diff --git "a/checkpoint-3564/trainer_state.json" "b/checkpoint-3564/trainer_state.json" --- "a/checkpoint-3564/trainer_state.json" +++ "b/checkpoint-3564/trainer_state.json" @@ -11,2029 +11,2029 @@ { "epoch": 1.0, "eval_Claim": { - "f1-score": 0.46243291592128805, - "precision": 0.6588785046728972, - "recall": 0.3562241616903996, - "support": 4354.0 + "f1-score": 0.5724454649827784, + "precision": 0.6002889477486154, + "recall": 0.5470704410796576, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.8342621259029929, - "precision": 0.8096955128205128, - "recall": 0.8603661132396765, - "support": 2349.0 + "f1-score": 0.7860447185813415, + "precision": 0.698526892771497, + "recall": 0.8986337593653592, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9993622448979592, - "precision": 0.9987253027405991, + "f1-score": 0.9987820830889982, + "precision": 0.9975671292124707, "recall": 1.0, - "support": 12536.0 + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.8941052999825083, - "precision": 0.8401157057392676, - "recall": 0.955510692388216, - "support": 13374.0 + "f1-score": 0.8965445389824622, + "precision": 0.9051258677512096, + "recall": 0.8881243979633962, + "support": 14534.0 }, - "eval_accuracy": 0.8857510808573268, - "eval_loss": 0.2754688858985901, + "eval_accuracy": 0.8791279948197712, + "eval_loss": 0.3157936632633209, "eval_macro avg": { - "f1-score": 0.7975406466761871, - "precision": 0.8268537564933192, - "recall": 0.793025241829573, - "support": 32613.0 - }, - "eval_runtime": 4.8781, - "eval_samples_per_second": 16.605, - "eval_steps_per_second": 2.255, + "f1-score": 0.8134542014088951, + "precision": 0.8003772093709481, + "recall": 0.8334571496021033, + "support": 32431.0 + }, + "eval_runtime": 4.8184, + "eval_samples_per_second": 16.603, + "eval_steps_per_second": 2.075, "eval_weighted avg": { - "f1-score": 0.8726239240080429, - "precision": 0.8746959682542021, - "recall": 0.8857510808573268, - "support": 32613.0 + "f1-score": 0.8781740995293698, + "precision": 0.8793943851252293, + "recall": 0.8791279948197712, + "support": 32431.0 }, "step": 81 }, { "epoch": 2.0, "eval_Claim": { - "f1-score": 0.6446312571646924, - "precision": 0.5518482172064115, - "recall": 0.77491961414791, - "support": 4354.0 + "f1-score": 0.6090644692195831, + "precision": 0.680108254397835, + "recall": 0.5514592933947773, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.8912341158733577, - "precision": 0.9019180470793374, - "recall": 0.8808003405704555, - "support": 2349.0 + "f1-score": 0.8585858585858586, + "precision": 0.8960229995208433, + "recall": 0.8241516086381666, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9998005504806733, - "precision": 0.9999202106439001, - "recall": 0.9996809189534142, - "support": 12536.0 + "f1-score": 0.999458141425088, + "precision": 0.9992776523702032, + "recall": 0.9996386956914461, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.8690409646250897, - "precision": 0.9324023303632625, - "recall": 0.8137430835950351, - "support": 13374.0 + "f1-score": 0.9105885478942475, + "precision": 0.8801849235905997, + "recall": 0.9431677445988716, + "support": 14534.0 }, - "eval_accuracy": 0.8848618649004998, - "eval_loss": 0.25178349018096924, + "eval_accuracy": 0.8990780426135487, + "eval_loss": 0.24451124668121338, "eval_macro avg": { - "f1-score": 0.8511767220359533, - "precision": 0.8465222013232279, - "recall": 0.8672859893167036, - "support": 32613.0 - }, - "eval_runtime": 4.907, - "eval_samples_per_second": 16.507, - "eval_steps_per_second": 2.242, + "f1-score": 0.8444242542811943, + "precision": 0.8638984574698704, + "recall": 0.8296043355808155, + "support": 32431.0 + }, + "eval_runtime": 4.8529, + "eval_samples_per_second": 16.485, + "eval_steps_per_second": 2.061, "eval_weighted avg": { - "f1-score": 0.8909418634778542, - "precision": 0.9053537287957648, - "recall": 0.8848618649004998, - "support": 32613.0 + "f1-score": 0.8949194640367888, + "precision": 0.8938342965082798, + "recall": 0.8990780426135487, + "support": 32431.0 }, "step": 162 }, { "epoch": 3.0, "eval_Claim": { - "f1-score": 0.6598764019226366, - "precision": 0.6576186131386861, - "recall": 0.6621497473587505, - "support": 4354.0 + "f1-score": 0.6542320859402667, + "precision": 0.7079713847726111, + "recall": 0.60807548825982, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9062119366626066, - "precision": 0.8661233993015134, - "recall": 0.9501915708812261, - "support": 2349.0 + "f1-score": 0.8915187376725839, + "precision": 0.8866608544027899, + "recall": 0.8964301454385192, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9998404722022812, - "precision": 0.9997607273887382, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9995936611133684, + "precision": 0.9992778479870013, + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9007852612503775, - "precision": 0.9097148086014946, - "recall": 0.8920293106026619, - "support": 13374.0 + "f1-score": 0.9157990498332153, + "precision": 0.8973258501155497, + "recall": 0.935048850970139, + "support": 14534.0 }, - "eval_accuracy": 0.9070002759635728, - "eval_loss": 0.23778581619262695, + "eval_accuracy": 0.9085442940396534, + "eval_loss": 0.22665414214134216, "eval_macro avg": { - "f1-score": 0.8666785180094754, - "precision": 0.8583043871076081, - "recall": 0.876072714645248, - "support": 32613.0 - }, - "eval_runtime": 4.9168, - "eval_samples_per_second": 16.474, - "eval_steps_per_second": 2.237, + "f1-score": 0.8652858836398586, + "precision": 0.872808984319488, + "recall": 0.8598660396478349, + "support": 32431.0 + }, + "eval_runtime": 4.8494, + "eval_samples_per_second": 16.497, + "eval_steps_per_second": 2.062, "eval_weighted avg": { - "f1-score": 0.907089072967282, - "precision": 0.9075314026721463, - "recall": 0.9070002759635728, - "support": 32613.0 + "f1-score": 0.9059516032151627, + "precision": 0.9047762338408377, + "recall": 0.9085442940396534, + "support": 32431.0 }, "step": 243 }, { "epoch": 4.0, "eval_Claim": { - "f1-score": 0.6604639922667954, - "precision": 0.696838347781744, - "recall": 0.6276986678915939, - "support": 4354.0 + "f1-score": 0.6861642294713161, + "precision": 0.6424741478360781, + "recall": 0.7362299758613122, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9140214420853479, - "precision": 0.9028239202657807, - "recall": 0.9255002128565347, - "support": 2349.0 + "f1-score": 0.8995515695067264, + "precision": 0.9155636695572797, + "recall": 0.884089907448215, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9994417862838917, - "precision": 0.9991230867346939, - "recall": 0.9997606892150607, - "support": 12536.0 + "f1-score": 0.9976481230212574, + "precision": 0.9990941208442794, + "recall": 0.9962063047601842, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9084203149780549, - "precision": 0.8963534463934785, - "recall": 0.9208165096455809, - "support": 13374.0 + "f1-score": 0.9062532879739067, + "precision": 0.9242435081193219, + "recall": 0.8889500481629283, + "support": 14534.0 }, - "eval_accuracy": 0.912366234323736, - "eval_loss": 0.2598011791706085, + "eval_accuracy": 0.9037649162838025, + "eval_loss": 0.2452327460050583, "eval_macro avg": { - "f1-score": 0.8705868839035225, - "precision": 0.8737847002939243, - "recall": 0.8684440199021926, - "support": 32613.0 - }, - "eval_runtime": 4.8872, - "eval_samples_per_second": 16.574, - "eval_steps_per_second": 2.251, + "f1-score": 0.8724043024933017, + "precision": 0.8703438615892398, + "recall": 0.8763690590581599, + "support": 32431.0 + }, + "eval_runtime": 4.8393, + "eval_samples_per_second": 16.531, + "eval_steps_per_second": 2.066, "eval_weighted avg": { - "f1-score": 0.9107077581074873, - "precision": 0.9096864919608297, - "recall": 0.912366234323736, - "support": 32613.0 + "f1-score": 0.9060584182508294, + "precision": 0.9095955973231696, + "recall": 0.9037649162838025, + "support": 32431.0 }, "step": 324 }, { "epoch": 5.0, "eval_Claim": { - "f1-score": 0.6731001206272618, - "precision": 0.6440713536201469, - "recall": 0.7048690858980248, - "support": 4354.0 + "f1-score": 