diff --git "a/checkpoint-3969/trainer_state.json" "b/checkpoint-3969/trainer_state.json" --- "a/checkpoint-3969/trainer_state.json" +++ "b/checkpoint-3969/trainer_state.json" @@ -11,2254 +11,2254 @@ { "epoch": 1.0, "eval_Claim": { - "f1-score": 0.5385059461955894, - "precision": 0.6267132491265789, - "recall": 0.4720647773279352, - "support": 4940.0 + "f1-score": 0.5254768748366867, + "precision": 0.5768789443488239, + "recall": 0.4824856046065259, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.7814784727863526, - "precision": 0.9588039867109635, - "recall": 0.6595063985374772, - "support": 2188.0 + "f1-score": 0.7984361424847959, + "precision": 0.7495921696574225, + "recall": 0.854089219330855, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9938608715302625, - "precision": 0.9996969237763298, - "recall": 0.9880925634688834, - "support": 13353.0 + "f1-score": 0.9990277532260916, + "precision": 0.9988511841640155, + "recall": 0.9992043847241867, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.8968837690149166, - "precision": 0.8455112497215416, - "recall": 0.954902824076986, - "support": 15899.0 + "f1-score": 0.8854183656825966, + "precision": 0.8719781543651113, + "recall": 0.8992793837488611, + "support": 12073.0 }, - "eval_accuracy": 0.8837548103353491, - "eval_loss": 0.2860471308231354, + "eval_accuracy": 0.8755765022723447, + "eval_loss": 0.28303390741348267, "eval_macro avg": { - "f1-score": 0.8026822648817803, - "precision": 0.8576813523338533, - "recall": 0.7686416408528204, - "support": 36380.0 - }, - "eval_runtime": 4.9572, - "eval_samples_per_second": 16.34, - "eval_steps_per_second": 2.219, + "f1-score": 0.8020897840575427, + "precision": 0.7993251131338432, + "recall": 0.8087646481026072, + "support": 29705.0 + }, + "eval_runtime": 4.8106, + "eval_samples_per_second": 16.63, + "eval_steps_per_second": 2.079, "eval_weighted avg": { - "f1-score": 0.8768739289107891, - "precision": 0.8792073379911217, - "recall": 0.8837548103353491, - "support": 36380.0 + "f1-score": 0.8718761173649596, + "precision": 0.8700202202343714, + "recall": 0.8755765022723447, + "support": 29705.0 }, "step": 81 }, { "epoch": 2.0, "eval_Claim": { - "f1-score": 0.6469459868753155, - "precision": 0.6453172205438067, - "recall": 0.648582995951417, - "support": 4940.0 + "f1-score": 0.5578782336033447, + "precision": 0.6124497991967871, + "recall": 0.5122360844529751, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.8435970110796187, - "precision": 0.9669226225634967, - "recall": 0.7481718464351006, - "support": 2188.0 + "f1-score": 0.784421283598464, + "precision": 0.9571619812583668, + "recall": 0.6644981412639405, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.998877245508982, - "precision": 0.9983541557567143, - "recall": 0.9994008836965476, - "support": 13353.0 + "f1-score": 0.9998674150351351, + "precision": 0.9997348652231551, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9099026477336144, - "precision": 0.8972179761540813, - "recall": 0.922951129001824, - "support": 15899.0 + "f1-score": 0.8977749872463995, + "precision": 0.8530201342281879, + "recall": 0.9474861260664292, + "support": 12073.0 }, - "eval_accuracy": 0.903243540406817, - "eval_loss": 0.24885737895965576, + "eval_accuracy": 0.8859114627167144, + "eval_loss": 0.27363118529319763, "eval_macro avg": { - "f1-score": 0.8498307227993828, - "precision": 0.8769529937545246, - "recall": 0.8297767137712222, - "support": 36380.0 - }, - "eval_runtime": 5.0097, - "eval_samples_per_second": 16.169, - "eval_steps_per_second": 2.196, + "f1-score": 0.8099854798708358, + "precision": 0.8555916949766242, + "recall": 0.7810550879458362, + "support": 29705.0 + }, + "eval_runtime": 4.7727, + "eval_samples_per_second": 16.762, + "eval_steps_per_second": 2.095, "eval_weighted avg": { - "f1-score": 0.9028656814459712, - "precision": 0.9043261520711511, - "recall": 0.903243540406817, - "support": 36380.0 + "f1-score": 0.8807489883812781, + "precision": 0.8826802296805741, + "recall": 0.8859114627167144, + "support": 29705.0 }, "step": 162 }, { "epoch": 3.0, "eval_Claim": { - "f1-score": 0.7040630102775911, - "precision": 0.6698958142935478, - "recall": 0.7419028340080972, - "support": 4940.0 + "f1-score": 0.64760736196319, + "precision": 0.6627322953289804, + "recall": 0.6331573896353166, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.8888411094388303, - "precision": 0.8392204628501827, - "recall": 0.9446983546617916, - "support": 2188.0 + "f1-score": 0.9016018306636155, + "precision": 0.8881875563570785, + "recall": 0.9154275092936803, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.998989104796136, - "precision": 0.9988769092542678, - "recall": 0.9991013255448213, - "support": 13353.0 + "f1-score": 0.9996904982977407, + "precision": 1.0, + "recall": 0.9993811881188119, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.914582593823615, - "precision": 0.9390987408880053, - "recall": 0.8913139191144097, - "support": 15899.0 + "f1-score": 0.8982820417748115, + "precision": 0.8936065573770492, + "recall": 0.9030067091857865, + "support": 12073.0 }, - "eval_accuracy": 0.9137987905442551, - "eval_loss": 0.24298855662345886, + "eval_accuracy": 0.9027436458508669, + "eval_loss": 0.22585685551166534, "eval_macro avg": { - "f1-score": 0.8766189545840432, - "precision": 0.8617729818215009, - "recall": 0.8942541083322799, - "support": 36380.0 - }, - "eval_runtime": 4.9876, - "eval_samples_per_second": 16.24, - "eval_steps_per_second": 2.205, + "f1-score": 0.8617954331748394, + "precision": 0.861131602265777, + "recall": 0.8627431990583988, + "support": 29705.0 + }, + "eval_runtime": 4.8003, + "eval_samples_per_second": 16.666, + "eval_steps_per_second": 2.083, "eval_weighted avg": { - "f1-score": 0.9154289662937856, - "precision": 0.9184781183611038, - "recall": 0.9137987905442551, - "support": 36380.0 + "f1-score": 0.9019670975035187, + "precision": 0.9013351218793044, + "recall": 0.9027436458508669, + "support": 29705.0 }, "step": 243 }, { "epoch": 4.0, "eval_Claim": { - "f1-score": 0.6981611893583725, - "precision": 0.6754352763058289, - "recall": 0.7224696356275304, - "support": 4940.0 + "f1-score": 0.6477602966229081, + "precision": 0.5561865427637239, + "recall": 0.7754318618042226, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.901231802911534, - "precision": 0.8836187966622749, - "recall": 0.9195612431444241, - "support": 2188.0 + "f1-score": 0.8984409799554566, + "precision": 0.8627031650983746, + "recall": 0.9372676579925651, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9989130842172332, - "precision": 0.9998499399759904, - "recall": 0.9979779824758481, - "support": 13353.0 + "f1-score": 1.0, + "precision": 1.0, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9146224912392481, - "precision": 0.9266670970240785, - "recall": 0.9028869740235235, - "support": 15899.0 + "f1-score": 0.861495720750997, + "precision": 0.9384029675907849, + "recall": 0.7962395427814131, + "support": 12073.0 }, - "eval_accuracy": 0.9142935678944475, - "eval_loss": 0.23980534076690674, + "eval_accuracy": 0.8811311227066151, + "eval_loss": 0.32341843843460083, "eval_macro avg": { - "f1-score": 0.8782321419315969, - "precision": 0.8713927774920431, - "recall": 0.8857239588178316, - "support": 36380.0 - }, - "eval_runtime": 4.9726, - "eval_samples_per_second": 16.289, - "eval_steps_per_second": 2.212, + "f1-score": 0.8519242493323405, + "precision": 0.8393231688632209, + "recall": 0.8772347656445502, + "support": 29705.0 + }, + "eval_runtime": 4.7708, + "eval_samples_per_second": 16.769, + "eval_steps_per_second": 2.096, "eval_weighted avg": { - "f1-score": 0.9153623106642749, - "precision": 0.9168247558035478, - "recall": 0.9142935678944475, - "support": 36380.0 + "f1-score": 0.8869263673393438, + "precision": 0.9027457246003854, + "recall": 0.8811311227066151, + "support": 29705.0 }, "step": 324 }, { "epoch": 5.0, "eval_Claim": { - "f1-score": 0.6650318783717509, - "precision": 0.6450999048525214, - "recall": 0.6862348178137652, - "support": 4940.0 + "f1-score": 0.6188788236838649, + "precision": 0.6833864888373442, + "recall": 0.5654990403071017, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9172031076581575, - "precision": 0.8916702632714717, - "recall": 0.9442413162705667, - "support": 2188.0 + "f1-score": 0.909733679000707, + "precision": 0.9230033476805356, + "recall": 0.8968401486988847, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9982383147794145, - "precision": 0.9992495872730002, - "recall": 0.9972290870965326, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.8998438545616774, - "precision": 0.9119622787753521, - "recall": 0.8880432731618341, - "support": 15899.0 + "f1-score": 0.8978936810431294, + "precision": 0.8706816059757236, + "recall": 0.9268615919821088, + "support": 12073.0 }, - "eval_accuracy": 0.9040956569543706, - "eval_loss": 0.2932502031326294, + "eval_accuracy": 0.9018347079616226, + "eval_loss": 0.3072384297847748, "eval_macro avg": { - "f1-score": 0.87007928884275, - "precision": 0.8619955085430864, - "recall": 0.8789371235856747, - "support": 36380.0 - }, - "eval_runtime": 4.9595, - "eval_samples_per_second": 16.332, - "eval_steps_per_second": 2.218, + "f1-score": 0.8566154962081685, + "precision": 0.8692457621526151, + "recall": 0.8473001952470238, + "support": 29705.0 + }, + "eval_runtime": 4.7922, + "eval_samples_per_second": 16.694, + "eval_steps_per_second": 2.087, "eval_weighted avg": { - "f1-score": 0.9051179642561333, - "precision": 0.9065430476942037, - "recall": 0.9040956569543706, - "support": 36380.0 + "f1-score": 0.8984684143294738, + "precision": 0.8974043833368577, + "recall": 0.9018347079616226, + "support": 29705.0 }, "step": 405 }, { "epoch": 6.0, "eval_Claim": { - "f1-score": 0.6622698072805139, - "precision": 0.5740163325909429, - "recall": 0.7825910931174089, - "support": 4940.0 + "f1-score": 0.620837537840565, + "precision": 0.6545212765957447, + "recall": 0.5904510556621881, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9223704351629127, - "precision": 0.8842767295597485, - "recall": 0.9638939670932358, - "support": 2188.0 + "f1-score": 0.8993613741466637, + "precision": 0.8547509418166597, + "recall": 0.9488847583643123, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9982004948639124, - "precision": 0.9993994444861497, - "recall": 0.9970044184827379, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.8780749379985253, - "precision": 0.9398091685199799, - "recall": 0.8239511918988616, - "support": 15899.0 + "f1-score": 0.8931567692054615, + "precision": 0.8869558114841134, + "recall": 0.8994450426571688, + "support": 12073.0 }, - "eval_accuracy": 0.8902693787795491, - "eval_loss": 0.3604726195335388, + "eval_accuracy": 0.8979633058407676, + "eval_loss": 0.3634909987449646, "eval_macro avg": { - "f1-score": 0.865228918826466, - "precision": 0.8493754187892053, - "recall": 0.8918601676480611, - "support": 36380.0 - }, - "eval_runtime": 4.991, - "eval_samples_per_second": 16.229, - "eval_steps_per_second": 2.204, + "f1-score": 0.8533278705744156, + "precision": 0.8490349090033436, + "recall": 0.8596952141709173, + "support": 29705.0 + }, + "eval_runtime": 4.7765, + "eval_samples_per_second": 16.749, + "eval_steps_per_second": 2.094, "eval_weighted avg": { - "f1-score": 0.8955262233990811, - "precision": 0.9086708334194257, - "recall": 0.8902693787795491, - "support": 36380.0 + "f1-score": 0.8960665959131486, + "precision": 0.8950239457358053, + "recall": 0.8979633058407676, + "support": 29705.0 }, "step": 486 }, { "epoch": 