diff --git "a/checkpoint-2673/trainer_state.json" "b/checkpoint-2673/trainer_state.json" --- "a/checkpoint-2673/trainer_state.json" +++ "b/checkpoint-2673/trainer_state.json" @@ -14,2111 +14,2111 @@ "f1-score": 0.0, "precision": 0.0, "recall": 0.0, - "support": 339.0 + "support": 271.0 }, "eval_B-MajorClaim": { "f1-score": 0.0, "precision": 0.0, "recall": 0.0, - "support": 160.0 + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.6898215147129765, - "precision": 0.6316254416961131, - "recall": 0.7598299681190224, - "support": 941.0 + "f1-score": 0.6746310611384398, + "precision": 0.6075949367088608, + "recall": 0.7582938388625592, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.5039833312905995, - "precision": 0.5940479630164692, - "recall": 0.43763303533418474, - "support": 4698.0 + "f1-score": 0.4376506024096386, + "precision": 0.550587343690792, + "recall": 0.3631592101974506, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.6970899470899471, - "precision": 0.6303827751196173, - "recall": 0.7795857988165681, - "support": 2028.0 + "f1-score": 0.620034542314335, + "precision": 0.5483008781977854, + "recall": 0.7133631395926477, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8719667429344848, - "precision": 0.8366728509585653, - "recall": 0.9103694233227912, - "support": 14861.0 + "f1-score": 0.8620734063103671, + "precision": 0.8396186972154862, + "recall": 0.8857621736062103, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.8820403825717322, - "precision": 0.8925603675823639, - "recall": 0.8717654922180846, - "support": 10473.0 + "f1-score": 0.8807729058286443, + "precision": 0.8630878069080317, + "recall": 0.8991979189247779, + "support": 9226.0 }, - "eval_accuracy": 0.8062985074626866, - "eval_loss": 0.5531178712844849, + "eval_accuracy": 0.785908251565951, + "eval_loss": 0.5655931830406189, "eval_macro avg": { - "f1-score": 0.5207002740856772, - "precision": 0.5121841997675898, - "recall": 0.5370262454015216, - "support": 33500.0 - }, - "eval_runtime": 5.0026, - "eval_samples_per_second": 16.192, - "eval_steps_per_second": 2.199, + "f1-score": 0.49645178828591785, + "precision": 0.48702709467442235, + "recall": 0.5171108973119495, + "support": 27619.0 + }, + "eval_runtime": 4.8252, + "eval_samples_per_second": 16.58, + "eval_steps_per_second": 2.072, "eval_weighted avg": { - "f1-score": 0.7948191296080505, - "precision": 0.789408749451204, - "recall": 0.8062985074626866, - "support": 33500.0 + "f1-score": 0.7721020318885458, + "precision": 0.766573839857488, + "recall": 0.785908251565951, + "support": 27619.0 }, "step": 81 }, { "epoch": 2.0, "eval_B-Claim": { - "f1-score": 0.4537205081669691, - "precision": 0.589622641509434, - "recall": 0.3687315634218289, - "support": 339.0 + "f1-score": 0.36036036036036034, + "precision": 0.4624277456647399, + "recall": 0.2952029520295203, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.42794759825327516, - "precision": 0.7101449275362319, - "recall": 0.30625, - "support": 160.0 + "f1-score": 0.5836909871244635, + "precision": 0.723404255319149, + "recall": 0.4892086330935252, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.7642352941176471, - "precision": 0.6858108108108109, - "recall": 0.8629117959617428, - "support": 941.0 + "f1-score": 0.7396491228070177, + "precision": 0.6654040404040404, + "recall": 0.8325434439178515, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.44524236983842014, - "precision": 0.6338969720802202, - "recall": 0.3431247339293316, - "support": 4698.0 + "f1-score": 0.46704216775764956, + "precision": 0.5973520249221184, + "recall": 0.3834041489627593, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.7363764327367787, - "precision": 0.6217317487266554, - "recall": 0.9028599605522682, - "support": 2028.0 + "f1-score": 0.7326456649010316, + "precision": 0.8348157560355781, + "recall": 0.6527570789865872, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8781162543828609, - "precision": 0.8411808209047208, - "recall": 0.9184442500504677, - "support": 14861.0 + "f1-score": 0.8679505932065659, + "precision": 0.8045344983428744, + "recall": 0.9422194777699365, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9036260459748006, - "precision": 0.9102800116267804, - "recall": 0.8970686527260575, - "support": 10473.0 + "f1-score": 0.9101698606271778, + "precision": 0.9143513454386348, + "recall": 0.9060264469976155, + "support": 9226.0 }, - "eval_accuracy": 0.820089552238806, - "eval_loss": 0.5071194171905518, + "eval_accuracy": 0.8169376154096818, + "eval_loss": 0.4831336438655853, "eval_macro avg": { - "f1-score": 0.6584663576386788, - "precision": 0.7132382761706934, - "recall": 0.6570558509488137, - "support": 33500.0 - }, - "eval_runtime": 4.9923, - "eval_samples_per_second": 16.225, - "eval_steps_per_second": 2.203, + "f1-score": 0.6659298223977524, + "precision": 0.7146128094467336, + "recall": 0.6430517402511137, + "support": 27619.0 + }, + "eval_runtime": 4.8125, + "eval_samples_per_second": 16.623, + "eval_steps_per_second": 2.078, "eval_weighted avg": { - "f1-score": 0.8071614798424686, - "precision": 0.8128932519203979, - "recall": 0.820089552238806, - "support": 33500.0 + "f1-score": 0.8047632099274016, + "precision": 0.8064582361050344, + "recall": 0.8169376154096818, + "support": 27619.0 }, "step": 162 }, { "epoch": 3.0, "eval_B-Claim": { - "f1-score": 0.5954198473282443, - "precision": 0.6170886075949367, - "recall": 0.5752212389380531, - "support": 339.0 + "f1-score": 0.5806451612903225, + "precision": 0.59765625, + "recall": 0.5645756457564576, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.7962382445141065, - "precision": 0.7987421383647799, - "recall": 0.79375, - "support": 160.0 + "f1-score": 0.7138047138047138, + "precision": 0.6708860759493671, + "recall": 0.762589928057554, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.7837431008529856, - "precision": 0.7423954372623575, - "recall": 0.8299681190223167, - "support": 941.0 + "f1-score": 0.7739984882842026, + "precision": 0.7420289855072464, + "recall": 0.8088467614533965, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6291688447464715, - "precision": 0.6182453256626258, - "recall": 0.640485312899106, - "support": 4698.0 + "f1-score": 0.6128912532475566, + "precision": 0.6068103870651641, + "recall": 0.6190952261934516, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8422354104846687, - "precision": 0.8447420634920635, - "recall": 0.8397435897435898, - "support": 2028.0 + "f1-score": 0.8072942336126171, + "precision": 0.8009779951100244, + "recall": 0.8137108792846498, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8773398679463618, - "precision": 0.8876110460712072, - "recall": 0.8673036807751834, - "support": 14861.0 + "f1-score": 0.8797559224694903, + "precision": 0.8951789627465303, + "recall": 0.8648553281580804, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9065678167474573, - "precision": 0.9024505629671681, - "recall": 0.910722811037907, - "support": 10473.0 + "f1-score": 0.9226235130211126, + "precision": 0.912356930902925, + "recall": 0.933123780619987, + "support": 9226.0 }, - "eval_accuracy": 0.843044776119403, - "eval_loss": 0.46074992418289185, + "eval_accuracy": 0.8435859372171332, + "eval_loss": 0.4283505082130432, "eval_macro avg": { - "f1-score": 0.7758161618028993, - "precision": 0.773039311630734, - "recall": 0.7795992503451652, - "support": 33500.0 - }, - "eval_runtime": 5.0136, - "eval_samples_per_second": 16.156, - "eval_steps_per_second": 2.194, + "f1-score": 0.7558590408185736, + "precision": 0.7465565124687511, + "recall": 0.7666853642176539, + "support": 27619.0 + }, + "eval_runtime": 4.8554, + "eval_samples_per_second": 16.476, + "eval_steps_per_second": 2.06, "eval_weighted avg": { - "f1-score": 0.8436796678417321, - "precision": 0.8446385621767526, - "recall": 0.843044776119403, - "support": 33500.0 + "f1-score": 0.8439412578936614, + "precision": 0.8447189682878564, + "recall": 0.8435859372171332, + "support": 27619.0 }, "step": 243 }, { "epoch": 4.0, "eval_B-Claim": { - "f1-score": 0.6172839506172839, - "precision": 0.6472491909385113, - "recall": 0.5899705014749262, - "support": 339.0 + "f1-score": 0.6033333333333333, + "precision": 0.5501519756838906, + "recall": 0.6678966789667896, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.7441860465116279, - "precision": 0.6956521739130435, - "recall": 0.8, - "support": 160.0 + "f1-score": 0.6847457627118644, + "precision": 0.6474358974358975, + "recall": 0.7266187050359713, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.7957208354559349, - "precision": 0.764187866927593, - "recall": 0.8299681190223167, - "support": 941.0 + "f1-score": 0.7818471337579618, + "precision": 0.7881219903691814, + "recall": 0.7756714060031595, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6001342582233162, - "precision": 0.6325471698113208, - "recall": 0.5708812260536399, - "support": 4698.0 + "f1-score": 0.6216657593903103, + "precision": 0.550733024691358, + "recall": 0.7135716070982254, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8068448195030473, - "precision": 0.7689901697944593, - "recall": 0.8486193293885601, - "support": 2028.0 + "f1-score": 0.7652218782249742, + "precision": 0.7960279119699409, + "recall": 0.7367113760556383, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.882309200297438, - "precision": 0.886383701188455, - "recall": 0.8782719870802772, - "support": 14861.0 + "f1-score": 0.8739165400768475, + "precision": 0.8853883758826725, + "recall": 0.8627381792519407, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9024700070571631, - "precision": 0.8895381190873678, - "recall": 0.9157834431395016, - "support": 10473.0 + "f1-score": 0.9062570845613239, + "precision": 0.9497505345687812, + "recall": 0.8665727292434424, + "support": 9226.0 }, - "eval_accuracy": 0.8404477611940299, - "eval_loss": 0.5207056403160095, + "eval_accuracy": 0.8286324631594192, + "eval_loss": 0.4826153814792633, "eval_macro avg": { - "f1-score": 0.7641355882379729, - "precision": 0.7549354845229643, - "recall": 0.776213515165603, - "support": 33500.0 - }, - "eval_runtime": 4.9749, - "eval_samples_per_second": 16.282, - "eval_steps_per_second": 2.211, + "f1-score": 0.7481410702938023, + "precision": 0.7382299586573888, + "recall": 0.7642543830935953, + "support": 27619.0 + }, + "eval_runtime": 4.8516, + "eval_samples_per_second": 16.49, + "eval_steps_per_second": 2.061, "eval_weighted avg": { - "f1-score": 0.8386979039247872, - "precision": 0.8379021545974175, - "recall": 0.8404477611940299, - "support": 33500.0 + "f1-score": 0.8345383371838407, + "precision": 0.8451795530099128, + "recall": 0.8286324631594192, + "support": 27619.0 }, "step": 324 }, { "epoch": 5.0, "eval_B-Claim": { - "f1-score": 0.6189735614307931, - "precision": 0.6546052631578947, - "recall": 0.5870206489675516, - "support": 339.0 + "f1-score": 0.5825242718446603, + "precision": 0.6147540983606558, + "recall": 0.5535055350553506, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8090614886731391, - "precision": 0.8389261744966443, - "recall": 0.78125, - "support": 160.0 + "f1-score": 0.7317073170731707, + "precision": 0.7094594594594594, + "recall": 0.7553956834532374, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.799186578546009, - "precision": 