{ "best_metric": null, "best_model_checkpoint": null, "epoch": 50.0, "eval_steps": 500, "global_step": 4050, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "eval_B-Claim": { "f1-score": 0.37160751565762, "precision": 0.42788461538461536, "recall": 0.3284132841328413, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.3578947368421052, "precision": 0.6666666666666666, "recall": 0.2446043165467626, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8640915593705293, "precision": 0.7895424836601307, "recall": 0.9541864139020537, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.5003402749421533, "precision": 0.5493126120741183, "recall": 0.4593851537115721, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.7718093699515347, "precision": 0.6502211636611093, "recall": 0.9493293591654247, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.875016720916752, "precision": 0.8846812731043188, "recall": 0.865561044460127, "support": 11336.0 }, "eval_O": { "f1-score": 0.9992483530087988, "precision": 0.9995577178239717, "recall": 0.998939179632249, "support": 11312.0 }, "eval_accuracy": 0.8614038040733883, "eval_loss": 0.31713685393333435, "eval_macro avg": { "f1-score": 0.6771440758127848, "precision": 0.7096952189107044, "recall": 0.685774107364433, "support": 29705.0 }, "eval_runtime": 4.8338, "eval_samples_per_second": 16.55, "eval_steps_per_second": 2.069, "eval_weighted avg": { "f1-score": 0.8576207231627551, "precision": 0.8601529227027923, "recall": 0.8614038040733883, "support": 29705.0 }, "step": 81 }, { "epoch": 2.0, "eval_B-Claim": { "f1-score": 0.4708624708624709, "precision": 0.6392405063291139, "recall": 0.3726937269372694, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.796875, "precision": 0.8717948717948718, "recall": 0.7338129496402878, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8736616702355461, "precision": 0.796875, "recall": 0.966824644549763, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.5100589925881107, "precision": 0.6459770114942529, "recall": 0.4213946513371657, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8401387776888176, "precision": 0.9077277970011534, "recall": 0.7819175360158966, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8912891699864469, "precision": 0.8338584492430646, "recall": 0.9572159491884262, "support": 11336.0 }, "eval_O": { "f1-score": 0.9996904982977407, "precision": 1.0, "recall": 0.9993811881188119, "support": 11312.0 }, "eval_accuracy": 0.8830499915839084, "eval_loss": 0.2966194748878479, "eval_macro avg": { "f1-score": 0.7689395113798762, "precision": 0.8136390908374939, "recall": 0.7476058065410885, "support": 29705.0 }, "eval_runtime": 4.8625, "eval_samples_per_second": 16.452, "eval_steps_per_second": 2.057, "eval_weighted avg": { "f1-score": 0.8731020208182413, "precision": 0.874440834821272, "recall": 0.8830499915839084, "support": 29705.0 }, "step": 162 }, { "epoch": 3.0, "eval_B-Claim": { "f1-score": 0.6085192697768763, "precision": 0.6756756756756757, "recall": 0.5535055350553506, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8571428571428571, "precision": 0.851063829787234, "recall": 0.8633093525179856, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8834729626808834, "precision": 0.8529411764705882, "recall": 0.9162717219589257, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.5764474423833614, "precision": 0.6584269662921348, "recall": 0.5126218445388653, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8581151832460733, "precision": 0.9070282235749861, "recall": 0.8142076502732241, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8959744247675935, "precision": 0.8563158317922328, "recall": 0.939484827099506, "support": 11336.0 }, "eval_O": { "f1-score": 0.9996020340481981, "precision": 1.0, "recall": 0.9992043847241867, "support": 11312.0 }, "eval_accuracy": 0.8918700555462044, "eval_loss": 0.2552729547023773, "eval_macro avg": { "f1-score": 0.811324882006549, "precision": 0.8287788147989789, "recall": 0.7998007594525776, "support": 29705.0 }, "eval_runtime": 4.8422, "eval_samples_per_second": 16.522, "eval_steps_per_second": 2.065, "eval_weighted avg": { "f1-score": 0.8867633844066056, "precision": 0.886070631898416, "recall": 0.8918700555462044, "support": 29705.0 }, "step": 243 }, { "epoch": 4.0, "eval_B-Claim": { "f1-score": 0.6722408026755852, "precision": 0.6146788990825688, "recall": 0.7416974169741697, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8664259927797834, "precision": 0.8695652173913043, "recall": 0.8633093525179856, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8687035507844755, "precision": 0.9100346020761245, "recall": 0.8309636650868878, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6483151400094921, "precision": 0.6171222046532641, "recall": 0.6828292926768308, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8696993060909791, "precision": 0.9009584664536742, "recall": 0.8405365126676602, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8906159274643798, "precision": 0.9020175517958925, "recall": 0.879498941425547, "support": 11336.0 }, "eval_O": { "f1-score": 0.999557835160948, "precision": 0.9999115357395613, "recall": 0.9992043847241867, "support": 11312.0 }, "eval_accuracy": 0.8935869382258879, "eval_loss": 0.29465603828430176, "eval_macro avg": { "f1-score": 0.8307940792808061, "precision": 0.8306126395989127, "recall": 0.8340056522961811, "support": 29705.0 }, "eval_runtime": 4.8494, "eval_samples_per_second": 16.497, "eval_steps_per_second": 2.062, "eval_weighted avg": { "f1-score": 0.8954766464091573, "precision": 0.8982496227307208, "recall": 0.8935869382258879, "support": 29705.0 }, "step": 324 }, { "epoch": 5.0, "eval_B-Claim": { "f1-score": 0.6438095238095239, "precision": 0.6653543307086615, "recall": 0.6236162361623616, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8785714285714286, "precision": 0.8723404255319149, "recall": 0.8848920863309353, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8783151326053042, "precision": 0.8674884437596302, "recall": 0.8894154818325435, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6283729628640129, "precision": 0.6748923959827834, "recall": 0.5878530367408148, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.885450461692039, "precision": 0.8896690070210632, "recall": 0.8812717337307501, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8989315871101154, "precision": 0.878494442573257, "recall": 0.9203422724064926, "support": 11336.0 }, "eval_O": { "f1-score": 0.9997347245556637, "precision": 1.0, "recall": 0.9994695898161244, "support": 11312.0 }, "eval_accuracy": 0.8995118666891095, "eval_loss": 0.3176642060279846, "eval_macro avg": { "f1-score": 0.830455117315441, "precision": 0.8354627207967587, "recall": 0.8266943481457174, "support": 29705.0 }, "eval_runtime": 4.8682, "eval_samples_per_second": 16.433, "eval_steps_per_second": 2.054, "eval_weighted avg": { "f1-score": 0.8971010593476484, "precision": 0.8958911872123141, "recall": 0.8995118666891095, "support": 29705.0 }, "step": 405 }, { "epoch": 6.0, "eval_B-Claim": { "f1-score": 0.6150712830957231, "precision": 0.6863636363636364, "recall": 0.5571955719557196, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8686131386861314, "precision": 0.8814814814814815, "recall": 0.8561151079136691, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8804841149773072, "precision": 0.8447024673439768, "recall": 0.919431279620853, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.5758052970651396, "precision": 0.6739276139410187, "recall": 0.5026243439140214, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8627552339105711, "precision": 0.8992456896551724, "recall": 0.829110779930452, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.894107779408681, "precision": 0.8521864257734432, "recall": 0.9403669724770642, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.8922403635751557, "eval_loss": 0.41071853041648865, "eval_macro avg": { "f1-score": 0.8138338353062219, "precision": 0.8339867592226755, "recall": 0.8006920079731114, "support": 29705.0 }, "eval_runtime": 4.8571, "eval_samples_per_second": 16.471, "eval_steps_per_second": 2.059, "eval_weighted avg": { "f1-score": 0.8864802913843919, "precision": 0.8861194550557427, "recall": 0.8922403635751557, "support": 29705.0 }, "step": 486 }, { "epoch": 