { + "epoch": 23.486692015209126, + "step": 3100, + "train/classification_loss": 0.271940678358078, + "train/contrastive_loss": 4.128767490386963, + "train/negative_loss": 2.0084855556488037, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 2.120281934738159, + "train/total_loss": 1.0976941585540771 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.1616610884666443, - "train/contrastive_loss": 3.1308951377868652, - "train/negative_loss": 1.524717092514038, + "train/classification_loss": 0.16765789687633514, + "train/contrastive_loss": 1.6871697902679443, + "train/negative_loss": 0.9942083358764648, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.6061781644821167, - "train/total_loss": 0.7878401279449463 + "train/positive_loss": 0.6929615139961243, + "train/total_loss": 0.5050918459892273 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.15682217478752136, - "train/contrastive_loss": 3.30527663230896, - "train/negative_loss": 1.6973477602005005, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.6079288721084595, - "train/total_loss": 0.8178775310516357 + "train/classification_loss": 0.1170414388179779, + "train/contrastive_loss": 0.24876238405704498, + "train/negative_loss": 0.23219887912273407, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.01656351052224636, + "train/total_loss": 0.16679391264915466 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.1452808976173401, - "train/contrastive_loss": 1.8239954710006714, - "train/negative_loss": 0.5416302680969238, - "train/num_negatives": 32, + "train/classification_loss": 0.14143285155296326, + "train/contrastive_loss": 1.9043775796890259, + "train/negative_loss": 1.8381905555725098, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.06618703156709671, + "train/total_loss": 0.522308349609375 + }, + { + "epoch": 23.486692015209126, + "step": 3100, + "train/classification_loss": 0.15256823599338531, + "train/contrastive_loss": 4.584046363830566, + "train/negative_loss": 4.580995559692383, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.2823652029037476, - "train/total_loss": 0.5100799798965454 + "train/positive_loss": 0.0030509408097714186, + "train/total_loss": 1.0693775415420532 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.11049263924360275, - "train/contrastive_loss": 1.1078393459320068, - "train/negative_loss": 0.6123132705688477, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.49552613496780396, - "train/total_loss": 0.3320605158805847 + "train/classification_loss": 0.1319369375705719, + "train/contrastive_loss": 2.539508819580078, + "train/negative_loss": 2.4032881259918213, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.13622072339057922, + "train/total_loss": 0.639838695526123 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.18916845321655273, - "train/contrastive_loss": 0.5569239854812622, - "train/negative_loss": 0.5569052696228027, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 1.871626045613084e-05, - "train/total_loss": 0.30055326223373413 + "train/classification_loss": 0.17261362075805664, + "train/contrastive_loss": 1.9881383180618286, + "train/negative_loss": 1.2049821615219116, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.783156156539917, + "train/total_loss": 0.5702413320541382 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.11572655290365219, - "train/contrastive_loss": 1.0184004306793213, - "train/negative_loss": 0.5984084010124207, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.41999197006225586, - "train/total_loss": 0.3194066286087036 + "train/classification_loss": 0.1730685532093048, + "train/contrastive_loss": 3.151801109313965, + "train/negative_loss": 1.12565016746521, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.026150941848755, + "train/total_loss": 0.8034287691116333 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.2711280286312103, - "train/contrastive_loss": 3.079716682434082, - "train/negative_loss": 1.6797378063201904, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 1.3999789953231812, - "train/total_loss": 0.8870713710784912 + "train/classification_loss": 0.19639505445957184, + "train/contrastive_loss": 2.1291651725769043, + "train/negative_loss": 1.9392566680908203, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.18990856409072876, + "train/total_loss": 0.6222280859947205 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.24621190130710602, - "train/contrastive_loss": 2.7158851623535156, - "train/negative_loss": 1.9604878425598145, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.7553973197937012, - "train/total_loss": 0.7893889546394348 + "train/classification_loss": 0.1104254275560379, + "train/contrastive_loss": 1.183456301689148, + "train/negative_loss": 1.1834315061569214, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 2.475624751241412e-05, + "train/total_loss": 0.34711670875549316 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.11391310393810272, - "train/contrastive_loss": 1.0831196308135986, - "train/negative_loss": 0.2585330009460449, + "train/classification_loss": 0.1313083916902542, + "train/contrastive_loss": 0.88410884141922, + "train/negative_loss": 0.8451991677284241, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 0.8245866298675537, - "train/total_loss": 0.33053702116012573 + "train/positive_loss": 0.0389096662402153, + "train/total_loss": 0.308130145072937 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.26211363077163696, - "train/contrastive_loss": 4.881839752197266, - "train/negative_loss": 3.284181833267212, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.5976577997207642, - "train/total_loss": 1.2384815216064453 + "train/classification_loss": 0.17915678024291992, + "train/contrastive_loss": 4.213595390319824, + "train/negative_loss": 2.0420477390289307, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 2.1715478897094727, + "train/total_loss": 1.0218758583068848 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.039828408509492874, - "train/contrastive_loss": 0.32205870747566223, - "train/negative_loss": 0.3046107590198517, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.017447957769036293, - "train/total_loss": 0.10424014925956726 + "train/classification_loss": 0.27718594670295715, + "train/contrastive_loss": 4.025005340576172, + "train/negative_loss": 2.7354655265808105, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.2895395755767822, + "train/total_loss": 1.0821870565414429 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.15078644454479218, - "train/contrastive_loss": 1.3343586921691895, - "train/negative_loss": 0.7054144740104675, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.6289442777633667, - "train/total_loss": 0.4176582098007202 + "train/classification_loss": 0.2558128833770752, + "train/contrastive_loss": 2.815211534500122, + "train/negative_loss": 1.454790472984314, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.360421061515808, + "train/total_loss": 0.8188552260398865 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.12821586430072784, - "train/contrastive_loss": 2.514204502105713, - "train/negative_loss": 1.0056064128875732, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 1.5085980892181396, - "train/total_loss": 0.6310567855834961 + "train/classification_loss": 0.22710184752941132, + "train/contrastive_loss": 0.713596761226654, + "train/negative_loss": 0.7131626605987549, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.0004340865998528898, + "train/total_loss": 0.3698211908340454 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.2981450855731964, - "train/contrastive_loss": 3.255282402038574, - "train/negative_loss": 2.079235315322876, - "train/num_negatives": 46, + "train/classification_loss": 0.23883533477783203, + "train/contrastive_loss": 2.08504319190979, + "train/negative_loss": 0.314080148935318, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.1760472059249878, - "train/total_loss": 0.9492015838623047 + "train/positive_loss": 1.7709629535675049, + "train/total_loss": 0.65584397315979 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.33994725346565247, - "train/contrastive_loss": 3.4029736518859863, - "train/negative_loss": 1.677606463432312, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.7253670692443848, - "train/total_loss": 1.020542025566101 + "train/classification_loss": 0.24670915305614471, + "train/contrastive_loss": 1.2970675230026245, + "train/negative_loss": 1.1285417079925537, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.1685258001089096, + "train/total_loss": 0.5061226487159729 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.20821544528007507, - "train/contrastive_loss": 3.4157299995422363, - "train/negative_loss": 1.8999602794647217, + "train/classification_loss": 0.18970893323421478, + "train/contrastive_loss": 1.3093541860580444, + "train/negative_loss": 1.309349775314331, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.5157697200775146, - "train/total_loss": 0.8913614749908447 + "train/num_positives": 8, + "train/positive_loss": 4.410760539030889e-06, + "train/total_loss": 0.451579749584198 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.18201956152915955, - "train/contrastive_loss": 1.2294206619262695, - "train/negative_loss": 0.8011401295661926, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.4282805621623993, - "train/total_loss": 0.4279037117958069 + "train/classification_loss": 0.16845521330833435, + "train/contrastive_loss": 6.597036361694336, + "train/negative_loss": 5.618363380432129, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.9786727428436279, + "train/total_loss": 1.4878625869750977 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.2088194638490677, - "train/contrastive_loss": 1.646306037902832, - "train/negative_loss": 1.0363829135894775, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.6099230647087097, - "train/total_loss": 0.5380806922912598 + "train/classification_loss": 0.2517804503440857, + "train/contrastive_loss": 3.5475969314575195, + "train/negative_loss": 3.064736843109131, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.4828599989414215, + "train/total_loss": 0.9612998366355896 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.22758091986179352, - "train/contrastive_loss": 5.684276580810547, - "train/negative_loss": 4.233237266540527, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.45103919506073, - "train/total_loss": 1.3644362688064575 + "train/classification_loss": 0.22155262529850006, + "train/contrastive_loss": 1.80209481716156, + "train/negative_loss": 1.0086891651153564, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 0.7934056520462036, + "train/total_loss": 0.5819715857505798 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.26538828015327454, - "train/contrastive_loss": 4.530345439910889, - "train/negative_loss": 1.9087587594985962, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.621586799621582, - "train/total_loss": 1.1714574098587036 + "train/classification_loss": 0.13832707703113556, + "train/contrastive_loss": 2.2231760025024414, + "train/negative_loss": 2.181638479232788, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.04153752699494362, + "train/total_loss": 0.5829622745513916 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.07371557503938675, - "train/contrastive_loss": 1.7133069038391113, - "train/negative_loss": 1.054520606994629, - "train/num_negatives": 42, + "train/classification_loss": 0.2803340256214142, + "train/contrastive_loss": 1.9823437929153442, + "train/negative_loss": 0.8339265584945679, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.6587862372398376, - "train/total_loss": 0.4163769483566284 + "train/positive_loss": 1.1484172344207764, + "train/total_loss": 0.6768027544021606 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.08110149949789047, - "train/contrastive_loss": 1.4886118173599243, - "train/negative_loss": 1.4602444171905518, - "train/num_negatives": 48, + "train/classification_loss": 0.21623481810092926, + "train/contrastive_loss": 0.3048136830329895, + "train/negative_loss": 0.27409207820892334, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.02836744859814644, - "train/total_loss": 0.3788238763809204 + "train/positive_loss": 0.03072160854935646, + "train/total_loss": 0.27719753980636597 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.18739557266235352, - "train/contrastive_loss": 1.7206175327301025, - "train/negative_loss": 