0.7660818713450293, - "recall": 0.8352816153028693, - "support": 941.0 + "f1-score": 0.7840735068912711, + "precision": 0.7607726597325408, + "recall": 0.8088467614533965, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6316756632594464, - "precision": 0.6859565976552756, - "recall": 0.5853554704129417, - "support": 4698.0 + "f1-score": 0.5830055074744296, + "precision": 0.6132413793103448, + "recall": 0.5556110972256936, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8593117408906882, - "precision": 0.8825363825363826, - "recall": 0.8372781065088757, - "support": 2028.0 + "f1-score": 0.803728638011393, + "precision": 0.8393726338561385, + "recall": 0.7709885742672627, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8880430502690643, - "precision": 0.8666026256804354, - "recall": 0.9105712939909831, - "support": 14861.0 + "f1-score": 0.8823829787234042, + "precision": 0.8523512002630713, + "recall": 0.9146083274523642, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9125369999045164, - "precision": 0.9125369999045164, - "recall": 0.9125369999045164, - "support": 10473.0 + "f1-score": 0.9254767941792524, + "precision": 0.9415657245401525, + "recall": 0.9099284630392369, + "support": 9226.0 }, - "eval_accuracy": 0.8551343283582089, - "eval_loss": 0.5152634382247925, + "eval_accuracy": 0.8438031789709982, + "eval_loss": 0.513404905796051, "eval_macro avg": { - "f1-score": 0.7883984404248082, - "precision": 0.8010351306823111, - "recall": 0.7784705907268197, - "support": 33500.0 - }, - "eval_runtime": 5.0012, - "eval_samples_per_second": 16.196, - "eval_steps_per_second": 2.199, + "f1-score": 0.7561284305996544, + "precision": 0.7616453079317661, + "recall": 0.7526977774209346, + "support": 27619.0 + }, + "eval_runtime": 4.8184, + "eval_samples_per_second": 16.603, + "eval_steps_per_second": 2.075, "eval_weighted avg": { - "f1-score": 0.8524125579848445, - "precision": 0.8514928929215189, - "recall": 0.8551343283582089, - "support": 33500.0 + "f1-score": 0.8417228378374912, + "precision": 0.8414191958613282, + "recall": 0.8438031789709982, + "support": 27619.0 }, "step": 405 }, { "epoch": 6.0, "eval_B-Claim": { - "f1-score": 0.6404341926729986, - "precision": 0.592964824120603, - "recall": 0.696165191740413, - "support": 339.0 + "f1-score": 0.5851851851851851, + "precision": 0.587360594795539, + "recall": 0.5830258302583026, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8, - "precision": 0.7878787878787878, - "recall": 0.8125, - "support": 160.0 + "f1-score": 0.7460815047021944, + "precision": 0.6611111111111111, + "recall": 0.8561151079136691, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.7957446808510639, - "precision": 0.7965921192758253, - "recall": 0.7948990435706695, - "support": 941.0 + "f1-score": 0.7778631821675633, + "precision": 0.7574850299401198, + "recall": 0.7993680884676145, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6506562412733874, - "precision": 0.5781637717121588, - "recall": 0.743933588761175, - "support": 4698.0 + "f1-score": 0.5999495204442201, + "precision": 0.6059138414478715, + "recall": 0.5941014746313422, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8647342995169083, - "precision": 0.8475378787878788, - "recall": 0.8826429980276134, - "support": 2028.0 + "f1-score": 0.800587227795449, + "precision": 0.7888138862102217, + "recall": 0.8127173373075013, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8677517085887408, - "precision": 0.9106099815157116, - "recall": 0.8287463831505282, - "support": 14861.0 + "f1-score": 0.8798354629791703, + "precision": 0.8729593608891977, + "recall": 0.8868207480592801, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9133676463514359, - "precision": 0.920317952694843, - "recall": 0.906521531557338, - "support": 10473.0 + "f1-score": 0.9269283557507548, + "precision": 0.9391478473690066, + "recall": 0.9150227617602428, + "support": 9226.0 }, - "eval_accuracy": 0.8420597014925373, - "eval_loss": 0.5665280818939209, + "eval_accuracy": 0.843296281545313, + "eval_loss": 0.5017187595367432, "eval_macro avg": { - "f1-score": 0.790384109893505, - "precision": 0.7762950451408297, - "recall": 0.8093441052582481, - "support": 33500.0 - }, - "eval_runtime": 5.0172, - "eval_samples_per_second": 16.144, - "eval_steps_per_second": 2.192, + "f1-score": 0.7594900627177908, + "precision": 0.744684524537581, + "recall": 0.7781673354854218, + "support": 27619.0 + }, + "eval_runtime": 4.8298, + "eval_samples_per_second": 16.564, + "eval_steps_per_second": 2.07, "eval_weighted avg": { - "f1-score": 0.8467380483132992, - "precision": 0.8562016100668557, - "recall": 0.8420597014925373, - "support": 33500.0 + "f1-score": 0.8433438519855446, + "precision": 0.8437360576786502, + "recall": 0.843296281545313, + "support": 27619.0 }, "step": 486 }, { "epoch": 6.17, - "grad_norm": 17.83205223083496, + "grad_norm": 7.375185489654541, "learning_rate": 1.7530864197530865e-05, - "loss": 0.3838, + "loss": 0.4085, "step": 500 }, { "epoch": 7.0, "eval_B-Claim": { - "f1-score": 0.6363636363636364, - "precision": 0.6542056074766355, - "recall": 0.6194690265486725, - "support": 339.0 + "f1-score": 0.5725338491295938, + "precision": 0.6016260162601627, + "recall": 0.5461254612546126, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8023598820058997, - "precision": 0.7597765363128491, - "recall": 0.85, - "support": 160.0 + "f1-score": 0.745644599303136, + "precision": 0.722972972972973, + "recall": 0.7697841726618705, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8014077425842132, - "precision": 0.7604961832061069, - "recall": 0.8469713071200851, - "support": 941.0 + "f1-score": 0.7915087187263078, + "precision": 0.760932944606414, + "recall": 0.8246445497630331, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6504799548277809, - "precision": 0.6928073129660813, - "recall": 0.6130268199233716, - "support": 4698.0 + "f1-score": 0.5842240562846706, + "precision": 0.6368731563421829, + "recall": 0.5396150962259435, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8663025409641415, - "precision": 0.8355474118185983, - "recall": 0.8994082840236687, - "support": 2028.0 + "f1-score": 0.7921810699588476, + "precision": 0.8213333333333334, + "recall": 0.7650273224043715, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8922586313494517, - "precision": 0.8638941398865785, - "recall": 0.9225489536370365, - "support": 14861.0 + "f1-score": 0.8817056396148556, + "precision": 0.8598256203890007, + "recall": 0.9047282992237121, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.90784071234232, - "precision": 0.9419010470129336, - "recall": 0.8761577389477705, - "support": 10473.0 + "f1-score": 0.9274176179194485, + "precision": 0.9214637277979885, + "recall": 0.9334489486234554, + "support": 9226.0 }, - "eval_accuracy": 0.8577014925373134, - "eval_loss": 0.6021790504455566, + "eval_accuracy": 0.8452152503711213, + "eval_loss": 0.6029371023178101, "eval_macro avg": { - "f1-score": 0.7938590143482062, - "precision": 0.7869468912399691, - "recall": 0.8039403043143721, - "support": 33500.0 - }, - "eval_runtime": 5.0234, - "eval_samples_per_second": 16.125, - "eval_steps_per_second": 2.19, + "f1-score": 0.7564593644195513, + "precision": 0.7607182531002935, + "recall": 0.7547676928795714, + "support": 27619.0 + }, + "eval_runtime": 4.8043, + "eval_samples_per_second": 16.652, + "eval_steps_per_second": 2.081, "eval_weighted avg": { - "f1-score": 0.856081111084223, - "precision": 0.8570485445028432, - "recall": 0.8577014925373134, - "support": 33500.0 + "f1-score": 0.8415705604110203, + "precision": 0.8398235103191511, + "recall": 0.8452152503711213, + "support": 27619.0 }, "step": 567 }, { "epoch": 8.0, "eval_B-Claim": { - "f1-score": 0.6373333333333333, - "precision": 0.5815085158150851, - "recall": 0.7050147492625368, - "support": 339.0 + "f1-score": 0.5641891891891891, + "precision": 0.5202492211838006, + "recall": 0.6162361623616236, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.7952522255192879, - "precision": 0.7570621468926554, - "recall": 0.8375, - "support": 160.0 + "f1-score": 0.7318840579710144, + "precision": 0.7372262773722628, + "recall": 0.7266187050359713, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.7883053600433134, - "precision": 0.8035320088300221, - "recall": 0.7736450584484591, - "support": 941.0 + "f1-score": 0.7673716012084594, + "precision": 0.7351664254703328, + "recall": 0.8025276461295419, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6543385490753911, - "precision": 0.5900461775269369, - "recall": 0.7343550446998723, - "support": 4698.0 + "f1-score": 0.5930426495115559, + "precision": 0.5665832005463237, + "recall": 0.6220944763809048, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8537494177922683, - "precision": 0.8089143865842895, - "recall": 0.9038461538461539, - "support": 2028.0 + "f1-score": 0.7816030123722432, + "precision": 0.852199413489736, + "recall": 0.7218082463984103, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8704377010208362, - "precision": 0.9058429746052535, - "recall": 0.8376959827737029, - "support": 14861.0 + "f1-score": 0.8746564286025915, + "precision": 0.8652567975830816, + "recall": 0.8842625264643613, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9111186539300781, - "precision": 0.925615763546798, - "recall": 0.8970686527260575, - "support": 10473.0 + "f1-score": 0.924221395658691, + "precision": 0.9473085239558439, + "recall": 0.9022328202904835, + "support": 9226.0 }, - "eval_accuracy": 0.8426268656716418, - "eval_loss": 0.6685047149658203, + "eval_accuracy": 0.835149715775372, + "eval_loss": 0.6411188840866089, "eval_macro avg": { - "f1-score": 0.7872193201020725, - "precision": 0.7675031391144344, - "recall": 0.8127322345366831, - "support": 33500.0 - }, - "eval_runtime": 4.992, - "eval_samples_per_second": 16.226, - "eval_steps_per_second": 2.204, + "f1-score": 0.7481383335019636, + "precision": 0.7462842656573401, + "recall": 0.7536829404373281, + "support": 27619.0 + }, + "eval_runtime": 4.8296, + "eval_samples_per_second": 16.564, + "eval_steps_per_second": 2.071, "eval_weighted avg": { - "f1-score": 0.8468148764239732, - "precision": 0.8550031822663273, - "recall": 0.8426268656716418, - "support": 33500.0 + "f1-score": 0.8374117728187233, + "precision": 0.8414359188593861, + "recall": 0.835149715775372, + "support": 27619.0 }, "step": 648 }, { "epoch": 9.0, "eval_B-Claim": { - "f1-score": 0.5807407407407408, - "precision": 0.5833333333333334, - "recall": 0.5781710914454278, - "support": 339.0 + "f1-score": 0.6241610738255032, + "precision": 0.5723076923076923, + "recall": 0.6863468634686347, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8012232415902141, - "precision": 0.7844311377245509, - "recall": 0.81875, - "support": 160.0 + "f1-score": 0.756578947368421, + "precision": 0.696969696969697, + "recall": 0.8273381294964028, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.7984111221449851, - "precision": 0.7493010251630942, - "recall": 0.8544102019128587, - "support": 941.0 + "f1-score": 0.780448717948718, + "precision": 0.791869918699187, + "recall": 0.7693522906793049, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.5917442137914578, - "precision": 