6.17, - "grad_norm": 17.04454231262207, + "grad_norm": 9.48759651184082, "learning_rate": 1.7530864197530865e-05, - "loss": 0.2104, + "loss": 0.214, "step": 500 }, { "epoch": 7.0, "eval_Claim": { - "f1-score": 0.7037529399734124, - "precision": 0.7110973341599504, - "recall": 0.6965587044534413, - "support": 4940.0 + "f1-score": 0.6812872467222885, + "precision": 0.6769303647560397, + "recall": 0.685700575815739, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9196885428253615, - "precision": 0.8959687906371911, - "recall": 0.9446983546617916, - "support": 2188.0 + "f1-score": 0.9276377217553688, + "precision": 0.9319887429643527, + "recall": 0.9233271375464684, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9987257327036954, - "precision": 0.9996248780853777, - "recall": 0.997828203399985, - "support": 13353.0 + "f1-score": 0.9999115983026874, + "precision": 0.9999115983026874, + "recall": 0.9999115983026874, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9188781285372909, - "precision": 0.9187048098082364, - "recall": 0.9190515126737531, - "support": 15899.0 + "f1-score": 0.9017086927670869, + "precision": 0.9029819752471135, + "recall": 0.9004389961070156, + "support": 12073.0 }, - "eval_accuracy": 0.9192963166575041, - "eval_loss": 0.36414769291877747, + "eval_accuracy": 0.9098468271334792, + "eval_loss": 0.3468942642211914, "eval_macro avg": { - "f1-score": 0.8852613360099401, - "precision": 0.8813489531726889, - "recall": 0.8895341937972427, - "support": 36380.0 - }, - "eval_runtime": 4.9835, - "eval_samples_per_second": 16.254, - "eval_steps_per_second": 2.207, + "f1-score": 0.8776363148868579, + "precision": 0.8779531703175483, + "recall": 0.8773445769429775, + "support": 29705.0 + }, + "eval_runtime": 4.7913, + "eval_samples_per_second": 16.697, + "eval_steps_per_second": 2.087, "eval_weighted avg": { - "f1-score": 0.9190227083446231, - "precision": 0.9188476996393494, - "recall": 0.9192963166575041, - "support": 36380.0 + "f1-score": 0.9100559053806798, + "precision": 0.910277290769933, + "recall": 0.9098468271334792, + "support": 29705.0 }, "step": 567 }, { "epoch": 8.0, "eval_Claim": { - "f1-score": 0.7025399811853246, - "precision": 0.6562390158172232, - "recall": 0.7558704453441295, - "support": 4940.0 + "f1-score": 0.6928905033731187, + "precision": 0.6105725260654838, + "recall": 0.8008637236084453, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9105835367206567, - "precision": 0.8848641655886158, - "recall": 0.9378427787934186, - "support": 2188.0 + "f1-score": 0.9302433371958285, + "precision": 0.9278779472954231, + "recall": 0.9326208178438662, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9986513823331086, - "precision": 0.9991005172026085, - "recall": 0.9982026510896428, - "support": 13353.0 + "f1-score": 0.9999115904871364, + "precision": 1.0, + "recall": 0.9998231966053748, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9083384525849526, - "precision": 0.9345974717232203, - "recall": 0.8835146864582678, - "support": 15899.0 + "f1-score": 0.8820387074174622, + "precision": 0.9356247097073851, + "recall": 0.8342582622380519, + "support": 12073.0 }, - "eval_accuracy": 0.911544804837823, - "eval_loss": 0.35222452878952026, + "eval_accuracy": 0.8997475172529877, + "eval_loss": 0.40174800157546997, "eval_macro avg": { - "f1-score": 0.8800283382060107, - "precision": 0.8687002925829169, - "recall": 0.8938576404213647, - "support": 36380.0 - }, - "eval_runtime": 4.9741, - "eval_samples_per_second": 16.285, - "eval_steps_per_second": 2.211, + "f1-score": 0.8762710346183864, + "precision": 0.868518795767073, + "recall": 0.8918915000739345, + "support": 29705.0 + }, + "eval_runtime": 4.7722, + "eval_samples_per_second": 16.764, + "eval_steps_per_second": 2.095, "eval_weighted avg": { - "f1-score": 0.9136769997620249, - "precision": 0.9174837257168467, - "recall": 0.911544804837823, - "support": 36380.0 + "f1-score": 0.9038782866839282, + "precision": 0.9139692560686062, + "recall": 0.8997475172529877, + "support": 29705.0 }, "step": 648 }, { "epoch": 9.0, "eval_Claim": { - "f1-score": 0.6566933693658913, - "precision": 0.6275594908688434, - "recall": 0.6886639676113361, - "support": 4940.0 + "f1-score": 0.6602241979126401, + "precision": 0.7130531589201224, + "recall": 0.6146833013435701, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.8702745722244328, - "precision": 0.7706131078224101, - "recall": 0.9995429616087751, - "support": 2188.0 + "f1-score": 0.9052154195011338, + "precision": 0.8839681133746679, + "recall": 0.9275092936802974, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9987648313807688, - "precision": 0.9983537862915295, - "recall": 0.9991762150827529, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9001826722338205, - "precision": 0.9350138917124077, - "recall": 0.8678533241084345, - "support": 15899.0 + "f1-score": 0.9072885349800925, + "precision": 0.8903596204449405, + "recall": 0.9248736850824153, + "support": 12073.0 }, - "eval_accuracy": 0.8996426608026388, - "eval_loss": 0.4436376690864563, + "eval_accuracy": 0.910149806429894, + "eval_loss": 0.4713464379310608, "eval_macro avg": { - "f1-score": 0.8564788613012284, - "precision": 0.8328850691737977, - "recall": 0.8888091171028246, - "support": 36380.0 - }, - "eval_runtime": 5.0402, - "eval_samples_per_second": 16.071, - "eval_steps_per_second": 2.182, + "f1-score": 0.8681709883747097, + "precision": 0.8718231247141469, + "recall": 0.8667665700265708, + "support": 29705.0 + }, + "eval_runtime": 4.7731, + "eval_samples_per_second": 16.761, + "eval_steps_per_second": 2.095, "eval_weighted avg": { - "f1-score": 0.9015045934020747, - "precision": 0.90662587513727, - "recall": 0.8996426608026388, - "support": 36380.0 + "f1-score": 0.9077607320175198, + "precision": 0.9067368029754924, + "recall": 0.910149806429894, + "support": 29705.0 }, "step": 729 }, { "epoch": 10.0, "eval_Claim": { - "f1-score": 0.7371103705155853, - "precision": 0.7145572025845686, - "recall": 0.7611336032388664, - "support": 4940.0 + "f1-score": 0.6952499153785399, + "precision": 0.6562300319488817, + "recall": 0.7392034548944337, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9365328549002018, - "precision": 0.9194187582562747, - "recall": 0.9542961608775137, - "support": 2188.0 + "f1-score": 0.9103908484270734, + "precision": 0.9344422700587084, + "recall": 0.887546468401487, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9991386090408599, - "precision": 0.9993257416841475, - "recall": 0.9989515464689583, - "support": 13353.0 + "f1-score": 1.0, + "precision": 1.0, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9225428371233836, - "precision": 0.9344473837021743, - "recall": 0.9109377948298635, - "support": 15899.0 + "f1-score": 0.9012517385257303, + "precision": 0.9174532349407929, + "recall": 0.8856125238134681, + "support": 12073.0 }, - "eval_accuracy": 0.9255085211654756, - "eval_loss": 0.44094258546829224, + "eval_accuracy": 0.9087695674128935, + "eval_loss": 0.4968256950378418, "eval_macro avg": { - "f1-score": 0.8988311678950076, - "precision": 0.8919372715567913, - "recall": 0.9063297763538004, - "support": 36380.0 - }, - "eval_runtime": 4.9458, - "eval_samples_per_second": 16.378, - "eval_steps_per_second": 2.224, + "f1-score": 0.8767231255828358, + "precision": 0.8770313842370958, + "recall": 0.8780906117773473, + "support": 29705.0 + }, + "eval_runtime": 4.7941, + "eval_samples_per_second": 16.687, + "eval_steps_per_second": 2.086, "eval_weighted avg": { - "f1-score": 0.9263184587909816, - "precision": 0.927497977075915, - "recall": 0.9255085211654756, - "support": 36380.0 + "f1-score": 0.9106135328171674, + "precision": 0.9134657277821738, + "recall": 0.9087695674128935, + "support": 29705.0 }, "step": 810 }, { "epoch": 11.0, "eval_Claim": { - "f1-score": 0.6750768578394996, - "precision": 0.7086579123080348, - "recall": 0.6445344129554655, - "support": 4940.0 + "f1-score": 0.6796779702366431, + "precision": 0.6913151364764268, + "recall": 0.668426103646833, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9008264462809917, - "precision": 0.8220211161387632, - "recall": 0.9963436928702011, - "support": 2188.0 + "f1-score": 0.9301903898458749, + "precision": 0.9079646017699115, + "recall": 0.9535315985130112, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9990631440884392, - "precision": 0.9998499849984999, - "recall": 0.9982775406275743, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9168605748066159, - "precision": 0.9167452681883921, - "recall": 0.9169759104346186, - "support": 15899.0 + "f1-score": 0.901923474663909, + "precision": 0.9008428358948934, + "recall": 0.9030067091857865, + "support": 12073.0 }, - "eval_accuracy": 0.9145959318306762, - "eval_loss": 0.4954843521118164, + "eval_accuracy": 0.9106884362901868, + "eval_loss": 0.5864725708961487, "eval_macro avg": { - "f1-score": 0.8729567557538865, - "precision": 0.8618185704084225, - "recall": 0.8890328892219649, - "support": 36380.0 - }, - "eval_runtime": 5.0503, - "eval_samples_per_second": 16.039, - "eval_steps_per_second": 2.178, + "f1-score": 0.8779369089628499, + "precision": 0.8750085450645222, + "recall": 0.8812411028364077, + "support": 29705.0 + }, + "eval_runtime": 4.8031, + "eval_samples_per_second": 16.656, + "eval_steps_per_second": 2.082, "eval_weighted avg": { - "f1-score": 0.9132365141301059, - "precision": 0.9132952764575458, - "recall": 0.9145959318306762, - "support": 36380.0 + "f1-score": 0.9101191594213591, + "precision": 0.9096858090555641, + "recall": 0.9106884362901868, + "support": 29705.0 }, "step": 891 }, { "epoch": 12.0, "eval_Claim": { - "f1-score": 0.6857200770244248, - "precision": 0.6866247209255125, - "recall": 0.6848178137651821, - "support": 4940.0 + "f1-score": 0.697112745636184, + "precision": 0.6425824731835661, + "recall": 0.7617562380038387, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9214571488734471, - "precision": 0.8543537680593518, - "recall": 1.0, - "support": 2188.0 + "f1-score": 0.9365296803652968, + "precision": 0.920556552962298, + "recall": 0.9530669144981413, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9987652935234033, - "precision": 0.9979811574697174, - "recall": 0.9995506627724107, - "support": 13353.0 + "f1-score": 1.0, + "precision": 1.0, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9127542413343095, - "precision": 0.9239592731022039, - "recall": 0.9018177243851815, - "support": 15899.0 + "f1-score": 0.8923037300940034, + "precision": 0.9260513186029936, + "recall": 0.8609293464756067, + "support": 12073.0 }, - "eval_accuracy": 0.91412864211105, - "eval_loss": 0.5210011005401611, + "eval_accuracy": 0.9066487123379903, + "eval_loss": 0.5556066632270813, "eval_macro avg": { - "f1-score": 0.8796741901888963, - "precision": 0.8657297298891964, - "recall": 0.8965465502306935, - "support": 36380.0 - }, - "eval_runtime": 4.9497, - "eval_samples_per_second": 16.365, - "eval_steps_per_second": 2.222, + "f1-score": 0.8814865390238711, + "precision": 0.8722975861872144, + "recall": 0.8939381247443967, + "support": 29705.0 + }, + "eval_runtime": 4.7999, + "eval_samples_per_second": 16.667, + "eval_steps_per_second": 2.083, "eval_weighted avg": { - "f1-score": 0.9140186385274314, - "precision": 0.9147147620844138, - "recall": 0.91412864211105, - "support": 36380.0 + "f1-score": 0.9091318205481447, + "precision": 0.9140393543072853, + "recall": 0.9066487123379903, + "support": 29705.0 }, "step": 972 }, { "epoch": 12.35, - "grad_norm": 3.2667109966278076, + "grad_norm": 3.987734794616699, "learning_rate": 