0.9177534580230713, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.8028640747070312, - "train/total_loss": 0.5315190553665161 + "train/classification_loss": 0.21694214642047882, + "train/contrastive_loss": 1.7727047204971313, + "train/negative_loss": 1.3621785640716553, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.4105261564254761, + "train/total_loss": 0.5714830756187439 + }, + { + "epoch": 23.486692015209126, + "step": 3100, + "train/classification_loss": 0.1999751329421997, + "train/contrastive_loss": 1.8136916160583496, + "train/negative_loss": 1.811160683631897, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.002530968515202403, + "train/total_loss": 0.5627135038375854 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.011393874883651733, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.011393874883651733 - }, - { - "epoch": 29.526315789473685, - "eval_exact_match_accuracy": 0.42105263157894735, - "eval_hamming_loss": 0.05713481564874754, - "eval_loss": 0.6318583488464355, - "eval_macro_f1": 0.448042420162728, - "eval_macro_precision": 0.6194419469729214, - "eval_macro_recall": 0.41129617146918185, - "eval_micro_f1": 0.6088631984585742, - "eval_micro_precision": 0.7085201793721974, - "eval_micro_recall": 0.5337837837837838, - "eval_runtime": 1.0672, - "eval_samples_per_second": 195.847, - "eval_steps_per_second": 25.301, + "train/classification_loss": 0.0835106298327446, + "train/contrastive_loss": 1.6874322891235352, + "train/negative_loss": 1.5227882862091064, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.16464398801326752, + "train/total_loss": 0.42099711298942566 + }, + { + "epoch": 23.486692015209126, + "eval_exact_match_accuracy": 0.3435114503816794, + "eval_hamming_loss": 0.05972159856308936, + "eval_loss": 0.6163199543952942, + "eval_macro_f1": 0.33192443362767415, + "eval_macro_precision": 0.419967566331173, + "eval_macro_recall": 0.2853545372963981, + "eval_micro_f1": 0.5907692307692308, + "eval_micro_precision": 0.7441860465116279, + "eval_micro_recall": 0.4897959183673469, + "eval_runtime": 1.3146, + "eval_samples_per_second": 199.3, + "eval_steps_per_second": 25.103, "step": 3100 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.041862331330776215, - "train/contrastive_loss": 0.1408112794160843, - "train/negative_loss": 0.14080506563186646, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 6.215948360477341e-06, - "train/total_loss": 0.07002458721399307 + "train/classification_loss": 0.0347839780151844, + "train/contrastive_loss": 0.11064280569553375, + "train/negative_loss": 0.07262276113033295, + "train/num_negatives": 26, + "train/num_positives": 30, + "train/positive_loss": 0.03802004083991051, + "train/total_loss": 0.05691254138946533 }, { - "epoch": 29.526315789473685, + "epoch": 23.486692015209126, "step": 3100, - "train/classification_loss": 0.08246800303459167, - "train/contrastive_loss": 1.1308801174163818, - "train/negative_loss": 1.1308616399765015, - "train/num_negatives": 54, - "train/num_positives": 2, - "train/positive_loss": 1.8477610865375027e-05, - "train/total_loss": 0.30864402651786804 + "train/classification_loss": 0.036561015993356705, + "train/contrastive_loss": 0.17340557277202606, + "train/negative_loss": 0.1733887493610382, + "train/num_negatives": 26, + "train/num_positives": 30, + "train/positive_loss": 1.6824837075546384e-05, + "train/total_loss": 0.07124213129281998 }, { - "epoch": 30.0, - "grad_norm": 1.6774592399597168, - "learning_rate": 1.883702290076336e-05, - "loss": 0.2199, + "epoch": 23.866920152091254, + "grad_norm": 5.928348541259766, + "learning_rate": 1.907556904400607e-05, + "loss": 0.3032, "step": 3150 }, { - "epoch": 30.0, + "epoch": 23.866920152091254, "step": 3150, - "train/classification_loss": 0.054636407643556595, - "train/contrastive_loss": 0.20507854223251343, - "train/negative_loss": 0.16818350553512573, + "train/classification_loss": 0.09806305170059204, + "train/contrastive_loss": 0.09896017611026764, + "train/negative_loss": 0.09893803298473358, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.036895036697387695, - "train/total_loss": 0.09565211832523346 + "train/num_positives": 8, + "train/positive_loss": 2.2143482055980712e-05, + "train/total_loss": 0.11785508692264557 }, { - "epoch": 30.0, + "epoch": 23.866920152091254, "step": 3150, - "train/classification_loss": 0.09567541629076004, - "train/contrastive_loss": 0.49416425824165344, - "train/negative_loss": 0.4940524995326996, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.0001117694191634655, - "train/total_loss": 0.19450826942920685 + "train/classification_loss": 0.04523998871445656, + "train/contrastive_loss": 0.22879329323768616, + "train/negative_loss": 0.22768887877464294, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.001104415743611753, + "train/total_loss": 0.09099864959716797 }, { - "epoch": 30.47846889952153, - "grad_norm": 3.2140984535217285, - "learning_rate": 1.881793893129771e-05, - "loss": 0.2122, + "epoch": 24.24334600760456, + "grad_norm": 6.138728618621826, + "learning_rate": 1.9060394537177543e-05, + "loss": 0.2699, "step": 3200 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.1330963522195816, - "train/contrastive_loss": 1.5445938110351562, - "train/negative_loss": 0.9737083315849304, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.5708855390548706, - "train/total_loss": 0.4420151114463806 + "train/classification_loss": 0.0461883619427681, + "train/contrastive_loss": 0.14900246262550354, + "train/negative_loss": 0.12669524550437927, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.022307224571704865, + "train/total_loss": 0.07598885893821716 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.20338734984397888, - "train/contrastive_loss": 2.9951157569885254, - "train/negative_loss": 1.731058955192566, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.26405668258667, - "train/total_loss": 0.8024104833602905 + "train/classification_loss": 0.240229994058609, + "train/contrastive_loss": 1.197166085243225, + "train/negative_loss": 1.1836270093917847, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.013539044186472893, + "train/total_loss": 0.4796631932258606 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.1464085429906845, - "train/contrastive_loss": 0.9245534539222717, - "train/negative_loss": 0.5968682765960693, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.3276851773262024, - "train/total_loss": 0.33131924271583557 + "train/classification_loss": 0.10546597093343735, + "train/contrastive_loss": 2.0022687911987305, + "train/negative_loss": 1.0830386877059937, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.9192301630973816, + "train/total_loss": 0.5059197545051575 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.15164007246494293, - "train/contrastive_loss": 3.828411102294922, - "train/negative_loss": 1.9765416383743286, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.8518694639205933, - "train/total_loss": 0.9173222780227661 + "train/classification_loss": 0.2509417235851288, + "train/contrastive_loss": 2.348822832107544, + "train/negative_loss": 1.2292232513427734, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.1195995807647705, + "train/total_loss": 0.7207062840461731 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.1310262829065323, - "train/contrastive_loss": 2.8596014976501465, - "train/negative_loss": 1.5958592891693115, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.2637420892715454, - "train/total_loss": 0.7029466032981873 + "train/classification_loss": 0.10379074513912201, + "train/contrastive_loss": 1.1358143091201782, + "train/negative_loss": 1.0140453577041626, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.12176891416311264, + "train/total_loss": 0.33095359802246094 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.13639581203460693, - "train/contrastive_loss": 1.5845136642456055, - "train/negative_loss": 0.36242082715034485, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.222092866897583, - "train/total_loss": 0.45329853892326355 + "train/classification_loss": 0.18862131237983704, + "train/contrastive_loss": 1.3446695804595947, + "train/negative_loss": 0.8760404586791992, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.4686291217803955, + "train/total_loss": 0.45755523443222046 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.1526937484741211, - "train/contrastive_loss": 1.9285571575164795, - "train/negative_loss": 0.9498431086540222, + "train/classification_loss": 0.25009021162986755, + "train/contrastive_loss": 2.6035518646240234, + "train/negative_loss": 1.343505620956421, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.978714108467102, - "train/total_loss": 0.538405179977417 + "train/num_positives": 8, + "train/positive_loss": 1.260046124458313, + "train/total_loss": 0.7708005905151367 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.24409045279026031, - "train/contrastive_loss": 0.9839267134666443, - "train/negative_loss": 0.9833745956420898, + "train/classification_loss": 0.17393480241298676, + "train/contrastive_loss": 3.1324586868286133, + "train/negative_loss": 1.364229679107666, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 1.7682288885116577, + "train/total_loss": 0.8004265427589417 + }, + { + "epoch": 24.24334600760456, + "step": 3200, + "train/classification_loss": 0.11859456449747086, + "train/contrastive_loss": 0.7748359441757202, + "train/negative_loss": 0.4586743116378784, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.3161616027355194, + "train/total_loss": 0.2735617458820343 + }, + { + "epoch": 24.24334600760456, + "step": 3200, + "train/classification_loss": 0.13792136311531067, + "train/contrastive_loss": 0.7221623063087463, + "train/negative_loss": 0.6551451683044434, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.0005521394778043032, - "train/total_loss": 0.4408757984638214 + "train/num_positives": 8, + "train/positive_loss": 0.06701715290546417, + "train/total_loss": 0.28235381841659546 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.14023295044898987, - "train/contrastive_loss": 1.0637485980987549, - "train/negative_loss": 0.6934577822685242, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.3702908754348755, - "train/total_loss": 0.35298267006874084 + "train/classification_loss": 0.1688239425420761, + "train/contrastive_loss": 4.3241868019104, + "train/negative_loss": 4.198012828826904, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.1261739283800125, + "train/total_loss": 1.0336612462997437 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.26318520307540894, - "train/contrastive_loss": 3.260310649871826, - "train/negative_loss": 1.2259471416473389, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 2.0343635082244873, - "train/total_loss": 0.9152473211288452 + "train/classification_loss": 0.14247582852840424, + "train/contrastive_loss": 2.1458189487457275, + "train/negative_loss": 1.6275535821914673, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.5182653069496155, + "train/total_loss": 0.5716395974159241 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.22227837145328522, - "train/contrastive_loss": 2.8919119834899902, - "train/negative_loss": 1.9863431453704834, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.9055688977241516, - "train/total_loss": 0.8006607890129089 + "train/classification_loss": 0.18234121799468994, + "train/contrastive_loss": 1.6596033573150635, + "train/negative_loss": 1.175521731376648, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.4840816855430603, + "train/total_loss": 0.5142619013786316 + }, + { + "epoch": 24.24334600760456, + "step": 3200, + "train/classification_loss": 0.19311180710792542, + "train/contrastive_loss": 3.455066680908203, + "train/negative_loss": 1.3178602457046509, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.137206554412842, + "train/total_loss": 0.8841251134872437 + }, + { + "epoch": 24.24334600760456, + "step": 3200, + "train/classification_loss": 0.21276018023490906, + "train/contrastive_loss": 3.0318875312805176, + "train/negative_loss": 1.7630462646484375, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 1.26884126663208, + "train/total_loss": 0.819137692451477 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.11705171316862106, - "train/contrastive_loss": 1.583751916885376, - "train/negative_loss": 0.4211987853050232, + "train/classification_loss": 0.0985073521733284, + "train/contrastive_loss": 0.7124829292297363, + "train/negative_loss": 