0.6731813246471227, - "recall": 0.5278842060451255, - "support": 4698.0 + "f1-score": 0.640754369825207, + "precision": 0.5933972310969116, + "recall": 0.69632591852037, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8581943081452406, - "precision": 0.85400390625, - "recall": 0.8624260355029586, - "support": 2028.0 + "f1-score": 0.797244094488189, + "precision": 0.7898586055582643, + "recall": 0.8047690014903129, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8840042543591067, - "precision": 0.84832364221205, - "recall": 0.922818114527959, - "support": 14861.0 + "f1-score": 0.8789423814465549, + "precision": 0.8996028447400019, + "recall": 0.8592095977417078, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.910873701156154, - "precision": 0.931178934769599, - "recall": 0.8914351188771126, - "support": 10473.0 + "f1-score": 0.9287169042769858, + "precision": 0.9435186220780674, + "recall": 0.9143724257533059, + "support": 9226.0 }, - "eval_accuracy": 0.8480597014925373, - "eval_loss": 0.7230669260025024, + "eval_accuracy": 0.8461566313045368, + "eval_loss": 0.6282346844673157, "eval_macro avg": { - "f1-score": 0.7750273688468428, - "precision": 0.7748219005856785, - "recall": 0.779413538330206, - "support": 33500.0 - }, - "eval_runtime": 5.0031, - "eval_samples_per_second": 16.19, - "eval_steps_per_second": 2.199, + "f1-score": 0.7724066413113684, + "precision": 0.7553606587785459, + "recall": 0.7939591753071484, + "support": 27619.0 + }, + "eval_runtime": 4.8069, + "eval_samples_per_second": 16.643, + "eval_steps_per_second": 2.08, "eval_weighted avg": { - "f1-score": 0.8439872109031008, - "precision": 0.8442403481354289, - "recall": 0.8480597014925373, - "support": 33500.0 + "f1-score": 0.8497367665000708, + "precision": 0.85521547416283, + "recall": 0.8461566313045368, + "support": 27619.0 }, "step": 729 }, { "epoch": 10.0, "eval_B-Claim": { - "f1-score": 0.6050198150594451, - "precision": 0.5478468899521531, - "recall": 0.6755162241887905, - "support": 339.0 + "f1-score": 0.5714285714285714, + "precision": 0.5602836879432624, + "recall": 0.5830258302583026, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8170731707317073, - "precision": 0.7976190476190477, - "recall": 0.8375, - "support": 160.0 + "f1-score": 0.7407407407407407, + "precision": 0.6962025316455697, + "recall": 0.7913669064748201, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.7826535880227156, - "precision": 0.7610441767068273, - "recall": 0.8055260361317748, - "support": 941.0 + "f1-score": 0.789156626506024, + "precision": 0.753956834532374, + "recall": 0.8278041074249605, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6194059405940595, - "precision": 0.5790447982228805, - "recall": 0.6658152405278842, - "support": 4698.0 + "f1-score": 0.6017810109118275, + "precision": 0.6039778449144008, + "recall": 0.5996000999750063, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8604878048780488, - "precision": 0.8513513513513513, - "recall": 0.8698224852071006, - "support": 2028.0 + "f1-score": 0.7822841502861408, + "precision": 0.7836490528414756, + "recall": 0.7809239940387481, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8745383703201763, - "precision": 0.8807151630954005, - "recall": 0.8684476145616042, - "support": 14861.0 + "f1-score": 0.8826781594183972, + "precision": 0.871336484744306, + "recall": 0.894318983768525, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9032226225139417, - "precision": 0.9347293156281921, - "recall": 0.8737706483338108, - "support": 10473.0 + "f1-score": 0.9204840581311821, + "precision": 0.9389020403562169, + "recall": 0.9027747669629308, + "support": 9226.0 }, - "eval_accuracy": 0.837910447761194, - "eval_loss": 0.7846710085868835, + "eval_accuracy": 0.8410876570476845, + "eval_loss": 0.7145851254463196, "eval_macro avg": { - "f1-score": 0.7803430445885849, - "precision": 0.7646215346536931, - "recall": 0.7994854641358522, - "support": 33500.0 - }, - "eval_runtime": 4.9942, - "eval_samples_per_second": 16.219, - "eval_steps_per_second": 2.203, + "f1-score": 0.7555076167746976, + "precision": 0.744044068139658, + "recall": 0.7685449555576133, + "support": 27619.0 + }, + "eval_runtime": 4.8288, + "eval_samples_per_second": 16.567, + "eval_steps_per_second": 2.071, "eval_weighted avg": { - "f1-score": 0.8412929574648306, - "precision": 0.8463911306722504, - "recall": 0.837910447761194, - "support": 33500.0 + "f1-score": 0.841386205332122, + "precision": 0.842161020451525, + "recall": 0.8410876570476845, + "support": 27619.0 }, "step": 810 }, { "epoch": 11.0, "eval_B-Claim": { - "f1-score": 0.6121842496285289, - "precision": 0.6167664670658682, - "recall": 0.6076696165191741, - "support": 339.0 + "f1-score": 0.6018348623853211, + "precision": 0.5985401459854015, + "recall": 0.6051660516605166, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8012048192771085, - "precision": 0.7732558139534884, - "recall": 0.83125, - "support": 160.0 + "f1-score": 0.7272727272727274, + "precision": 0.768, + "recall": 0.6906474820143885, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8081841432225064, - "precision": 0.7790927021696252, - "recall": 0.8395324123273114, - "support": 941.0 + "f1-score": 0.7833733013589128, + "precision": 0.7928802588996764, + "recall": 0.7740916271721959, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6122952658739061, - "precision": 0.6472960151802657, - "recall": 0.5808854831843338, - "support": 4698.0 + "f1-score": 0.6083602886290123, + "precision": 0.6056477582363141, + "recall": 0.6110972256935766, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8465227817745803, - "precision": 0.8239962651727357, - "recall": 0.8703155818540433, - "support": 2028.0 + "f1-score": 0.7648026315789475, + "precision": 0.8532110091743119, + "recall": 0.6929955290611028, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8846026818151793, - "precision": 0.8686514886164624, - "recall": 0.9011506628086939, - "support": 14861.0 + "f1-score": 0.882720702345586, + "precision": 0.8874030489435678, + "recall": 0.8780875088214538, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9093722797175742, - "precision": 0.9213130818226359, - "recall": 0.8977370380979662, - "support": 10473.0 + "f1-score": 0.921590369079677, + "precision": 0.8984865643982292, + "recall": 0.9459137220897463, + "support": 9226.0 }, - "eval_accuracy": 0.8482686567164179, - "eval_loss": 0.7823007106781006, + "eval_accuracy": 0.8425721423657627, + "eval_loss": 0.7725272178649902, "eval_macro avg": { - "f1-score": 0.7820523173299119, - "precision": 0.7757674048544404, - "recall": 0.7897915421130747, - "support": 33500.0 - }, - "eval_runtime": 4.9917, - "eval_samples_per_second": 16.227, - "eval_steps_per_second": 2.204, + "f1-score": 0.7557078403785977, + "precision": 0.7720241122339289, + "recall": 0.7425713066447115, + "support": 27619.0 + }, + "eval_runtime": 4.7985, + "eval_samples_per_second": 16.672, + "eval_steps_per_second": 2.084, "eval_weighted avg": { - "f1-score": 0.8465514092815755, - "precision": 0.845848769512072, - "recall": 0.8482686567164179, - "support": 33500.0 + "f1-score": 0.8415502116531838, + "precision": 0.8421955187218679, + "recall": 0.8425721423657627, + "support": 27619.0 }, "step": 891 }, { "epoch": 12.0, "eval_B-Claim": { - "f1-score": 0.6290801186943621, - "precision": 0.6328358208955224, - "recall": 0.6253687315634219, - "support": 339.0 + "f1-score": 0.5907473309608542, + "precision": 0.570446735395189, + "recall": 0.6125461254612546, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.7954545454545454, - "precision": 0.7291666666666666, - "recall": 0.875, - "support": 160.0 + "f1-score": 0.7328767123287672, + "precision": 0.6993464052287581, + "recall": 0.7697841726618705, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8107549120992762, - "precision": 0.7895266868076536, - "recall": 0.8331562167906482, - "support": 941.0 + "f1-score": 0.7870370370370371, + "precision": 0.7692307692307693, + "recall": 0.8056872037914692, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6450192625206385, - "precision": 0.667882379758377, - "recall": 0.6236696466581524, - "support": 4698.0 + "f1-score": 0.6039273805113003, + "precision": 0.596923828125, + "recall": 0.6110972256935766, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8621825023518345, - "precision": 0.8241906474820144, - "recall": 0.9038461538461539, - "support": 2028.0 + "f1-score": 0.7876195269250126, + "precision": 0.7980622131565528, + "recall": 0.7774465971187282, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.890260173448966, - "precision": 0.8826641973675508, - "recall": 0.8979880223403539, - "support": 14861.0 + "f1-score": 0.8808608098180718, + "precision": 0.8633629817873782, + "recall": 0.8990825688073395, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9123196448390677, - "precision": 0.9222439024390244, - "recall": 0.902606702950444, - "support": 10473.0 + "f1-score": 0.9129559185500112, + "precision": 0.9433526011560693, + "recall": 0.8844569694342077, + "support": 9226.0 }, - "eval_accuracy": 0.8566268656716418, - "eval_loss": 0.8441444635391235, + "eval_accuracy": 0.8380100655345958, + "eval_loss": 0.8826749920845032, "eval_macro avg": { - "f1-score": 0.7921530227726701, - "precision": 0.7783586144881156, - "recall": 0.8088050677355964, - "support": 33500.0 - }, - "eval_runtime": 4.9988, - "eval_samples_per_second": 16.204, - "eval_steps_per_second": 2.201, + "f1-score": 0.7565749594472935, + "precision": 0.7486750762971024, + "recall": 0.765728694709778, + "support": 27619.0 + }, + "eval_runtime": 4.8258, + "eval_samples_per_second": 16.577, + "eval_steps_per_second": 2.072, "eval_weighted avg": { - "f1-score": 0.8557354611316912, - "precision": 0.8554998717241106, - "recall": 0.8566268656716418, - "support": 33500.0 + "f1-score": 0.8389267473809466, + "precision": 0.8408692792556519, + "recall": 0.8380100655345958, + "support": 27619.0 }, "step": 972 }, { "epoch": 12.35, - "grad_norm": 18.23494529724121, + "grad_norm": 6.464661598205566, "learning_rate": 1.506172839506173e-05, - "loss": 0.0667, + "loss": 0.0768, "step": 1000 }, { "epoch": 13.0, "eval_B-Claim": { - "f1-score": 0.6439716312056738, - "precision": 0.6202185792349727, - "recall": 0.6696165191740413, - "support": 339.0 + "f1-score": 0.6059544658493871, + "precision": 0.5766666666666667, + "recall": 0.6383763837638377, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8048048048048049, - "precision": 0.7745664739884393, - "recall": 0.8375, - "support": 160.0 + "f1-score": 0.75, + "precision": 0.6909090909090909, + "recall": 0.8201438848920863, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8029273392577103, - "precision": 0.7901234567901234, - "recall": 0.8161530286928799, - "support": 941.0 + "f1-score": 0.7835703001579779, + "precision": 0.7835703001579779, + "recall": 0.7835703001579779, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6348773841961852, - "precision": 0.6253096614368291, - "recall": 0.6447424435930184, - "support": 4698.0 + "f1-score": 0.6174981923355025, + "precision": 0.5962299278566442, + "recall": 0.6403399150212447, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8476758045292015, - "precision": 0.8204891555145363, - "recall": 