1.506172839506173e-05, - "loss": 0.0286, + "loss": 0.0252, "step": 1000 }, { "epoch": 13.0, "eval_Claim": { - "f1-score": 0.7024380291747424, - "precision": 0.7079991774624718, - "recall": 0.6969635627530364, - "support": 4940.0 + "f1-score": 0.6895428929242329, + "precision": 0.7212470526591565, + "recall": 0.6605086372360844, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9075119409465915, - "precision": 0.8643507030603805, - "recall": 0.9552102376599635, - "support": 2188.0 + "f1-score": 0.9265742214139577, + "precision": 0.9069870939029817, + "recall": 0.9470260223048327, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9986895802912875, - "precision": 0.9985774183887391, - "recall": 0.9988017673930952, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9200429808482398, - "precision": 0.9246014101505431, - "recall": 0.9155292785709793, - "support": 15899.0 + "f1-score": 0.9083234293676489, + "precision": 0.8989292667099286, + "recall": 0.917916010933488, + "support": 12073.0 }, - "eval_accuracy": 0.9188015393073117, - "eval_loss": 0.5909315347671509, + "eval_accuracy": 0.9151657970038714, + "eval_loss": 0.5547633767127991, "eval_macro avg": { - "f1-score": 0.8821706328152152, - "precision": 0.8738821772655336, - "recall": 0.8916262115942686, - "support": 36380.0 - }, - "eval_runtime": 4.9699, - "eval_samples_per_second": 16.298, - "eval_steps_per_second": 2.213, + "f1-score": 0.8810990862027029, + "precision": 0.8817687548472308, + "recall": 0.8813626676186013, + "support": 29705.0 + }, + "eval_runtime": 4.8002, + "eval_samples_per_second": 16.666, + "eval_steps_per_second": 2.083, "eval_weighted avg": { - "f1-score": 0.9186076225687217, - "precision": 0.918717904416961, - "recall": 0.9188015393073117, - "support": 36380.0 + "f1-score": 0.9138424940934561, + "precision": 0.9130371003853032, + "recall": 0.9151657970038714, + "support": 29705.0 }, "step": 1053 }, { "epoch": 14.0, "eval_Claim": { - "f1-score": 0.7095000509113125, - "precision": 0.7137881581643106, - "recall": 0.7052631578947368, - "support": 4940.0 + "f1-score": 0.6662641873943491, + "precision": 0.6706368497812348, + "recall": 0.6619481765834933, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.921028466483012, - "precision": 0.9252767527675276, - "recall": 0.916819012797075, - "support": 2188.0 + "f1-score": 0.9252051582649472, + "precision": 0.9337434926644581, + "recall": 0.91682156133829, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9994754982766372, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9989515464689583, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9206986297074409, - "precision": 0.9180215107553776, - "recall": 0.9233914082646707, - "support": 15899.0 + "f1-score": 0.8968191757085687, + "precision": 0.8933914187078744, + "recall": 0.9002733371987078, + "support": 12073.0 }, - "eval_accuracy": 0.9211105002748763, - "eval_loss": 0.559357225894928, + "eval_accuracy": 0.9060090893788925, + "eval_loss": 0.598924994468689, "eval_macro avg": { - "f1-score": 0.8876756613446006, - "precision": 0.889271605421804, - "recall": 0.8861062813563602, - "support": 36380.0 - }, - "eval_runtime": 4.9811, - "eval_samples_per_second": 16.262, - "eval_steps_per_second": 2.208, + "f1-score": 0.8720721303419663, + "precision": 0.8744429402883919, + "recall": 0.8697607687801228, + "support": 29705.0 + }, + "eval_runtime": 4.8166, + "eval_samples_per_second": 16.609, + "eval_steps_per_second": 2.076, "eval_weighted avg": { - "f1-score": 0.9209544908843665, - "precision": 0.9208148168193181, - "recall": 0.9211105002748763, - "support": 36380.0 + "f1-score": 0.9058181633386757, + "precision": 0.9056571278963901, + "recall": 0.9060090893788925, + "support": 29705.0 }, "step": 1134 }, { "epoch": 15.0, "eval_Claim": { - "f1-score": 0.6915445526531356, - "precision": 0.6282454109475774, - "recall": 0.7690283400809717, - "support": 4940.0 + "f1-score": 0.6859867017940033, + "precision": 0.7189061267420458, + "recall": 0.6559500959692899, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9327135900841081, - "precision": 0.9042918454935622, - "recall": 0.9629798903107861, - "support": 2188.0 + "f1-score": 0.9230404925408477, + "precision": 0.9410912602607436, + "recall": 0.9056691449814126, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9995130539011873, - "precision": 0.9998501199040767, - "recall": 0.9991762150827529, - "support": 13353.0 + "f1-score": 1.0, + "precision": 1.0, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.8984226716408141, - "precision": 0.936421311139914, - "recall": 0.8633876344424177, - "support": 15899.0 + "f1-score": 0.9089134677944047, + "precision": 0.89272306094736, + "recall": 0.9257019796239543, + "support": 12073.0 }, - "eval_accuracy": 0.9064046179219352, - "eval_loss": 0.6213898658752441, + "eval_accuracy": 0.9146944958761152, + "eval_loss": 0.5877691507339478, "eval_macro avg": { - "f1-score": 0.8805484670698113, - "precision": 0.8672021718712826, - "recall": 0.898643019979232, - "support": 36380.0 - }, - "eval_runtime": 4.9756, - "eval_samples_per_second": 16.28, - "eval_steps_per_second": 2.211, + "f1-score": 0.8794851655323139, + "precision": 0.8881801119875374, + "recall": 0.8718303051436642, + "support": 29705.0 + }, + "eval_runtime": 4.7894, + "eval_samples_per_second": 16.704, + "eval_steps_per_second": 2.088, "eval_weighted avg": { - "f1-score": 0.9094977265082566, - "precision": 0.9159231436204942, - "recall": 0.9064046179219352, - "support": 36380.0 + "f1-score": 0.9133441511431463, + "precision": 0.9126906124611831, + "recall": 0.9146944958761152, + "support": 29705.0 }, "step": 1215 }, { "epoch": 16.0, "eval_Claim": { - "f1-score": 0.6819089456869009, - "precision": 0.6727738376674547, - "recall": 0.6912955465587044, - "support": 4940.0 + "f1-score": 0.6963696369636965, + "precision": 0.6844300278035218, + "recall": 0.7087332053742802, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9290586630286495, - "precision": 0.9244343891402715, - "recall": 0.9337294332723949, - "support": 2188.0 + "f1-score": 0.9235896215186861, + "precision": 0.9468033186920449, + "recall": 0.9014869888475836, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9998127410958391, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9996255523103422, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9087691578448412, - "precision": 0.9131843007748, - "recall": 0.9043965029247123, - "support": 15899.0 + "f1-score": 0.9048587195552051, + "precision": 0.9065513801130695, + "recall": 0.9031723680940943, + "support": 12073.0 }, - "eval_accuracy": 0.9121770203408466, - "eval_loss": 0.6214332580566406, + "eval_accuracy": 0.9126409695337485, + "eval_loss": 0.5437762141227722, "eval_macro avg": { - "f1-score": 0.8798873769140577, - "precision": 0.8775981318956315, - "recall": 0.8822617587665385, - "support": 36380.0 - }, - "eval_runtime": 4.962, - "eval_samples_per_second": 16.324, - "eval_steps_per_second": 2.217, + "f1-score": 0.881204494509397, + "precision": 0.884446181652159, + "recall": 0.8783481405789896, + "support": 29705.0 + }, + "eval_runtime": 4.8031, + "eval_samples_per_second": 16.656, + "eval_steps_per_second": 2.082, "eval_weighted avg": { - "f1-score": 0.9126011797368843, - "precision": 0.9130808795913876, - "recall": 0.9121770203408466, - "support": 36380.0 + "f1-score": 0.913192823886985, + "precision": 0.9138872213369954, + "recall": 0.9126409695337485, + "support": 29705.0 }, "step": 1296 }, { "epoch": 17.0, "eval_Claim": { - "f1-score": 0.6977198697068404, - "precision": 0.7524590163934426, - "recall": 0.6504048582995952, - "support": 4940.0 + "f1-score": 0.647066981001248, + "precision": 0.7666776207689779, + "recall": 0.5597408829174664, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.935002195871761, - "precision": 0.8998309382924767, - "recall": 0.9730347349177331, - "support": 2188.0 + "f1-score": 0.9200805910006717, + "precision": 0.8876889848812095, + "recall": 0.9549256505576208, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9994757732344791, - "precision": 0.9994757732344791, - "recall": 0.9994757732344791, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9219572623103128, - "precision": 0.908120309926179, - "recall": 0.9362224039247752, - "support": 15899.0 + "f1-score": 0.9095471382482972, + "precision": 0.8760165720423507, + "recall": 0.9457467075291974, + "support": 12073.0 }, - "eval_accuracy": 0.9228422210005498, - "eval_loss": 0.6075455546379089, + "eval_accuracy": 0.912910284463895, + "eval_loss": 0.6229106783866882, "eval_macro avg": { - "f1-score": 0.8885387752808483, - "precision": 0.8899715094616443, - "recall": 0.8897844425941457, - "support": 36380.0 - }, - "eval_runtime": 5.0046, - "eval_samples_per_second": 16.185, - "eval_steps_per_second": 2.198, + "f1-score": 0.8691626278387974, + "precision": 0.8825736959523487, + "recall": 0.8651033102510712, + "support": 29705.0 + }, + "eval_runtime": 4.8073, + "eval_samples_per_second": 16.641, + "eval_steps_per_second": 2.08, "eval_weighted avg": { - "f1-score": 0.9207454500932067, - "precision": 0.9200160099363349, - "recall": 0.9228422210005498, - "support": 36380.0 + "f1-score": 0.9079094842894391, + "precision": 0.9087011338660376, + "recall": 0.912910284463895, + "support": 29705.0 }, "step": 1377 }, { "epoch": 18.0, "eval_Claim": { - "f1-score": 0.6748943740271293, - "precision": 0.7486433152442032, - "recall": 0.6143724696356275, - "support": 4940.0 + "f1-score": 0.6821297989031078, + "precision": 0.6511780104712042, + "recall": 0.7161708253358925, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9180620674257648, - "precision": 0.8974247053688346, - "recall": 0.9396709323583181, - "support": 2188.0 + "f1-score": 0.922899353647276, + "precision": 0.9169724770642201, + "recall": 0.9289033457249071, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9997753145596164, - "precision": 0.9998501984870047, - "recall": 0.9997004418482738, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9206027683147655, - "precision": 0.898945097099017, - "recall": 0.9433297691678721, - "support": 15899.0 + "f1-score": 0.8941394877853255, + "precision": 0.9112487100103199, + "recall": 0.8776608962146939, + "support": 12073.0 }, - "eval_accuracy": 0.9191313908741067, - "eval_loss": 0.6335331201553345, + "eval_accuracy": 0.9053021376872581, + "eval_loss": 0.5977877378463745, "eval_macro avg": { - "f1-score": 0.878333631081819, - "precision": 0.8862158290497649, - "recall": 0.874268403252523, - "support": 36380.0 - }, - "eval_runtime": 4.9489, - "eval_samples_per_second": 16.367, - "eval_steps_per_second": 2.223, + "f1-score": 0.8747811103601705, + "precision": 0.8698277009156502, + "recall": 0.8806837668188734, + "support": 29705.0 + }, + "eval_runtime": 4.7819, + "eval_samples_per_second": 16.73, + "eval_steps_per_second": 2.091, "eval_weighted avg": { - "f1-score": 0.9161451676737934, - "precision": 0.9154807870211, - "recall": 0.9191313908741067, - "support": 36380.0 + "f1-score": 0.9067713337488226, + "precision": 0.9089358856298488, + "recall": 0.9053021376872581, + "support": 29705.0 }, "step": 1458 }, { "epoch": 18.52, - "grad_norm": 0.10011027753353119, + "grad_norm": 3.641792058944702, "learning_rate": 1.2592592592592593e-05, - "loss": 0.0091, + "loss": 0.0101, "step": 1500 }, { "epoch": 19.0, "eval_Claim": { - "f1-score": 0.7046035805626598, - "precision": 0.7439243924392439, - "recall": 0.6692307692307692, - "support": 4940.0 + "f1-score": 0.6483875179527353, + "precision": 0.711257519335434, + "recall": 