0.6640249759846301, + "precision": 0.7332272606735614, + "recall": 0.6067588325652842, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.8871466886734063, - "precision": 0.8606885508406725, - "recall": 0.9152830991911451, - "support": 2349.0 + "f1-score": 0.8848007008322383, + "precision": 0.8794079233783195, + "recall": 0.8902600264433671, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9998005345673595, - "precision": 1.0, - "recall": 0.9996011486917677, - "support": 12536.0 + "f1-score": 0.999367774566474, + "precision": 0.9992775219001174, + "recall": 0.9994580435371692, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9039056236399038, - "precision": 0.9234729698104376, - "recall": 0.8851502916105877, - "support": 13374.0 + "f1-score": 0.9201314377682404, + "precision": 0.8973839110529758, + "recall": 0.9440621989816981, + "support": 14534.0 }, - "eval_accuracy": 0.9072455769171802, - "eval_loss": 0.3166182339191437, + "eval_accuracy": 0.9118127717307515, + "eval_loss": 0.28411003947257996, "eval_macro avg": { - "f1-score": 0.8659882418769828, - "precision": 0.8570582185678143, - "recall": 0.8762259063478812, - "support": 32613.0 - }, - "eval_runtime": 4.9018, - "eval_samples_per_second": 16.525, - "eval_steps_per_second": 2.244, + "f1-score": 0.8670812222878956, + "precision": 0.8773241542512435, + "recall": 0.8601347753818797, + "support": 32431.0 + }, + "eval_runtime": 4.8228, + "eval_samples_per_second": 16.588, + "eval_steps_per_second": 2.073, "eval_weighted avg": { - "f1-score": 0.9087455557232214, - "precision": 0.9110652677714914, - "recall": 0.9072455769171802, - "support": 32613.0 + "f1-score": 0.9087220731552942, + "precision": 0.9078434958291443, + "recall": 0.9118127717307515, + "support": 32431.0 }, "step": 405 }, { "epoch": 6.0, "eval_Claim": { - "f1-score": 0.634920634920635, - "precision": 0.7171775592828225, - "recall": 0.5695911805236564, - "support": 4354.0 + "f1-score": 0.6496087442553721, + "precision": 0.7484258729250143, + "recall": 0.5738424402018872, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9061357171881348, - "precision": 0.8666925767586475, - "recall": 0.9493401447424435, - "support": 2349.0 + "f1-score": 0.8865800865800866, + "precision": 0.8711186729051468, + "recall": 0.9026002644336713, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9998404722022812, - "precision": 0.9997607273887382, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9986892655367232, + "precision": 0.9994572100597069, + "recall": 0.9979225002258152, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9096214165876431, - "precision": 0.8879236684705212, - "recall": 0.9324061612083147, - "support": 13374.0 + "f1-score": 0.9191778088205947, + "precision": 0.889647180015452, + "recall": 0.9507362047612495, + "support": 14534.0 }, - "eval_accuracy": 0.9111397295556987, - "eval_loss": 0.32181796431541443, + "eval_accuracy": 0.9105177145323918, + "eval_loss": 0.3082066774368286, "eval_macro avg": { - "f1-score": 0.8626295602246735, - "precision": 0.8678886329751823, - "recall": 0.8628144290531921, - "support": 32613.0 - }, - "eval_runtime": 4.8856, - "eval_samples_per_second": 16.579, - "eval_steps_per_second": 2.252, + "f1-score": 0.8635139762981943, + "precision": 0.87716223397633, + "recall": 0.8562753524056559, + "support": 32431.0 + }, + "eval_runtime": 4.8389, + "eval_samples_per_second": 16.533, + "eval_steps_per_second": 2.067, "eval_weighted avg": { - "f1-score": 0.907375409471386, - "precision": 0.9065876667768511, - "recall": 0.9111397295556987, - "support": 32613.0 + "f1-score": 0.9061618943658084, + "precision": 0.9059932736781678, + "recall": 0.9105177145323918, + "support": 32431.0 }, "step": 486 }, { "epoch": 6.17, - "grad_norm": 7.612313747406006, + "grad_norm": 0.8612700700759888, "learning_rate": 1.7530864197530865e-05, - "loss": 0.2088, + "loss": 0.2223, "step": 500 }, { "epoch": 7.0, "eval_Claim": { - "f1-score": 0.6538414409998775, - "precision": 0.7008142894667717, - "recall": 0.6127698667891593, - "support": 4354.0 + "f1-score": 