6.17, "grad_norm": 5.239528656005859, "learning_rate": 1.7530864197530865e-05, "loss": 0.2363, "step": 500 }, { "epoch": 7.0, "eval_B-Claim": { "f1-score": 0.6954954954954955, "precision": 0.6795774647887324, "recall": 0.7121771217712177, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8920863309352518, "precision": 0.8920863309352518, "recall": 0.8920863309352518, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8876494023904382, "precision": 0.8954983922829582, "recall": 0.8799368088467614, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6831545741324923, "precision": 0.6898572884811417, "recall": 0.6765808547863035, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8984707946853848, "precision": 0.9068825910931174, "recall": 0.8902136115250869, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9068164859763859, "precision": 0.9024198479951079, "recall": 0.9112561750176429, "support": 11336.0 }, "eval_O": { "f1-score": 0.9999558011049724, "precision": 0.9999116061168567, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9094428547382596, "eval_loss": 0.3520519435405731, "eval_macro avg": { "f1-score": 0.8519469835314888, "precision": 0.8523190745275951, "recall": 0.8517501289831806, "support": 29705.0 }, "eval_runtime": 4.8905, "eval_samples_per_second": 16.358, "eval_steps_per_second": 2.045, "eval_weighted avg": { "f1-score": 0.9091888981291354, "precision": 0.908989097041669, "recall": 0.9094428547382596, "support": 29705.0 }, "step": 567 }, { "epoch": 8.0, "eval_B-Claim": { "f1-score": 0.6666666666666667, "precision": 0.658273381294964, "recall": 0.6752767527675276, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8819444444444444, "precision": 0.8523489932885906, "recall": 0.9136690647482014, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8759007205764612, "precision": 0.887987012987013, "recall": 0.8641390205371248, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.642274412855377, "precision": 0.6353631694790902, "recall": 0.6493376655836041, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.87683284457478, "precision": 0.862914862914863, "recall": 0.8912071535022354, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8911585772014743, "precision": 0.8972547616918537, "recall": 0.8851446718419196, "support": 11336.0 }, "eval_O": { "f1-score": 0.9998673915926268, "precision": 0.9999115904871364, "recall": 0.9998231966053748, "support": 11312.0 }, "eval_accuracy": 0.8952364921730348, "eval_loss": 0.4631531238555908, "eval_macro avg": { "f1-score": 0.8335207225588329, "precision": 0.827721967449073, "recall": 0.8397996465122839, "support": 29705.0 }, "eval_runtime": 4.8517, "eval_samples_per_second": 16.489, "eval_steps_per_second": 2.061, "eval_weighted avg": { "f1-score": 0.8956465277692122, "precision": 0.8961582200263131, "recall": 0.8952364921730348, "support": 29705.0 }, "step": 648 }, { "epoch": 9.0, "eval_B-Claim": { "f1-score": 0.6604127579737337, "precision": 0.6717557251908397, "recall": 0.6494464944649446, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8661417322834645, "precision": 0.9565217391304348, "recall": 0.7913669064748201, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8852963818321786, "precision": 0.8633633633633634, "recall": 0.9083728278041074, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6148568412719356, "precision": 0.6512017887087759, "recall": 0.5823544113971507, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8500414250207126, "precision": 0.957089552238806, "recall": 0.7645305514157973, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8985408601693112, "precision": 0.8677183468901487, "recall": 0.9316337332392378, "support": 11336.0 }, "eval_O": { "f1-score": 0.9995136401821638, "precision": 0.9998230871295887, "recall": 0.9992043847241867, "support": 11312.0 }, "eval_accuracy": 0.8952701565393032, "eval_loss": 0.5578464865684509, "eval_macro avg": { "f1-score": 0.8249719483905, "precision": 0.8524962289502797, "recall": 0.8038441870743206, "support": 29705.0 }, "eval_runtime": 4.8325, "eval_samples_per_second": 16.555, "eval_steps_per_second": 2.069, "eval_weighted avg": { "f1-score": 0.8928897918536849, "precision": 0.8934537922744324, "recall": 0.8952701565393032, "support": 29705.0 }, "step": 729 }, { "epoch": 10.0, "eval_B-Claim": { "f1-score": 0.6775956284153005, "precision": 0.6690647482014388, "recall": 0.6863468634686347, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8832116788321168, "precision": 0.8962962962962963, "recall": 0.8705035971223022, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8820269200316706, "precision": 0.8841269841269841, "recall": 0.8799368088467614, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6348430262480701, "precision": 0.6542031291434632, "recall": 0.6165958510372407, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8864321608040201, "precision": 0.896797153024911, "recall": 0.8763040238450075, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8934983440822729, "precision": 0.8828797795384086, "recall": 0.9043754410726887, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.897458340346743, "eval_loss": 0.6301500201225281, "eval_macro avg": { "f1-score": 0.8368011083447787, "precision": 0.8404811557616432, "recall": 0.8334375121989479, "support": 29705.0 }, "eval_runtime": 4.8426, "eval_samples_per_second": 16.52, "eval_steps_per_second": 2.065, "eval_weighted avg": { "f1-score": 0.8964756773362035, "precision": 0.895761611933671, "recall": 0.897458340346743, "support": 29705.0 }, "step": 810 }, { "epoch": 11.0, "eval_B-Claim": { "f1-score": 0.6483300589390962, "precision": 0.6932773109243697, "recall": 0.6088560885608856, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8805970149253732, "precision": 0.9147286821705426, "recall": 0.8489208633093526, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8848207475209764, "precision": 0.855457227138643, "recall": 0.9162717219589257, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6216666666666667, "precision": 0.6995936230071897, "recall": 0.55936015996001, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8927083333333333, "precision": 0.9381499726327313, "recall": 0.851465474416294, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9015977681968045, "precision": 0.8655250770978737, "recall": 0.9408080451658434, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9019357010604275, "eval_loss": 0.5967662334442139, "eval_macro avg": { "f1-score": 0.8328172270831785, "precision": 0.8523902704244785, "recall": 0.8179546219101873, "support": 29705.0 }, "eval_runtime": 4.8502, "eval_samples_per_second": 16.494, "eval_steps_per_second": 2.062, "eval_weighted avg": { "f1-score": 0.8979974574260295, "precision": 0.8977509549706758, "recall": 0.9019357010604275, "support": 29705.0 }, "step": 891 }, { "epoch": 12.0, "eval_B-Claim": { "f1-score": 0.6921898928024502, "precision": 0.5916230366492147, "recall": 0.8339483394833949, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8978102189781022, "precision": 0.9111111111111111, "recall": 0.8848920863309353, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8482758620689655, "precision": 0.9335863377609108, "recall": 0.7772511848341233, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6750635055038103, "precision": 0.5854598861758766, "recall": 0.7970507373156711, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8948170731707318, "precision": 0.9157566302652106, "recall": 0.8748137108792846, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8742200328407225, "precision": 0.9336606874436316, "recall": 0.821894848270995, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.8894125568086181, "eval_loss": 0.6401852965354919, "eval_macro avg": { "f1-score": 0.8403395121949689, "precision": 0.8387425270579937, "recall": 0.8556929867306292, "support": 29705.0 }, "eval_runtime": 4.8727, "eval_samples_per_second": 16.418, "eval_steps_per_second": 2.052, "eval_weighted avg": { "f1-score": 0.8945864954779644, "precision": 0.9075830699664568, "recall": 0.8894125568086181, "support": 29705.0 }, "step": 972 }, { "epoch": 12.35, "grad_norm": 23.30899429321289, "learning_rate": 1.506172839506173e-05, "loss": 0.0272, "step": 1000 }, { "epoch": 13.0, "eval_B-Claim": { "f1-score": 0.706766917293233, "precision": 0.7203065134099617, "recall": 