0.5957293666026872, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9264422045893201, - "precision": 0.8727272727272727, - "recall": 0.9872029250457038, - "support": 2188.0 + "f1-score": 0.9248662479646429, + "precision": 0.9259431765253843, + "recall": 0.9237918215613383, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.999213041034289, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9984273197034375, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.921880854252529, - "precision": 0.9153078306156612, - "recall": 0.9285489653437323, - "support": 15899.0 + "f1-score": 0.9020460770098276, + "precision": 0.8779302234417875, + "recall": 0.92752422761534, + "support": 12073.0 }, - "eval_accuracy": 0.9225123694337548, - "eval_loss": 0.6328368782997131, + "eval_accuracy": 0.9082982662851372, + "eval_loss": 0.6089487075805664, "eval_macro avg": { - "f1-score": 0.8880349201096995, - "precision": 0.8829898739455444, - "recall": 0.8958524948309107, - "support": 36380.0 - }, - "eval_runtime": 4.9991, - "eval_samples_per_second": 16.203, - "eval_steps_per_second": 2.2, + "f1-score": 0.8688249607318015, + "precision": 0.8787827298256514, + "recall": 0.8617613539448413, + "support": 29705.0 + }, + "eval_runtime": 4.7853, + "eval_samples_per_second": 16.718, + "eval_steps_per_second": 2.09, "eval_weighted avg": { - "f1-score": 0.9210355324440019, - "precision": 0.9205605544347316, - "recall": 0.9225123694337548, - "support": 36380.0 + "f1-score": 0.9054096491562553, + "precision": 0.9045077476547858, + "recall": 0.9082982662851372, + "support": 29705.0 }, "step": 1539 }, { "epoch": 20.0, "eval_Claim": { - "f1-score": 0.6881697835679765, - "precision": 0.7153778942769768, - "recall": 0.6629554655870445, - "support": 4940.0 + "f1-score": 0.7207543032562219, + "precision": 0.66363819907127, + "recall": 0.7886276391554703, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9309440559440559, - "precision": 0.8919597989949749, - "recall": 0.973491773308958, - "support": 2188.0 + "f1-score": 0.9401589527816735, + "precision": 0.9459078080903104, + "recall": 0.9344795539033457, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9997753313862053, - "precision": 0.9997753313862053, - "recall": 0.9997753313862053, - "support": 13353.0 + "f1-score": 1.0, + "precision": 1.0, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9161451814768461, - "precision": 0.9115248116555632, - "recall": 0.92081262972514, - "support": 15899.0 + "f1-score": 0.8998161371702227, + "precision": 0.9299982322785929, + "recall": 0.8715315166073055, + "support": 12073.0 }, - "eval_accuracy": 0.9179494227597581, - "eval_loss": 0.6887677907943726, + "eval_accuracy": 0.9133815855916513, + "eval_loss": 0.6035876274108887, "eval_macro avg": { - "f1-score": 0.883758588093771, - "precision": 0.87965945907843, - "recall": 0.8892588000018369, - "support": 36380.0 - }, - "eval_runtime": 5.0229, - "eval_samples_per_second": 16.126, - "eval_steps_per_second": 2.19, + "f1-score": 0.8901823483020296, + "precision": 0.8848860598600434, + "recall": 0.8986596774165303, + "support": 29705.0 + }, + "eval_runtime": 4.7726, + "eval_samples_per_second": 16.762, + "eval_steps_per_second": 2.095, "eval_weighted avg": { - "f1-score": 0.916774507023963, - "precision": 0.9161052176591827, - "recall": 0.9179494227597581, - "support": 36380.0 + "f1-score": 0.9157652323317351, + "precision": 0.9204344815700675, + "recall": 0.9133815855916513, + "support": 29705.0 }, "step": 1620 }, { "epoch": 21.0, "eval_Claim": { - "f1-score": 0.6423871114136601, - "precision": 0.7307692307692307, - "recall": 0.573076923076923, - "support": 4940.0 + "f1-score": 0.6858804196688157, + "precision": 0.7248196633716271, + "recall": 0.650911708253359, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.8879392212725545, - "precision": 0.9239130434782609, - "recall": 0.8546617915904936, - "support": 2188.0 + "f1-score": 0.9371935559187485, + "precision": 0.9418113561708118, + "recall": 0.9326208178438662, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9984999624990625, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9970044184827379, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9172614007499698, - "precision": 0.883336245558856, - "recall": 0.9538964714761935, - "support": 15899.0 + "f1-score": 0.907693558880937, + "precision": 0.8915248821790878, + "recall": 0.9244595378116458, + "support": 12073.0 }, - "eval_accuracy": 0.9120395821880154, - "eval_loss": 0.7241741418838501, + "eval_accuracy": 0.9154351119340178, + "eval_loss": 0.6373494267463684, "eval_macro avg": { - "f1-score": 0.8615219239838117, - "precision": 0.8845046299515869, - "recall": 0.844659901156587, - "support": 36380.0 - }, - "eval_runtime": 4.9949, - "eval_samples_per_second": 16.217, - "eval_steps_per_second": 2.202, + "f1-score": 0.8826918836171254, + "precision": 0.8895389754303817, + "recall": 0.8769980159772177, + "support": 29705.0 + }, + "eval_runtime": 4.7918, + "eval_samples_per_second": 16.695, + "eval_steps_per_second": 2.087, "eval_weighted avg": { - "f1-score": 0.9079909938510604, - "precision": 0.9078802833224489, - "recall": 0.9120395821880154, - "support": 36380.0 + "f1-score": 0.913858759733591, + "precision": 0.9130855511853444, + "recall": 0.9154351119340178, + "support": 29705.0 }, "step": 1701 }, { "epoch": 22.0, "eval_Claim": { - "f1-score": 0.6720596061734965, - "precision": 0.7086419753086419, - "recall": 0.6390688259109312, - "support": 4940.0 + "f1-score": 0.6833652924256951, + "precision": 0.6827107279693486, + "recall": 0.684021113243762, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9051685393258427, - "precision": 0.8903625110521662, - "recall": 0.9204753199268738, - "support": 2188.0 + "f1-score": 0.9286995515695068, + "precision": 0.8973136915077989, + "recall": 0.962360594795539, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9997004867090977, - "precision": 0.9995507973347308, - "recall": 0.9998502209241369, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9166278528178853, - "precision": 0.9051882742548755, - "recall": 0.9283602742310837, - "support": 15899.0 + "f1-score": 0.9007964638672282, + "precision": 0.9070372858582466, + "recall": 0.8946409343162428, + "support": 12073.0 }, - "eval_accuracy": 0.9148433205057724, - "eval_loss": 0.7163732051849365, + "eval_accuracy": 0.9101161420636257, + "eval_loss": 0.6425191760063171, "eval_macro avg": { - "f1-score": 0.8733891212565805, - "precision": 0.8759358894876037, - "recall": 0.8719386602482564, - "support": 36380.0 - }, - "eval_runtime": 4.9476, - "eval_samples_per_second": 16.371, - "eval_steps_per_second": 2.223, + "f1-score": 0.8782042772418507, + "precision": 0.8717433278630627, + "recall": 0.8852556605888859, + "support": 29705.0 + }, + "eval_runtime": 4.8051, + "eval_samples_per_second": 16.649, + "eval_steps_per_second": 2.081, "eval_weighted avg": { - "f1-score": 0.9132201772820274, - "precision": 0.9122428450081295, - "recall": 0.9148433205057724, - "support": 36380.0 + "f1-score": 0.9100704832242509, + "precision": 0.910224494827858, + "recall": 0.9101161420636257, + "support": 29705.0 }, "step": 1782 }, { "epoch": 23.0, "eval_Claim": { - "f1-score": 0.7022826614861584, - "precision": 0.6750700280112045, - "recall": 0.7317813765182186, - "support": 4940.0 + "f1-score": 0.7114282372207277, + "precision": 0.6941337594156586, + "recall": 0.7296065259117083, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9080485115766262, - "precision": 0.8772901576480614, - "recall": 0.9410420475319927, - "support": 2188.0 + "f1-score": 0.9322964318389753, + "precision": 0.918018018018018, + "recall": 0.9470260223048327, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9988753936122357, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9977533138620535, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9134190823574582, - "precision": 0.9295994789970694, - "recall": 0.8977923139820114, - "support": 15899.0 + "f1-score": 0.9069348418185628, + "precision": 0.9177408412483039, + "recall": 0.8963803528534747, + "support": 12073.0 }, - "eval_accuracy": 0.9145409565695437, - "eval_loss": 0.695353627204895, + "eval_accuracy": 0.9161083992593839, + "eval_loss": 0.6442272663116455, "eval_macro avg": { - "f1-score": 0.8806564122581195, - "precision": 0.8704899161640838, - "recall": 0.892092262973569, - "support": 36380.0 - }, - "eval_runtime": 4.9626, - "eval_samples_per_second": 16.322, - "eval_steps_per_second": 2.217, + "f1-score": 0.8876648777195664, + "precision": 0.8824731546704951, + "recall": 0.893253225267504, + "support": 29705.0 + }, + "eval_runtime": 4.7909, + "eval_samples_per_second": 16.698, + "eval_steps_per_second": 2.087, "eval_weighted avg": { - "f1-score": 0.9157921828580727, - "precision": 0.9177311412832249, - "recall": 0.9145409565695437, - "support": 36380.0 + "f1-score": 0.9167803117094421, + "precision": 0.917711141572463, + "recall": 0.9161083992593839, + "support": 29705.0 }, "step": 1863 }, { "epoch": 24.0, "eval_Claim": { - "f1-score": 0.6945771284068559, - "precision": 0.6463308349311487, - "recall": 0.7506072874493928, - "support": 4940.0 + "f1-score": 0.70479277610558, + "precision": 0.6809843400447427, + "recall": 0.7303262955854126, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9287416071041802, - "precision": 0.8826677645121449, - "recall": 0.979890310786106, - "support": 2188.0 + "f1-score": 0.9265043638033992, + "precision": 0.915985467756585, + "recall": 0.9372676579925651, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9992880428673137, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9985770987793006, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9030182916923878, - "precision": 0.9339381720430108, - "recall": 0.8740801308258381, - "support": 15899.0 + "f1-score": 0.904177523745482, + "precision": 0.917754457810767, + "recall": 0.8909964383334714, + "support": 12073.0 }, - "eval_accuracy": 0.9093732820230896, - "eval_loss": 0.7689566016197205, + "eval_accuracy": 0.9133142568591146, + "eval_loss": 0.6410267353057861, "eval_macro avg": { - "f1-score": 0.8814062675176844, - "precision": 0.8657341928715762, - "recall": 0.9007887069601594, - "support": 36380.0 - }, - "eval_runtime": 4.9669, - "eval_samples_per_second": 16.308, - "eval_steps_per_second": 2.215, + "f1-score": 0.8838686659136152, + "precision": 0.8786810664030237, + "recall": 0.8896475979778622, + "support": 29705.0 + }, + "eval_runtime": 4.8171, + "eval_samples_per_second": 16.607, + "eval_steps_per_second": 2.076, "eval_weighted avg": { - "f1-score": 0.911596446033489, - "precision": 0.9160482240413489, - "recall": 0.9093732820230896, - "support": 36380.0 + "f1-score": 0.9143090027231838, + "precision": 0.9157243906772613, + "recall": 0.9133142568591146, + "support": 29705.0 }, "step": 1944 }, { "epoch": 24.69, - "grad_norm": 0.003183299908414483, + "grad_norm": 0.002705144230276346, "learning_rate": 1.0123456790123458e-05, - "loss": 0.0045, + "loss": 0.0041, "step": 2000 }, { "epoch": 25.0, "eval_Claim": { - "f1-score": 0.6596491228070176, - "precision": 0.7481386392811297, - "recall": 0.5898785425101215, - "support": 4940.0 + "f1-score": 0.6708741775211494, + "precision": 0.761817627325404, + "recall": 0.5993282149712092, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9256909832351607, - "precision": 0.9177897574123989, - "recall": 0.9337294332723949, - "support": 2188.0 + "f1-score": 0.9283402681460935, + "precision": 0.9236430542778289, + "recall": 0.9330855018587361, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9998876614866129, - "precision": 0.9999251048532055, - "recall": 