0.7096551060676575, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 1.1625531911849976, - "train/total_loss": 0.4338020980358124 + "train/positive_loss": 0.0028278306126594543, + "train/total_loss": 0.24100393056869507 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.28097274899482727, - "train/contrastive_loss": 4.874276638031006, - "train/negative_loss": 3.5287346839904785, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.3455418348312378, - "train/total_loss": 1.2558281421661377 + "train/classification_loss": 0.15517030656337738, + "train/contrastive_loss": 1.213793396949768, + "train/negative_loss": 0.8954811096191406, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.31831225752830505, + "train/total_loss": 0.39792898297309875 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.038338251411914825, - "train/contrastive_loss": 0.10258130729198456, - "train/negative_loss": 0.09295916557312012, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.009622139856219292, - "train/total_loss": 0.05885451287031174 + "train/classification_loss": 0.1850353479385376, + "train/contrastive_loss": 2.4885902404785156, + "train/negative_loss": 1.549621343612671, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.9389688968658447, + "train/total_loss": 0.6827534437179565 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.19519929587841034, - "train/contrastive_loss": 1.5279150009155273, - "train/negative_loss": 0.8495047688484192, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.6784101724624634, - "train/total_loss": 0.500782310962677 + "train/classification_loss": 0.2600933015346527, + "train/contrastive_loss": 3.240817070007324, + "train/negative_loss": 1.885237693786621, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.3555793762207031, + "train/total_loss": 0.9082567691802979 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.12025267630815506, - "train/contrastive_loss": 2.117055892944336, - "train/negative_loss": 0.8996873497962952, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 1.2173686027526855, - "train/total_loss": 0.5436638593673706 + "train/classification_loss": 0.25313451886177063, + "train/contrastive_loss": 3.00954532623291, + "train/negative_loss": 1.3742566108703613, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.6352885961532593, + "train/total_loss": 0.8550436496734619 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.3341538608074188, - "train/contrastive_loss": 3.493695020675659, - "train/negative_loss": 2.1067862510681152, - "train/num_negatives": 46, + "train/classification_loss": 0.20847252011299133, + "train/contrastive_loss": 0.48768851161003113, + "train/negative_loss": 0.45423221588134766, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.03345630317926407, + "train/total_loss": 0.3060102164745331 + }, + { + "epoch": 24.24334600760456, + "step": 3200, + "train/classification_loss": 0.25195956230163574, + "train/contrastive_loss": 2.857666492462158, + "train/negative_loss": 0.5311806797981262, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.386908769607544, - "train/total_loss": 1.0328928232192993 + "train/positive_loss": 2.3264858722686768, + "train/total_loss": 0.8234928846359253 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.32442739605903625, - "train/contrastive_loss": 3.2090256214141846, - "train/negative_loss": 2.253246545791626, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 0.9557790756225586, - "train/total_loss": 0.9662325382232666 + "train/classification_loss": 0.2503430247306824, + "train/contrastive_loss": 1.890232801437378, + "train/negative_loss": 1.209954023361206, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.6802787780761719, + "train/total_loss": 0.6283895969390869 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.21873056888580322, - "train/contrastive_loss": 3.3451757431030273, - "train/negative_loss": 1.9225412607192993, + "train/classification_loss": 0.2253788411617279, + "train/contrastive_loss": 0.35783010721206665, + "train/negative_loss": 0.357825368642807, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 1.4226346015930176, - "train/total_loss": 0.8877657055854797 + "train/num_positives": 8, + "train/positive_loss": 4.738589723274345e-06, + "train/total_loss": 0.29694485664367676 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.16060107946395874, - "train/contrastive_loss": 0.8471282720565796, - "train/negative_loss": 0.6935101747512817, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.15361812710762024, - "train/total_loss": 0.3300267457962036 + "train/classification_loss": 0.17557424306869507, + "train/contrastive_loss": 5.388922691345215, + "train/negative_loss": 4.436996936798096, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.9519259929656982, + "train/total_loss": 1.2533588409423828 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.20182284712791443, - "train/contrastive_loss": 1.3535045385360718, - "train/negative_loss": 0.940479040145874, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.41302546858787537, - "train/total_loss": 0.4725237488746643 + "train/classification_loss": 0.24908564984798431, + "train/contrastive_loss": 3.547696590423584, + "train/negative_loss": 2.5069704055786133, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 1.0407261848449707, + "train/total_loss": 0.9586250185966492 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.2428281456232071, - "train/contrastive_loss": 7.003986358642578, - "train/negative_loss": 5.30624532699585, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.6977410316467285, - "train/total_loss": 1.6436253786087036 + "train/classification_loss": 0.2300679087638855, + "train/contrastive_loss": 3.013021945953369, + "train/negative_loss": 1.0855708122253418, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.9274510145187378, + "train/total_loss": 0.8326722979545593 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.26693281531333923, - "train/contrastive_loss": 3.5833821296691895, - "train/negative_loss": 1.547589659690857, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.035792350769043, - "train/total_loss": 0.9836091995239258 + "train/classification_loss": 0.12677784264087677, + "train/contrastive_loss": 2.2214553356170654, + "train/negative_loss": 2.075814723968506, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.14564067125320435, + "train/total_loss": 0.5710689425468445 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.05743790790438652, - "train/contrastive_loss": 1.3192987442016602, - "train/negative_loss": 1.2005095481872559, - "train/num_negatives": 42, + "train/classification_loss": 0.27104201912879944, + "train/contrastive_loss": 2.2794997692108154, + "train/negative_loss": 0.8221967816352844, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.1187891736626625, - "train/total_loss": 0.32129764556884766 + "train/positive_loss": 1.4573029279708862, + "train/total_loss": 0.7269419431686401 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.1238444522023201, - "train/contrastive_loss": 1.6248847246170044, - "train/negative_loss": 1.3049777746200562, - "train/num_negatives": 48, + "train/classification_loss": 0.20599576830863953, + "train/contrastive_loss": 0.5194090604782104, + "train/negative_loss": 0.5187619924545288, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.31990692019462585, - "train/total_loss": 0.44882139563560486 + "train/positive_loss": 0.0006470971275120974, + "train/total_loss": 0.30987757444381714 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.20514675974845886, - "train/contrastive_loss": 1.8072960376739502, - "train/negative_loss": 0.8515953421592712, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.955700695514679, - "train/total_loss": 0.5666059851646423 + "train/classification_loss": 0.22401738166809082, + "train/contrastive_loss": 1.8550976514816284, + "train/negative_loss": 0.7668852806091309, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.0882123708724976, + "train/total_loss": 0.5950369238853455 + }, + { + "epoch": 24.24334600760456, + "step": 3200, + "train/classification_loss": 0.19081558287143707, + "train/contrastive_loss": 1.3705555200576782, + "train/negative_loss": 0.9166907668113708, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.4538647532463074, + "train/total_loss": 0.46492671966552734 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.012021834030747414, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.012021834030747414 - }, - { - "epoch": 30.47846889952153, - "eval_exact_match_accuracy": 0.40669856459330145, - "eval_hamming_loss": 0.05741626794258373, - "eval_loss": 0.6562801599502563, - "eval_macro_f1": 0.4094777659621332, - "eval_macro_precision": 0.5730951103891282, - "eval_macro_recall": 0.35266956931316795, - "eval_micro_f1": 0.5984251968503937, - "eval_micro_precision": 0.7169811320754716, - "eval_micro_recall": 0.5135135135135135, - "eval_runtime": 1.0625, - "eval_samples_per_second": 196.705, - "eval_steps_per_second": 25.412, + "train/classification_loss": 0.08456185460090637, + "train/contrastive_loss": 1.737784504890442, + "train/negative_loss": 1.7237993478775024, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.013985160738229752, + "train/total_loss": 0.4321187734603882 + }, + { + "epoch": 24.24334600760456, + "eval_exact_match_accuracy": 0.3435114503816794, + "eval_hamming_loss": 0.06039515042658285, + "eval_loss": 0.6014404892921448, + "eval_macro_f1": 0.3766636471948195, + "eval_macro_precision": 0.4697778271909459, + "eval_macro_recall": 0.32145519556373153, + "eval_micro_f1": 0.5803432137285491, + "eval_micro_precision": 0.7469879518072289, + "eval_micro_recall": 0.4744897959183674, + "eval_runtime": 1.3235, + "eval_samples_per_second": 197.963, + "eval_steps_per_second": 24.934, "step": 3200 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.06776297837495804, - "train/contrastive_loss": 0.2266082912683487, - "train/negative_loss": 0.22660650312900543, + "train/classification_loss": 0.037417229264974594, + "train/contrastive_loss": 0.08196588605642319, + "train/negative_loss": 0.07999131083488464, "train/num_negatives": 44, "train/num_positives": 12, - "train/positive_loss": 1.7881409348774469e-06, - "train/total_loss": 0.11308463662862778 + "train/positive_loss": 0.001974578481167555, + "train/total_loss": 0.05381040647625923 }, { - "epoch": 30.47846889952153, + "epoch": 24.24334600760456, "step": 3200, - "train/classification_loss": 0.04088152199983597, - "train/contrastive_loss": 0.09430741518735886, - "train/negative_loss": 0.09380242973566055, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.0005049830069765449, - "train/total_loss": 0.0597430057823658 + "train/classification_loss": 0.0441744327545166, + "train/contrastive_loss": 0.3444546163082123, + "train/negative_loss": 0.34444329142570496, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.1325066225253977e-05, + "train/total_loss": 0.11306535452604294 }, { - "epoch": 30.95693779904306, - "grad_norm": 4.12062406539917, - "learning_rate": 1.8798854961832063e-05, - "loss": 0.2159, + "epoch": 24.623574144486692, + "grad_norm": 3.2997684478759766, + "learning_rate": 1.9045220030349014e-05, + "loss": 0.2593, "step": 3250 }, { - "epoch": 30.95693779904306, + "epoch": 24.623574144486692, "step": 3250, - "train/classification_loss": 0.054366353899240494, - "train/contrastive_loss": 0.24535155296325684, - "train/negative_loss": 0.2449103593826294, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.00044119232916273177, - "train/total_loss": 0.1034366637468338 + "train/classification_loss": 0.09509336203336716, + "train/contrastive_loss": 0.42795538902282715, + "train/negative_loss": 0.4274183213710785, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.0005370574072003365, + "train/total_loss": 0.18068444728851318 }, { - "epoch": 30.95693779904306, + "epoch": 24.623574144486692, "step": 3250, - "train/classification_loss": 0.03396584838628769, - "train/contrastive_loss": 0.44770997762680054, - "train/negative_loss": 0.446166455745697, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 0.0015435114037245512, - "train/total_loss": 0.12350784242153168 + "train/classification_loss": 0.01662762649357319, + "train/contrastive_loss": 0.04431380331516266, + "train/negative_loss": 0.044308602809906006, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 5.199381121201441e-06, + "train/total_loss": 0.02549038827419281 }, { - "epoch": 31.43062200956938, - "grad_norm": 