0.6977381768334475, + "precision": 0.629950495049505, + "recall": 0.7818740399385561, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.8866711022399646, - "precision": 0.9254629629629629, - "recall": 0.8510004257130694, - "support": 2349.0 + "f1-score": 0.8892209178228387, + "precision": 0.8621688741721855, + "recall": 0.9180255619215514, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.999880349379811, - "precision": 0.9998404722022812, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9988698521766648, + "precision": 0.9998190045248869, + "recall": 0.9979225002258152, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9127824473310774, - "precision": 0.8890070167977886, - "recall": 0.9378645132346344, - "support": 13374.0 + "f1-score": 0.9034227633516503, + "precision": 0.9449996243143738, + "recall": 0.8653502132929682, + "support": 14534.0 }, - "eval_accuracy": 0.9120596081317266, - "eval_loss": 0.4053190350532532, + "eval_accuracy": 0.9025623631710401, + "eval_loss": 0.3512111008167267, "eval_macro avg": { - "f1-score": 0.8632938349876826, - "precision": 0.8787811853574512, - "recall": 0.8503887588688042, - "support": 32613.0 - }, - "eval_runtime": 4.9061, - "eval_samples_per_second": 16.51, - "eval_steps_per_second": 2.242, + "f1-score": 0.8723129275461503, + "precision": 0.8592344995152377, + "recall": 0.8907930788447227, + "support": 32431.0 + }, + "eval_runtime": 4.8174, + "eval_samples_per_second": 16.607, + "eval_steps_per_second": 2.076, "eval_weighted avg": { - "f1-score": 0.909811074225195, - "precision": 0.9091110268457285, - "recall": 0.9120596081317266, - "support": 32613.0 + "f1-score": 0.9061105026169647, + "precision": 0.9136494810618672, + "recall": 0.9025623631710401, + "support": 32431.0 }, "step": 567 }, { "epoch": 8.0, "eval_Claim": { - "f1-score": 0.668355545764952, - "precision": 0.6421164021164021, - "recall": 0.6968305006890216, - "support": 4354.0 + "f1-score": 0.6987032799389779, + "precision": 0.6177710335525206, + "recall": 0.80403774412991, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.8960263537162857, - "precision": 0.8676236044657097, - "recall": 0.9263516389953171, - "support": 2349.0 + "f1-score": 0.9015364639688379, + "precision": 0.8856292517006803, + "recall": 0.9180255619215514, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, - "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.998915499322187, + "precision": 0.9994574554661362, + "recall": 0.9983741306115076, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9012129071630178, - "precision": 0.9198069137340392, - "recall": 0.8833557649170032, - "support": 13374.0 + "f1-score": 0.8976577489773016, + "precision": 0.9472075788830315, + "recall": 0.8530342644832806, + "support": 14534.0 }, - "eval_accuracy": 0.9063870235795541, - "eval_loss": 0.4514281749725342, + "eval_accuracy": 0.9003114304215103, + "eval_loss": 0.3705728054046631, "eval_macro avg": { - "f1-score": 0.8663987016610639, - "precision": 0.8573867300790377, - "recall": 0.8766344761503355, - "support": 32613.0 - }, - "eval_runtime": 4.8816, - "eval_samples_per_second": 16.593, - "eval_steps_per_second": 2.253, + "f1-score": 0.8742032480518261, + "precision": 0.8625163299005921, + "recall": 0.8933679252865624, + "support": 32431.0 + }, + "eval_runtime": 4.8542, + "eval_samples_per_second": 16.48, + "eval_steps_per_second": 2.06, "eval_weighted avg": { - "f1-score": 0.9077241398073883, - "precision": 0.9098003963445191, - "recall": 0.9063870235795541, - "support": 32613.0 + "f1-score": 0.9045397397569994, + "precision": 0.9144455555782818, + "recall": 0.9003114304215103, + "support": 32431.0 }, "step": 648 }, { "epoch": 9.0, "eval_Claim": { - "f1-score": 0.6913841807909605, - "precision": 0.7090777402221149, - "recall": 0.674552135966927, - "support": 4354.0 + "f1-score": 0.6657139454567107, + "precision": 0.7278645833333334, + "recall": 0.6133421110379635, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9055304497525284, - "precision": 0.9155787641427328, - "recall": 0.8957002979991485, - "support": 