0.9998502209241369, - "support": 13353.0 + "f1-score": 1.0, + "precision": 1.0, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9174541242455649, - "precision": 0.8901046903649376, - "recall": 0.9465375180828983, - "support": 15899.0 + "f1-score": 0.9124055491144605, + "precision": 0.881839258114374, + "recall": 0.94516690135012, + "support": 12073.0 }, - "eval_accuracy": 0.9169048927982408, - "eval_loss": 0.7267045378684998, + "eval_accuracy": 0.9166470291196768, + "eval_loss": 0.6537314653396606, "eval_macro avg": { - "f1-score": 0.8756704729435891, - "precision": 0.888989547977918, - "recall": 0.867498928697388, - "support": 36380.0 - }, - "eval_runtime": 4.9577, - "eval_samples_per_second": 16.338, - "eval_steps_per_second": 2.219, + "f1-score": 0.8779049986954259, + "precision": 0.8918249849294018, + "recall": 0.8693951545450163, + "support": 29705.0 + }, + "eval_runtime": 4.8009, + "eval_samples_per_second": 16.664, + "eval_steps_per_second": 2.083, "eval_weighted avg": { - "f1-score": 0.9131990545133639, - "precision": 0.9128010792876335, - "recall": 0.9169048927982408, - "support": 36380.0 + "f1-score": 0.9130268986169812, + "precision": 0.9130241066053866, + "recall": 0.9166470291196768, + "support": 29705.0 }, "step": 2025 }, { "epoch": 26.0, "eval_Claim": { - "f1-score": 0.7098703888334996, - "precision": 0.6994106090373281, - "recall": 0.7206477732793523, - "support": 4940.0 + "f1-score": 0.7055278711322285, + "precision": 0.682358215646716, + "recall": 0.7303262955854126, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9220324134910205, - "precision": 0.8851976450798991, - "recall": 0.9620658135283364, - "support": 2188.0 + "f1-score": 0.9258656751103881, + "precision": 0.9260808926080892, + "recall": 0.9256505576208178, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9995504607777028, - "precision": 1.0, - "recall": 0.9991013255448213, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9171909755322529, - "precision": 0.9268511977393873, - "recall": 0.9077300459148374, - "support": 15899.0 + "f1-score": 0.9045403093950447, + "precision": 0.9157894736842105, + "recall": 0.8935641514122422, + "support": 12073.0 }, - "eval_accuracy": 0.9191313908741067, - "eval_loss": 0.6802815794944763, + "eval_accuracy": 0.9135162430567244, + "eval_loss": 0.6640132665634155, "eval_macro avg": { - "f1-score": 0.887161059658619, - "precision": 0.8778648629641536, - "recall": 0.8973862395668368, - "support": 36380.0 - }, - "eval_runtime": 4.9696, - "eval_samples_per_second": 16.299, - "eval_steps_per_second": 2.213, + "f1-score": 0.8839724141856584, + "precision": 0.8810350470139681, + "recall": 0.8873852511546182, + "support": 29705.0 + }, + "eval_runtime": 4.8275, + "eval_samples_per_second": 16.572, + "eval_steps_per_second": 2.071, "eval_weighted avg": { - "f1-score": 0.9195597378864155, - "precision": 0.9203108314716255, - "recall": 0.9191313908741067, - "support": 36380.0 + "f1-score": 0.9144964914035518, + "precision": 0.9158162439956733, + "recall": 0.9135162430567244, + "support": 29705.0 }, "step": 2106 }, { "epoch": 27.0, "eval_Claim": { - "f1-score": 0.698643105080467, - "precision": 0.7271731990365666, - "recall": 0.6722672064777327, - "support": 4940.0 + "f1-score": 0.6693952997095327, + "precision": 0.7442748091603053, + "recall": 0.6082053742802304, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9165368350767862, - "precision": 0.8932754880694144, - "recall": 0.9410420475319927, - "support": 2188.0 + "f1-score": 0.9346269346269347, + "precision": 0.9292604501607717, + "recall": 0.9400557620817844, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9991755359016639, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9983524301655059, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9206260132186058, - "precision": 0.9127155838536193, - "recall": 0.9286747594188314, - "support": 15899.0 + "f1-score": 0.9072057227826226, + "precision": 0.8811085089773615, + "recall": 0.9348960490350369, + "support": 12073.0 }, - "eval_accuracy": 0.9201759208356239, - "eval_loss": 0.6692858934402466, + "eval_accuracy": 0.9142231947483589, + "eval_loss": 0.6806138753890991, "eval_macro avg": { - "f1-score": 0.8837453723193808, - "precision": 0.8832910677399001, - "recall": 0.8850841108985157, - "support": 36380.0 - }, - "eval_runtime": 4.9586, - "eval_samples_per_second": 16.335, - "eval_steps_per_second": 2.218, + "f1-score": 0.8778069892797724, + "precision": 0.8886609420746096, + "recall": 0.8707892963492629, + "support": 29705.0 + }, + "eval_runtime": 4.796, + "eval_samples_per_second": 16.68, + "eval_steps_per_second": 2.085, "eval_weighted avg": { - "f1-score": 0.9190682641369721, - "precision": 0.918388879571886, - "recall": 0.9201759208356239, - "support": 36380.0 + "f1-score": 0.9111614699094798, + "precision": 0.9106725777549173, + "recall": 0.9142231947483589, + "support": 29705.0 }, "step": 2187 }, { "epoch": 28.0, "eval_Claim": { - "f1-score": 0.6998441558441558, - "precision": 0.7188900747065101, - "recall": 0.6817813765182186, - "support": 4940.0 + "f1-score": 0.7062782871068818, + "precision": 0.6839739267251068, + "recall": 0.7300863723608445, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9112884834663626, - "precision": 0.9094219390077378, - "recall": 0.9131627056672761, - "support": 2188.0 + "f1-score": 0.9229723606705936, + "precision": 0.9005305039787799, + "recall": 0.9465613382899628, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9997003745318352, - "precision": 0.9999250767962838, - "recall": 0.9994757732344791, - "support": 13353.0 + "f1-score": 1.0, + "precision": 1.0, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9211232449297972, - "precision": 0.9139372175097518, - "recall": 0.9284231712686333, - "support": 15899.0 + "f1-score": 0.9056619659019154, + "precision": 0.9208183530217429, + "recall": 0.8909964383334714, + "support": 12073.0 }, - "eval_accuracy": 0.9200934579439253, - "eval_loss": 0.6845241189002991, + "eval_accuracy": 0.9139538798182124, + "eval_loss": 0.6814476251602173, "eval_macro avg": { - "f1-score": 0.8829890646930376, - "precision": 0.8855435770050709, - "recall": 0.8807107566721517, - "support": 36380.0 - }, - "eval_runtime": 4.9593, - "eval_samples_per_second": 16.333, - "eval_steps_per_second": 2.218, + "f1-score": 0.8837281534198478, + "precision": 0.8763306959314074, + "recall": 0.8919110372460697, + "support": 29705.0 + }, + "eval_runtime": 4.7745, + "eval_samples_per_second": 16.756, + "eval_steps_per_second": 2.094, "eval_weighted avg": { - "f1-score": 0.9193256433193232, - "precision": 0.9187416037176308, - "recall": 0.9200934579439253, - "support": 36380.0 + "f1-score": 0.914864882516695, + "precision": 0.9162694814739634, + "recall": 0.9139538798182124, + "support": 29705.0 }, "step": 2268 }, { "epoch": 29.0, "eval_Claim": { - "f1-score": 0.6987782896054476, - "precision": 0.6914387633769322, - "recall": 0.7062753036437247, - "support": 4940.0 + "f1-score": 0.6726794438181135, + "precision": 0.7038007863695938, + "recall": 0.644193857965451, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9134720700985761, - "precision": 0.8771560790912916, - "recall": 0.9529250457038391, - "support": 2188.0 + "f1-score": 0.9280182232346242, + "precision": 0.9101876675603218, + "recall": 0.9465613382899628, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9993630811884155, - "precision": 0.9999250262408157, - "recall": 0.9988017673930952, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9151595913446284, - "precision": 0.9233625712273513, - "recall": 0.907101075539342, - "support": 15899.0 + "f1-score": 0.9038177945272816, + "precision": 0.8940756949509685, + "recall": 0.9137745382257931, + "support": 12073.0 }, - "eval_accuracy": 0.916245189664651, - "eval_loss": 0.7010424137115479, + "eval_accuracy": 0.9111597374179431, + "eval_loss": 0.7108751535415649, "eval_macro avg": { - "f1-score": 0.8816932580592669, - "precision": 0.8729706099840976, - "recall": 0.8912757980700003, - "support": 36380.0 - }, - "eval_runtime": 4.9512, - "eval_samples_per_second": 16.36, - "eval_steps_per_second": 2.222, + "f1-score": 0.8761178156712479, + "precision": 0.8769939387494352, + "recall": 0.8761324336203018, + "support": 29705.0 + }, + "eval_runtime": 4.7653, + "eval_samples_per_second": 16.788, + "eval_steps_per_second": 2.098, "eval_weighted avg": { - "f1-score": 0.9165821661881187, - "precision": 0.9171925615027781, - "recall": 0.916245189664651, - "support": 36380.0 + "f1-score": 0.9097497186891812, + "precision": 0.9088483922476083, + "recall": 0.9111597374179431, + "support": 29705.0 }, "step": 2349 }, { "epoch": 30.0, "eval_Claim": { - "f1-score": 0.6948434432158727, - "precision": 0.7097318978256281, - "recall": 0.6805668016194332, - "support": 4940.0 + "f1-score": 0.6894300778307808, + "precision": 0.7230121116377041, + "recall": 0.6588291746641075, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9181940098345999, - "precision": 0.8985126859142607, - "recall": 0.9387568555758684, - "support": 2188.0 + "f1-score": 0.934747521328107, + "precision": 0.9276887871853547, + "recall": 0.9419144981412639, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9995504944560982, - "precision": 0.9999250543356066, - "recall": 0.9991762150827529, - "support": 13353.0 + "f1-score": 1.0, + "precision": 1.0, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9180584714693073, - "precision": 0.91476208317722, - "recall": 0.9213787030630858, - "support": 15899.0 + "f1-score": 0.9087121676265164, + "precision": 0.8963738920225625, + "recall": 0.9213948480079516, + "support": 12073.0 }, - "eval_accuracy": 0.9182792743265531, - "eval_loss": 0.7106152176856995, + "eval_accuracy": 0.9159737417943107, + "eval_loss": 0.6745549440383911, "eval_macro avg": { - "f1-score": 0.8826616047439695, - "precision": 0.8807329303131789, - "recall": 0.8849696438352851, - "support": 36380.0 - }, - "eval_runtime": 4.9623, - "eval_samples_per_second": 16.323, - "eval_steps_per_second": 2.217, + "f1-score": 0.883222441696351, + "precision": 0.8867686977114053, + "recall": 0.8805346302033308, + "support": 29705.0 + }, + "eval_runtime": 4.7909, + "eval_samples_per_second": 16.698, + "eval_steps_per_second": 2.087, "eval_weighted avg": { - "f1-score": 0.917667523182169, - "precision": 0.9172023898575309, - "recall": 0.9182792743265531, - "support": 36380.0 + "f1-score": 0.9145936115149541, + "precision": 0.9137795909684305, + "recall": 0.9159737417943107, + "support": 29705.0 }, "step": 2430 }, { "epoch": 30.86, - "grad_norm": 0.0027945609763264656, + "grad_norm": 0.013120048679411411, "learning_rate": 7.654320987654322e-06, - "loss": 0.0031, + "loss": 0.0026, "step": 2500 }, { "epoch": 31.0, "eval_Claim": { - "f1-score": 0.7131089189464049, - "precision": 0.7165338238299611, - "recall": 0.7097165991902834, - "support": 4940.0 + "f1-score": 0.6823113802674742, + "precision": 0.71953166577967, + "recall": 0.6487523992322457, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9162036008001779, - "precision": 0.8918217221981826, - "recall": 0.9419561243144424, - "support": 2188.0 + "f1-score": 0.9337597076290544, + "precision": 0.9182389937106918, + "recall": 0.949814126394052, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9994380549207659, - "precision": 0.9999250374812594, - "recall": 0.9989515464689583, - "support": 13353.0 + "f1-score": 1.0, + "precision": 1.0, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9217583110130771, - "precision": 0.9235918161151806, - "recall": 0.9199320711994465, - "support": 15899.0 + "f1-score": 0.9071154317457725, + "precision": 0.8948343943911677, + "recall": 0.9197382589248737, + "support": 12073.0 }, - "eval_accuracy": 0.9217152281473338, - "eval_loss": 0.6842674016952515, + "eval_accuracy": 0.914458845312237, + "eval_loss": 0.6794646382331848, "eval_macro avg": { - "f1-score": 0.8876272214201064, - "precision": 0.882968099906146, - "recall": 0.8926390852932827, - "support": 36380.0 - }, - "eval_runtime": 4.9684, - "eval_samples_per_second": 16.303, - "eval_steps_per_second": 2.214, + "f1-score": 0.8807966299105753, + "precision": 0.8831512634703823, + "recall": 0.8795761961377929, + "support": 29705.0 + }, + "eval_runtime": 4.7931, + "eval_samples_per_second": 16.691, + "eval_steps_per_second": 2.086, "eval_weighted avg": { - "f1-score": 0.9216037183150064, - "precision": 0.9215824169266671, - "recall": 0.9217152281473338, - "support": 36380.0 + "f1-score": 0.9128742410785817, + "precision": 0.9119809439797892, + "recall": 0.914458845312237, + "support": 29705.0 }, "step": 2511 }, { "epoch": 32.0, "eval_Claim": { - "f1-score": 0.6948140387637506, - "precision": 0.720086862106406, - "recall": 0.671255060728745, - "support": 4940.0 + "f1-score": 0.6683053040103493, + "precision": 0.7251544076361595, + "recall": 0.619721689059501, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9145506734378449, - "precision": 0.8846646732165742, - "recall": 0.946526508226691, - "support": 2188.0 + "f1-score": 0.9353594919482876, + "precision": 0.9136021267168808, + "recall": 0.95817843866171, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9991005172026085, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9982026510896428, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9190726159230096, - "precision": 0.9131946600434647, - "recall": 0.9250267312409586, - "support": 15899.0 + "f1-score": 0.904937720615085, + "precision": 0.886909495784953, + "recall": 0.9237140727242608, + "support": 12073.0 }, - "eval_accuracy": 0.918719076415613, - "eval_loss": 0.7216777205467224, + "eval_accuracy": 0.9126073051674802, + "eval_loss": 0.7192577123641968, "eval_macro avg": { - "f1-score": 0.8818844613318033, - "precision": 0.8794865488416111, - "recall": 0.8852527378215094, - "support": 36380.0 - }, - "eval_runtime": 4.9621, - "eval_samples_per_second": 16.324, - "eval_steps_per_second": 2.217, + "f1-score": 0.8771506291434306, + "precision": 0.8814165075344984, + "recall": 0.875403550111368, + "support": 29705.0 + }, + "eval_runtime": 4.7949, + "eval_samples_per_second": 16.684, + "eval_steps_per_second": 2.086, "eval_weighted avg": { - "f1-score": 0.9177224560676552, - "precision": 0.9171181227002353, - "recall": 0.918719076415613, - "support": 36380.0 + "f1-score": 0.9101398160166225, + "precision": 0.9092130513494017, + "recall": 0.9126073051674802, + "support": 29705.0 }, "step": 2592 }, { "epoch": 33.0, "eval_Claim": { - "f1-score": 0.7047761512982157, - "precision": 0.7463226974428604, - "recall": 0.6676113360323886, - "support": 4940.0 + "f1-score": 0.7084387215451827, + "precision": 0.6801324503311258, + "recall": 0.7392034548944337, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.921081556386019, - "precision": 0.8873358746293943, - "recall": 0.9574954296160878, - "support": 2188.0 + "f1-score": 0.937269372693727, + "precision": 0.9304029304029304, + "recall": 0.9442379182156134, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9991380279578758, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9982775406275743, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9233112623954739, - "precision": 0.9127842655193608, - "recall": 0.9340839046480911, - "support": 15899.0 + "f1-score": 0.9026608285618052, + "precision": 0.9178868053771727, + "recall": 0.8879317485297772, + "support": 12073.0 }, - "eval_accuracy": 0.922869708631116, - "eval_loss": 0.7047613263130188, + "eval_accuracy": 0.9138192223531392, + "eval_loss": 0.699444055557251, "eval_macro avg": { - "f1-score": 0.8870767495093961, - "precision": 0.8866107093979039, - "recall": 0.8893670527310353, - "support": 36380.0 - }, - "eval_runtime": 4.9399, - "eval_samples_per_second": 16.397, - "eval_steps_per_second": 2.227, + "f1-score": 0.8870922307001787, + "precision": 0.8821055465278073, + "recall": 0.8928432804099561, + "support": 29705.0 + }, + "eval_runtime": 4.7863, + "eval_samples_per_second": 16.715, + "eval_steps_per_second": 2.089, "eval_weighted avg": { - "f1-score": 0.9213341528568705, - "precision": 0.9206619586736987, - "recall": 0.922869708631116, - "support": 36380.0 + "f1-score": 0.9149840250686382, + "precision": 0.9167031328236271, + "recall": 0.9138192223531392, + "support": 29705.0 }, "step": 2673 }, { "epoch": 34.0, "eval_Claim": { - "f1-score": 0.704132582598534, - "precision": 0.7408897831433042, - "recall": 0.6708502024291498, - "support": 4940.0 + "f1-score": 0.6974015088013411, + "precision": 0.6961510877360746, + "recall": 0.6986564299424184, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9225707391971923, - "precision": 0.8869675242513707, - "recall": 0.9611517367458866, - "support": 2188.0 + "f1-score": 0.9402335699564919, + "precision": 0.92686230248307, + "recall": 0.953996282527881, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9989880439263896, - "precision": 0.9999249699879952, - "recall": 0.9980528720137797, - "support": 13353.0 + "f1-score": 1.0, + "precision": 1.0, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9222287974584982, - "precision": 0.9134378084896347, - "recall": 0.9311906409208126, - "support": 15899.0 + "f1-score": 0.9036895462855244, + "precision": 0.9066277615673197, + "recall": 0.9007703139236313, + "support": 12073.0 }, - "eval_accuracy": 0.9221825178669598, - "eval_loss": 0.7385492324829102, + "eval_accuracy": 0.9140548729170174, + "eval_loss": 0.7247604131698608, "eval_macro avg": { - "f1-score": 0.8869800407951536, - "precision": 0.8853050214680762, - "recall": 0.8903113630274071, - "support": 36380.0 - }, - "eval_runtime": 4.9565, - "eval_samples_per_second": 16.342, - "eval_steps_per_second": 2.219, + "f1-score": 0.8853311562608394, + "precision": 0.882410287946616, + "recall": 0.8883557565984826, + "support": 29705.0 + }, + "eval_runtime": 4.8212, + "eval_samples_per_second": 16.593, + "eval_steps_per_second": 2.074, "eval_weighted avg": { - "f1-score": 0.9208082115651994, - "precision": 0.9201601515452535, - "recall": 0.9221825178669598, - "support": 36380.0 + "f1-score": 0.9140682047983671, + "precision": 0.9141182418458097, + "recall": 0.9140548729170174, + "support": 29705.0 }, "step": 2754 }, { "epoch": 35.0, "eval_Claim": { - "f1-score": 0.7115286429081996, - "precision": 0.7179064496187925, - "recall": 0.7052631578947368, - "support": 4940.0 + "f1-score": 0.6959401447320005, + "precision": 0.7119196988707653, + "recall": 0.680662188099808, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9221824686940967, - "precision": 0.9028021015761821, - "recall": 0.9424131627056673, - "support": 2188.0 + "f1-score": 0.9385984934946359, + "precision": 0.9223867205024675, + "recall": 0.9553903345724907, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9993630811884155, - "precision": 0.9999250262408157, - "recall": 0.9988017673930952, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9208904540309396, - "precision": 0.9207167557371896, - "recall": 0.9210642178753381, - "support": 15899.0 + "f1-score": 0.9068492020947589, + "precision": 0.9029397273772376, + "recall": 0.9107926778762528, + "support": 12073.0 }, - "eval_accuracy": 0.9215777899945025, - "eval_loss": 0.7338384389877319, + "eval_accuracy": 0.9157044268641643, + "eval_loss": 0.7355715036392212, "eval_macro avg": { - "f1-score": 0.8884911617054129, - "precision": 0.8853375832932451, - "recall": 0.8918855764672093, - "support": 36380.0 - }, - "eval_runtime": 4.9467, - "eval_samples_per_second": 16.375, - "eval_steps_per_second": 2.224, + "f1-score": 0.885335910356592, + "precision": 0.8842894382168318, + "recall": 0.8867113001371378, + "support": 29705.0 + }, + "eval_runtime": 4.8984, + "eval_samples_per_second": 16.332, + "eval_steps_per_second": 2.041, "eval_weighted avg": { - "f1-score": 0.9213419265864765, - "precision": 0.9211727167186562, - "recall": 0.9215777899945025, - "support": 36380.0 + "f1-score": 0.9150120491578154, + "precision": 0.914473958742095, + "recall": 0.9157044268641643, + "support": 29705.0 }, "step": 2835 }, { "epoch": 36.0, "eval_Claim": { - "f1-score": 0.7146180690189995, - "precision": 0.6856398809523809, - "recall": 0.7461538461538462, - "support": 4940.0 + "f1-score": 0.6910893483254744, + "precision": 0.6965634901291737, + "recall": 0.685700575815739, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9238578680203046, - "precision": 0.893299189073837, - "recall": 0.956581352833638, - "support": 2188.0 + "f1-score": 0.9263347600810993, + "precision": 0.8989943156974202, + "recall": 0.9553903345724907, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9992505433560669, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.998502209241369, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9148493291062222, - "precision": 0.9318893528183716, - "recall": 0.8984212843575068, - "support": 15899.0 + "f1-score": 0.9052998837016115, + "precision": 0.9079396817462301, + "recall": 0.9026753913691709, + "support": 12073.0 }, - "eval_accuracy": 0.9179769103903244, - "eval_loss": 0.7659569978713989, + "eval_accuracy": 0.9131122706615048, + "eval_loss": 0.7339844703674316, "eval_macro avg": { - "f1-score": 0.8881439523753982, - "precision": 0.8777071057111474, - "recall": 0.89991467314659, - "support": 36380.0 - }, - "eval_runtime": 4.9595, - "eval_samples_per_second": 16.332, - "eval_steps_per_second": 2.218, + "f1-score": 0.8806809980270462, + "precision": 0.8758743718932059, + "recall": 0.8859415754393501, + "support": 29705.0 + }, + "eval_runtime": 4.803, + "eval_samples_per_second": 16.656, + "eval_steps_per_second": 2.082, "eval_weighted avg": { - "f1-score": 0.9191807659449058, - "precision": 0.9211299741082354, - "recall": 0.9179769103903244, - "support": 36380.0 + "f1-score": 0.912830106158716, + "precision": 0.9126903946124048, + "recall": 0.9131122706615048, + "support": 29705.0 }, "step": 2916 }, { "epoch": 37.0, "eval_Claim": { - "f1-score": 0.7063829787234043, - "precision": 0.7443946188340808, - "recall": 0.6720647773279352, - "support": 4940.0 + "f1-score": 0.6958525345622121, + "precision": 0.6693262411347518, + "recall": 0.7245681381957774, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9240397065170479, - "precision": 0.875306623058054, - "recall": 0.9785191956124314, - "support": 2188.0 + "f1-score": 0.9284262813276133, + "precision": 0.9029424681598595, + "recall": 0.9553903345724907, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9987627938364638, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9975286452482588, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9224097588369263, - "precision": 0.9151293797201931, - "recall": 0.9298069060947229, - "support": 15899.0 + "f1-score": 0.8996916839126579, + "precision": 0.917873147190624, + "recall": 0.8822165161931583, + "support": 12073.0 }, - "eval_accuracy": 0.9225948323254536, - "eval_loss": 0.760911226272583, + "eval_accuracy": 0.9102507995286989, + "eval_loss": 0.7738711833953857, "eval_macro avg": { - "f1-score": 0.8878988094784606, - "precision": 0.883707655403082, - "recall": 0.8944798810708371, - "support": 36380.0 - }, - "eval_runtime": 4.9543, - "eval_samples_per_second": 16.349, - "eval_steps_per_second": 2.22, + "f1-score": 