5.788070201873779, - "learning_rate": 1.8779770992366414e-05, - "loss": 0.2117, + "epoch": 25.0, + "grad_norm": 0.1426517218351364, + "learning_rate": 1.9030045523520486e-05, + "loss": 0.2653, "step": 3300 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, + "step": 3300, + "train/classification_loss": 0.03253503516316414, + "train/contrastive_loss": 0.08125778287649155, + "train/negative_loss": 0.06892689317464828, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.012330886907875538, + "train/total_loss": 0.04878659173846245 + }, + { + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.14800579845905304, - "train/contrastive_loss": 2.1537089347839355, - "train/negative_loss": 0.809990644454956, + "train/classification_loss": 0.2432728409767151, + "train/contrastive_loss": 1.5128800868988037, + "train/negative_loss": 1.5122932195663452, "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 1.343718409538269, - "train/total_loss": 0.578747570514679 + "train/num_positives": 6, + "train/positive_loss": 0.0005868220468983054, + "train/total_loss": 0.5458488464355469 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.21009111404418945, - "train/contrastive_loss": 3.1362881660461426, - "train/negative_loss": 1.9414223432540894, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.1948657035827637, - "train/total_loss": 0.8373487591743469 + "train/classification_loss": 0.09974425286054611, + "train/contrastive_loss": 1.8228647708892822, + "train/negative_loss": 1.4262115955352783, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.3966531455516815, + "train/total_loss": 0.4643172323703766 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.15864136815071106, - "train/contrastive_loss": 1.5626497268676758, - "train/negative_loss": 0.4516650140285492, - "train/num_negatives": 44, + "train/classification_loss": 0.24507825076580048, + "train/contrastive_loss": 2.4550557136535645, + "train/negative_loss": 1.281431794166565, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.1736239194869995, + "train/total_loss": 0.7360894083976746 + }, + { + "epoch": 25.0, + "step": 3300, + "train/classification_loss": 0.09520050883293152, + "train/contrastive_loss": 1.0583842992782593, + "train/negative_loss": 1.0576406717300415, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.0007436311570927501, + "train/total_loss": 0.30687737464904785 + }, + { + "epoch": 25.0, + "step": 3300, + "train/classification_loss": 0.1952214241027832, + "train/contrastive_loss": 1.332282543182373, + "train/negative_loss": 0.8030473589897156, + "train/num_negatives": 40, "train/num_positives": 12, - "train/positive_loss": 1.1109846830368042, - "train/total_loss": 0.4711713194847107 + "train/positive_loss": 0.5292352437973022, + "train/total_loss": 0.4616779386997223 + }, + { + "epoch": 25.0, + "step": 3300, + "train/classification_loss": 0.27844029664993286, + "train/contrastive_loss": 3.964850425720215, + "train/negative_loss": 1.8056200742721558, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 2.1592302322387695, + "train/total_loss": 1.0714104175567627 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.1556483954191208, - "train/contrastive_loss": 3.069450616836548, - "train/negative_loss": 1.4610708951950073, + "train/classification_loss": 0.19268262386322021, + "train/contrastive_loss": 1.325662612915039, + "train/negative_loss": 1.0957167148590088, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 1.6083797216415405, - "train/total_loss": 0.7695385217666626 + "train/positive_loss": 0.22994588315486908, + "train/total_loss": 0.45781514048576355 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.1448250114917755, - "train/contrastive_loss": 3.2379918098449707, - "train/negative_loss": 1.7954978942871094, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.4424937963485718, - "train/total_loss": 0.7924233675003052 + "train/classification_loss": 0.10857563465833664, + "train/contrastive_loss": 0.09426319599151611, + "train/negative_loss": 0.09015125781297684, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.004111934453248978, + "train/total_loss": 0.12742827832698822 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.13619111478328705, - "train/contrastive_loss": 1.4320874214172363, - "train/negative_loss": 0.16865532100200653, - "train/num_negatives": 32, + "train/classification_loss": 0.15868255496025085, + "train/contrastive_loss": 1.71653413772583, + "train/negative_loss": 1.7140077352523804, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.002526361495256424, + "train/total_loss": 0.5019893646240234 + }, + { + "epoch": 25.0, + "step": 3300, + "train/classification_loss": 0.171363964676857, + "train/contrastive_loss": 6.332603454589844, + "train/negative_loss": 6.332584857940674, + "train/num_negatives": 34, "train/num_positives": 22, - "train/positive_loss": 1.2634321451187134, - "train/total_loss": 0.4226086139678955 + "train/positive_loss": 1.877716749731917e-05, + "train/total_loss": 1.4378846883773804 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.16267898678779602, - "train/contrastive_loss": 2.2757370471954346, - "train/negative_loss": 1.2414798736572266, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 1.034257173538208, - "train/total_loss": 0.6178264021873474 + "train/classification_loss": 0.15402153134346008, + "train/contrastive_loss": 4.169260025024414, + "train/negative_loss": 3.527508497238159, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.641751766204834, + "train/total_loss": 0.9878735542297363 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.221414253115654, - "train/contrastive_loss": 0.7670682668685913, - "train/negative_loss": 0.7666692733764648, - "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.0003989888937212527, - "train/total_loss": 0.37482792139053345 + "train/classification_loss": 0.19588702917099, + "train/contrastive_loss": 2.297722339630127, + "train/negative_loss": 1.2917360067367554, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.005986213684082, + "train/total_loss": 0.6554315090179443 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.11663338541984558, - "train/contrastive_loss": 1.2876126766204834, - "train/negative_loss": 0.7544316649436951, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.5331810712814331, - "train/total_loss": 0.3741559386253357 + "train/classification_loss": 0.17782172560691833, + "train/contrastive_loss": 3.47976016998291, + "train/negative_loss": 1.472051978111267, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.0077083110809326, + "train/total_loss": 0.8737738132476807 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.2755032479763031, - "train/contrastive_loss": 2.9959495067596436, - "train/negative_loss": 1.573377013206482, - "train/num_negatives": 46, - "train/num_positives": 6, - "train/positive_loss": 1.4225724935531616, - "train/total_loss": 0.8746931552886963 + "train/classification_loss": 0.23833522200584412, + "train/contrastive_loss": 3.1224875450134277, + "train/negative_loss": 2.08335018157959, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 1.0391372442245483, + "train/total_loss": 0.86283278465271 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.22606584429740906, - "train/contrastive_loss": 2.8755135536193848, - "train/negative_loss": 1.8919392824172974, - "train/num_negatives": 46, - "train/num_positives": 8, - "train/positive_loss": 0.983574390411377, - "train/total_loss": 0.8011685609817505 + "train/classification_loss": 0.10661076009273529, + "train/contrastive_loss": 0.6379455924034119, + "train/negative_loss": 0.637262761592865, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0006828454206697643, + "train/total_loss": 0.2341998815536499 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.11842040717601776, - "train/contrastive_loss": 1.5844272375106812, - "train/negative_loss": 0.2698725461959839, + "train/classification_loss": 0.15502777695655823, + "train/contrastive_loss": 0.7489840984344482, + "train/negative_loss": 0.6939836144447327, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 1.3145546913146973, - "train/total_loss": 0.4353058338165283 + "train/positive_loss": 0.055000465363264084, + "train/total_loss": 0.3048245906829834 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.2656199038028717, - "train/contrastive_loss": 4.086475372314453, - "train/negative_loss": 2.529242515563965, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.5572329759597778, - "train/total_loss": 1.08291494846344 + "train/classification_loss": 0.19260847568511963, + "train/contrastive_loss": 4.057049751281738, + "train/negative_loss": 1.9627246856689453, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 2.094325065612793, + "train/total_loss": 1.0040184259414673 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.03537563979625702, - "train/contrastive_loss": 0.1012759879231453, - "train/negative_loss": 0.09287260472774506, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.008403382264077663, - "train/total_loss": 0.05563084036111832 + "train/classification_loss": 0.2805325388908386, + "train/contrastive_loss": 3.977631092071533, + "train/negative_loss": 2.4782581329345703, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.499372959136963, + "train/total_loss": 1.0760587453842163 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.18172809481620789, - "train/contrastive_loss": 1.5021051168441772, - "train/negative_loss": 0.8312801122665405, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.6708250045776367, - "train/total_loss": 0.4821491241455078 + "train/classification_loss": 0.28130894899368286, + "train/contrastive_loss": 2.761672019958496, + "train/negative_loss": 1.4260400533676147, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.3356318473815918, + "train/total_loss": 0.83364337682724 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.1431398242712021, - "train/contrastive_loss": 2.570267915725708, - "train/negative_loss": 0.8159906268119812, - "train/num_negatives": 38, - "train/num_positives": 14, - "train/positive_loss": 1.754277229309082, - "train/total_loss": 0.6571934223175049 + "train/classification_loss": 0.20854602754116058, + "train/contrastive_loss": 0.31044256687164307, + "train/negative_loss": 0.3068552017211914, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.0035873630549758673, + "train/total_loss": 0.2706345319747925 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.3330380618572235, - "train/contrastive_loss": 2.982614040374756, - "train/negative_loss": 1.8325543403625488, - "train/num_negatives": 46, + "train/classification_loss": 0.23820187151432037, + "train/contrastive_loss": 1.8961248397827148, + "train/negative_loss": 0.5929375290870667, + "train/num_negatives": 36, "train/num_positives": 10, - "train/positive_loss": 1.150059700012207, - "train/total_loss": 0.9295608997344971 + "train/positive_loss": 1.3031872510910034, + "train/total_loss": 0.617426872253418 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.31108030676841736, - "train/contrastive_loss": 2.671219825744629, - "train/negative_loss": 1.1259933710098267, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.5452265739440918, - "train/total_loss": 0.8453242778778076 + "train/classification_loss": 0.2612606883049011, + "train/contrastive_loss": 1.2913511991500854, + "train/negative_loss": 1.0744271278381348, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.21692410111427307, + "train/total_loss": 0.5195308923721313 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.2029123157262802, - "train/contrastive_loss": 3.9709529876708984, - "train/negative_loss": 1.6683111190795898, + "train/classification_loss": 0.206011101603508, + "train/contrastive_loss": 2.398674964904785, + "train/negative_loss": 2.3986732959747314, "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 2.3026418685913086, - "train/total_loss": 0.9971029162406921 + "train/num_positives": 8, + "train/positive_loss": 1.6093268868644373e-06, + "train/total_loss": 0.6857460737228394 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.1808260828256607, - "train/contrastive_loss": 1.626341700553894, - "train/negative_loss": 1.2797354459762573, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.3466062843799591, - "train/total_loss": 0.5060944557189941 + "train/classification_loss": 0.1616557091474533, + "train/contrastive_loss": 5.75839900970459, + "train/negative_loss": 