0.8767258382642998, - "support": 2028.0 + "f1-score": 0.7900677200902934, + "precision": 0.7978723404255319, + "recall": 0.7824143070044709, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8825089822369967, - "precision": 0.8807640750670241, - "recall": 0.8842608169033039, - "support": 14861.0 + "f1-score": 0.8801810597319606, + "precision": 0.8856045722450437, + "recall": 0.8748235709244884, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9050703813745069, - "precision": 0.9237422946907934, - "recall": 0.8871383557719851, - "support": 10473.0 + "f1-score": 0.9129007550060182, + "precision": 0.9216747680070703, + "recall": 0.9042922176457836, + "support": 9226.0 }, - "eval_accuracy": 0.8468059701492537, - "eval_loss": 0.8703527450561523, + "eval_accuracy": 0.8392773090988088, + "eval_loss": 0.8458374738693237, "eval_macro avg": { - "f1-score": 0.7888337610864399, - "precision": 0.7764590995318168, - "recall": 0.8023052860570755, - "support": 33500.0 - }, - "eval_runtime": 5.0165, - "eval_samples_per_second": 16.147, - "eval_steps_per_second": 2.193, + "f1-score": 0.7628817847387342, + "precision": 0.7503610951811466, + "recall": 0.7777086542014128, + "support": 27619.0 + }, + "eval_runtime": 4.8414, + "eval_samples_per_second": 16.524, + "eval_steps_per_second": 2.065, "eval_weighted avg": { - "f1-score": 0.8477056223787199, - "precision": 0.8490368858423804, - "recall": 0.8468059701492537, - "support": 33500.0 + "f1-score": 0.8409296175505384, + "precision": 0.8429896387826172, + "recall": 0.8392773090988088, + "support": 27619.0 }, "step": 1053 }, { "epoch": 14.0, "eval_B-Claim": { - "f1-score": 0.6222222222222222, - "precision": 0.5879265091863517, - "recall": 0.6607669616519174, - "support": 339.0 + "f1-score": 0.5909090909090908, + "precision": 0.5614617940199336, + "recall": 0.6236162361623616, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8253012048192772, - "precision": 0.7965116279069767, - "recall": 0.85625, - "support": 160.0 + "f1-score": 0.7424749163879598, + "precision": 0.69375, + "recall": 0.7985611510791367, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.7978552278820376, - "precision": 0.8051948051948052, - "recall": 0.7906482465462275, - "support": 941.0 + "f1-score": 0.7811764705882354, + "precision": 0.7757009345794392, + "recall": 0.7867298578199052, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6353440436495907, - "precision": 0.604731679169071, - "recall": 0.669220945083014, - "support": 4698.0 + "f1-score": 0.6013087736306351, + "precision": 0.5836273817925194, + "recall": 0.620094976255936, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8582976699952449, - "precision": 0.8287419651056015, - "recall": 0.8900394477317555, - "support": 2028.0 + "f1-score": 0.7706057596822246, + "precision": 0.7702233250620347, + "recall": 0.7709885742672627, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8795296048133461, - "precision": 0.8938920158432354, - "recall": 0.8656214252069174, - "support": 14861.0 + "f1-score": 0.8808171531721921, + "precision": 0.8792300254900237, + "recall": 0.882410021171489, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9139328444615978, - "precision": 0.9236470014627011, - "recall": 0.9044208918170534, - "support": 10473.0 + "f1-score": 0.9170672412840489, + "precision": 0.9353093654908148, + "recall": 0.8995230869282462, + "support": 9226.0 }, - "eval_accuracy": 0.8474626865671642, - "eval_loss": 0.8902716040611267, + "eval_accuracy": 0.8368514428473153, + "eval_loss": 0.9143030047416687, "eval_macro avg": { - "f1-score": 0.7903546882633309, - "precision": 0.7772350862669632, - "recall": 0.8052811311481264, - "support": 33500.0 - }, - "eval_runtime": 5.0016, - "eval_samples_per_second": 16.195, - "eval_steps_per_second": 2.199, + "f1-score": 0.7549084865220552, + "precision": 0.7427575466335379, + "recall": 0.7688462719549054, + "support": 27619.0 + }, + "eval_runtime": 4.8212, + "eval_samples_per_second": 16.594, + "eval_steps_per_second": 2.074, "eval_weighted avg": { - "f1-score": 0.8495983708816418, - "precision": 0.8526459607911416, - "recall": 0.8474626865671642, - "support": 33500.0 + "f1-score": 0.838578387953211, + "precision": 0.8407717357848979, + "recall": 0.8368514428473153, + "support": 27619.0 }, "step": 1134 }, { "epoch": 15.0, "eval_B-Claim": { - "f1-score": 0.6417910447761194, - "precision": 0.649546827794562, - "recall": 0.6342182890855457, - "support": 339.0 + "f1-score": 0.5709090909090909, + "precision": 0.5627240143369175, + "recall": 0.5793357933579336, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8260869565217391, - "precision": 0.8209876543209876, - "recall": 0.83125, - "support": 160.0 + "f1-score": 0.7446808510638296, + "precision": 0.7342657342657343, + "recall": 0.7553956834532374, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8086680761099365, - "precision": 0.804416403785489, - "recall": 0.8129649309245484, - "support": 941.0 + "f1-score": 0.7841269841269841, + "precision": 0.7878787878787878, + "recall": 0.7804107424960506, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6283666377063423, - "precision": 0.6414634146341464, - "recall": 0.6157939548744147, - "support": 4698.0 + "f1-score": 0.5929925907321362, + "precision": 0.5959111559818273, + "recall": 0.5901024743814046, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8459692898272553, - "precision": 0.8238317757009346, - "recall": 0.8693293885601578, - "support": 2028.0 + "f1-score": 0.7486910994764399, + "precision": 0.7913669064748201, + "recall": 0.7103825136612022, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.886482474783254, - "precision": 0.888851305641997, - "recall": 0.8841262364578427, - "support": 14861.0 + "f1-score": 0.882824477638092, + "precision": 0.8767945706081963, + "recall": 0.8889378969654199, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9091340000948002, - "precision": 0.9026731927710844, - "recall": 0.9156879595149432, - "support": 10473.0 + "f1-score": 0.9183122909247868, + "precision": 0.9142672969488612, + "recall": 0.9223932365055278, + "support": 9226.0 }, - "eval_accuracy": 0.8506865671641791, - "eval_loss": 0.9465591907501221, + "eval_accuracy": 0.8376117889858431, + "eval_loss": 0.9049480557441711, "eval_macro avg": { - "f1-score": 0.7923569256884925, - "precision": 0.7902529392356001, - "recall": 0.7947672513453503, - "support": 33500.0 - }, - "eval_runtime": 5.0023, - "eval_samples_per_second": 16.192, - "eval_steps_per_second": 2.199, + "f1-score": 0.74893391212448, + "precision": 0.7518869237850208, + "recall": 0.746708334402968, + "support": 27619.0 + }, + "eval_runtime": 4.8265, + "eval_samples_per_second": 16.575, + "eval_steps_per_second": 2.072, "eval_weighted avg": { - "f1-score": 0.8499631749642652, - "precision": 0.849425426847339, - "recall": 0.8506865671641791, - "support": 33500.0 + "f1-score": 0.8368987072750449, + "precision": 0.8365590307236803, + "recall": 0.8376117889858431, + "support": 27619.0 }, "step": 1215 }, { "epoch": 16.0, "eval_B-Claim": { - "f1-score": 0.6467065868263473, - "precision": 0.6565349544072948, - "recall": 0.6371681415929203, - "support": 339.0 + "f1-score": 0.5880077369439072, + "precision": 0.6178861788617886, + "recall": 0.5608856088560885, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8227848101265822, - "precision": 0.8333333333333334, - "recall": 0.8125, - "support": 160.0 + "f1-score": 0.7475409836065575, + "precision": 0.6867469879518072, + "recall": 0.8201438848920863, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8090005232862376, - "precision": 0.7969072164948454, - "recall": 0.8214665249734325, - "support": 941.0 + "f1-score": 0.7921259842519685, + "precision": 0.7896389324960753, + "recall": 0.7946287519747235, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6444299106169611, - "precision": 0.6604110813226095, - "recall": 0.6292039165602384, - "support": 4698.0 + "f1-score": 0.5849176980987624, + "precision": 0.5974973931178311, + "recall": 0.5728567858035492, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8560079443892751, - "precision": 0.862, - "recall": 0.8500986193293886, - "support": 2028.0 + "f1-score": 0.7757869249394673, + "precision": 0.7567312234293812, + "recall": 0.7958271236959762, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8901776005799202, - "precision": 0.8721590909090909, - "recall": 0.9089563286454478, - "support": 14861.0 + "f1-score": 0.8798468592797044, + "precision": 0.8881100026961445, + "recall": 0.8717360621030346, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9152476403619733, - "precision": 0.9330423569090368, - "recall": 0.8981189725961998, - "support": 10473.0 + "f1-score": 0.92327420388972, + "precision": 0.9104320337197049, + "recall": 0.9364838499891611, + "support": 9226.0 }, - "eval_accuracy": 0.8571044776119403, - "eval_loss": 0.9341827034950256, + "eval_accuracy": 0.8394583438936963, + "eval_loss": 0.9981638789176941, "eval_macro avg": { - "f1-score": 0.797765002312471, - "precision": 0.8020554333394587, - "recall": 0.7939303576710898, - "support": 33500.0 - }, - "eval_runtime": 4.9904, - "eval_samples_per_second": 16.231, - "eval_steps_per_second": 2.204, + "f1-score": 0.7559286272871554, + "precision": 0.7495775360389619, + "recall": 0.7646517239020884, + "support": 27619.0 + }, + "eval_runtime": 4.8136, + "eval_samples_per_second": 16.62, + "eval_steps_per_second": 2.077, "eval_weighted avg": { - "f1-score": 0.8564173816715166, - "precision": 0.856401327986896, - "recall": 0.8571044776119403, - "support": 33500.0 + "f1-score": 0.8385046755256232, + "precision": 0.8379700346267476, + "recall": 0.8394583438936963, + "support": 27619.0 }, "step": 1296 }, { "epoch": 17.0, "eval_B-Claim": { - "f1-score": 0.6514935988620199, - "precision": 0.6291208791208791, - "recall": 0.6755162241887905, - "support": 339.0 + "f1-score": 0.5753424657534246, + "precision": 0.6125, + "recall": 0.5424354243542435, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8395061728395061, - "precision": 0.8292682926829268, - "recall": 0.85, - "support": 160.0 + "f1-score": 0.7160493827160496, + "precision": 0.6270270270270271, + "recall": 0.8345323741007195, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8132436627004656, - "precision": 0.7923387096774194, - "recall": 0.8352816153028693, - "support": 941.0 + "f1-score": 0.7974882260596545, + "precision": 0.7925117004680188, + "recall": 0.8025276461295419, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6634417455743105, - "precision": 0.6422877640494221, - "recall": 0.6860366113239676, - "support": 4698.0 + "f1-score": 0.5781887408213218, + "precision": 0.6340590515955861, + "recall": 0.5313671582104474, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8681209717402082, - "precision": 0.8728813559322034, - "recall": 0.8634122287968442, - "support": 2028.0 + "f1-score": 0.7680890538033396, + "precision": 0.7203131796433232, + "recall": 0.8226527570789866, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8910224018864746, - "precision": 0.8921343766864543, - "recall": 0.8899131956126775, - "support": 14861.0 + "f1-score": 0.8883316761056591, + "precision": 0.8710470538363714, + "recall": 0.9063161609033169, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9171560300897841, - "precision": 