0.6937269372693727, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8958333333333333, "precision": 0.8657718120805369, "recall": 0.9280575539568345, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8957345971563981, "precision": 0.8957345971563981, "recall": 0.8957345971563981, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6850271528316524, "precision": 0.7096169300830432, "recall": 0.6620844788802799, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8991678903573177, "precision": 0.8861553304389773, "recall": 0.912568306010929, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9089637690660374, "precision": 0.9007362494586401, "recall": 0.9173429781227946, "support": 11336.0 }, "eval_O": { "f1-score": 0.9999557971975424, "precision": 1.0, "recall": 0.9999115983026874, "support": 11312.0 }, "eval_accuracy": 0.9116310385456994, "eval_loss": 0.6060317754745483, "eval_macro avg": { "f1-score": 0.855921351033645, "precision": 0.8540459189467938, "recall": 0.8584894928141853, "support": 29705.0 }, "eval_runtime": 4.8344, "eval_samples_per_second": 16.548, "eval_steps_per_second": 2.068, "eval_weighted avg": { "f1-score": 0.9106004556040976, "precision": 0.9098905735839876, "recall": 0.9116310385456994, "support": 29705.0 }, "step": 1053 }, { "epoch": 14.0, "eval_B-Claim": { "f1-score": 0.6832740213523131, "precision": 0.6597938144329897, "recall": 0.7084870848708487, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9044117647058822, "precision": 0.924812030075188, "recall": 0.8848920863309353, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8753993610223643, "precision": 0.8852988691437803, "recall": 0.8657187993680885, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6549304444170873, "precision": 0.6453178068898593, "recall": 0.664833791552112, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8969913309535952, "precision": 0.9214248297537978, "recall": 0.8738201689021361, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8920768042374753, "precision": 0.8927467090732397, "recall": 0.891407904022583, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.8988049149974752, "eval_loss": 0.6456504464149475, "eval_macro avg": { "f1-score": 0.8438691038126739, "precision": 0.8470562941955507, "recall": 0.8413085478638147, "support": 29705.0 }, "eval_runtime": 4.8614, "eval_samples_per_second": 16.456, "eval_steps_per_second": 2.057, "eval_weighted avg": { "f1-score": 0.8993641205442222, "precision": 0.900073004604436, "recall": 0.8988049149974752, "support": 29705.0 }, "step": 1134 }, { "epoch": 15.0, "eval_B-Claim": { "f1-score": 0.6766917293233083, "precision": 0.6896551724137931, "recall": 0.6642066420664207, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.899628252788104, "precision": 0.9307692307692308, "recall": 0.8705035971223022, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8849144634525662, "precision": 0.8713629402756509, "recall": 0.8988941548183255, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6276567124024752, "precision": 0.6795805418001748, "recall": 0.583104223944014, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9034411915767849, "precision": 0.9351408825093036, "recall": 0.8738201689021361, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8967347113450592, "precision": 0.8708336796608761, "recall": 0.9242237120677488, "support": 11336.0 }, "eval_O": { "f1-score": 0.9998232278592893, "precision": 0.9996465182043125, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9005554620434271, "eval_loss": 0.7282584309577942, "eval_macro avg": { "f1-score": 0.8412700412496553, "precision": 0.8538555665190488, "recall": 0.8306789284172782, "support": 29705.0 }, "eval_runtime": 4.8599, "eval_samples_per_second": 16.461, "eval_steps_per_second": 2.058, "eval_weighted avg": { "f1-score": 0.8979582328721724, "precision": 0.8971236682980019, "recall": 0.9005554620434271, "support": 29705.0 }, "step": 1215 }, { "epoch": 16.0, "eval_B-Claim": { "f1-score": 0.7162629757785466, "precision": 0.6742671009771987, "recall": 0.7638376383763837, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.909090909090909, "precision": 0.9191176470588235, "recall": 0.8992805755395683, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.884022708840227, "precision": 0.9083333333333333, "recall": 0.8609794628751974, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6889387321449651, "precision": 0.6527964205816554, "recall": 0.7293176705823544, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9093224656138562, "precision": 0.9330893883951908, "recall": 0.886736214605067, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8969196969017621, "precision": 0.9120087535333272, "recall": 0.8823218066337333, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9053694664197947, "eval_loss": 0.6488288640975952, "eval_macro avg": { "f1-score": 0.8577939269100382, "precision": 0.8570875205542184, "recall": 0.8603533383731864, "support": 29705.0 }, "eval_runtime": 4.862, "eval_samples_per_second": 16.454, "eval_steps_per_second": 2.057, "eval_weighted avg": { "f1-score": 0.9071351271559185, "precision": 0.9098177877443702, "recall": 0.9053694664197947, "support": 29705.0 }, "step": 1296 }, { "epoch": 17.0, "eval_B-Claim": { "f1-score": 0.7306397306397308, "precision": 0.6718266253869969, "recall": 0.8007380073800738, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.924187725631769, "precision": 0.927536231884058, "recall": 0.920863309352518, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8808545603944125, "precision": 0.9178082191780822, "recall": 0.8467614533965245, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.7016976556184318, "precision": 0.652211249463289, "recall": 0.7593101724568858, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9132860040567952, "precision": 0.9326773692387365, "recall": 0.8946845504222554, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8968950846383634, "precision": 0.9209890314184793, "recall": 0.874029640084686, "support": 11336.0 }, "eval_O": { "f1-score": 0.9999558011049724, "precision": 0.9999116061168567, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9069180272681366, "eval_loss": 0.616534411907196, "eval_macro avg": { "f1-score": 0.8639309374406394, "precision": 0.8604229046694998, "recall": 0.8709124475847062, "support": 29705.0 }, "eval_runtime": 4.847, "eval_samples_per_second": 16.505, "eval_steps_per_second": 2.063, "eval_weighted avg": { "f1-score": 0.9092303028261727, "precision": 0.913323469327906, "recall": 0.9069180272681366, "support": 29705.0 }, "step": 1377 }, { "epoch": 18.0, "eval_B-Claim": { "f1-score": 0.7090909090909091, "precision": 0.6989247311827957, "recall": 0.7195571955719557, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9257950530035336, "precision": 0.9097222222222222, "recall": 0.9424460431654677, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.885873902633679, "precision": 0.8951612903225806, "recall": 0.8767772511848341, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6834259727134917, "precision": 0.6909323116219668, "recall": 0.6760809797550612, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9266351653817458, "precision": 0.9277888446215139, "recall": 0.9254843517138599, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9006238467621475, "precision": 0.8970768422895151, "recall": 0.9041990119971771, "support": 11336.0 }, "eval_O": { "f1-score": 0.9999558011049724, "precision": 0.9999116061168567, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9093081972731863, "eval_loss": 0.638508677482605, "eval_macro avg": { "f1-score": 0.8616286643843541, "precision": 0.8599311211967786, "recall": 0.8635064047697651, "support": 29705.0 }, "eval_runtime": 4.9839, "eval_samples_per_second": 16.052, "eval_steps_per_second": 2.006, "eval_weighted avg": { "f1-score": 0.9090147511474005, "precision": 0.9087634870703264, "recall": 0.9093081972731863, "support": 29705.0 }, "step": 1458 }, { "epoch": 18.52, "grad_norm": 0.05384668707847595, "learning_rate": 1.2592592592592593e-05, "loss": 0.0098, "step": 1500 }, { "epoch": 19.0, "eval_B-Claim": { "f1-score": 0.7074074074074074, "precision": 0.7100371747211895, "recall": 0.7047970479704797, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9310344827586207, "precision": 0.8940397350993378, "recall": 