4.912726402282715, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.8456724882125854, + "train/total_loss": 1.3133355379104614 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.19602756202220917, - "train/contrastive_loss": 1.2590303421020508, - "train/negative_loss": 0.5354171991348267, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.7236132025718689, - "train/total_loss": 0.4478336572647095 + "train/classification_loss": 0.2748758792877197, + "train/contrastive_loss": 3.9220967292785645, + "train/negative_loss": 2.653684616088867, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 1.2684119939804077, + "train/total_loss": 1.0592951774597168 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.24868392944335938, - "train/contrastive_loss": 6.708365440368652, - "train/negative_loss": 5.020604133605957, - "train/num_negatives": 40, - "train/num_positives": 16, - "train/positive_loss": 1.6877614259719849, - "train/total_loss": 1.5903570652008057 + "train/classification_loss": 0.2040204107761383, + "train/contrastive_loss": 2.4594550132751465, + "train/negative_loss": 1.1005076169967651, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.3589472770690918, + "train/total_loss": 0.6959114074707031 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.26906827092170715, - "train/contrastive_loss": 4.042838096618652, - "train/negative_loss": 1.5821844339370728, - "train/num_negatives": 36, - "train/num_positives": 18, - "train/positive_loss": 2.460653781890869, - "train/total_loss": 1.0776358842849731 + "train/classification_loss": 0.12629291415214539, + "train/contrastive_loss": 2.4487836360931396, + "train/negative_loss": 1.9097634553909302, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.5390202403068542, + "train/total_loss": 0.6160496473312378 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.08090478926897049, - "train/contrastive_loss": 1.882480502128601, - "train/negative_loss": 1.2227306365966797, - "train/num_negatives": 42, + "train/classification_loss": 0.28995922207832336, + "train/contrastive_loss": 2.622593879699707, + "train/negative_loss": 0.9086761474609375, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 0.6597498655319214, - "train/total_loss": 0.4574008882045746 + "train/positive_loss": 1.71391761302948, + "train/total_loss": 0.8144780397415161 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.10236773639917374, - "train/contrastive_loss": 1.733987808227539, - "train/negative_loss": 0.9823006987571716, - "train/num_negatives": 48, + "train/classification_loss": 0.22247469425201416, + "train/contrastive_loss": 0.40787750482559204, + "train/negative_loss": 0.39307525753974915, + "train/num_negatives": 46, "train/num_positives": 8, - "train/positive_loss": 0.7516871094703674, - "train/total_loss": 0.4491652846336365 + "train/positive_loss": 0.014802239835262299, + "train/total_loss": 0.3040502071380615 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.17099466919898987, - "train/contrastive_loss": 1.2642008066177368, - "train/negative_loss": 0.8069105744361877, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 0.45729026198387146, - "train/total_loss": 0.42383483052253723 + "train/classification_loss": 0.2023748755455017, + "train/contrastive_loss": 1.6303668022155762, + "train/negative_loss": 1.095678687095642, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.5346881747245789, + "train/total_loss": 0.528448224067688 + }, + { + "epoch": 25.0, + "step": 3300, + "train/classification_loss": 0.20820797979831696, + "train/contrastive_loss": 2.984541893005371, + "train/negative_loss": 2.5536351203918457, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.43090689182281494, + "train/total_loss": 0.8051163554191589 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.0186111181974411, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.0186111181974411 - }, - { - "epoch": 31.43062200956938, - "eval_exact_match_accuracy": 0.40669856459330145, - "eval_hamming_loss": 0.05797917253025612, - "eval_loss": 0.6642809510231018, - "eval_macro_f1": 0.45601288872366735, - "eval_macro_precision": 0.5312114092949007, - "eval_macro_recall": 0.4173924281536737, - "eval_micro_f1": 0.5896414342629482, - "eval_micro_precision": 0.7184466019417476, - "eval_micro_recall": 0.5, - "eval_runtime": 1.0643, - "eval_samples_per_second": 196.374, - "eval_steps_per_second": 25.369, + "train/classification_loss": 0.09071245789527893, + "train/contrastive_loss": 1.7642064094543457, + "train/negative_loss": 1.7287695407867432, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.035436857491731644, + "train/total_loss": 0.44355374574661255 + }, + { + "epoch": 25.0, + "eval_exact_match_accuracy": 0.3435114503816794, + "eval_hamming_loss": 0.06219128872923215, + "eval_loss": 0.6581822633743286, + "eval_macro_f1": 0.34106648690226693, + "eval_macro_precision": 0.4240140182693978, + "eval_macro_recall": 0.29166747861416104, + "eval_micro_f1": 0.5745007680491552, + "eval_micro_precision": 0.722007722007722, + "eval_micro_recall": 0.4770408163265306, + "eval_runtime": 1.3099, + "eval_samples_per_second": 200.019, + "eval_steps_per_second": 25.193, "step": 3300 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.04426412656903267, - "train/contrastive_loss": 0.10552118718624115, - "train/negative_loss": 0.06692909449338913, - "train/num_negatives": 42, - "train/num_positives": 14, - "train/positive_loss": 0.03859209269285202, - "train/total_loss": 0.06536836177110672 + "train/classification_loss": 0.06349444389343262, + "train/contrastive_loss": 0.6426239609718323, + "train/negative_loss": 0.6425623297691345, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 6.163310172269121e-05, + "train/total_loss": 0.1920192390680313 }, { - "epoch": 31.43062200956938, + "epoch": 25.0, "step": 3300, - "train/classification_loss": 0.050094205886125565, - "train/contrastive_loss": 0.06330926716327667, - "train/negative_loss": 0.06313195824623108, - "train/num_negatives": 24, - "train/num_positives": 28, - "train/positive_loss": 0.00017730613762978464, - "train/total_loss": 0.06275606155395508 + "train/classification_loss": 0.03974666818976402, + "train/contrastive_loss": 0.09562747180461884, + "train/negative_loss": 0.09390008449554443, + "train/num_negatives": 38, + "train/num_positives": 18, + "train/positive_loss": 0.0017273901030421257, + "train/total_loss": 0.05887216329574585 }, { - "epoch": 31.90909090909091, - "grad_norm": 4.555282115936279, - "learning_rate": 1.8760687022900765e-05, - "loss": 0.2034, + "epoch": 25.380228136882128, + "grad_norm": 6.141340255737305, + "learning_rate": 1.9014871016691957e-05, + "loss": 0.2328, "step": 3350 }, { - "epoch": 31.90909090909091, + "epoch": 25.380228136882128, "step": 3350, - "train/classification_loss": 0.039144426584243774, - "train/contrastive_loss": 0.1515885293483734, - "train/negative_loss": 0.1513158231973648, - "train/num_negatives": 46, + "train/classification_loss": 0.07157076150178909, + "train/contrastive_loss": 0.16674461960792542, + "train/negative_loss": 0.11869807541370392, + "train/num_negatives": 42, "train/num_positives": 10, - "train/positive_loss": 0.00027269942802377045, - "train/total_loss": 0.0694621354341507 + "train/positive_loss": 0.048046547919511795, + "train/total_loss": 0.1049196869134903 }, { - "epoch": 31.90909090909091, + "epoch": 25.380228136882128, "step": 3350, - "train/classification_loss": 0.022538412362337112, - "train/contrastive_loss": 0.20771776139736176, - "train/negative_loss": 0.14565926790237427, - "train/num_negatives": 50, - "train/num_positives": 6, - "train/positive_loss": 0.06205849349498749, - "train/total_loss": 0.06408196687698364 + "train/classification_loss": 0.0491863377392292, + "train/contrastive_loss": 0.5353805422782898, + "train/negative_loss": 0.5353769063949585, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 3.6358901525090914e-06, + "train/total_loss": 0.15626244246959686 }, { - "epoch": 32.38277511961722, - "grad_norm": 2.1454997062683105, - "learning_rate": 1.8741603053435116e-05, - "loss": 0.2065, + "epoch": 25.76045627376426, + "grad_norm": 2.8165411949157715, + "learning_rate": 1.899969650986343e-05, + "loss": 0.2535, "step": 3400 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.14062102138996124, - "train/contrastive_loss": 1.666820764541626, - "train/negative_loss": 1.1983513832092285, - "train/num_negatives": 48, - "train/num_positives": 8, - "train/positive_loss": 0.46846944093704224, - "train/total_loss": 0.4739851951599121 + "train/classification_loss": 0.03099321573972702, + "train/contrastive_loss": 0.1604263037443161, + "train/negative_loss": 0.1431298702955246, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.017296431586146355, + "train/total_loss": 0.06307847797870636 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.20427674055099487, - "train/contrastive_loss": 2.9264791011810303, - "train/negative_loss": 1.471085786819458, - "train/num_negatives": 24, - "train/num_positives": 32, - "train/positive_loss": 1.4553933143615723, - "train/total_loss": 0.7895725965499878 + "train/classification_loss": 0.2451237589120865, + "train/contrastive_loss": 1.5617382526397705, + "train/negative_loss": 1.5582208633422852, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.00351735926233232, + "train/total_loss": 0.5574713945388794 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.13334983587265015, - "train/contrastive_loss": 1.1195950508117676, - "train/negative_loss": 0.7226179242134094, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.39697718620300293, - "train/total_loss": 0.3572688698768616 + "train/classification_loss": 0.09470326453447342, + "train/contrastive_loss": 2.4531636238098145, + "train/negative_loss": 2.4402997493743896, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.012863829731941223, + "train/total_loss": 0.5853360295295715 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.14801658689975739, - "train/contrastive_loss": 3.1553831100463867, - "train/negative_loss": 1.333030343055725, - "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.822352647781372, - "train/total_loss": 0.7790932059288025 + "train/classification_loss": 0.21301519870758057, + "train/contrastive_loss": 2.0606114864349365, + "train/negative_loss": 1.1346051692962646, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.9260063767433167, + "train/total_loss": 0.6251375079154968 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.1479511260986328, - "train/contrastive_loss": 3.318699359893799, - "train/negative_loss": 1.9247477054595947, - "train/num_negatives": 34, - "train/num_positives": 22, - "train/positive_loss": 1.3939515352249146, - "train/total_loss": 0.8116909861564636 + "train/classification_loss": 0.09756490588188171, + "train/contrastive_loss": 0.8785727620124817, + "train/negative_loss": 0.8765426278114319, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.002030152129009366, + "train/total_loss": 0.27327945828437805 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.1451767235994339, - "train/contrastive_loss": 1.7957091331481934, - "train/negative_loss": 0.5530140995979309, - "train/num_negatives": 32, - "train/num_positives": 22, - "train/positive_loss": 1.2426949739456177, - "train/total_loss": 0.5043185353279114 + "train/classification_loss": 0.19722296297550201, + "train/contrastive_loss": 1.68013334274292, + "train/negative_loss": 1.3542821407318115, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.3258512318134308, + "train/total_loss": 0.5332496166229248 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.13333527743816376, - "train/contrastive_loss": 1.2141146659851074, - "train/negative_loss": 0.8395195007324219, + "train/classification_loss": 0.2297293096780777, + "train/contrastive_loss": 2.748241901397705, + "train/negative_loss": 1.436108112335205, "train/num_negatives": 48, - "train/num_positives": 6, - "train/positive_loss": 0.37459516525268555, - "train/total_loss": 0.376158207654953 + "train/num_positives": 8, + "train/positive_loss": 1.3121336698532104, + "train/total_loss": 0.7793776988983154 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.2217167764902115, - "train/contrastive_loss": 1.5513319969177246, - "train/negative_loss": 1.5513280630111694, - "train/num_negatives": 48, + "train/classification_loss": 0.20641794800758362, + "train/contrastive_loss": 2.657620906829834, + "train/negative_loss": 1.0356405973434448, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 1.6219803094863892, + "train/total_loss": 