0.9325898144492697, - "recall": 0.9022247684522104, - "support": 10473.0 + "f1-score": 0.9186122627100153, + "precision": 0.9246650560070283, + "recall": 0.9126381964014741, + "support": 9226.0 }, - "eval_accuracy": 0.8596716417910448, - "eval_loss": 0.9319809079170227, + "eval_accuracy": 0.8417031753503024, + "eval_loss": 1.0078351497650146, "eval_macro avg": { - "f1-score": 0.8062835119561098, - "precision": 0.7986601703712249, - "recall": 0.8146263776681942, - "support": 33500.0 - }, - "eval_runtime": 4.9891, - "eval_samples_per_second": 16.236, - "eval_steps_per_second": 2.205, + "f1-score": 0.7488716868527806, + "precision": 0.7403032955110508, + "recall": 0.7646385310255328, + "support": 27619.0 + }, + "eval_runtime": 4.8211, + "eval_samples_per_second": 16.594, + "eval_steps_per_second": 2.074, "eval_weighted avg": { - "f1-score": 0.861035739499467, - "precision": 0.8628131055121488, - "recall": 0.8596716417910448, - "support": 33500.0 + "f1-score": 0.838734195768893, + "precision": 0.8380757474281803, + "recall": 0.8417031753503024, + "support": 27619.0 }, "step": 1377 }, { "epoch": 18.0, "eval_B-Claim": { - "f1-score": 0.6266866566716641, - "precision": 0.6371951219512195, - "recall": 0.616519174041298, - "support": 339.0 + "f1-score": 0.5923076923076923, + "precision": 0.6184738955823293, + "recall": 0.5682656826568265, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8246153846153846, - "precision": 0.8121212121212121, - "recall": 0.8375, - "support": 160.0 + "f1-score": 0.745644599303136, + "precision": 0.722972972972973, + "recall": 0.7697841726618705, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8112033195020746, - "precision": 0.7922998986828774, - "recall": 0.8310308182784272, - "support": 941.0 + "f1-score": 0.7912087912087912, + "precision": 0.7862714508580343, + "recall": 0.7962085308056872, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6293847360753111, - "precision": 0.6646153846153846, - "recall": 0.5977011494252874, - "support": 4698.0 + "f1-score": 0.6042421107087429, + "precision": 0.6261056017153578, + "recall": 0.5838540364908773, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8671295187933924, - "precision": 0.842717543043276, - "recall": 0.8929980276134122, - "support": 2028.0 + "f1-score": 0.7803540264273249, + "precision": 0.7832832832832833, + "recall": 0.7774465971187282, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8906353189803197, - "precision": 0.8744083511638462, - "recall": 0.9074759437453738, - "support": 14861.0 + "f1-score": 0.8878702702702702, + "precision": 0.8708117736873356, + "recall": 0.9056104446012703, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9144762272902976, - "precision": 0.9256578303824708, - "recall": 0.9035615391960279, - "support": 10473.0 + "f1-score": 0.9169446115151184, + "precision": 0.925190334326382, + "recall": 0.9088445696943421, + "support": 9226.0 }, - "eval_accuracy": 0.8565074626865672, - "eval_loss": 0.9800853729248047, + "eval_accuracy": 0.8442376624787284, + "eval_loss": 0.9712997674942017, "eval_macro avg": { - "f1-score": 0.794875880275492, - "precision": 0.792716477422898, - "recall": 0.7981123788999752, - "support": 33500.0 - }, - "eval_runtime": 5.0389, - "eval_samples_per_second": 16.075, - "eval_steps_per_second": 2.183, + "f1-score": 0.7597960145344393, + "precision": 0.7618727589179565, + "recall": 0.7585734334328001, + "support": 27619.0 + }, + "eval_runtime": 4.8169, + "eval_samples_per_second": 16.608, + "eval_steps_per_second": 2.076, "eval_weighted avg": { - "f1-score": 0.8548106483514697, - "precision": 0.8540863869670992, - "recall": 0.8565074626865672, - "support": 33500.0 + "f1-score": 0.8428273600970289, + "precision": 0.8419904974860213, + "recall": 0.8442376624787284, + "support": 27619.0 }, "step": 1458 }, { "epoch": 18.52, - "grad_norm": 5.693321704864502, + "grad_norm": 7.254913806915283, "learning_rate": 1.2592592592592593e-05, - "loss": 0.0243, + "loss": 0.0261, "step": 1500 }, { "epoch": 19.0, "eval_B-Claim": { - "f1-score": 0.6505681818181819, - "precision": 0.6273972602739726, - "recall": 0.6755162241887905, - "support": 339.0 + "f1-score": 0.588235294117647, + "precision": 0.5689655172413793, + "recall": 0.6088560885608856, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8301886792452831, - "precision": 0.8354430379746836, - "recall": 0.825, - "support": 160.0 + "f1-score": 0.7224080267558529, + "precision": 0.675, + "recall": 0.7769784172661871, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8123044838373306, - "precision": 0.797338792221085, - "recall": 0.8278427205100957, - "support": 941.0 + "f1-score": 0.7804107424960506, + "precision": 0.7804107424960506, + "recall": 0.7804107424960506, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6452893252559013, - "precision": 0.6335110746513536, - "recall": 0.6575138356747552, - "support": 4698.0 + "f1-score": 0.6187416656564432, + "precision": 0.60075329566855, + "recall": 0.6378405398650338, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8693540553666828, - "precision": 0.8564593301435407, - "recall": 0.8826429980276134, - "support": 2028.0 + "f1-score": 0.7764764515325194, + "precision": 0.779, + "recall": 0.7739692001987084, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.885918725529904, - "precision": 0.8887384031963161, - "recall": 0.8831168831168831, - "support": 14861.0 + "f1-score": 0.8802663706992231, + "precision": 0.886048797926535, + "recall": 0.8745589273112209, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9220829315332689, - "precision": 0.93133339826629, - "recall": 0.9130144180273083, - "support": 10473.0 + "f1-score": 0.9251841746248295, + "precision": 0.931640839652709, + "recall": 0.9188163884673748, + "support": 9226.0 }, - "eval_accuracy": 0.8568656716417911, - "eval_loss": 1.0031511783599854, + "eval_accuracy": 0.8424635214888302, + "eval_loss": 0.9876435995101929, "eval_macro avg": { - "f1-score": 0.8022437689409362, - "precision": 0.795745899532463, - "recall": 0.8092352970779207, - "support": 33500.0 - }, - "eval_runtime": 4.9883, - "eval_samples_per_second": 16.238, - "eval_steps_per_second": 2.205, + "f1-score": 0.7559603894117952, + "precision": 0.7459741704264606, + "recall": 0.7673471863093516, + "support": 27619.0 + }, + "eval_runtime": 4.8824, + "eval_samples_per_second": 16.385, + "eval_steps_per_second": 2.048, "eval_weighted avg": { - "f1-score": 0.8577606960594096, - "precision": 0.8588413433126966, - "recall": 0.8568656716417911, - "support": 33500.0 + "f1-score": 0.8438722529775301, + "precision": 0.8455528133195762, + "recall": 0.8424635214888302, + "support": 27619.0 }, "step": 1539 }, { "epoch": 20.0, "eval_B-Claim": { - "f1-score": 0.5980707395498392, - "precision": 0.657243816254417, - "recall": 0.5486725663716814, - "support": 339.0 + "f1-score": 0.6126760563380281, + "precision": 0.5858585858585859, + "recall": 0.6420664206642066, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8275862068965517, - "precision": 0.8301886792452831, - "recall": 0.825, - "support": 160.0 + "f1-score": 0.7375415282392027, + "precision": 0.6851851851851852, + "recall": 0.7985611510791367, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8045634920634921, - "precision": 0.7544186046511628, - "recall": 0.8618490967056323, - "support": 941.0 + "f1-score": 0.7827476038338658, + "precision": 0.7915993537964459, + "recall": 0.7740916271721959, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6099424139146786, - "precision": 0.6809236420886906, - "recall": 0.5523627075351213, - "support": 4698.0 + "f1-score": 0.6214689265536723, + "precision": 0.6243693239152371, + "recall": 0.6185953511622094, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8622423328305682, - "precision": 0.8794871794871795, - "recall": 0.8456607495069034, - "support": 2028.0 + "f1-score": 0.7863746958637469, + "precision": 0.7706247019551741, + "recall": 0.8027819175360159, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8888383132295087, - "precision": 0.8599018621036739, - "recall": 0.9197900545050804, - "support": 14861.0 + "f1-score": 0.8849448123620308, + "precision": 0.8858051971009369, + "recall": 0.8840860973888497, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9156209433145824, - "precision": 0.9221382916908774, - "recall": 0.9091950730449728, - "support": 10473.0 + "f1-score": 0.9158329708568943, + "precision": 0.9188304603971198, + "recall": 0.9128549750704531, + "support": 9226.0 }, - "eval_accuracy": 0.8546268656716418, - "eval_loss": 1.0620654821395874, + "eval_accuracy": 0.8439842137658858, + "eval_loss": 1.0289872884750366, "eval_macro avg": { - "f1-score": 0.7866949202570316, - "precision": 0.7977574393601834, - "recall": 0.7803614639527703, - "support": 33500.0 - }, - "eval_runtime": 5.0619, - "eval_samples_per_second": 16.002, - "eval_steps_per_second": 2.173, + "f1-score": 0.7630837991496344, + "precision": 0.7517532583155264, + "recall": 0.7761482200104383, + "support": 27619.0 + }, + "eval_runtime": 4.857, + "eval_samples_per_second": 16.471, + "eval_steps_per_second": 2.059, "eval_weighted avg": { - "f1-score": 0.8508870221979551, - "precision": 0.8502889230298556, - "recall": 0.8546268656716418, - "support": 33500.0 + "f1-score": 0.8441548041890129, + "precision": 0.844457711736205, + "recall": 0.8439842137658858, + "support": 27619.0 }, "step": 1620 }, { "epoch": 21.0, "eval_B-Claim": { - "f1-score": 0.6402266288951841, - "precision": 0.6158038147138964, - "recall": 0.6666666666666666, - "support": 339.0 + "f1-score": 0.6, + "precision": 0.5913978494623656, + "recall": 0.6088560885608856, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8109756097560975, - "precision": 0.7916666666666666, - "recall": 0.83125, - "support": 160.0 + "f1-score": 0.754325259515571, + "precision": 0.7266666666666667, + "recall": 0.7841726618705036, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8087373468300479, - "precision": 0.8108974358974359, - "recall": 0.8065887353878852, - "support": 941.0 + "f1-score": 0.7965435978004713, + "precision": 0.7921875, + "recall": 0.8009478672985783, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6453424245462748, - "precision": 0.6100473933649289, - "recall": 0.6849723286504895, - "support": 4698.0 + "f1-score": 0.6180417495029821, + "precision": 0.614529280948851, + "recall": 0.6215946013496626, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8580842551188516, - "precision": 0.8208014407924359, - "recall": 0.8989151873767258, - "support": 2028.0 + "f1-score": 0.7913561847988076, + "precision": 0.7913561847988078, + "recall": 0.7913561847988078, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8847649587227945, - "precision": 0.8972531654327821, - "recall": 0.8726196083709037, - "support": 14861.0 + "f1-score": 0.8852830847073192, + "precision": 0.8755823986194996, + "recall": 0.8952011291460833, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9148056244830438, - "precision": 0.9326388888888889, - "recall": 0.8976415544734078, - "support": 10473.0 + "f1-score": 0.9224318658280921, + "precision": 0.9393258426966292, + "recall": 0.9061348363321049, + "support": 9226.0 }, - "eval_accuracy": 0.8515820895522388, - "eval_loss": 1.061930775642395, + "eval_accuracy": 