0.9712230215827338, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8844621513944222, "precision": 0.8922829581993569, "recall": 0.8767772511848341, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6726757661601854, "precision": 0.6937583001328022, "recall": 0.6528367908022994, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9193273214720935, "precision": 0.9023923444976076, "recall": 0.936910084451068, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9006964825441325, "precision": 0.8945445053510833, "recall": 0.9069336626676077, "support": 11336.0 }, "eval_O": { "f1-score": 0.9998674150351351, "precision": 0.9997348652231551, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9079952869887224, "eval_loss": 0.6490303874015808, "eval_macro avg": { "f1-score": 0.859353003824571, "precision": 0.8552556976035047, "recall": 0.8642111226655746, "support": 29705.0 }, "eval_runtime": 4.8703, "eval_samples_per_second": 16.426, "eval_steps_per_second": 2.053, "eval_weighted avg": { "f1-score": 0.9070447031096275, "precision": 0.9063560601392924, "recall": 0.9079952869887224, "support": 29705.0 }, "step": 1539 }, { "epoch": 20.0, "eval_B-Claim": { "f1-score": 0.7035714285714285, "precision": 0.6816608996539792, "recall": 0.7269372693726938, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9059233449477351, "precision": 0.8783783783783784, "recall": 0.935251798561151, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8838709677419355, "precision": 0.9028006589785832, "recall": 0.8657187993680885, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6907709278097806, "precision": 0.6777296777296777, "recall": 0.7043239190202449, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9150943396226415, "precision": 0.9146401985111663, "recall": 0.9155489319423745, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9017412935323382, "precision": 0.9081961345740873, "recall": 0.8953775582215949, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9088705605116983, "eval_loss": 0.6806873083114624, "eval_macro avg": { "f1-score": 0.857281757460837, "precision": 0.851915135403696, "recall": 0.8633083252123068, "support": 29705.0 }, "eval_runtime": 4.8892, "eval_samples_per_second": 16.362, "eval_steps_per_second": 2.045, "eval_weighted avg": { "f1-score": 0.9094792195246973, "precision": 0.9102297948919554, "recall": 0.9088705605116983, "support": 29705.0 }, "step": 1620 }, { "epoch": 21.0, "eval_B-Claim": { "f1-score": 0.7094594594594594, "precision": 0.6542056074766355, "recall": 0.7749077490774908, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9070631970260222, "precision": 0.9384615384615385, "recall": 0.8776978417266187, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8767346938775511, "precision": 0.9070945945945946, "recall": 0.8483412322274881, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6910236955760477, "precision": 0.648269820411739, "recall": 0.7398150462384404, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8995290423861854, "precision": 0.9502487562189055, "recall": 0.8539493293591655, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8971761541909459, "precision": 0.9119737561509021, "recall": 0.8828510938602682, "support": 11336.0 }, "eval_O": { "f1-score": 0.9999558011049724, "precision": 0.9999116061168567, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9044941928968188, "eval_loss": 0.7068688869476318, "eval_macro avg": { "f1-score": 0.8544202919458835, "precision": 0.858595097061596, "recall": 0.8539374703556388, "support": 29705.0 }, "eval_runtime": 4.841, "eval_samples_per_second": 16.525, "eval_steps_per_second": 2.066, "eval_weighted avg": { "f1-score": 0.9066064646845915, "precision": 0.9102050037942175, "recall": 0.9044941928968188, "support": 29705.0 }, "step": 1701 }, { "epoch": 22.0, "eval_B-Claim": { "f1-score": 0.7012987012987013, "precision": 0.7052238805970149, "recall": 0.6974169741697417, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9310344827586207, "precision": 0.8940397350993378, "recall": 0.9712230215827338, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8862370723945903, "precision": 0.8926282051282052, "recall": 0.8799368088467614, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.676991150442478, "precision": 0.7062177572631008, "recall": 0.6500874781304674, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9239260859131269, "precision": 0.893686165273909, "recall": 0.9562841530054644, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.904022057858112, "precision": 0.897072874142274, "recall": 0.9110797459421313, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9105201144588453, "eval_loss": 0.6818939447402954, "eval_macro avg": { "f1-score": 0.8605013643808042, "precision": 0.8555526596434059, "recall": 0.8665754545253285, "support": 29705.0 }, "eval_runtime": 4.8684, "eval_samples_per_second": 16.432, "eval_steps_per_second": 2.054, "eval_weighted avg": { "f1-score": 0.9092392748602662, "precision": 0.9084735380390377, "recall": 0.9105201144588453, "support": 29705.0 }, "step": 1782 }, { "epoch": 23.0, "eval_B-Claim": { "f1-score": 0.7009174311926606, "precision": 0.6970802919708029, "recall": 0.7047970479704797, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9181494661921709, "precision": 0.9084507042253521, "recall": 0.9280575539568345, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.885193982581156, "precision": 0.8873015873015873, "recall": 0.8830963665086888, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6707700037683707, "precision": 0.6742424242424242, "recall": 0.6673331667083229, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9126604580921219, "precision": 0.925, "recall": 0.9006458022851466, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8992663532926242, "precision": 0.895685656777807, "recall": 0.9028757939308398, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9058744319138192, "eval_loss": 0.7413836717605591, "eval_macro avg": { "f1-score": 0.8552796707313007, "precision": 0.8553943806454248, "recall": 0.8552579616229018, "support": 29705.0 }, "eval_runtime": 4.8821, "eval_samples_per_second": 16.386, "eval_steps_per_second": 2.048, "eval_weighted avg": { "f1-score": 0.9057370428806488, "precision": 0.905639012166318, "recall": 0.9058744319138192, "support": 29705.0 }, "step": 1863 }, { "epoch": 24.0, "eval_B-Claim": { "f1-score": 0.7047970479704797, "precision": 0.7047970479704797, "recall": 0.7047970479704797, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9333333333333332, "precision": 0.910958904109589, "recall": 0.9568345323741008, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8864177918983319, "precision": 0.8913738019169329, "recall": 0.8815165876777251, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6843109810408449, "precision": 0.6969932607568688, "recall": 0.6720819795051237, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9312638580931263, "precision": 0.9237536656891495, "recall": 0.9388971684053651, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9026424370116758, "precision": 0.8983050847457628, "recall": 0.9070218772053634, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9107894293889918, "eval_loss": 0.7259247303009033, "eval_macro avg": { "f1-score": 0.8632522070496845, "precision": 0.8608831093126833, "recall": 0.8658784561625941, "support": 29705.0 }, "eval_runtime": 4.8619, "eval_samples_per_second": 16.454, "eval_steps_per_second": 2.057, "eval_weighted avg": { "f1-score": 0.910242479146784, "precision": 0.909787426705373, "recall": 0.9107894293889918, "support": 29705.0 }, "step": 1944 }, { "epoch": 24.69, "grad_norm": 0.0542856864631176, "learning_rate": 1.0123456790123458e-05, "loss": 0.004, "step": 2000 }, { "epoch": 25.0, "eval_B-Claim": { "f1-score": 0.7201426024955437, "precision": 0.696551724137931, "recall": 0.7453874538745388, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9285714285714285, "precision": 0.9219858156028369, "recall": 0.935251798561151, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8883534136546184, "precision": 0.9035947712418301, "recall": 0.8736176935229067, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6924489046628319, "precision": 0.6784172661870503, "recall": 0.707073231692077, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9278195488721803, "precision": 0.936267071320182, "recall": 