0.737942099571228 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.09455323219299316, + "train/contrastive_loss": 0.059598419815301895, + "train/negative_loss": 0.05849968269467354, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.0010987359564751387, + "train/total_loss": 0.1064729169011116 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.1705746352672577, + "train/contrastive_loss": 2.756265163421631, + "train/negative_loss": 2.4538519382476807, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.30241310596466064, + "train/total_loss": 0.7218276262283325 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.19520694017410278, + "train/contrastive_loss": 4.845227241516113, + "train/negative_loss": 4.842982769012451, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.0022444853093475103, + "train/total_loss": 1.1642524003982544 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.16895650327205658, + "train/contrastive_loss": 4.367084503173828, + "train/negative_loss": 3.909681558609009, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.45740288496017456, + "train/total_loss": 1.0423734188079834 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.1809540092945099, + "train/contrastive_loss": 2.1995387077331543, + "train/negative_loss": 1.3106061220169067, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.8889327049255371, + "train/total_loss": 0.6208617687225342 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.17229358851909637, + "train/contrastive_loss": 3.231588840484619, + "train/negative_loss": 1.2036175727844238, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 2.0279712677001953, + "train/total_loss": 0.8186113834381104 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.21193243563175201, + "train/contrastive_loss": 1.9270035028457642, + "train/negative_loss": 1.7921016216278076, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.13490186631679535, + "train/total_loss": 0.5973331332206726 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.10786862671375275, + "train/contrastive_loss": 0.37207528948783875, + "train/negative_loss": 0.3720712661743164, + "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 3.8941807360970415e-06, - "train/total_loss": 0.5319831967353821 + "train/positive_loss": 4.013390025647823e-06, + "train/total_loss": 0.1822836846113205 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.16727688908576965, - "train/contrastive_loss": 1.0471539497375488, - "train/negative_loss": 0.6750437021255493, + "train/classification_loss": 0.15325023233890533, + "train/contrastive_loss": 1.064897060394287, + "train/negative_loss": 0.9129491448402405, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 0.3721102476119995, - "train/total_loss": 0.37670767307281494 + "train/positive_loss": 0.15194785594940186, + "train/total_loss": 0.3662296533584595 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.27526921033859253, - "train/contrastive_loss": 2.588313102722168, - "train/negative_loss": 0.9335647225379944, - "train/num_negatives": 46, + "train/classification_loss": 0.22160863876342773, + "train/contrastive_loss": 3.997860908508301, + "train/negative_loss": 1.794972538948059, + "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 1.6547484397888184, - "train/total_loss": 0.792931854724884 + "train/positive_loss": 2.202888250350952, + "train/total_loss": 1.0211808681488037 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.2721159756183624, - "train/contrastive_loss": 2.8597500324249268, - "train/negative_loss": 2.1921215057373047, - "train/num_negatives": 46, + "train/classification_loss": 0.2862570881843567, + "train/contrastive_loss": 3.9449095726013184, + "train/negative_loss": 2.6396265029907227, + "train/num_negatives": 48, "train/num_positives": 8, - "train/positive_loss": 0.6676285266876221, - "train/total_loss": 0.8440660238265991 + "train/positive_loss": 1.3052830696105957, + "train/total_loss": 1.0752389430999756 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.10629865527153015, - "train/contrastive_loss": 1.1325740814208984, - "train/negative_loss": 0.32362380623817444, + "train/classification_loss": 0.2698604166507721, + "train/contrastive_loss": 2.7238426208496094, + "train/negative_loss": 1.337735891342163, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.3861066102981567, + "train/total_loss": 0.8146289587020874 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.2149481326341629, + "train/contrastive_loss": 0.30424925684928894, + "train/negative_loss": 0.26369011402130127, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.04055915027856827, + "train/total_loss": 0.2757979929447174 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.21533192694187164, + "train/contrastive_loss": 1.2119178771972656, + "train/negative_loss": 0.4567946791648865, + "train/num_negatives": 36, + "train/num_positives": 10, + "train/positive_loss": 0.7551231980323792, + "train/total_loss": 0.4577155113220215 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.2572571337223053, + "train/contrastive_loss": 1.4418656826019287, + "train/negative_loss": 1.0049817562103271, "train/num_negatives": 50, "train/num_positives": 6, - "train/positive_loss": 0.8089503049850464, - "train/total_loss": 0.33281347155570984 + "train/positive_loss": 0.4368838965892792, + "train/total_loss": 0.5456302762031555 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.2793283462524414, - "train/contrastive_loss": 3.9349493980407715, - "train/negative_loss": 2.7556097507476807, - "train/num_negatives": 38, - "train/num_positives": 18, - "train/positive_loss": 1.1793395280838013, - "train/total_loss": 1.0663182735443115 + "train/classification_loss": 0.22255632281303406, + "train/contrastive_loss": 2.05981707572937, + "train/negative_loss": 2.059807538986206, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 9.50702087720856e-06, + "train/total_loss": 0.6345197558403015 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.039096470922231674, - "train/contrastive_loss": 0.12449292838573456, - "train/negative_loss": 0.1223239004611969, + "train/classification_loss": 0.18448933959007263, + "train/contrastive_loss": 6.19025993347168, + "train/negative_loss": 5.32510232925415, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.8651578426361084, + "train/total_loss": 1.4225413799285889 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.2878333032131195, + "train/contrastive_loss": 3.437633752822876, + "train/negative_loss": 2.8843636512756348, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.002169025596231222, - "train/total_loss": 0.06399505585432053 + "train/positive_loss": 0.553270161151886, + "train/total_loss": 0.9753600358963013 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.17956577241420746, - "train/contrastive_loss": 1.1348330974578857, - "train/negative_loss": 0.6755077242851257, - "train/num_negatives": 44, - "train/num_positives": 12, - "train/positive_loss": 0.45932537317276, - "train/total_loss": 0.4065324068069458 + "train/classification_loss": 0.20732049643993378, + "train/contrastive_loss": 2.7313613891601562, + "train/negative_loss": 1.0555006265640259, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.6758607625961304, + "train/total_loss": 0.7535927891731262 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.11751966923475266, - "train/contrastive_loss": 2.3910632133483887, - "train/negative_loss": 1.3013373613357544, - "train/num_negatives": 38, + "train/classification_loss": 0.15583164989948273, + "train/contrastive_loss": 2.633301019668579, + "train/negative_loss": 1.8983875513076782, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.7349135279655457, + "train/total_loss": 0.6824918389320374 + }, + { + "epoch": 25.76045627376426, + "step": 3400, + "train/classification_loss": 0.3014976978302002, + "train/contrastive_loss": 2.6790504455566406, + "train/negative_loss": 0.8961038589477539, + "train/num_negatives": 36, "train/num_positives": 14, - "train/positive_loss": 1.0897258520126343, - "train/total_loss": 0.5957323312759399 + "train/positive_loss": 1.7829465866088867, + "train/total_loss": 0.8373078107833862 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.3677001893520355, - "train/contrastive_loss": 3.3496382236480713, - "train/negative_loss": 2.043771743774414, + "train/classification_loss": 0.21236392855644226, + "train/contrastive_loss": 0.3641359210014343, + "train/negative_loss": 0.35586291551589966, "train/num_negatives": 46, - "train/num_positives": 10, - "train/positive_loss": 1.3058664798736572, - "train/total_loss": 1.0376278162002563 + "train/num_positives": 8, + "train/positive_loss": 0.008272991515696049, + "train/total_loss": 0.2851911187171936 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.3521638810634613, - "train/contrastive_loss": 3.366544246673584, - "train/negative_loss": 1.990777611732483, + "train/classification_loss": 0.2167302370071411, + "train/contrastive_loss": 2.2479119300842285, + "train/negative_loss": 1.401942491531372, "train/num_negatives": 40, "train/num_positives": 16, - "train/positive_loss": 1.3757667541503906, - "train/total_loss": 1.0254727602005005 + "train/positive_loss": 0.8459693193435669, + "train/total_loss": 0.6663126349449158 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.24825702607631683, - "train/contrastive_loss": 4.009066581726074, - "train/negative_loss": 2.162397623062134, - "train/num_negatives": 44, + "train/classification_loss": 0.24383023381233215, + "train/contrastive_loss": 3.4067745208740234, + "train/negative_loss": 2.6854748725891113, + "train/num_negatives": 42, "train/num_positives": 12, - "train/positive_loss": 1.8466689586639404, - "train/total_loss": 1.0500702857971191 + "train/positive_loss": 0.7212997674942017, + "train/total_loss": 0.9251852035522461 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.15585412085056305, - "train/contrastive_loss": 0.6032668352127075, - "train/negative_loss": 0.5751528143882751, - "train/num_negatives": 42, - "train/num_positives": 10, - "train/positive_loss": 0.028113991022109985, - "train/total_loss": 0.27650749683380127 + "train/classification_loss": 0.09650248289108276, + "train/contrastive_loss": 1.7968575954437256, + "train/negative_loss": 1.7833348512649536, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.01352268923074007, + "train/total_loss": 0.4558739960193634 + }, + { + "epoch": 25.76045627376426, + "eval_exact_match_accuracy": 0.33587786259541985, + "eval_hamming_loss": 0.0648854961832061, + "eval_loss": 0.6561746597290039, + "eval_macro_f1": 0.3615832829629011, + "eval_macro_precision": 0.45288020024001907, + "eval_macro_recall": 0.3370549254499709, + "eval_micro_f1": 0.5853658536585366, + "eval_micro_precision": 0.6688524590163935, + "eval_micro_recall": 0.5204081632653061, + "eval_runtime": 1.3236, + "eval_samples_per_second": 197.951, + "eval_steps_per_second": 24.933, + "step": 3400 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.2183021456003189, - "train/contrastive_loss": 1.296947956085205, - "train/negative_loss": 0.8836459517478943, - "train/num_negatives": 44, - "train/num_positives": 10, - "train/positive_loss": 0.4133020341396332, - "train/total_loss": 0.47769176959991455 + "train/classification_loss": 0.06349218636751175, + "train/contrastive_loss": 0.07242412120103836, + "train/negative_loss": 0.07192068547010422, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.0005034365458413959, + "train/total_loss": 0.0779770091176033 }, { - "epoch": 32.38277511961722, + "epoch": 25.76045627376426, "step": 3400, - "train/classification_loss": 0.24354805052280426, - "train/contrastive_loss": 6.3578057289123535, - "train/negative_loss": 4.961342811584473, + "train/classification_loss": 0.03888005390763283, + "train/contrastive_loss": 0.051010072231292725, + "train/negative_loss": 0.05100360885262489, + "train/num_negatives": 38, + "train/num_positives": 18, + "train/positive_loss": 6.463827503466746e-06, + "train/total_loss": 0.04908207058906555 + }, + { + "epoch": 26.136882129277566, + "grad_norm": 5.621964931488037, + "learning_rate": 1.8984522003034903e-05, + "loss": 0.26, + "step": 3450 + }, + { + "epoch": 26.136882129277566, + "step": 3450, + "train/classification_loss": 0.05482238158583641, + "train/contrastive_loss": 0.2221813201904297, + "train/negative_loss": 0.18056347966194153, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.04161784425377846, + "train/total_loss": 0.09925864636898041 + }, + { + "epoch": 