0.8461204243455592, + "eval_loss": 1.0996745824813843, "eval_macro avg": { - "f1-score": 0.7947052640503277, - "precision": 0.782729829393862, - "recall": 0.8083791544180112, - "support": 33500.0 - }, - "eval_runtime": 5.0268, - "eval_samples_per_second": 16.114, - "eval_steps_per_second": 2.188, + "f1-score": 0.7668545345933204, + "precision": 0.7615779604561171, + "recall": 0.7726090527652323, + "support": 27619.0 + }, + "eval_runtime": 4.8412, + "eval_samples_per_second": 16.525, + "eval_steps_per_second": 2.066, "eval_weighted avg": { - "f1-score": 0.8540023792481587, - "precision": 0.8576320297856588, - "recall": 0.8515820895522388, - "support": 33500.0 + "f1-score": 0.8466408555926697, + "precision": 0.8474703689459093, + "recall": 0.8461204243455592, + "support": 27619.0 }, "step": 1701 }, { "epoch": 22.0, "eval_B-Claim": { - "f1-score": 0.6100151745068285, - "precision": 0.628125, - "recall": 0.5929203539823009, - "support": 339.0 + "f1-score": 0.5984251968503936, + "precision": 0.6413502109704642, + "recall": 0.5608856088560885, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.7964601769911505, - "precision": 0.7541899441340782, - "recall": 0.84375, - "support": 160.0 + "f1-score": 0.7266666666666667, + "precision": 0.6770186335403726, + "recall": 0.7841726618705036, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8071428571428572, - "precision": 0.7762512266928361, - "recall": 0.8405951115834219, - "support": 941.0 + "f1-score": 0.7909162098668755, + "precision": 0.7841614906832298, + "recall": 0.7977883096366508, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6242526790750141, - "precision": 0.664026877849772, - "recall": 0.5889740315027672, - "support": 4698.0 + "f1-score": 0.6076327583865272, + "precision": 0.665377751338489, + "recall": 0.5591102224443889, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8485707536026459, - "precision": 0.8145124716553288, - "recall": 0.8856015779092702, - "support": 2028.0 + "f1-score": 0.7673170731707317, + "precision": 0.753713464302827, + "recall": 0.7814207650273224, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8863839211926877, - "precision": 0.8711500974658869, - "recall": 0.9021600161496535, - "support": 14861.0 + "f1-score": 0.8918528587205845, + "precision": 0.8600688186137965, + "recall": 0.9260762173606211, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9082298361764848, - "precision": 0.9194716242661448, - "recall": 0.8972596199751742, - "support": 10473.0 + "f1-score": 0.9196605686577033, + "precision": 0.9353284017036539, + "recall": 0.9045089963147627, + "support": 9226.0 }, - "eval_accuracy": 0.8505671641791045, - "eval_loss": 1.0729957818984985, + "eval_accuracy": 0.847930772294435, + "eval_loss": 1.1058261394500732, "eval_macro avg": { - "f1-score": 0.7830079140982384, - "precision": 0.7753896060091495, - "recall": 0.7930372444432268, - "support": 33500.0 - }, - "eval_runtime": 5.0353, - "eval_samples_per_second": 16.086, - "eval_steps_per_second": 2.185, + "f1-score": 0.757495904617069, + "precision": 0.7595741101646905, + "recall": 0.7591375402157626, + "support": 27619.0 + }, + "eval_runtime": 4.8079, + "eval_samples_per_second": 16.639, + "eval_steps_per_second": 2.08, "eval_weighted avg": { - "f1-score": 0.8487114418737095, - "precision": 0.8480977508780456, - "recall": 0.8505671641791045, - "support": 33500.0 + "f1-score": 0.8448680329971596, + "precision": 0.8444464777692308, + "recall": 0.847930772294435, + "support": 27619.0 }, "step": 1782 }, { "epoch": 23.0, "eval_B-Claim": { - "f1-score": 0.6436781609195402, - "precision": 0.6274509803921569, - "recall": 0.6607669616519174, - "support": 339.0 + "f1-score": 0.6037099494097807, + "precision": 0.5559006211180124, + "recall": 0.6605166051660517, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.7938461538461539, - "precision": 0.7818181818181819, - "recall": 0.80625, - "support": 160.0 + "f1-score": 0.7397260273972602, + "precision": 0.7058823529411765, + "recall": 0.7769784172661871, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8104166666666667, - "precision": 0.7946884576098059, - "recall": 0.8267800212539851, - "support": 941.0 + "f1-score": 0.7720706260032103, + "precision": 0.7846655791190864, + "recall": 0.7598736176935229, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6443532259762131, - "precision": 0.6373100145742244, - "recall": 0.651553852703278, - "support": 4698.0 + "f1-score": 0.6147824084734056, + "precision": 0.5699039487726788, + "recall": 0.6673331667083229, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8590571907154823, - "precision": 0.8344955834495583, - "recall": 0.8851084812623274, - "support": 2028.0 + "f1-score": 0.7702471250305848, + "precision": 0.7589199614271939, + "recall": 0.7819175360158966, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8891257995735607, - "precision": 0.8805014846585285, - "recall": 0.8979207321176234, - "support": 14861.0 + "f1-score": 0.8730755460078767, + "precision": 0.8860828488372093, + "recall": 0.8604446012702893, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.908608751166331, - "precision": 0.9353892821031344, - "recall": 0.8833190107896496, - "support": 10473.0 + "f1-score": 0.9182879377431906, + "precision": 0.9424920127795527, + "recall": 0.8952959028831563, + "support": 9226.0 }, - "eval_accuracy": 0.8531940298507462, - "eval_loss": 1.0359262228012085, + "eval_accuracy": 0.8337014374162715, + "eval_loss": 1.1438162326812744, "eval_macro avg": { - "f1-score": 0.7927265641234211, - "precision": 0.7845219978007986, - "recall": 0.8016712942541115, - "support": 33500.0 - }, - "eval_runtime": 5.0006, - "eval_samples_per_second": 16.198, - "eval_steps_per_second": 2.2, + "f1-score": 0.75598566000933, + "precision": 0.7434067607135587, + "recall": 0.771765692429061, + "support": 27619.0 + }, + "eval_runtime": 4.8472, + "eval_samples_per_second": 16.504, + "eval_steps_per_second": 2.063, "eval_weighted avg": { - "f1-score": 0.853920052577175, - "precision": 0.8553284369175107, - "recall": 0.8531940298507462, - "support": 33500.0 + "f1-score": 0.8376374525692796, + "precision": 0.8433838333277589, + "recall": 0.8337014374162715, + "support": 27619.0 }, "step": 1863 }, { "epoch": 24.0, "eval_B-Claim": { - "f1-score": 0.6414565826330533, - "precision": 0.6106666666666667, - "recall": 0.6755162241887905, - "support": 339.0 + "f1-score": 0.6189624329159212, + "precision": 0.6006944444444444, + "recall": 0.6383763837638377, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8206686930091186, - "precision": 0.7988165680473372, - "recall": 0.84375, - "support": 160.0 + "f1-score": 0.7397260273972602, + "precision": 0.7058823529411765, + "recall": 0.7769784172661871, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8077935755660874, - "precision": 0.8006263048016702, - "recall": 0.8150903294367694, - "support": 941.0 + "f1-score": 0.7880690737833596, + "precision": 0.7831513260530422, + "recall": 0.7930489731437599, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6328502415458938, - "precision": 0.6245854063018242, - "recall": 0.6413367390378885, - "support": 4698.0 + "f1-score": 0.6222925285957654, + "precision": 0.6063552288356652, + "recall": 0.6390902274431393, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8545850471812243, - "precision": 0.8389548693586698, - "recall": 0.8708086785009862, - "support": 2028.0 + "f1-score": 0.7763733592610598, + "precision": 0.7601142313184198, + "recall": 0.7933432687531048, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8874008927976548, - "precision": 0.878735897605067, - "recall": 0.8962384765493574, - "support": 14861.0 + "f1-score": 0.8828359526752604, + "precision": 0.8836161187698833, + "recall": 0.8820571630204658, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9130242825607064, - "precision": 0.9388619854721549, - "recall": 0.888570610140361, - "support": 10473.0 + "f1-score": 0.9172044790115285, + "precision": 0.9338425249915758, + "recall": 0.9011489269455886, + "support": 9226.0 }, - "eval_accuracy": 0.8517910447761194, - "eval_loss": 1.0978286266326904, + "eval_accuracy": 0.8418117962272349, + "eval_loss": 1.1221617460250854, "eval_macro avg": { - "f1-score": 0.7939684736133913, - "precision": 0.7844639568933414, - "recall": 0.8044730082648791, - "support": 33500.0 - }, - "eval_runtime": 5.0476, - "eval_samples_per_second": 16.047, - "eval_steps_per_second": 2.179, + "f1-score": 0.7636376933771649, + "precision": 0.753379461050601, + "recall": 0.7748633371908691, + "support": 27619.0 + }, + "eval_runtime": 4.8078, + "eval_samples_per_second": 16.64, + "eval_steps_per_second": 2.08, "eval_weighted avg": { - "f1-score": 0.8526833258684211, - "precision": 0.8541944616635844, - "recall": 0.8517910447761194, - "support": 33500.0 + "f1-score": 0.8433323915719101, + "precision": 0.8452544012114124, + "recall": 0.8418117962272349, + "support": 27619.0 }, "step": 1944 }, { "epoch": 24.69, - "grad_norm": 0.4236430525779724, + "grad_norm": 0.008052258752286434, "learning_rate": 1.0123456790123458e-05, - "loss": 0.0113, + "loss": 0.0118, "step": 2000 }, { "epoch": 25.0, "eval_B-Claim": { - "f1-score": 0.6367583212735166, - "precision": 0.625, - "recall": 0.6489675516224189, - "support": 339.0 + "f1-score": 0.6019417475728155, + "precision": 0.6352459016393442, + "recall": 0.5719557195571956, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8260869565217391, - "precision": 0.8209876543209876, - "recall": 0.83125, - "support": 160.0 + "f1-score": 0.7574750830564784, + "precision": 0.7037037037037037, + "recall": 0.8201438848920863, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8119614799797263, - "precision": 0.7761627906976745, - "recall": 0.8512221041445271, - "support": 941.0 + "f1-score": 0.7897196261682242, + "precision": 0.7788018433179723, + "recall": 0.8009478672985783, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6425783428125348, - "precision": 0.6748653080346685, - "recall": 0.6132396764580673, - "support": 4698.0 + "f1-score": 0.6033311125916057, + "precision": 0.6461187214611872, + "recall": 0.5658585353661585, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.859186189889026, - "precision": 0.8593981253083375, - "recall": 0.8589743589743589, - "support": 2028.0 + "f1-score": 0.7857317371121426, + "precision": 0.7730769230769231, + "recall": 0.7988077496274217, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8904676552061268, - "precision": 0.8633720930232558, - "recall": 0.9193190229459659, - "support": 14861.0 + "f1-score": 0.8852726492056677, + "precision": 0.8623891584406893, + "recall": 0.9094036697247706, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9072733540158565, - "precision": 0.9367500508440105, - "recall": 0.8795951494318724, - "support": 10473.0 + "f1-score": 0.9221917808219178, + "precision": 0.9325132978723404, + "recall": 0.9120962497290267, + "support": 9226.0 }, - "eval_accuracy": 0.8552537313432835, - "eval_loss": 1.0907410383224487, + "eval_accuracy": 0.8462290452224918, + "eval_loss": 1.1378806829452515, "eval_macro avg": { - "f1-score": 0.7963303285283609, - "precision": 0.7937908603184193, - "recall": 0.8003668376538873, - "support": 33500.0 - }, - "eval_runtime": 5.0312, - "eval_samples_per_second": 16.1, - "eval_steps_per_second": 