0.8809926249506208, + "precision": 0.8725354641213088, + "recall": 0.8905437472403566, + "support": 29705.0 + }, + "eval_runtime": 4.8009, + "eval_samples_per_second": 16.663, + "eval_steps_per_second": 2.083, "eval_weighted avg": { - "f1-score": 0.9211985468554017, - "precision": 0.9207015480061224, - "recall": 0.9225948323254536, - "support": 36380.0 + "f1-score": 0.9113706251927232, + "precision": 0.9131919363932695, + "recall": 0.9102507995286989, + "support": 29705.0 }, "step": 2997 }, { "epoch": 37.04, - "grad_norm": 0.003850826993584633, + "grad_norm": 29.786388397216797, "learning_rate": 5.185185185185185e-06, - "loss": 0.0007, + "loss": 0.0015, "step": 3000 }, { "epoch": 38.0, "eval_Claim": { - "f1-score": 0.7178156146179401, - "precision": 0.7367860187553282, - "recall": 0.6997975708502024, - "support": 4940.0 + "f1-score": 0.6688368598973332, + "precision": 0.6993977481015973, + "recall": 0.6408349328214972, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9259421560035057, - "precision": 0.8893097643097643, - "recall": 0.9657221206581352, - "support": 2188.0 + "f1-score": 0.9244361776331085, + "precision": 0.9250814332247557, + "recall": 0.9237918215613383, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9997003745318352, - "precision": 0.9999250767962838, - "recall": 0.9994757732344791, - "support": 13353.0 + "f1-score": 1.0, + "precision": 1.0, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9237383881496359, - "precision": 0.9218290009395553, - "recall": 0.9256557016164538, - "support": 15899.0 + "f1-score": 0.9031757694505674, + "precision": 0.8903822937625755, + "recall": 0.9163422513045639, + "support": 12073.0 }, - "eval_accuracy": 0.9244914788345244, - "eval_loss": 0.7453881502151489, + "eval_accuracy": 0.9100824776973574, + "eval_loss": 0.757270336151123, "eval_macro avg": { - "f1-score": 0.8917991333257291, - "precision": 0.8869624652002329, - "recall": 0.8976627915898177, - "support": 36380.0 - }, - "eval_runtime": 4.9642, - "eval_samples_per_second": 16.317, - "eval_steps_per_second": 2.216, + "f1-score": 0.8741122017452523, + "precision": 0.8787153687722321, + "recall": 0.8702422514218499, + "support": 29705.0 + }, + "eval_runtime": 4.7997, + "eval_samples_per_second": 16.668, + "eval_steps_per_second": 2.083, "eval_weighted avg": { - "f1-score": 0.9237901678906805, - "precision": 0.9234109822253945, - "recall": 0.9244914788345244, - "support": 36380.0 + "f1-score": 0.9087069433056804, + "precision": 0.9078421306508232, + "recall": 0.9100824776973574, + "support": 29705.0 }, "step": 3078 }, { "epoch": 39.0, "eval_Claim": { - "f1-score": 0.704778156996587, - "precision": 0.7448151487826871, - "recall": 0.668825910931174, - "support": 4940.0 + "f1-score": 0.6469723299024195, + "precision": 0.7304557802595835, + "recall": 0.5806142034548945, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9222684475585724, - "precision": 0.8852459016393442, - "recall": 0.9625228519195612, - "support": 2188.0 + "f1-score": 0.9280862028578122, + "precision": 0.9357581483230987, + "recall": 0.9205390334572491, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9992880428673137, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9985770987793006, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9231870525988172, - "precision": 0.9137453021994948, - "recall": 0.9328259638971005, - "support": 15899.0 + "f1-score": 0.9053363157053842, + "precision": 0.8742575021214225, + "recall": 0.9387062039261161, + "support": 12073.0 }, - "eval_accuracy": 0.9228971962616822, - "eval_loss": 0.7312874794006348, + "eval_accuracy": 0.910486450092577, + "eval_loss": 0.7571966052055359, "eval_macro avg": { - "f1-score": 0.8873804250053225, - "precision": 0.8859515881553816, - "recall": 0.8906879563817841, - "support": 36380.0 - }, - "eval_runtime": 4.966, - "eval_samples_per_second": 16.311, - "eval_steps_per_second": 2.215, + "f1-score": 0.870098712116404, + "precision": 0.8851178576760261, + "recall": 0.859964860209565, + "support": 29705.0 + }, + "eval_runtime": 4.8778, + "eval_samples_per_second": 16.401, + "eval_steps_per_second": 2.05, "eval_weighted avg": { - "f1-score": 0.9214065872594044, - "precision": 0.9207515510567105, - "recall": 0.9228971962616822, - "support": 36380.0 + "f1-score": 0.9067816030666352, + "precision": 0.9064198636736301, + "recall": 0.910486450092577, + "support": 29705.0 }, "step": 3159 }, { "epoch": 40.0, "eval_Claim": { - "f1-score": 0.7213450891941768, - "precision": 0.7307852098047362, - "recall": 0.7121457489878542, - "support": 4940.0 + "f1-score": 0.6899975179945396, + "precision": 0.7146529562982005, + "recall": 0.6669865642994242, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9265486725663717, - "precision": 0.8979416809605489, - "recall": 0.9570383912248629, - "support": 2188.0 + "f1-score": 0.9303297724105899, + "precision": 0.9298978644382544, + "recall": 0.9307620817843866, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9992880428673137, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9985770987793006, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9231736847070662, - "precision": 0.9231446540880504, - "recall": 0.9232027171520222, - "support": 15899.0 + "f1-score": 0.9086069936942102, + "precision": 0.8984533160579804, + "recall": 0.9189927938374887, + "support": 12073.0 }, - "eval_accuracy": 0.9242440901594282, - "eval_loss": 0.7378404140472412, + "eval_accuracy": 0.9153341188352129, + "eval_loss": 0.7418585419654846, "eval_macro avg": { - "f1-score": 0.8925888723337321, - "precision": 0.8879678862133338, - "recall": 0.89774098903601, - "support": 36380.0 - }, - "eval_runtime": 4.9789, - "eval_samples_per_second": 16.269, - "eval_steps_per_second": 2.209, + "f1-score": 0.8822335710248349, + "precision": 0.8857510341986088, + "recall": 0.8791853599803248, + "support": 29705.0 + }, + "eval_runtime": 4.7965, + "eval_samples_per_second": 16.679, + "eval_steps_per_second": 2.085, "eval_weighted avg": { - "f1-score": 0.9239077758592451, - "precision": 0.9237177622243814, - "recall": 0.9242440901594282, - "support": 36380.0 + "f1-score": 0.9143104379767389, + "precision": 0.913611870422152, + "recall": 0.9153341188352129, + "support": 29705.0 }, "step": 3240 }, { "epoch": 41.0, "eval_Claim": { - "f1-score": 0.7108195355618034, - "precision": 0.731932232468368, - "recall": 0.6908906882591093, - "support": 4940.0 + "f1-score": 0.7027151935297515, + "precision": 0.6777356808558057, + "recall": 0.7296065259117083, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9284467713787086, - "precision": 0.8881469115191987, - "recall": 0.9725776965265083, - "support": 2188.0 + "f1-score": 0.9321642824180896, + "precision": 0.925756186984418, + "recall": 0.9386617100371747, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.999213041034289, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9984273197034375, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9213497240341193, - "precision": 0.9187566451935706, - "recall": 0.9239574816026165, - "support": 15899.0 + "f1-score": 0.902130520653864, + "precision": 0.9155578300921188, + "recall": 0.8890913608879317, + "support": 12073.0 }, - "eval_accuracy": 0.9225673446948873, - "eval_loss": 0.7535675168037415, + "eval_accuracy": 0.9125399764349437, + "eval_loss": 0.7465766668319702, "eval_macro avg": { - "f1-score": 0.8899572680022301, - "precision": 0.8847089472952844, - "recall": 0.8964632965229179, - "support": 36380.0 - }, - "eval_runtime": 4.9535, - "eval_samples_per_second": 16.352, - "eval_steps_per_second": 2.221, + "f1-score": 0.8842524991504263, + "precision": 0.8797624244830856, + "recall": 0.8893398992092036, + "support": 29705.0 + }, + "eval_runtime": 4.7973, + "eval_samples_per_second": 16.676, + "eval_steps_per_second": 2.085, "eval_weighted avg": { - "f1-score": 0.921768033007181, - "precision": 0.9213667557100144, - "recall": 0.9225673446948873, - "support": 36380.0 + "f1-score": 0.9135955643241823, + "precision": 0.9150836328867065, + "recall": 0.9125399764349437, + "support": 29705.0 }, "step": 3321 }, { "epoch": 42.0, "eval_Claim": { - "f1-score": 0.7140475197999167, - "precision": 0.7358247422680413, - "recall": 0.6935222672064777, - "support": 4940.0 + "f1-score": 0.6649142571035175, + "precision": 0.6951059931954986, + "recall": 0.6372360844529751, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9298360655737705, - "precision": 0.8910766652702137, - "recall": 0.9721206581352834, - "support": 2188.0 + "f1-score": 0.9264330471106985, + "precision": 0.9253592953175707, + "recall": 0.9275092936802974, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9992880428673137, + "f1-score": 1.0, "precision": 1.0, - "recall": 0.9985770987793006, - "support": 13353.0 + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9222619271519027, - "precision": 0.919265137786665, - "recall": 0.9252783193911567, - "support": 15899.0 + "f1-score": 0.9014211042143091, + "precision": 0.8890052356020942, + "recall": 0.9141886854965626, + "support": 12073.0 }, - "eval_accuracy": 0.9235294117647059, - "eval_loss": 0.7416272759437561, + "eval_accuracy": 0.9089715536105033, + "eval_loss": 0.7535542845726013, "eval_macro avg": { - "f1-score": 0.8913583888482258, - "precision": 0.88654163633123, - "recall": 0.8973745858780546, - "support": 36380.0 - }, - "eval_runtime": 4.9829, - "eval_samples_per_second": 16.256, - "eval_steps_per_second": 2.208, + "f1-score": 0.8731921021071313, + "precision": 0.8773676310287909, + "recall": 0.8697335159074588, + "support": 29705.0 + }, + "eval_runtime": 4.7844, + "eval_samples_per_second": 16.721, + "eval_steps_per_second": 2.09, "eval_weighted avg": { - "f1-score": 0.9227160988312902, - "precision": 0.9222937437076839, - "recall": 0.9235294117647059, - "support": 36380.0 + "f1-score": 0.9075880670651082, + "precision": 0.9067003936235091, + "recall": 0.9089715536105033, + "support": 29705.0 }, "step": 3402 }, { "epoch": 43.0, "eval_Claim": { - "f1-score": 0.7126268438121143, - "precision": 0.7373890452478892, - "recall": 0.6894736842105263, - "support": 4940.0 + "f1-score": 0.6895224347407675, + "precision": 0.6914354644149577, + "recall": 0.6876199616122841, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9299279633267845, - "precision": 0.8900961136648559, - "recall": 0.973491773308958, - "support": 2188.0 + "f1-score": 0.9361702127659575, + "precision": 0.9220369535826949, + "recall": 0.9507434944237918, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9991380279578758, - "precision": 1.0, - "recall": 0.9982775406275743, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9221279393806557, - "precision": 0.9181319366504551, - "recall": 0.9261588779168501, - "support": 15899.0 + "f1-score": 0.9023691962989088, + "precision": 0.9040571998669771, + "recall": 0.9006874844694773, + "support": 12073.0 }, - "eval_accuracy": 0.9233369983507421, - "eval_loss": 0.7578195333480835, + "eval_accuracy": 0.9122369971385289, + "eval_loss": 0.756969690322876, "eval_macro avg": { - "f1-score": 0.8909551936193576, - "precision": 0.8864042738908001, - "recall": 0.8968504690159773, - "support": 36380.0 - }, - "eval_runtime": 4.9557, - "eval_samples_per_second": 16.345, - "eval_steps_per_second": 2.22, + "f1-score": 0.8820044112276516, + "precision": 0.8793603059953716, + "recall": 0.8847627351263884, + "support": 29705.0 + }, + "eval_runtime": 4.8137, + "eval_samples_per_second": 16.619, + "eval_steps_per_second": 2.077, "eval_weighted avg": { - "f1-score": 0.9224150958693075, - "precision": 0.9219519472520303, - "recall": 0.9233369983507421, - "support": 