0.9195230998509687, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9005723412751232, "precision": 0.9059180576631259, "recall": 0.8952893436838391, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9098131627672109, "eval_loss": 0.7058032751083374, "eval_macro avg": { "f1-score": 0.8654154627902466, "precision": 0.8632478151647079, "recall": 0.8680203744550686, "support": 29705.0 }, "eval_runtime": 4.8778, "eval_samples_per_second": 16.401, "eval_steps_per_second": 2.05, "eval_weighted avg": { "f1-score": 0.91047415131532, "precision": 0.9112754497242, "recall": 0.9098131627672109, "support": 29705.0 }, "step": 2025 }, { "epoch": 26.0, "eval_B-Claim": { "f1-score": 0.7129798903107861, "precision": 0.7065217391304348, "recall": 0.7195571955719557, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9214285714285715, "precision": 0.9148936170212766, "recall": 0.9280575539568345, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8883610451306412, "precision": 0.8904761904761904, "recall": 0.8862559241706162, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6902947154471545, "precision": 0.7018858176181865, "recall": 0.6790802299425144, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9235352532274081, "precision": 0.9230769230769231, "recall": 0.9239940387481371, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9056852079312161, "precision": 0.9007853403141362, "recall": 0.9106386732533521, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9122033327722605, "eval_loss": 0.6844227313995361, "eval_macro avg": { "f1-score": 0.8631835262108254, "precision": 0.8625199468053067, "recall": 0.8639405165204871, "support": 29705.0 }, "eval_runtime": 4.8434, "eval_samples_per_second": 16.517, "eval_steps_per_second": 2.065, "eval_weighted avg": { "f1-score": 0.9117462313290554, "precision": 0.9113620822382424, "recall": 0.9122033327722605, "support": 29705.0 }, "step": 2106 }, { "epoch": 27.0, "eval_B-Claim": { "f1-score": 0.6977611940298508, "precision": 0.7056603773584905, "recall": 0.6900369003690037, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9022556390977442, "precision": 0.9448818897637795, "recall": 0.8633093525179856, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8909657320872274, "precision": 0.8786482334869432, "recall": 0.9036334913112164, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6695436118739375, "precision": 0.7021393307734504, "recall": 0.6398400399900025, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9034411915767849, "precision": 0.9351408825093036, "recall": 0.8738201689021361, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9045152378485712, "precision": 0.8856297548605241, "recall": 0.9242237120677488, "support": 11336.0 }, "eval_O": { "f1-score": 0.9996904982977407, "precision": 1.0, "recall": 0.9993811881188119, "support": 11312.0 }, "eval_accuracy": 0.9082646019188689, "eval_loss": 0.7212072610855103, "eval_macro avg": { "f1-score": 0.8525961578302652, "precision": 0.8645857812503559, "recall": 0.8420349790395578, "support": 29705.0 }, "eval_runtime": 4.85, "eval_samples_per_second": 16.495, "eval_steps_per_second": 2.062, "eval_weighted avg": { "f1-score": 0.9068521423522986, "precision": 0.9063105146188606, "recall": 0.9082646019188689, "support": 29705.0 }, "step": 2187 }, { "epoch": 28.0, "eval_B-Claim": { "f1-score": 0.7015706806282722, "precision": 0.6655629139072847, "recall": 0.7416974169741697, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8803088803088803, "precision": 0.95, "recall": 0.8201438848920863, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.886762360446571, "precision": 0.895330112721417, "recall": 0.8783570300157978, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.675830190974334, "precision": 0.6582938388625592, "recall": 0.6943264183954011, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8754676643506146, "precision": 0.9473684210526315, "recall": 0.8137108792846498, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9025816950345251, "precision": 0.9000087711604245, "recall": 0.9051693719124911, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9042248779666723, "eval_loss": 0.7449725866317749, "eval_macro avg": { "f1-score": 0.8460744959633139, "precision": 0.8595091511006167, "recall": 0.8362007144963709, "support": 29705.0 }, "eval_runtime": 4.8267, "eval_samples_per_second": 16.574, "eval_steps_per_second": 2.072, "eval_weighted avg": { "f1-score": 0.9050256946747467, "precision": 0.9067344629444252, "recall": 0.9042248779666723, "support": 29705.0 }, "step": 2268 }, { "epoch": 29.0, "eval_B-Claim": { "f1-score": 0.6972477064220184, "precision": 0.6934306569343066, "recall": 0.7011070110701108, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9163636363636365, "precision": 0.9264705882352942, "recall": 0.9064748201438849, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8855564325177585, "precision": 0.8848580441640379, "recall": 0.8862559241706162, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6666666666666666, "precision": 0.6897381079636558, "recall": 0.6450887278180455, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9041095890410958, "precision": 0.9237947122861586, "recall": 0.8852459016393442, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9035369774919614, "precision": 0.8903065593423531, "recall": 0.917166549047283, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9072210065645514, "eval_loss": 0.7131910920143127, "eval_macro avg": { "f1-score": 0.8533544297861624, "precision": 0.8583712384179724, "recall": 0.8487627048413263, "support": 29705.0 }, "eval_runtime": 4.8395, "eval_samples_per_second": 16.531, "eval_steps_per_second": 2.066, "eval_weighted avg": { "f1-score": 0.9062005389193937, "precision": 0.9055906528951113, "recall": 0.9072210065645514, "support": 29705.0 }, "step": 2349 }, { "epoch": 30.0, "eval_B-Claim": { "f1-score": 0.707182320441989, "precision": 0.7058823529411765, "recall": 0.7084870848708487, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9090909090909091, "precision": 0.96, "recall": 0.8633093525179856, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8928850664581705, "precision": 0.8839009287925697, "recall": 0.9020537124802528, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6665808776219276, "precision": 0.6870026525198939, "recall": 0.6473381654586353, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8990971853425385, "precision": 0.9657729606389047, "recall": 0.8410332836562344, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9038552864482148, "precision": 0.8850934302866322, "recall": 0.9234297812279464, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9071200134657466, "eval_loss": 0.784762442111969, "eval_macro avg": { "f1-score": 0.8540988064862499, "precision": 0.8696646178827396, "recall": 0.8408073400302719, "support": 29705.0 }, "eval_runtime": 4.8562, "eval_samples_per_second": 16.474, "eval_steps_per_second": 2.059, "eval_weighted avg": { "f1-score": 0.906183556468886, "precision": 0.9063275917791694, "recall": 0.9071200134657466, "support": 29705.0 }, "step": 2430 }, { "epoch": 30.86, "grad_norm": 0.020902352407574654, "learning_rate": 7.654320987654322e-06, "loss": 0.0031, "step": 2500 }, { "epoch": 31.0, "eval_B-Claim": { "f1-score": 0.696461824953445, "precision": 0.7030075187969925, "recall": 0.6900369003690037, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8897058823529412, "precision": 0.9097744360902256, "recall": 0.8705035971223022, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.894283476898982, "precision": 0.8866459627329193, "recall": 0.9020537124802528, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6603250099088387, "precision": 0.7003923766816144, "recall": 0.6245938515371158, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.898928024502297, "precision": 0.9244094488188976, "recall": 0.8748137108792846, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9050101236376169, "precision": 0.8843984171086975, "recall": 0.926605504587156, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9074229927621612, "eval_loss": 0.7240239381790161, "eval_macro avg": { "f1-score": 0.8492449060363031, "precision": 0.8583754514613353, "recall": 0.841229610996445, "support": 29705.0 }, "eval_runtime": 4.8477, "eval_samples_per_second": 16.503, "eval_steps_per_second": 2.063, "eval_weighted