2349.0 + "f1-score": 0.8882521489971346, + "precision": 0.8884479717813051, + "recall": 0.8880564125165271, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9998404340194671, - "precision": 1.0, - "recall": 0.9996809189534142, - "support": 12536.0 + "f1-score": 0.9987339482727438, + "precision": 0.9999094612947035, + "recall": 0.9975611959172613, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9162317231167869, - "precision": 0.9072648632798181, - "recall": 0.9253775983251085, - "support": 13374.0 + "f1-score": 0.9178854152690192, + "precision": 0.8955360649299646, + "recall": 0.9413788358332187, + "support": 14534.0 }, - "eval_accuracy": 0.9183147824487168, - "eval_loss": 0.4471658766269684, + "eval_accuracy": 0.9107335573987851, + "eval_loss": 0.41826051473617554, "eval_macro avg": { - "f1-score": 0.8782466969199356, - "precision": 0.8829803419111664, - "recall": 0.8738277378111496, - "support": 32613.0 - }, - "eval_runtime": 4.8813, - "eval_samples_per_second": 16.594, - "eval_steps_per_second": 2.253, + "f1-score": 0.8676463644989021, + "precision": 0.8779395203348266, + "recall": 0.8600846388262426, + "support": 32431.0 + }, + "eval_runtime": 4.809, + "eval_samples_per_second": 16.635, + "eval_steps_per_second": 2.079, "eval_weighted avg": { - "f1-score": 0.9175807345372852, - "precision": 0.9170508471898524, - "recall": 0.9183147824487168, - "support": 32613.0 + "f1-score": 0.9079779452489347, + "precision": 0.9071100542045434, + "recall": 0.9107335573987851, + "support": 32431.0 }, "step": 729 }, { "epoch": 10.0, "eval_Claim": { - "f1-score": 0.6829727187206021, - "precision": 0.6997590361445784, - "recall": 0.6669728984841525, - "support": 4354.0 + "f1-score": 0.6738416222952305, + "precision": 0.6964871194379391, + "recall": 0.652622339258284, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.908567906307787, - "precision": 0.8780778395552026, - "recall": 0.9412515964240102, - "support": 2349.0 + "f1-score": 0.8951019550136641, + "precision": 0.8557073954983923, + "recall": 0.9382988100484795, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999202233745512, - "precision": 1.0, - "recall": 0.9998404594767071, - "support": 12536.0 + "f1-score": 0.9989605459393501, + "precision": 0.9996382054992764, + "recall": 0.998283804534369, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9131976852716073, - "precision": 0.9119379613749907, - "recall": 0.914460894272469, - "support": 13374.0 + "f1-score": 0.9109121470961545, + "precision": 0.9083259218717931, + "recall": 0.9135131415990092, + "support": 14534.0 }, - "eval_accuracy": 0.9161683991046515, - "eval_loss": 0.45130759477615356, + "eval_accuracy": 0.9075267490980852, + "eval_loss": 0.5459077954292297, "eval_macro avg": { - "f1-score": 0.8761646334186369, - "precision": 0.8724437092686929, - "recall": 0.8806314621643347, - "support": 32613.0 - }, - "eval_runtime": 4.8986, - "eval_samples_per_second": 16.535, - "eval_steps_per_second": 2.246, + "f1-score": 0.8697040675860997, + "precision": 0.8650396605768502, + "recall": 0.8756795238600354, + "support": 32431.0 + }, + "eval_runtime": 4.8478, + "eval_samples_per_second": 16.502, + "eval_steps_per_second": 2.063, "eval_weighted avg": { - "f1-score": 0.9154630053129839, - "precision": 0.9150220459300829, - "recall": 0.9161683991046515, - "support": 32613.0 + "f1-score": 0.9065515080853328, + "precision": 0.9060496255197702, + "recall": 0.9075267490980852, + "support": 32431.0 }, "step": 810 }, { "epoch": 11.0, "eval_Claim": { - "f1-score": 0.6209771986970684, - "precision": 0.717554953327311, - "recall": 0.5473128158015618, - "support": 4354.0 + "f1-score": 0.6581417175035196, + "precision": 0.7070834383665239, + "recall": 0.6155365371955234, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.8710538016896399, - "precision": 0.9115867845509539, - "recall": 0.8339719029374202, - "support": 2349.0 + "f1-score": 0.881979417560762, + "precision": 0.876414273281114, + "recall": 0.8876156897311591, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, - "precision": 