36380.0 + "f1-score": 0.9121148336942613, + "precision": 0.9120285878532032, + "recall": 0.9122369971385289, + "support": 29705.0 }, "step": 3483 }, { "epoch": 43.21, - "grad_norm": 0.03563033416867256, + "grad_norm": 0.007461093831807375, "learning_rate": 2.7160493827160496e-06, - "loss": 0.0003, + "loss": 0.0013, "step": 3500 }, { "epoch": 44.0, "eval_Claim": { - "f1-score": 0.7031551835157759, - "precision": 0.7482868889904066, - "recall": 0.6631578947368421, - "support": 4940.0 + "f1-score": 0.6884735202492213, + "precision": 0.6876495931067497, + "recall": 0.689299424184261, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.924901185770751, - "precision": 0.8901098901098901, - "recall": 0.9625228519195612, - "support": 2188.0 + "f1-score": 0.9315510486287162, + "precision": 0.9240969364426155, + "recall": 0.9391263940520446, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9992505433560669, - "precision": 1.0, - "recall": 0.998502209241369, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9228619340413639, - "precision": 0.9114273446604919, - "recall": 0.9345870809484873, - "support": 15899.0 + "f1-score": 0.9021576763485477, + "precision": 0.9038829300740001, + "recall": 0.9004389961070156, + "support": 12073.0 }, - "eval_accuracy": 0.922869708631116, - "eval_loss": 0.7405058741569519, + "eval_accuracy": 0.9115300454468944, + "eval_loss": 0.7517338991165161, "eval_macro avg": { - "f1-score": 0.8875422116709895, - "precision": 0.8874560309401971, - "recall": 0.8896925092115648, - "support": 36380.0 - }, - "eval_runtime": 4.9702, - "eval_samples_per_second": 16.297, - "eval_steps_per_second": 2.213, + "f1-score": 0.8805345115828644, + "precision": 0.8788852664350555, + "recall": 0.8822162035858303, + "support": 29705.0 + }, + "eval_runtime": 4.7861, + "eval_samples_per_second": 16.715, + "eval_steps_per_second": 2.089, "eval_weighted avg": { - "f1-score": 0.9211886969706305, - "precision": 0.9205025020321663, - "recall": 0.922869708631116, - "support": 36380.0 + "f1-score": 0.9115470505875514, + "precision": 0.9115757890749278, + "recall": 0.9115300454468944, + "support": 29705.0 }, "step": 3564 }, { "epoch": 45.0, "eval_Claim": { - "f1-score": 0.7141807494489346, - "precision": 0.7416612164813604, - "recall": 0.6886639676113361, - "support": 4940.0 + "f1-score": 0.6683679979584024, + "precision": 0.713818479149632, + "recall": 0.628358925143954, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9250714128762909, - "precision": 0.8908167583580194, - "recall": 0.9620658135283364, - "support": 2188.0 + "f1-score": 0.9285714285714285, + "precision": 0.9268518518518518, + "recall": 0.9302973977695167, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9991005172026085, - "precision": 1.0, - "recall": 0.9982026510896428, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9235625000000001, - "precision": 0.9177690826656729, - "recall": 0.9294295238694258, - "support": 15899.0 + "f1-score": 0.9047735021919143, + "precision": 0.8871288704927167, + "recall": 0.9231342665451835, + "support": 12073.0 }, - "eval_accuracy": 0.9239417262231996, - "eval_loss": 0.7358143329620361, + "eval_accuracy": 0.9115637098131628, + "eval_loss": 0.7444580793380737, "eval_macro avg": { - "f1-score": 0.8904787948819586, - "precision": 0.8875617643762631, - "recall": 0.8945904890246853, - "support": 36380.0 - }, - "eval_runtime": 4.9567, - "eval_samples_per_second": 16.341, - "eval_steps_per_second": 2.219, + "f1-score": 0.8754171824566794, + "precision": 0.8819277019027643, + "recall": 0.8704476473646635, + "support": 29705.0 + }, + "eval_runtime": 4.7766, + "eval_samples_per_second": 16.748, + "eval_steps_per_second": 2.094, "eval_weighted avg": { - "f1-score": 0.9229471838196122, - "precision": 0.9224168257835843, - "recall": 0.9239417262231996, - "support": 36380.0 + "f1-score": 0.9095732719689872, + "precision": 0.9086378572002458, + "recall": 0.9115637098131628, + "support": 29705.0 }, "step": 3645 }, { "epoch": 46.0, "eval_Claim": { - "f1-score": 0.7153789705419855, - "precision": 0.7419004131332898, - "recall": 0.6906882591093118, - "support": 4940.0 + "f1-score": 0.693688194359663, + "precision": 0.7061894108873975, + "recall": 0.6816218809980806, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9296449575255936, - "precision": 0.8880565959217644, - "recall": 0.9753199268738574, - "support": 2188.0 + "f1-score": 0.9338862013360976, + "precision": 0.9259936043855642, + "recall": 0.9419144981412639, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9989879680647701, - "precision": 1.0, - "recall": 0.9979779824758481, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9230384025539106, - "precision": 0.9186394218788936, - "recall": 0.9274797157053902, - "support": 15899.0 + "f1-score": 0.9061971714839401, + "precision": 0.9022167487684729, + "recall": 0.9102128716971755, + "support": 12073.0 }, - "eval_accuracy": 0.9240791643760308, - "eval_loss": 0.7376746535301208, + "eval_accuracy": 0.9146271671435785, + "eval_loss": 0.7456544637680054, "eval_macro avg": { - "f1-score": 0.891762574671565, - "precision": 0.887149107733487, - "recall": 0.8978664710411017, - "support": 36380.0 - }, - "eval_runtime": 4.9655, - "eval_samples_per_second": 16.313, - "eval_steps_per_second": 2.215, + "f1-score": 0.8834318420711683, + "precision": 0.8835778425395728, + "recall": 0.88343731270913, + "support": 29705.0 + }, + "eval_runtime": 4.8002, + "eval_samples_per_second": 16.666, + "eval_steps_per_second": 2.083, "eval_weighted avg": { - "f1-score": 0.9231146009158029, - "precision": 0.9226636624850962, - "recall": 0.9240791643760308, - "support": 36380.0 + "f1-score": 0.9140896809557788, + "precision": 0.9136373875607494, + "recall": 0.9146271671435785, + "support": 29705.0 }, "step": 3726 }, { "epoch": 47.0, "eval_Claim": { - "f1-score": 0.7173056994818653, - "precision": 0.7348195329087048, - "recall": 0.7006072874493927, - "support": 4940.0 + "f1-score": 0.6868585732165207, + "precision": 0.717948717948718, + "recall": 0.6583493282149712, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9317934190455436, - "precision": 0.8904623073719283, - "recall": 0.9771480804387569, - "support": 2188.0 + "f1-score": 0.9339796860572483, + "precision": 0.9279816513761467, + "recall": 0.9400557620817844, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.999213041034289, - "precision": 1.0, - "recall": 0.9984273197034375, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9222892323156174, - "precision": 0.9211896843822551, - "recall": 0.9233914082646707, - "support": 15899.0 + "f1-score": 0.9071659240485632, + "precision": 0.8955609362389023, + "recall": 0.9190756232916425, + "support": 12073.0 }, - "eval_accuracy": 0.9239142385926333, - "eval_loss": 0.7394932508468628, + "eval_accuracy": 0.9148291533411883, + "eval_loss": 0.7466372847557068, "eval_macro avg": { - "f1-score": 0.8926503479693289, - "precision": 0.8866178811657222, - "recall": 0.8998935239640644, - "support": 36380.0 - }, - "eval_runtime": 4.9785, - "eval_samples_per_second": 16.27, - "eval_steps_per_second": 2.21, + "f1-score": 0.8819899961068263, + "precision": 0.885350727920156, + "recall": 0.8793701783970995, + "support": 29705.0 + }, + "eval_runtime": 4.8142, + "eval_samples_per_second": 16.618, + "eval_steps_per_second": 2.077, "eval_weighted avg": { - "f1-score": 0.9232606486484036, - "precision": 0.9229613747414034, - "recall": 0.9239142385926333, - "support": 36380.0 + "f1-score": 0.913531898357159, + "precision": 0.9127262764442979, + "recall": 0.9148291533411883, + "support": 29705.0 }, "step": 3807 }, { "epoch": 48.0, "eval_Claim": { - "f1-score": 0.7136500051743766, - "precision": 0.730044463264874, - "recall": 0.6979757085020243, - "support": 4940.0 + "f1-score": 0.6945298094652735, + "precision": 0.7121250315099571, + "recall": 0.6777831094049904, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9288503253796095, - "precision": 0.8839801816680429, - "recall": 0.9785191956124314, - "support": 2188.0 + "f1-score": 0.9340405904059041, + "precision": 0.9271978021978022, + "recall": 0.9409851301115242, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9992505433560669, - "precision": 1.0, - "recall": 0.998502209241369, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9214804565894154, - "precision": 0.9213935354043516, - "recall": 0.9215673941757343, - "support": 15899.0 + "f1-score": 0.9072962079460393, + "precision": 0.9010701740053917, + "recall": 0.9136088793174852, + "support": 12073.0 }, - "eval_accuracy": 0.922869708631116, - "eval_loss": 0.7407149076461792, + "eval_accuracy": 0.9154014475677495, + "eval_loss": 0.746742308139801, "eval_macro avg": { - "f1-score": 0.8908078326248671, - "precision": 0.8838545450843172, - "recall": 0.8991411268828897, - "support": 36380.0 - }, - "eval_runtime": 4.9928, - "eval_samples_per_second": 16.223, - "eval_steps_per_second": 2.203, + "f1-score": 0.8839556022305474, + "precision": 0.8850761534575019, + "recall": 0.8830942797085, + "support": 29705.0 + }, + "eval_runtime": 4.8237, + "eval_samples_per_second": 16.585, + "eval_steps_per_second": 2.073, "eval_weighted avg": { - "f1-score": 0.9222475487146972, - "precision": 0.9220122074054958, - "recall": 0.922869708631116, - "support": 36380.0 + "f1-score": 0.9146656366617318, + "precision": 0.914091469477332, + "recall": 0.9154014475677495, + "support": 29705.0 }, "step": 3888 }, { "epoch": 49.0, "eval_Claim": { - "f1-score": 0.714640712362808, - "precision": 0.7314540059347181, - "recall": 0.698582995951417, - "support": 4940.0 + "f1-score": 0.6939069823338541, + "precision": 0.695160125210691, + "recall": 0.692658349328215, + "support": 4168.0 }, "eval_MajorClaim": { - "f1-score": 0.9282462605679601, - "precision": 0.8828865979381443, - "recall": 0.9785191956124314, - "support": 2188.0 + "f1-score": 0.9330863625839315, + "precision": 0.9298569450853715, + "recall": 0.9363382899628253, + "support": 2152.0 }, "eval_O": { - "f1-score": 0.9992505433560669, - "precision": 1.0, - "recall": 0.998502209241369, - "support": 13353.0 + "f1-score": 0.9999558011049724, + "precision": 0.9999116061168567, + "recall": 1.0, + "support": 11312.0 }, "eval_Premise": { - "f1-score": 0.9219256045027199, - "precision": 0.9217806841046278, - "recall": 0.9220705704761306, - "support": 15899.0 + "f1-score": 0.9042037688962519, + "precision": 0.9042412193505633, + "recall": 0.904166321543941, + "support": 12073.0 }, - "eval_accuracy": 0.9231720725673447, - "eval_loss": 0.7421937584877014, + "eval_accuracy": 0.9133142568591146, + "eval_loss": 0.7449847459793091, "eval_macro avg": { - "f1-score": 0.8910157801973887, - "precision": 0.8840303219943725, - "recall": 0.899418742820337, - "support": 36380.0 - }, - "eval_runtime": 4.9931, - "eval_samples_per_second": 16.222, - "eval_steps_per_second": 2.203, + "f1-score": 0.8827882287297525, + "precision": 0.8822924739408706, + "recall": 0.8832907402087453, + "support": 29705.0 + }, + "eval_runtime": 4.8111, + "eval_samples_per_second": 16.628, + "eval_steps_per_second": 2.079, "eval_weighted avg": { - "f1-score": 0.9225402866579788, - "precision": 0.9223070302964718, - "recall": 0.9231720725673447, - "support": 36380.0 + "f1-score": 0.9132522564764188, + "precision": 0.9131925223805796, + "recall": 0.9133142568591146, + "support": 29705.0 }, "step": 3969 } @@ -2268,7 +2268,7 @@ "num_input_tokens_seen": 0, "num_train_epochs": 50, "save_steps": 500, - "total_flos": 7023767009502600.0, + "total_flos": 7045647903613200.0, "train_batch_size": 4, "trial_name": null, "trial_params": null