avg": { "f1-score": 0.9056114779535739, "precision": 0.9048600287675556, "recall": 0.9074229927621612, "support": 29705.0 }, "step": 2511 }, { "epoch": 32.0, "eval_B-Claim": { "f1-score": 0.7124773960216998, "precision": 0.6985815602836879, "recall": 0.7269372693726938, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9, "precision": 0.8936170212765957, "recall": 0.9064748201438849, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8906624102154829, "precision": 0.9, "recall": 0.8815165876777251, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6955112219451373, "precision": 0.6939537198308037, "recall": 0.6970757310672332, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8990129081245255, "precision": 0.9164086687306502, "recall": 0.8822652757078987, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9081789783976418, "precision": 0.9059071359606776, "recall": 0.9104622441778405, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9115973741794311, "eval_loss": 0.7000880241394043, "eval_macro avg": { "f1-score": 0.8579775592434981, "precision": 0.8583525865832021, "recall": 0.8578188468781823, "support": 29705.0 }, "eval_runtime": 4.8892, "eval_samples_per_second": 16.363, "eval_steps_per_second": 2.045, "eval_weighted avg": { "f1-score": 0.9116829140602968, "precision": 0.9118273403549036, "recall": 0.9115973741794311, "support": 29705.0 }, "step": 2592 }, { "epoch": 33.0, "eval_B-Claim": { "f1-score": 0.7093235831809871, "precision": 0.7028985507246377, "recall": 0.7158671586715867, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9154929577464789, "precision": 0.896551724137931, "recall": 0.935251798561151, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8892430278884461, "precision": 0.8971061093247589, "recall": 0.8815165876777251, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6861111111111111, "precision": 0.693289104363358, "recall": 0.6790802299425144, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9068664169787766, "precision": 0.9116465863453815, "recall": 0.9021361152508693, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9057299670691549, "precision": 0.9016522423288749, "recall": 0.9098447424135497, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9103181282612355, "eval_loss": 0.7453714609146118, "eval_macro avg": { "f1-score": 0.8589667234249935, "precision": 0.8575920453178488, "recall": 0.8605280903596281, "support": 29705.0 }, "eval_runtime": 4.8419, "eval_samples_per_second": 16.523, "eval_steps_per_second": 2.065, "eval_weighted avg": { "f1-score": 0.91002789457842, "precision": 0.9097828132642751, "recall": 0.9103181282612355, "support": 29705.0 }, "step": 2673 }, { "epoch": 34.0, "eval_B-Claim": { "f1-score": 0.6943942133815552, "precision": 0.6808510638297872, "recall": 0.7084870848708487, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.8992805755395683, "precision": 0.8992805755395683, "recall": 0.8992805755395683, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8844621513944222, "precision": 0.8922829581993569, "recall": 0.8767772511848341, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6702770780856424, "precision": 0.6755521706016756, "recall": 0.665083729067733, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8947235546579146, "precision": 0.9096509240246407, "recall": 0.8802781917536016, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9017062151848765, "precision": 0.8967111576376167, "recall": 0.9067572335920959, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9054367951523312, "eval_loss": 0.8030840754508972, "eval_macro avg": { "f1-score": 0.8492633983205684, "precision": 0.8506184071189493, "recall": 0.8480948665726687, "support": 29705.0 }, "eval_runtime": 4.8456, "eval_samples_per_second": 16.51, "eval_steps_per_second": 2.064, "eval_weighted avg": { "f1-score": 0.905222828949398, "precision": 0.9050818036848003, "recall": 0.9054367951523312, "support": 29705.0 }, "step": 2754 }, { "epoch": 35.0, "eval_B-Claim": { "f1-score": 0.7187499999999999, "precision": 0.6786885245901639, "recall": 0.7638376383763837, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.909090909090909, "precision": 0.9191176470588235, "recall": 0.8992805755395683, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.88582995951417, "precision": 0.9086378737541528, "recall": 0.8641390205371248, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.697302458820721, "precision": 0.667352067626228, "recall": 0.7300674831292177, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9053747161241484, "precision": 0.92, "recall": 0.8912071535022354, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9020081398989221, "precision": 0.9148144788170189, "recall": 0.8895553987297107, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.908601245581552, "eval_loss": 0.7647125124931335, "eval_macro avg": { "f1-score": 0.8597651690641243, "precision": 0.858372941692341, "recall": 0.8625838956877485, "support": 29705.0 }, "eval_runtime": 4.8579, "eval_samples_per_second": 16.468, "eval_steps_per_second": 2.058, "eval_weighted avg": { "f1-score": 0.9099971710462852, "precision": 0.9120088292083405, "recall": 0.908601245581552, "support": 29705.0 }, "step": 2835 }, { "epoch": 36.0, "eval_B-Claim": { "f1-score": 0.7014388489208633, "precision": 0.6842105263157895, "recall": 0.7195571955719557, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9057971014492754, "precision": 0.9124087591240876, "recall": 0.8992805755395683, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8851674641148325, "precision": 0.893719806763285, "recall": 0.8767772511848341, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6858477806467478, "precision": 0.6749757986447241, "recall": 0.6970757310672332, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9008179959100204, "precision": 0.9278567667193259, "recall": 0.8753104818678589, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9030236150960054, "precision": 0.9037017404364344, "recall": 0.9023465067043048, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9078269651573809, "eval_loss": 0.771583080291748, "eval_macro avg": { "f1-score": 0.8545846865911064, "precision": 0.8566961997148067, "recall": 0.8529068202765364, "support": 29705.0 }, "eval_runtime": 4.8879, "eval_samples_per_second": 16.367, "eval_steps_per_second": 2.046, "eval_weighted avg": { "f1-score": 0.9083455992865349, "precision": 0.9090283547955545, "recall": 0.9078269651573809, "support": 29705.0 }, "step": 2916 }, { "epoch": 37.0, "eval_B-Claim": { "f1-score": 0.7112676056338029, "precision": 0.6801346801346801, "recall": 0.7453874538745388, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9187279151943463, "precision": 0.9027777777777778, "recall": 0.935251798561151, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8825910931174089, "precision": 0.9053156146179402, "recall": 0.8609794628751974, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6840438489646773, "precision": 0.6671418389166073, "recall": 0.701824543864034, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9142150803461063, "precision": 0.9099409448818898, "recall": 0.9185295578738202, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8983738026286477, "precision": 0.9075524349626429, "recall": 0.889378969654199, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.906514054872917, "eval_loss": 0.8116065263748169, "eval_macro avg": { "f1-score": 0.8584599065549984, "precision": 0.8532661844702197, "recall": 0.8644788266718487, "support": 29705.0 }, "eval_runtime": 4.9036, "eval_samples_per_second": 16.315, "eval_steps_per_second": 2.039, "eval_weighted avg": { "f1-score": 0.907331314133137, "precision": 0.9083934494452071, "recall": 0.906514054872917, "support": 29705.0 }, "step": 2997 }, { "epoch": 37.04, "grad_norm": 0.0017636946868151426, "learning_rate": 5.185185185185185e-06, "loss": 0.0017, "step": 3000 }, { "epoch": 38.0, "eval_B-Claim": { "f1-score": 0.6705426356589148, "precision": 0.7061224489795919, "recall": 0.6383763837638377, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.903225806451613, "precision": 0.9, "recall": 0.9064748201438849, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8861347792408986, "precision": 0.8693009118541033, "recall": 0.9036334913112164, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6457135214658286, "precision": 