{ "best_metric": null, "best_model_checkpoint": null, "epoch": 20.0, "eval_steps": 500, "global_step": 1620, "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 } ], "logging_steps": 500, "max_steps": 4050, "num_input_tokens_seen": 0, "num_train_epochs": 50, "save_steps": 500, "total_flos": 2875816254360000.0, "train_batch_size": 4, "trial_name": null, "trial_params": null }