2.186, + "f1-score": 0.7636662480755503, + "precision": 0.7616927927874515, + "recall": 0.7684590965993198, + "support": 27619.0 + }, + "eval_runtime": 4.8628, + "eval_samples_per_second": 16.452, + "eval_steps_per_second": 2.056, "eval_weighted avg": { - "f1-score": 0.8539840908916618, - "precision": 0.8545711875497904, - "recall": 0.8552537313432835, - "support": 33500.0 + "f1-score": 0.8438939014101086, + "precision": 0.8430313935317453, + "recall": 0.8462290452224918, + "support": 27619.0 }, "step": 2025 }, { "epoch": 26.0, "eval_B-Claim": { - "f1-score": 0.6366279069767441, - "precision": 0.6275071633237822, - "recall": 0.6460176991150443, - "support": 339.0 + "f1-score": 0.6040515653775324, + "precision": 0.6029411764705882, + "recall": 0.6051660516605166, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8459214501510574, - "precision": 0.8187134502923976, - "recall": 0.875, - "support": 160.0 + "f1-score": 0.7389830508474576, + "precision": 0.6987179487179487, + "recall": 0.7841726618705036, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8115485564304463, - "precision": 0.8018672199170125, - "recall": 0.8214665249734325, - "support": 941.0 + "f1-score": 0.7824037706205812, + "precision": 0.778125, + "recall": 0.7867298578199052, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6410673608851286, - "precision": 0.6536164565361645, - "recall": 0.6289910600255428, - "support": 4698.0 + "f1-score": 0.6289062499999999, + "precision": 0.614650441422095, + "recall": 0.64383904023994, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8678606001936108, - "precision": 0.8521863117870723, - "recall": 0.8841222879684418, - "support": 2028.0 + "f1-score": 0.7797139141742523, + "precision": 0.8182314410480349, + "recall": 0.7446597118728266, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8923780184835536, - "precision": 0.8787839248434238, - "recall": 0.9063993001816836, - "support": 14861.0 + "f1-score": 0.8828411672199895, + "precision": 0.8744375216337833, + "recall": 0.891407904022583, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9141020646669263, - "precision": 0.932723839809202, - "recall": 0.896209300105032, - "support": 10473.0 + "f1-score": 0.9202110121991427, + "precision": 0.9332367365135978, + "recall": 0.9075438976804683, + "support": 9226.0 }, - "eval_accuracy": 0.8577910447761194, - "eval_loss": 1.082922101020813, + "eval_accuracy": 0.844491111191571, + "eval_loss": 1.116464376449585, "eval_macro avg": { - "f1-score": 0.8013579939696381, - "precision": 0.7950569095012936, - "recall": 0.8083151674813109, - "support": 33500.0 - }, - "eval_runtime": 5.0123, - "eval_samples_per_second": 16.16, - "eval_steps_per_second": 2.195, + "f1-score": 0.762444390062708, + "precision": 0.760048609400864, + "recall": 0.7662170178809634, + "support": 27619.0 + }, + "eval_runtime": 4.8569, + "eval_samples_per_second": 16.471, + "eval_steps_per_second": 2.059, "eval_weighted avg": { - "f1-score": 0.8573614298108096, - "precision": 0.8574693960861622, - "recall": 0.8577910447761194, - "support": 33500.0 + "f1-score": 0.8452605268375321, + "precision": 0.8465930758524871, + "recall": 0.844491111191571, + "support": 27619.0 }, "step": 2106 }, { "epoch": 27.0, "eval_B-Claim": { - "f1-score": 0.6215022091310751, - "precision": 0.6205882352941177, - "recall": 0.6224188790560472, - "support": 339.0 + "f1-score": 0.5996409335727109, + "precision": 0.583916083916084, + "recall": 0.6162361623616236, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.84375, - "precision": 0.84375, - "recall": 0.84375, - "support": 160.0 + "f1-score": 0.7414965986394558, + "precision": 0.7032258064516129, + "recall": 0.7841726618705036, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8073394495412843, - "precision": 0.7757100881488737, - "recall": 0.8416578108395324, - "support": 941.0 + "f1-score": 0.7847113884555382, + "precision": 0.7750385208012327, + "recall": 0.7946287519747235, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6341463414634146, - "precision": 0.6431698774080561, - "recall": 0.6253724989357173, - "support": 4698.0 + "f1-score": 0.6112161310649024, + "precision": 0.6164209456024403, + "recall": 0.6060984753811547, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.872308462694041, - "precision": 0.8860630722278738, - "recall": 0.8589743589743589, - "support": 2028.0 + "f1-score": 0.785261102977062, + "precision": 0.7717026378896883, + "recall": 0.799304520615996, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8874136659290551, - "precision": 0.8843301035750084, - "recall": 0.8905188076172532, - "support": 14861.0 + "f1-score": 0.8810789780832576, + "precision": 0.8638636941595321, + "recall": 0.8989943542695836, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9142366248269939, - "precision": 0.9139312977099237, - "recall": 0.9145421560202426, - "support": 10473.0 + "f1-score": 0.9194492446624672, + "precision": 0.9465166991851257, + "recall": 0.893886841534793, + "support": 9226.0 }, - "eval_accuracy": 0.8546268656716418, - "eval_loss": 1.1081665754318237, + "eval_accuracy": 0.8418480031862123, + "eval_loss": 1.133851170539856, "eval_macro avg": { - "f1-score": 0.7972423933694092, - "precision": 0.7953632391948362, - "recall": 0.7996049302061646, - "support": 33500.0 - }, - "eval_runtime": 4.9873, - "eval_samples_per_second": 16.241, - "eval_steps_per_second": 2.206, + "f1-score": 0.7604077682079133, + "precision": 0.7515263411436737, + "recall": 0.7704745382869111, + "support": 27619.0 + }, + "eval_runtime": 4.8334, + "eval_samples_per_second": 16.551, + "eval_steps_per_second": 2.069, "eval_weighted avg": { - "f1-score": 0.8542182509443944, - "precision": 0.8539553059521852, - "recall": 0.8546268656716418, - "support": 33500.0 + "f1-score": 0.8421466887502327, + "precision": 0.8433197634457057, + "recall": 0.8418480031862123, + "support": 27619.0 }, "step": 2187 }, { "epoch": 28.0, "eval_B-Claim": { - "f1-score": 0.62882096069869, - "precision": 0.6206896551724138, - "recall": 0.6371681415929203, - "support": 339.0 + "f1-score": 0.608, + "precision": 0.6637554585152838, + "recall": 0.5608856088560885, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8414634146341463, - "precision": 0.8214285714285714, - "recall": 0.8625, - "support": 160.0 + "f1-score": 0.7241379310344827, + "precision": 0.695364238410596, + "recall": 0.7553956834532374, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8027571580063626, - "precision": 0.801058201058201, - "recall": 0.8044633368756642, - "support": 941.0 + "f1-score": 0.7953307392996108, + "precision": 0.7837423312883436, + "recall": 0.8072669826224329, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6291229186552126, - "precision": 0.6269287677023885, - "recall": 0.6313324819071946, - "support": 4698.0 + "f1-score": 0.6028865979381443, + "precision": 0.6698228466707391, + "recall": 0.5481129717570608, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8659150875509715, - "precision": 0.8430639887902849, - "recall": 0.8900394477317555, - "support": 2028.0 + "f1-score": 0.774845869297164, + "precision": 0.7693437806072478, + "recall": 0.7804272230501739, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8836598666935972, - "precision": 0.8841360727517683, - "recall": 0.8831841733396137, - "support": 14861.0 + "f1-score": 0.8894579598804951, + "precision": 0.8615842566561932, + "recall": 0.9191954834156669, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9132002885308969, - "precision": 0.9198798682425886, - "recall": 0.9066170151818963, - "support": 10473.0 + "f1-score": 0.923653752105635, + "precision": 0.9261196469434456, + "recall": 0.9212009538261435, + "support": 9226.0 }, - "eval_accuracy": 0.8508059701492537, - "eval_loss": 1.1174240112304688, + "eval_accuracy": 0.8490893949817155, + "eval_loss": 1.1788508892059326, "eval_macro avg": { - "f1-score": 0.7949913849671253, - "precision": 0.7881693035923166, - "recall": 0.8021863709470064, - "support": 33500.0 - }, - "eval_runtime": 5.0241, - "eval_samples_per_second": 16.122, - "eval_steps_per_second": 2.189, + "f1-score": 0.7597589785079332, + "precision": 0.7671046512988356, + "recall": 0.756069272425829, + "support": 27619.0 + }, + "eval_runtime": 4.8209, + "eval_samples_per_second": 16.595, + "eval_steps_per_second": 2.074, "eval_weighted avg": { - "f1-score": 0.8510719435168806, - "precision": 0.8514546653411631, - "recall": 0.8508059701492537, - "support": 33500.0 + "f1-score": 0.8452625545373849, + "precision": 0.8440780437434046, + "recall": 0.8490893949817155, + "support": 27619.0 }, "step": 2268 }, { "epoch": 29.0, "eval_B-Claim": { - "f1-score": 0.6284875183553598, - "precision": 0.6257309941520468, - "recall": 0.6312684365781711, - "support": 339.0 + "f1-score": 0.6103646833013436, + "precision": 0.636, + "recall": 0.5867158671586716, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.828828828828829, - "precision": 0.7976878612716763, - "recall": 0.8625, - "support": 160.0 + "f1-score": 0.7733333333333333, + "precision": 0.7204968944099379, + "recall": 0.8345323741007195, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8067226890756303, - "precision": 0.7975077881619937, - "recall": 0.8161530286928799, - "support": 941.0 + "f1-score": 0.7955449482895783, + "precision": 0.8012820512820513, + "recall": 0.7898894154818326, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6283974777125462, - "precision": 0.6422222222222222, - "recall": 0.6151553852703278, - "support": 4698.0 + "f1-score": 0.6074445617740233, + "precision": 0.6436363636363637, + "recall": 0.575106223444139, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8468342644320298, - "precision": 0.8020282186948854, - "recall": 0.8969428007889546, - "support": 2028.0 + "f1-score": 0.8131604226705091, + "precision": 0.7870757787075778, + "recall": 0.8410332836562344, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8888888888888888, - "precision": 0.8860657929927788, - "recall": 0.8917300316264046, - "support": 14861.0 + "f1-score": 0.8914829226113272, + "precision": 0.8741733084619298, + "recall": 0.9094918842625265, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9130037070916182, - "precision": 0.9207613128762867, - "recall": 0.9053757280626372, - "support": 10473.0 + "f1-score": 0.9266265718972116, + "precision": 0.9349073256840247, + "recall": 0.9184912204639063, + "support": 9226.0 }, - "eval_accuracy": 0.8526268656716418, - "eval_loss": 1.1392927169799805, + "eval_accuracy": 0.852782504797422, + "eval_loss": 1.1232454776763916, "eval_macro avg": { - "f1-score": 0.7915947677692717, - "precision": 0.7817148843388414, - "recall": 0.8027322015741964, - "support": 33500.0 - }, - "eval_runtime": 5.0048, - "eval_samples_per_second": 16.185, - "eval_steps_per_second": 2.198, + "f1-score": 0.7739939205539039, + "precision": 0.7710816745974122, + "recall": 0.7793228955097186, + "support": 27619.0 + }, + "eval_runtime": 4.7958, + "eval_samples_per_second": 16.681, + "eval_steps_per_second": 2.085, "eval_weighted avg": { - "f1-score": 0.8521209782825445, - "precision": 0.8520847134451853, - "recall": 0.8526268656716418, - "support": 33500.0 + "f1-score": 0.850815268485991, + "precision": 0.8499356548650282, + "recall": 0.852782504797422, + "support": 