0.6944764096662831, "recall": 0.6033491627093227, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9037149355572404, "precision": 0.9197530864197531, "recall": 0.8882265275707899, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9012292615834264, "precision": 0.8787929589270747, "recall": 0.9248412138320395, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.904527857263087, "eval_loss": 0.8267531394958496, "eval_macro avg": { "f1-score": 0.8443658485654174, "precision": 0.8526351165495436, "recall": 0.8378430856187273, "support": 29705.0 }, "eval_runtime": 4.8546, "eval_samples_per_second": 16.479, "eval_steps_per_second": 2.06, "eval_weighted avg": { "f1-score": 0.902178139390374, "precision": 0.9012215694769297, "recall": 0.904527857263087, "support": 29705.0 }, "step": 3078 }, { "epoch": 39.0, "eval_B-Claim": { "f1-score": 0.7102473498233215, "precision": 0.6813559322033899, "recall": 0.7416974169741697, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9044117647058822, "precision": 0.924812030075188, "recall": 0.8848920863309353, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8862179487179487, "precision": 0.8991869918699187, "recall": 0.8736176935229067, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6807317073170731, "precision": 0.6646820671588474, "recall": 0.6975756060984754, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.8934088568486096, "precision": 0.9273115980758952, "recall": 0.8618976651763537, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9012203749557836, "precision": 0.9034574468085106, "recall": 0.8989943542695836, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9057734388150143, "eval_loss": 0.8176060914993286, "eval_macro avg": { "f1-score": 0.8537482860526598, "precision": 0.8572580094559642, "recall": 0.8512392603389178, "support": 29705.0 }, "eval_runtime": 4.8519, "eval_samples_per_second": 16.488, "eval_steps_per_second": 2.061, "eval_weighted avg": { "f1-score": 0.9065625311881947, "precision": 0.907660209066212, "recall": 0.9057734388150143, "support": 29705.0 }, "step": 3159 }, { "epoch": 40.0, "eval_B-Claim": { "f1-score": 0.7125220458553791, "precision": 0.6824324324324325, "recall": 0.7453874538745388, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.911032028469751, "precision": 0.9014084507042254, "recall": 0.920863309352518, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8852988691437802, "precision": 0.9057851239669421, "recall": 0.8657187993680885, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6844162188568637, "precision": 0.6692142345354669, "recall": 0.7003249187703074, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9068198850861853, "precision": 0.9120603015075377, "recall": 0.9016393442622951, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9002621173752722, "precision": 0.9068289626778842, "recall": 0.8937896965419901, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9068843629018684, "eval_loss": 0.8002561330795288, "eval_macro avg": { "f1-score": 0.857193023541033, "precision": 0.8539613579749269, "recall": 0.8611033603099626, "support": 29705.0 }, "eval_runtime": 4.8617, "eval_samples_per_second": 16.455, "eval_steps_per_second": 2.057, "eval_weighted avg": { "f1-score": 0.9076340748507997, "precision": 0.9085646728677234, "recall": 0.9068843629018684, "support": 29705.0 }, "step": 3240 }, { "epoch": 41.0, "eval_B-Claim": { "f1-score": 0.7071428571428571, "precision": 0.6851211072664359, "recall": 0.7306273062730627, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9071428571428571, "precision": 0.900709219858156, "recall": 0.9136690647482014, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8860353130016052, "precision": 0.9004893964110929, "recall": 0.8720379146919431, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6871921182266011, "precision": 0.6773488710852149, "recall": 0.6973256685828543, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9089548515839362, "precision": 0.9128256513026052, "recall": 0.905116741182315, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9021310531212617, "precision": 0.906186025812194, "recall": 0.8981122088920254, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9083319306514055, "eval_loss": 0.8057775497436523, "eval_macro avg": { "f1-score": 0.856942721459874, "precision": 0.854668610247957, "recall": 0.8595555577672004, "support": 29705.0 }, "eval_runtime": 4.8716, "eval_samples_per_second": 16.422, "eval_steps_per_second": 2.053, "eval_weighted avg": { "f1-score": 0.9088142845986974, "precision": 0.9093752515552543, "recall": 0.9083319306514055, "support": 29705.0 }, "step": 3321 }, { "epoch": 42.0, "eval_B-Claim": { "f1-score": 0.703971119133574, "precision": 0.6890459363957597, "recall": 0.7195571955719557, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.911032028469751, "precision": 0.9014084507042254, "recall": 0.920863309352518, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8856914468425259, "precision": 0.8964401294498382, "recall": 0.8751974723538705, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6838693592620294, "precision": 0.6821686147724446, "recall": 0.6855786053486629, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9073657927590512, "precision": 0.9121485943775101, "recall": 0.9026328862394436, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9030123935959069, "precision": 0.9029725676986857, "recall": 0.9030522230063515, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9084665881164787, "eval_loss": 0.8089765310287476, "eval_macro avg": { "f1-score": 0.8564203057232626, "precision": 0.8548834704854948, "recall": 0.858125955981829, "support": 29705.0 }, "eval_runtime": 4.8242, "eval_samples_per_second": 16.583, "eval_steps_per_second": 2.073, "eval_weighted avg": { "f1-score": 0.9085773254377499, "precision": 0.9087050193897211, "recall": 0.9084665881164787, "support": 29705.0 }, "step": 3402 }, { "epoch": 43.0, "eval_B-Claim": { "f1-score": 0.7043795620437956, "precision": 0.6967509025270758, "recall": 0.7121771217712177, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9078014184397164, "precision": 0.8951048951048951, "recall": 0.920863309352518, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8885350318471338, "precision": 0.8956661316211878, "recall": 0.8815165876777251, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6697614442295293, "precision": 0.6917954182205648, "recall": 0.649087728067983, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9136352388022766, "precision": 0.9102564102564102, "recall": 0.9170392449080974, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9019385260216556, "precision": 0.8928941908713693, "recall": 0.9111679604798871, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9076923076923077, "eval_loss": 0.8238369822502136, "eval_macro avg": { "f1-score": 0.855150174483444, "precision": 0.8546382783716434, "recall": 0.8559788503224899, "support": 29705.0 }, "eval_runtime": 4.8549, "eval_samples_per_second": 16.478, "eval_steps_per_second": 2.06, "eval_weighted avg": { "f1-score": 0.9067413680757126, "precision": 0.9060516379796367, "recall": 0.9076923076923077, "support": 29705.0 }, "step": 3483 }, { "epoch": 43.21, "grad_norm": 0.007966211065649986, "learning_rate": 2.7160493827160496e-06, "loss": 0.0007, "step": 3500 }, { "epoch": 44.0, "eval_B-Claim": { "f1-score": 0.6909090909090909, "precision": 0.6810035842293907, "recall": 0.7011070110701108, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.909090909090909, "precision": 0.9191176470588235, "recall": 0.8992805755395683, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8834258524980174, "precision": 0.8869426751592356, "recall": 0.8799368088467614, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6563347539713289, "precision": 0.6790486370924639, "recall": 0.6350912271932017, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9042687547360444, "precision": 0.9198355601233299, "recall": 0.8892200695479384, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8994695190886163, "precision": 0.8868976161893328, "recall": 0.9124029640084685, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9041575492341357, "eval_loss": 0.8489497303962708, "eval_macro avg": { "f1-score": 0.8490712686134296, "precision": 0.8532636742646538, "recall": 0.8452912366008641, "support": 