26.136882129277566, + "step": 3450, + "train/classification_loss": 0.07460784912109375, + "train/contrastive_loss": 0.2857626676559448, + "train/negative_loss": 0.16137027740478516, + "train/num_negatives": 38, + "train/num_positives": 14, + "train/positive_loss": 0.12439237534999847, + "train/total_loss": 0.1317603886127472 + }, + { + "epoch": 26.517110266159698, + "grad_norm": 5.829709053039551, + "learning_rate": 1.8969347496206375e-05, + "loss": 0.2628, + "step": 3500 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.04087063670158386, + "train/contrastive_loss": 0.2730056643486023, + "train/negative_loss": 0.20240381360054016, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.07060185819864273, + "train/total_loss": 0.09547176957130432 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.24416328966617584, + "train/contrastive_loss": 1.2197999954223633, + "train/negative_loss": 1.2050063610076904, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.01479368470609188, + "train/total_loss": 0.4881232976913452 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.09027385711669922, + "train/contrastive_loss": 1.0879557132720947, + "train/negative_loss": 1.0485284328460693, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.039427295327186584, + "train/total_loss": 0.3078650236129761 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.24930012226104736, + "train/contrastive_loss": 2.0239694118499756, + "train/negative_loss": 1.0361988544464111, "train/num_negatives": 40, "train/num_positives": 16, - "train/positive_loss": 1.3964629173278809, - "train/total_loss": 1.5151091814041138 + "train/positive_loss": 0.9877704977989197, + "train/total_loss": 0.6540939807891846 }, { - "epoch": 32.38277511961722, - "step": 3400, - "train/classification_loss": 0.28151726722717285, - "train/contrastive_loss": 4.085949897766113, - "train/negative_loss": 1.8635660409927368, + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.1405484825372696, + "train/contrastive_loss": 1.657487392425537, + "train/negative_loss": 1.183213472366333, "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.4742739796638489, + "train/total_loss": 0.4720459580421448 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.15809781849384308, + "train/contrastive_loss": 0.9944461584091187, + "train/negative_loss": 0.8628441095352173, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.13160207867622375, + "train/total_loss": 0.3569870591163635 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.22484783828258514, + "train/contrastive_loss": 3.2822723388671875, + "train/negative_loss": 1.1114423274993896, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 2.170830011367798, + "train/total_loss": 0.8813023567199707 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.19673900306224823, + "train/contrastive_loss": 1.2354463338851929, + "train/negative_loss": 1.0188989639282227, + "train/num_negatives": 46, + "train/num_positives": 10, + "train/positive_loss": 0.2165474146604538, + "train/total_loss": 0.443828284740448 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.10185237228870392, + "train/contrastive_loss": 0.14335335791110992, + "train/negative_loss": 0.11734618991613388, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.026007171720266342, + "train/total_loss": 0.13052304089069366 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.1786959022283554, + "train/contrastive_loss": 1.9147863388061523, + "train/negative_loss": 1.8889577388763428, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.0258286502212286, + "train/total_loss": 0.561653196811676 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.1728069931268692, + "train/contrastive_loss": 4.631826400756836, + "train/negative_loss": 4.62503719329834, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.006789043080061674, + "train/total_loss": 1.0991723537445068 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.1350773572921753, + "train/contrastive_loss": 2.317467212677002, + "train/negative_loss": 1.8343220949172974, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.48314520716667175, + "train/total_loss": 0.5985708236694336 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.2011203020811081, + "train/contrastive_loss": 1.8068182468414307, + "train/negative_loss": 1.2571401596069336, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.5496780872344971, + "train/total_loss": 0.5624839663505554 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.19098064303398132, + "train/contrastive_loss": 2.7464306354522705, + "train/negative_loss": 1.1040246486663818, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.6424059867858887, + "train/total_loss": 0.7402667999267578 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.2268064171075821, + "train/contrastive_loss": 2.5167148113250732, + "train/negative_loss": 2.0372207164764404, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.4794940650463104, + "train/total_loss": 0.7301493883132935 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.10732728987932205, + "train/contrastive_loss": 0.676042914390564, + "train/negative_loss": 0.6755276322364807, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.0005152623634785414, + "train/total_loss": 0.24253588914871216 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.15172827243804932, + "train/contrastive_loss": 1.4607198238372803, + "train/negative_loss": 0.8999800086021423, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.5607398748397827, + "train/total_loss": 0.44387224316596985 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.22986359894275665, + "train/contrastive_loss": 3.2647483348846436, + "train/negative_loss": 2.350944995880127, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.9138032793998718, + "train/total_loss": 0.8828132748603821 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.31553730368614197, + "train/contrastive_loss": 3.7270119190216064, + "train/negative_loss": 2.609485626220703, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.1175262928009033, + "train/total_loss": 1.0609396696090698 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.2617214620113373, + "train/contrastive_loss": 2.7027387619018555, + "train/negative_loss": 1.3698372840881348, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.3329014778137207, + "train/total_loss": 0.8022692203521729 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.2243420034646988, + "train/contrastive_loss": 0.5735950469970703, + "train/negative_loss": 0.43930789828300476, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.13428717851638794, + "train/total_loss": 0.33906102180480957 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.22474873065948486, + "train/contrastive_loss": 2.1106088161468506, + "train/negative_loss": 0.47914591431617737, + "train/num_negatives": 36, + "train/num_positives": 10, + "train/positive_loss": 1.6314629316329956, + "train/total_loss": 0.646870493888855 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.259309321641922, + "train/contrastive_loss": 1.1438089609146118, + "train/negative_loss": 0.9350510239601135, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.20875796675682068, + "train/total_loss": 0.48807111382484436 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.2109253853559494, + "train/contrastive_loss": 0.852957010269165, + "train/negative_loss": 0.8529542088508606, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 2.8014246709062718e-06, + "train/total_loss": 0.38151678442955017 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.17987607419490814, + "train/contrastive_loss": 5.514058589935303, + "train/negative_loss": 4.61842679977417, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.8956317901611328, + "train/total_loss": 1.282687783241272 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.30330607295036316, + "train/contrastive_loss": 4.087950706481934, + "train/negative_loss": 2.486027479171753, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 1.6019234657287598, + "train/total_loss": 1.1208962202072144 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.19975897669792175, + "train/contrastive_loss": 2.5329947471618652, + "train/negative_loss": 1.1732195615768433, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.3597753047943115, + "train/total_loss": 0.7063579559326172 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.12747396528720856, + "train/contrastive_loss": 1.9260979890823364, + "train/negative_loss": 1.8662116527557373, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.05988633632659912, + "train/total_loss": 0.5126935839653015 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.28724098205566406, + "train/contrastive_loss": 2.3989384174346924, + "train/negative_loss": 0.8391770720481873, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 1.5597612857818604, + "train/total_loss": 0.7670286893844604 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.23326809704303741, + "train/contrastive_loss": 0.9924879670143127, + "train/negative_loss": 0.8530234694480896, + "train/num_negatives": 46, + "train/num_positives": 8, + "train/positive_loss": 0.13946451246738434, + "train/total_loss": 0.43176567554473877 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.1934327781200409, + "train/contrastive_loss": 1.6941725015640259, + "train/negative_loss": 0.956020176410675, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.7381523251533508, + "train/total_loss": 0.5322672724723816 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.1920088678598404, + "train/contrastive_loss": 2.1642441749572754, + "train/negative_loss": 1.8980878591537476, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.2661561965942383, + "train/total_loss": 0.6248577237129211 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.10704698413610458, + "train/contrastive_loss": 2.0262694358825684, + "train/negative_loss": 1.9549133777618408, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 0.07135593891143799, + "train/total_loss": 0.5123008489608765 + }, + { + "epoch": 26.517110266159698, + "eval_exact_match_accuracy": 0.31297709923664124, + "eval_hamming_loss": 0.06511001347103727, + "eval_loss": 0.6037638783454895, + "eval_macro_f1": 0.3615136781901614, + "eval_macro_precision": 0.44078315806145874, + "eval_macro_recall": 0.34336205505116246, + "eval_micro_f1": 0.5833333333333334, + "eval_micro_precision": 0.6677631578947368, + "eval_micro_recall": 0.5178571428571429, + "eval_runtime": 1.3159, + "eval_samples_per_second": 199.108, + "eval_steps_per_second": 25.078, + "step": 3500 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.050277478992938995, + "train/contrastive_loss": 0.24744358658790588, + "train/negative_loss": 0.24671988189220428, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.0007237021345645189, + "train/total_loss": 0.09976619482040405 + }, + { + "epoch": 26.517110266159698, + "step": 3500, + "train/classification_loss": 0.0420842245221138, + "train/contrastive_loss": 0.12675541639328003, + "train/negative_loss": 0.12668782472610474, + "train/num_negatives": 38, "train/num_positives": 18, - "train/positive_loss": 2.222383737564087, - "train/total_loss": 1.0987071990966797 + "train/positive_loss": 6.75850169500336e-05, + "train/total_loss": 0.06743530929088593 }, { - "epoch": 32.38277511961722, - "step": 3400, - "train/classification_loss": 0.06852437555789948, - "train/contrastive_loss": 2.2804019451141357, - "train/negative_loss": 2.030466318130493, + "epoch": 26.897338403041825, + "grad_norm": 5.520751476287842, + "learning_rate": 1.8954172989377846e-05, + "loss": 0.2578, + "step": 3550 + }, + { + "epoch": 26.897338403041825, + "step": 3550, + "train/classification_loss": 0.06911422312259674, + "train/contrastive_loss": 0.38478147983551025, + "train/negative_loss": 0.3839743435382843, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.0008071434567682445, + "train/total_loss": 0.14607051014900208 + }, + { + "epoch": 26.897338403041825, + "step": 3550, + "train/classification_loss": 0.04467397183179855, + "train/contrastive_loss": 0.13470162451267242, + "train/negative_loss": 0.13461384177207947, "train/num_negatives": 42, "train/num_positives": 14, - "train/positive_loss": 0.24993567168712616, - "train/total_loss": 0.5246047973632812 + "train/positive_loss": 8.778787014307454e-05, + "train/total_loss": 0.07161429524421692 }, { - "epoch": 32.38277511961722, - "step": 3400, - "train/classification_loss": 