27619.0 }, "step": 2349 }, { "epoch": 30.0, "eval_B-Claim": { - "f1-score": 0.616519174041298, - "precision": 0.616519174041298, - "recall": 0.616519174041298, - "support": 339.0 + "f1-score": 0.5788423153692615, + "precision": 0.6304347826086957, + "recall": 0.5350553505535055, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8359133126934984, - "precision": 0.8282208588957055, - "recall": 0.84375, - "support": 160.0 + "f1-score": 0.7422680412371134, + "precision": 0.7105263157894737, + "recall": 0.7769784172661871, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.807161125319693, - "precision": 0.7781065088757396, - "recall": 0.8384697130712009, - "support": 941.0 + "f1-score": 0.7920184190330007, + "precision": 0.7701492537313432, + "recall": 0.8151658767772512, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6277404921700224, - "precision": 0.6614804337576615, - "recall": 0.5972754363558961, - "support": 4698.0 + "f1-score": 0.5994579945799458, + "precision": 0.6546315477952057, + "recall": 0.5528617845538615, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8676361843726892, - "precision": 0.8674223755544603, - "recall": 0.8678500986193294, - "support": 2028.0 + "f1-score": 0.7890781563126252, + "precision": 0.7958564931783729, + "recall": 0.7824143070044709, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8910416393011953, - "precision": 0.870181514976589, - "recall": 0.9129264517865554, - "support": 14861.0 + "f1-score": 0.8883685286103541, + "precision": 0.8585418038183015, + "recall": 0.9203422724064926, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9146880310755039, - "precision": 0.9305473226635053, - "recall": 0.8993602597154589, - "support": 10473.0 + "f1-score": 0.9204178745282503, + "precision": 0.9290051893562989, + "recall": 0.9119878603945372, + "support": 9226.0 }, - "eval_accuracy": 0.8562686567164179, - "eval_loss": 1.1244691610336304, + "eval_accuracy": 0.8473514609507947, + "eval_loss": 1.144108533859253, "eval_macro avg": { - "f1-score": 0.7943857084248428, - "precision": 0.7932111698235657, - "recall": 0.7965930190842484, - "support": 33500.0 - }, - "eval_runtime": 5.0345, - "eval_samples_per_second": 16.089, - "eval_steps_per_second": 2.185, + "f1-score": 0.7586359042386501, + "precision": 0.7641636266110988, + "recall": 0.7564008384223293, + "support": 27619.0 + }, + "eval_runtime": 4.786, + "eval_samples_per_second": 16.715, + "eval_steps_per_second": 2.089, "eval_weighted avg": { - "f1-score": 0.8546947257966622, - "precision": 0.8542648006571733, - "recall": 0.8562686567164179, - "support": 33500.0 + "f1-score": 0.8440043412194941, + "precision": 0.8429627129938011, + "recall": 0.8473514609507947, + "support": 27619.0 }, "step": 2430 }, { "epoch": 30.86, - "grad_norm": 0.8852996230125427, + "grad_norm": 0.6330540180206299, "learning_rate": 7.654320987654322e-06, - "loss": 0.0072, + "loss": 0.0077, "step": 2500 }, { "epoch": 31.0, "eval_B-Claim": { - "f1-score": 0.6339969372128637, - "precision": 0.6592356687898089, - "recall": 0.6106194690265486, - "support": 339.0 + "f1-score": 0.5895953757225434, + "precision": 0.6169354838709677, + "recall": 0.5645756457564576, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.85, - "precision": 0.85, - "recall": 0.85, - "support": 160.0 + "f1-score": 0.7357142857142858, + "precision": 0.7304964539007093, + "recall": 0.7410071942446043, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.818833162743091, - "precision": 0.7897334649555775, - "recall": 0.8501594048884166, - "support": 941.0 + "f1-score": 0.7916666666666666, + "precision": 0.7737556561085973, + "recall": 0.8104265402843602, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6385407341092211, - "precision": 0.673194903256253, - "recall": 0.60727969348659, - "support": 4698.0 + "f1-score": 0.6163372859025034, + "precision": 0.6517135692393424, + "recall": 0.5846038490377405, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8638589618021547, - "precision": 0.857976653696498, - "recall": 0.8698224852071006, - "support": 2028.0 + "f1-score": 0.7895956734483647, + "precision": 0.8197860962566845, + "recall": 0.7615499254843517, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8960573476702509, - "precision": 0.8762057877813505, - "recall": 0.9168292847049324, - "support": 14861.0 + "f1-score": 0.8899611426619412, + "precision": 0.8624513779690475, + "recall": 0.9192836979534227, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.915512062784614, - "precision": 0.9291965778345953, - "recall": 0.9022247684522104, - "support": 10473.0 + "f1-score": 0.9214837543148321, + "precision": 0.9317451523545707, + "recall": 0.9114459137220897, + "support": 9226.0 }, - "eval_accuracy": 0.8607164179104477, - "eval_loss": 1.112532138824463, + "eval_accuracy": 0.8498135341612658, + "eval_loss": 1.1577571630477905, "eval_macro avg": { - "f1-score": 0.8023998866174564, - "precision": 0.8050775794734405, - "recall": 0.8009907293951141, - "support": 33500.0 - }, - "eval_runtime": 5.0139, - "eval_samples_per_second": 16.155, - "eval_steps_per_second": 2.194, + "f1-score": 0.7620505977758768, + "precision": 0.7695548270999886, + "recall": 0.7561275380690038, + "support": 27619.0 + }, + "eval_runtime": 4.8274, + "eval_samples_per_second": 16.572, + "eval_steps_per_second": 2.071, "eval_weighted avg": { - "f1-score": 0.8590353250300645, - "precision": 0.8584488466661996, - "recall": 0.8607164179104477, - "support": 33500.0 + "f1-score": 0.8475614553244858, + "precision": 0.8468547323259233, + "recall": 0.8498135341612658, + "support": 27619.0 }, "step": 2511 }, { "epoch": 32.0, "eval_B-Claim": { - "f1-score": 0.6291793313069909, - "precision": 0.6489028213166145, - "recall": 0.6106194690265486, - "support": 339.0 + "f1-score": 0.620817843866171, + "precision": 0.6254681647940075, + "recall": 0.6162361623616236, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8571428571428572, - "precision": 0.8518518518518519, - "recall": 0.8625, - "support": 160.0 + "f1-score": 0.7516778523489932, + "precision": 0.7044025157232704, + "recall": 0.8057553956834532, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8206438426162495, - "precision": 0.7903543307086615, - "recall": 0.8533475026567482, - "support": 941.0 + "f1-score": 0.8068535825545171, + "precision": 0.7956989247311828, + "recall": 0.8183254344391785, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.634185303514377, - "precision": 0.6834727004426955, - "recall": 0.5915283099191145, - "support": 4698.0 + "f1-score": 0.6207257958862832, + "precision": 0.6522577092511013, + "recall": 0.5921019745063734, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8680641183723797, - "precision": 0.8682782437099161, - "recall": 0.8678500986193294, - "support": 2028.0 + "f1-score": 0.7961023142509135, + "precision": 0.7810707456978967, + "recall": 0.8117237953303527, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8952249408050513, - "precision": 0.875474368045282, - "recall": 0.9158872215867034, - "support": 14861.0 + "f1-score": 0.8936483212184148, + "precision": 0.876953125, + "recall": 0.9109915314043754, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9166826646189288, - "precision": 0.921547814339477, - "recall": 0.9118686145326077, - "support": 10473.0 + "f1-score": 0.9254433982920954, + "precision": 0.9348595443485954, + "recall": 0.9162150444396271, + "support": 9226.0 }, - "eval_accuracy": 0.8611343283582089, - "eval_loss": 1.1487747430801392, + "eval_accuracy": 0.853760092689815, + "eval_loss": 1.1496995687484741, "eval_macro avg": { - "f1-score": 0.8030175797681192, - "precision": 0.8056974472020713, - "recall": 0.8019430309058647, - "support": 33500.0 - }, - "eval_runtime": 5.0225, - "eval_samples_per_second": 16.127, - "eval_steps_per_second": 2.19, + "f1-score": 0.7736098726310556, + "precision": 0.7672443899351507, + "recall": 0.7816213340235691, + "support": 27619.0 + }, + "eval_runtime": 4.8513, + "eval_samples_per_second": 16.49, + "eval_steps_per_second": 2.061, "eval_weighted avg": { - "f1-score": 0.8587120145129986, - "precision": 0.8577198700280971, - "recall": 0.8611343283582089, - "support": 33500.0 + "f1-score": 0.8522422681785793, + "precision": 0.8515595976922847, + "recall": 0.853760092689815, + "support": 27619.0 }, "step": 2592 }, { "epoch": 33.0, "eval_B-Claim": { - "f1-score": 0.6415620641562065, - "precision": 0.6084656084656085, - "recall": 0.6784660766961652, - "support": 339.0 + "f1-score": 0.5996055226824456, + "precision": 0.6440677966101694, + "recall": 0.5608856088560885, + "support": 271.0 }, "eval_B-MajorClaim": { - "f1-score": 0.84472049689441, - "precision": 0.8395061728395061, - "recall": 0.85, - "support": 160.0 + "f1-score": 0.7645051194539249, + "precision": 0.7272727272727273, + "recall": 0.8057553956834532, + "support": 139.0 }, "eval_B-Premise": { - "f1-score": 0.8122041031036298, - "precision": 0.8041666666666667, - "recall": 0.820403825717322, - "support": 941.0 + "f1-score": 0.8009223674096849, + "precision": 0.7799401197604791, + "recall": 0.8230647709320695, + "support": 633.0 }, "eval_I-Claim": { - "f1-score": 0.6475124115348051, - "precision": 0.6426923883413713, - "recall": 0.6524052788420605, - "support": 4698.0 + "f1-score": 0.6167238195726722, + "precision": 0.6528902541189612, + "recall": 0.5843539115221195, + "support": 4001.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8639523336643495, - "precision": 0.87, - "recall": 0.8579881656804734, - "support": 2028.0 + "f1-score": 0.8000990834778301, + "precision": 0.7979249011857708, + "recall": 0.8022851465474417, + "support": 2013.0 }, "eval_I-Premise": { - "f1-score": 0.8912810912008557, - "precision": 0.8856554381768653, - "recall": 0.8969786689993944, - "support": 14861.0 + "f1-score": 0.8918248674533948, + "precision": 0.8653335545967474, + "recall": 0.9199894142554693, + "support": 11336.0 }, "eval_O": { - "f1-score": 0.9161865104343194, - "precision": 0.9293713163064833, - "recall": 0.9033705719469111, - "support": 10473.0 + "f1-score": 0.9253171538885825, + "precision": 0.9420485175202157, + "recall": 0.9091697376978105, + "support": 9226.0 }, - "eval_accuracy": 0.8577313432835821, - "eval_loss": 1.128581166267395, + "eval_accuracy": 0.8528549187153771, + "eval_loss": 1.1728317737579346, "eval_macro avg": { - "f1-score": 0.8024884301412252, - "precision": 0.7971225129709287, - "recall": 0.8085160839831894, - "support": 33500.0 - }, - "eval_runtime": 4.9952, - "eval_samples_per_second": 16.216, - "eval_steps_per_second": 2.202, + "f1-score": 0.7712854191340764, + "precision": 0.7727825530092959, + "recall": 0.7722148550706361, + "support": 27619.0 + }, + "eval_runtime": 4.8324, + "eval_samples_per_second": 16.555, + "eval_steps_per_second": 2.069, "eval_weighted avg": { - "f1-score": 0.8582563326104159, - "precision": 0.8589872168689389, - "recall": 0.8577313432835821, - "support": 33500.0 + "f1-score": 0.8508837171222798, + "precision": 0.8504483474473299, + "recall": 0.8528549187153771, + "support": 27619.0 }, "step": 2673 } @@ -2128,7 +2128,7 @@ "num_input_tokens_seen": 0, "num_train_epochs": 50, "save_steps": 500, - "total_flos": 4730360494167000.0, + "total_flos": 4745096819694000.0, "train_batch_size": 4, "trial_name": null, "trial_params": null