29705.0 }, "eval_runtime": 4.8348, "eval_samples_per_second": 16.547, "eval_steps_per_second": 2.068, "eval_weighted avg": { "f1-score": 0.9031302267949082, "precision": 0.9024783031240742, "recall": 0.9041575492341357, "support": 29705.0 }, "step": 3564 }, { "epoch": 45.0, "eval_B-Claim": { "f1-score": 0.7079646017699115, "precision": 0.6802721088435374, "recall": 0.7380073800738007, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.909090909090909, "precision": 0.9191176470588235, "recall": 0.8992805755395683, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8860353130016052, "precision": 0.9004893964110929, "recall": 0.8720379146919431, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6735205616850551, "precision": 0.6757232704402516, "recall": 0.6713321669582605, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9035175879396985, "precision": 0.9140823589222166, "recall": 0.8931942374565326, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9010728104115371, "precision": 0.8982293127629734, "recall": 0.9039343683839096, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9062447399427705, "eval_loss": 0.8230426907539368, "eval_macro avg": { "f1-score": 0.8544573976998165, "precision": 0.8554162992055565, "recall": 0.8539695204434308, "support": 29705.0 }, "eval_runtime": 4.8598, "eval_samples_per_second": 16.461, "eval_steps_per_second": 2.058, "eval_weighted avg": { "f1-score": 0.9062171502179094, "precision": 0.9062469269614183, "recall": 0.9062447399427705, "support": 29705.0 }, "step": 3645 }, { "epoch": 46.0, "eval_B-Claim": { "f1-score": 0.6994535519125684, "precision": 0.6906474820143885, "recall": 0.7084870848708487, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.911032028469751, "precision": 0.9014084507042254, "recall": 0.920863309352518, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8853503184713376, "precision": 0.8924558587479936, "recall": 0.8783570300157978, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6642938858522944, "precision": 0.6758923952405587, "recall": 0.6530867283179205, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9045601794168949, "precision": 0.9075, "recall": 0.9016393442622951, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8994741454864154, "precision": 0.8936781609195402, "recall": 0.9053458009880028, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9048645009257701, "eval_loss": 0.8292382955551147, "eval_macro avg": { "f1-score": 0.8520234442298945, "precision": 0.8516546210895294, "recall": 0.852539899686769, "support": 29705.0 }, "eval_runtime": 4.8644, "eval_samples_per_second": 16.446, "eval_steps_per_second": 2.056, "eval_weighted avg": { "f1-score": 0.9043518097227135, "precision": 0.9039274333076133, "recall": 0.9048645009257701, "support": 29705.0 }, "step": 3726 }, { "epoch": 47.0, "eval_B-Claim": { "f1-score": 0.6935779816513761, "precision": 0.6897810218978102, "recall": 0.6974169741697417, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9214285714285715, "precision": 0.9148936170212766, "recall": 0.9280575539568345, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8818398096748612, "precision": 0.8853503184713376, "recall": 0.8783570300157978, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6593998461143883, "precision": 0.6771135106663155, "recall": 0.6425893526618346, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9073657927590512, "precision": 0.9121485943775101, "recall": 0.9026328862394436, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8988950517535048, "precision": 0.8901479110803564, "recall": 0.9078158080451658, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9043931997980138, "eval_loss": 0.8381735682487488, "eval_macro avg": { "f1-score": 0.8517867219116791, "precision": 0.8527764247878009, "recall": 0.8509813721555455, "support": 29705.0 }, "eval_runtime": 4.8478, "eval_samples_per_second": 16.502, "eval_steps_per_second": 2.063, "eval_weighted avg": { "f1-score": 0.9035819975016209, "precision": 0.902963494965743, "recall": 0.9043931997980138, "support": 29705.0 }, "step": 3807 }, { "epoch": 48.0, "eval_B-Claim": { "f1-score": 0.6927374301675978, "precision": 0.6992481203007519, "recall": 0.6863468634686347, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9214285714285715, "precision": 0.9148936170212766, "recall": 0.9280575539568345, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8841607565011821, "precision": 0.8820754716981132, "recall": 0.8862559241706162, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6538918990352848, "precision": 0.6937745372966909, "recall": 0.6183454136465883, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9070929070929071, "precision": 0.9121044701155199, "recall": 0.9021361152508693, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.9015521639500194, "precision": 0.884083778512677, "recall": 0.9197247706422018, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9057061100824777, "eval_loss": 0.8515612483024597, "eval_macro avg": { "f1-score": 0.8515519611679376, "precision": 0.8551685707064328, "recall": 0.8486952344479636, "support": 29705.0 }, "eval_runtime": 4.8633, "eval_samples_per_second": 16.45, "eval_steps_per_second": 2.056, "eval_weighted avg": { "f1-score": 0.9038774285280207, "precision": 0.9029069908176629, "recall": 0.9057061100824777, "support": 29705.0 }, "step": 3888 }, { "epoch": 49.0, "eval_B-Claim": { "f1-score": 0.6935779816513761, "precision": 0.6897810218978102, "recall": 0.6974169741697417, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9148936170212766, "precision": 0.9020979020979021, "recall": 0.9280575539568345, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8832406671961874, "precision": 0.8881789137380192, "recall": 0.8783570300157978, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6624122527121888, "precision": 0.6768388106416275, "recall": 0.6485878530367408, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.906686626746507, "precision": 0.9107769423558897, "recall": 0.9026328862394436, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8990681191757448, "precision": 0.8918496658276192, "recall": 0.9064043754410727, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9046625147281603, "eval_loss": 0.8405541181564331, "eval_macro avg": { "f1-score": 0.8514113235004687, "precision": 0.8513604652226955, "recall": 0.8516366675513759, "support": 29705.0 }, "eval_runtime": 4.9166, "eval_samples_per_second": 16.272, "eval_steps_per_second": 2.034, "eval_weighted avg": { "f1-score": 0.904007035510759, "precision": 0.9034833662980484, "recall": 0.9046625147281603, "support": 29705.0 }, "step": 3969 }, { "epoch": 49.38, "grad_norm": 0.0013613604241982102, "learning_rate": 2.469135802469136e-07, "loss": 0.0004, "step": 4000 }, { "epoch": 50.0, "eval_B-Claim": { "f1-score": 0.6886446886446888, "precision": 0.6836363636363636, "recall": 0.6937269372693727, "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.9116607773851589, "precision": 0.8958333333333334, "recall": 0.9280575539568345, "support": 139.0 }, "eval_B-Premise": { "f1-score": 0.8814638027048528, "precision": 0.8878205128205128, "recall": 0.8751974723538705, "support": 633.0 }, "eval_I-Claim": { "f1-score": 0.6650541056651814, "precision": 0.677737415672029, "recall": 0.6528367908022994, "support": 4001.0 }, "eval_I-MajorClaim": { "f1-score": 0.9078226208271052, "precision": 0.9105447276361819, "recall": 0.905116741182315, "support": 2013.0 }, "eval_I-Premise": { "f1-score": 0.8993035784678727, "precision": 0.8930839495432796, "recall": 0.9056104446012703, "support": 11336.0 }, "eval_O": { "f1-score": 1.0, "precision": 1.0, "recall": 1.0, "support": 11312.0 }, "eval_accuracy": 0.9049991583908433, "eval_loss": 0.8387188911437988, "eval_macro avg": { "f1-score": 0.8505642248135514, "precision": 0.8498080432345286, "recall": 0.8515065628808518, "support": 29705.0 }, "eval_runtime": 4.8522, "eval_samples_per_second": 16.487, "eval_steps_per_second": 2.061, "eval_weighted avg": { "f1-score": 0.9044317092676175, "precision": 0.903966681075651, "recall": 0.9049991583908433, "support": 29705.0 }, "step": 4050 } ], "logging_steps": 500, "max_steps": 4050, "num_input_tokens_seen": 0, "num_train_epochs": 50, "save_steps": 500, "total_flos": 7189540635900000.0, "train_batch_size": 4, "trial_name": null, "trial_params": null }