0.11203252524137497, - "train/contrastive_loss": 1.7833894491195679, - "train/negative_loss": 1.2253434658050537, + "epoch": 27.273764258555133, + "grad_norm": 1.2910661697387695, + "learning_rate": 1.8938998482549317e-05, + "loss": 0.1961, + "step": 3600 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.03896690905094147, + "train/contrastive_loss": 0.3950559198856354, + "train/negative_loss": 0.2890613079071045, + "train/num_negatives": 30, + "train/num_positives": 26, + "train/positive_loss": 0.10599461942911148, + "train/total_loss": 0.11797809600830078 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.2626723647117615, + "train/contrastive_loss": 1.518162727355957, + "train/negative_loss": 1.5147135257720947, + "train/num_negatives": 48, + "train/num_positives": 6, + "train/positive_loss": 0.0034492413979023695, + "train/total_loss": 0.5663049221038818 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.09132181853055954, + "train/contrastive_loss": 2.2070178985595703, + "train/negative_loss": 1.600876808166504, + "train/num_negatives": 32, + "train/num_positives": 24, + "train/positive_loss": 0.6061410307884216, + "train/total_loss": 0.5327253937721252 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.2697763442993164, + "train/contrastive_loss": 2.473684787750244, + "train/negative_loss": 1.2972780466079712, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.176406741142273, + "train/total_loss": 0.7645133137702942 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.13061441481113434, + "train/contrastive_loss": 1.7107596397399902, + "train/negative_loss": 1.160955786705017, + "train/num_negatives": 36, + "train/num_positives": 20, + "train/positive_loss": 0.5498038530349731, + "train/total_loss": 0.47276633977890015 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.18963827192783356, + "train/contrastive_loss": 1.145474910736084, + "train/negative_loss": 0.7887651920318604, + "train/num_negatives": 40, + "train/num_positives": 12, + "train/positive_loss": 0.35670968890190125, + "train/total_loss": 0.41873323917388916 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.2434290498495102, + "train/contrastive_loss": 3.3733468055725098, + "train/negative_loss": 1.2152137756347656, "train/num_negatives": 48, "train/num_positives": 8, - "train/positive_loss": 0.5580459833145142, - "train/total_loss": 0.46871042251586914 + "train/positive_loss": 2.158133029937744, + "train/total_loss": 0.9180984497070312 }, { - "epoch": 32.38277511961722, - "step": 3400, - "train/classification_loss": 0.22319838404655457, - "train/contrastive_loss": 1.6257950067520142, - "train/negative_loss": 0.9708728790283203, + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.2135680913925171, + "train/contrastive_loss": 1.2590053081512451, + "train/negative_loss": 0.7274684309959412, "train/num_negatives": 46, "train/num_positives": 10, - "train/positive_loss": 0.6549221277236938, - "train/total_loss": 0.548357367515564 + "train/positive_loss": 0.5315368175506592, + "train/total_loss": 0.46536916494369507 }, { - "epoch": 32.38277511961722, - "step": 3400, - "train/classification_loss": 0.009490761905908585, - "train/contrastive_loss": 0.0, - "train/negative_loss": 0.0, - "train/num_negatives": 0, - "train/num_positives": 0, - "train/positive_loss": 0.0, - "train/total_loss": 0.009490761905908585 - }, - { - "epoch": 32.38277511961722, - "eval_exact_match_accuracy": 0.4258373205741627, - "eval_hamming_loss": 0.056290458767238954, - "eval_loss": 0.6555870771408081, - "eval_macro_f1": 0.45208023609527814, - "eval_macro_precision": 0.5340377160965396, - "eval_macro_recall": 0.42312821741887485, - "eval_micro_f1": 0.6138996138996139, - "eval_micro_precision": 0.7162162162162162, - "eval_micro_recall": 0.5371621621621622, - "eval_runtime": 1.0625, - "eval_samples_per_second": 196.703, - "eval_steps_per_second": 25.411, - "step": 3400 + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.11367911845445633, + "train/contrastive_loss": 0.08795370906591415, + "train/negative_loss": 0.08716916292905807, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 0.0007845453801564872, + "train/total_loss": 0.13126985728740692 }, { - "epoch": 32.38277511961722, - "step": 3400, + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.17117154598236084, + "train/contrastive_loss": 1.8075507879257202, + "train/negative_loss": 1.763033151626587, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 0.04451759159564972, + "train/total_loss": 0.5326817035675049 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.18474672734737396, + "train/contrastive_loss": 4.208568572998047, + "train/negative_loss": 4.080986022949219, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.12758265435695648, + "train/total_loss": 1.0264604091644287 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.13995344936847687, + "train/contrastive_loss": 3.058405876159668, + "train/negative_loss": 2.5021185874938965, + "train/num_negatives": 34, + "train/num_positives": 16, + "train/positive_loss": 0.5562872290611267, + "train/total_loss": 0.7516345977783203 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.19392699003219604, + "train/contrastive_loss": 2.261110544204712, + "train/negative_loss": 1.2504570484161377, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 1.0106534957885742, + "train/total_loss": 0.6461490988731384 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.19774414598941803, + "train/contrastive_loss": 2.6791462898254395, + "train/negative_loss": 1.2418702840805054, + "train/num_negatives": 52, + "train/num_positives": 4, + "train/positive_loss": 1.437276005744934, + "train/total_loss": 0.7335733771324158 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.20452485978603363, + "train/contrastive_loss": 2.163522958755493, + "train/negative_loss": 1.8638108968734741, + "train/num_negatives": 44, + "train/num_positives": 12, + "train/positive_loss": 0.29971206188201904, + "train/total_loss": 0.6372294425964355 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.11192843317985535, + "train/contrastive_loss": 0.5887251496315002, + "train/negative_loss": 0.5886683464050293, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 5.6825520005077124e-05, + "train/total_loss": 0.22967347502708435 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.17801986634731293, + "train/contrastive_loss": 0.8904207944869995, + "train/negative_loss": 0.7229050993919373, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.16751568019390106, + "train/total_loss": 0.3561040163040161 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.18601901829242706, + "train/contrastive_loss": 2.4977288246154785, + "train/negative_loss": 1.539307713508606, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.9584212303161621, + "train/total_loss": 0.6855648159980774 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.29851365089416504, + "train/contrastive_loss": 3.9171507358551025, + "train/negative_loss": 2.472288131713867, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.4448626041412354, + "train/total_loss": 1.0819437503814697 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.2816876471042633, + "train/contrastive_loss": 2.650508165359497, + "train/negative_loss": 1.3075979948043823, + "train/num_negatives": 48, + "train/num_positives": 8, + "train/positive_loss": 1.3429101705551147, + "train/total_loss": 0.8117892742156982 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.22853101789951324, + "train/contrastive_loss": 0.9125323295593262, + "train/negative_loss": 0.663948655128479, + "train/num_negatives": 40, + "train/num_positives": 6, + "train/positive_loss": 0.24858370423316956, + "train/total_loss": 0.41103750467300415 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.22826716303825378, + "train/contrastive_loss": 1.779207468032837, + "train/negative_loss": 0.2818395495414734, + "train/num_negatives": 36, + "train/num_positives": 10, + "train/positive_loss": 1.4973678588867188, + "train/total_loss": 0.5841086506843567 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.26458168029785156, + "train/contrastive_loss": 1.8742969036102295, + "train/negative_loss": 1.2306065559387207, + "train/num_negatives": 50, + "train/num_positives": 6, + "train/positive_loss": 0.6436903476715088, + "train/total_loss": 0.6394410729408264 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.19689883291721344, + "train/contrastive_loss": 0.7221878170967102, + "train/negative_loss": 0.722175121307373, + "train/num_negatives": 44, + "train/num_positives": 8, + "train/positive_loss": 1.2695951227215119e-05, + "train/total_loss": 0.3413363993167877 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.16559024155139923, + "train/contrastive_loss": 5.641915321350098, + "train/negative_loss": 4.857276439666748, + "train/num_negatives": 34, + "train/num_positives": 22, + "train/positive_loss": 0.7846390604972839, + "train/total_loss": 1.2939733266830444 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.2922292947769165, + "train/contrastive_loss": 3.6553738117218018, + "train/negative_loss": 2.413935899734497, + "train/num_negatives": 42, + "train/num_positives": 14, + "train/positive_loss": 1.2414379119873047, + "train/total_loss": 1.0233039855957031 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.23776789009571075, + "train/contrastive_loss": 3.039091110229492, + "train/negative_loss": 1.0969626903533936, + "train/num_negatives": 30, + "train/num_positives": 20, + "train/positive_loss": 1.9421285390853882, + "train/total_loss": 0.8455861210823059 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.11832771450281143, + "train/contrastive_loss": 2.55261492729187, + "train/negative_loss": 1.888369083404541, + "train/num_negatives": 54, + "train/num_positives": 2, + "train/positive_loss": 0.6642459034919739, + "train/total_loss": 0.6288507580757141 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.31666257977485657, + "train/contrastive_loss": 2.6672844886779785, + "train/negative_loss": 0.8443609476089478, + "train/num_negatives": 36, + "train/num_positives": 14, + "train/positive_loss": 1.8229234218597412, + "train/total_loss": 0.8501194715499878 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.23869599401950836, + "train/contrastive_loss": 0.30176228284835815, + "train/negative_loss": 0.29503902792930603, + "train/num_negatives": 46, + "train/num_positives": 8, + "train/positive_loss": 0.006723259575664997, + "train/total_loss": 0.29904845356941223 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.21165335178375244, + "train/contrastive_loss": 1.5867743492126465, + "train/negative_loss": 0.8900071382522583, + "train/num_negatives": 40, + "train/num_positives": 16, + "train/positive_loss": 0.6967671513557434, + "train/total_loss": 0.5290082693099976 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.1919238269329071, + "train/contrastive_loss": 1.1165754795074463, + "train/negative_loss": 1.1130186319351196, + "train/num_negatives": 42, + "train/num_positives": 12, + "train/positive_loss": 0.003556900192052126, + "train/total_loss": 0.4152389168739319 + }, + { + "epoch": 27.273764258555133, + "step": 3600, + "train/classification_loss": 0.10730393975973129, + "train/contrastive_loss": 3.5370101928710938, + "train/negative_loss": 2.5218238830566406, + "train/num_negatives": 28, + "train/num_positives": 2, + "train/positive_loss": 1.0151864290237427, + "train/total_loss": 0.8147059679031372 + }, + { + "epoch": 27.273764258555133, + "eval_exact_match_accuracy": 0.3435114503816794, + "eval_hamming_loss": 0.061068702290076333, + "eval_loss": 0.6214548349380493, + "eval_macro_f1": 0.34450747222511713, + "eval_macro_precision": 0.4155203008644811, + "eval_macro_recall": 0.30412920801370735, + "eval_micro_f1": 0.5866261398176292, + "eval_micro_precision": 0.7255639097744361, + "eval_micro_recall": 0.4923469387755102, + "eval_runtime": 1.3221, + "eval_samples_per_second": 198.171, + "eval_steps_per_second": 24.96, + "step": 3600 + }, + { + "epoch": 27.273764258555133, + "step": 3600, "total_flos": 0.0, - "train_loss": 0.6106048311906702, - "train_runtime": 911.6953, - "train_samples_per_second": 915.328, - "train_steps_per_second": 57.585 + "train_loss": 0.6497067387898763, + "train_runtime": 976.094, + "train_samples_per_second": 1074.179, + "train_steps_per_second": 67.616 } ], "logging_steps": 50, - "max_steps": 52500, + "max_steps": 66000, "num_input_tokens_seen": 0, "num_train_epochs": 500, "save_steps": 2000,