1.0, + "f1-score": 0.9999548390010388, + "precision": 0.9999096820809249, "recall": 1.0, - "support": 12536.0 + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9144776812837283, - "precision": 0.8758814267132197, - "recall": 0.9566322715717063, - "support": 13374.0 + "f1-score": 0.9173079519373565, + "precision": 0.9002915065588976, + "recall": 0.9349800467868447, + "support": 14534.0 }, - "eval_accuracy": 0.9098212369300586, - "eval_loss": 0.5217961668968201, + "eval_accuracy": 0.9089759797724399, + "eval_loss": 0.5726844668388367, "eval_macro avg": { - "f1-score": 0.8516271704176092, - "precision": 0.8762557911478711, - "recall": 0.8344792475776721, - "support": 32613.0 - }, - "eval_runtime": 4.8915, - "eval_samples_per_second": 16.559, - "eval_steps_per_second": 2.249, + "f1-score": 0.8643459815006692, + "precision": 0.8709247250718651, + "recall": 0.8595330684283817, + "support": 32431.0 + }, + "eval_runtime": 4.8447, + "eval_samples_per_second": 16.513, + "eval_steps_per_second": 2.064, "eval_weighted avg": { - "f1-score": 0.9050398495319223, - "precision": 0.9050252912813879, - "recall": 0.9098212369300586, - "support": 32613.0 + "f1-score": 0.9066330640790278, + "precision": 0.9054793272287636, + "recall": 0.9089759797724399, + "support": 32431.0 }, "step": 891 }, { "epoch": 12.0, "eval_Claim": { - "f1-score": 0.6767220902612826, - "precision": 0.7006886374815543, - "recall": 0.6543408360128617, - "support": 4354.0 + "f1-score": 0.7079179557694512, + "precision": 0.7407673860911271, + "recall": 0.6778582400702217, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9093525179856115, - "precision": 0.8791732909379968, - "recall": 0.9416773094934014, - "support": 2349.0 + "f1-score": 0.8947480331703169, + "precision": 0.8644207066557108, + "recall": 0.9272807404142794, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999202233745512, - "precision": 1.0, - "recall": 0.9998404594767071, - "support": 12536.0 + "f1-score": 0.999186477447347, + "precision": 0.9999095431931253, + "recall": 0.998464456688646, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9131033456142309, - "precision": 0.9089427280136326, - "recall": 0.9173022282039779, - "support": 13374.0 + "f1-score": 0.9243840851702722, + "precision": 0.9169374492282697, + "recall": 0.9319526627218935, + "support": 14534.0 }, - "eval_accuracy": 0.9156777971974366, - "eval_loss": 0.48604533076286316, + "eval_accuracy": 0.9186272393697388, + "eval_loss": 0.5660926699638367, "eval_macro avg": { - "f1-score": 0.8747745443089191, - "precision": 0.872201164108296, - "recall": 0.878290208296737, - "support": 32613.0 - }, - "eval_runtime": 4.8984, - "eval_samples_per_second": 16.536, - "eval_steps_per_second": 2.246, + "f1-score": 0.8815591378893468, + "precision": 0.8805087712920582, + "recall": 0.8838890249737601, + "support": 32431.0 + }, + "eval_runtime": 4.8659, + "eval_samples_per_second": 16.441, + "eval_steps_per_second": 2.055, "eval_weighted avg": { - "f1-score": 0.9146463407295841, - "precision": 0.9139967630227935, - "recall": 0.9156777971974366, - "support": 32613.0 + "f1-score": 0.917429533390557, + "precision": 0.9168331226787283, + "recall": 0.9186272393697388, + "support": 32431.0 }, "step": 972 }, { "epoch": 12.35, - "grad_norm": 1.4397014379501343, + "grad_norm": 2.9101791381835938, "learning_rate": 1.506172839506173e-05, - "loss": 0.0253, + "loss": 0.0294, "step": 1000 }, { "epoch": 13.0, "eval_Claim": { - "f1-score": 0.6805793134907059, - "precision": 0.6758776896942242, - "recall": 0.6853468075333027, - "support": 4354.0 + "f1-score": 0.6744648496900223, + "precision": 0.7221943887775552, + "recall": 0.6326530612244898, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.903767491926803, - "precision": 0.914198606271777, - "recall": 0.8935717326521925, - "support": 2349.0 + "f1-score": 0.9044719714465672, + "precision": 0.863672814755413, + "recall": 0.9493168796826796, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9998005982053838, - "precision": 0.9996809952946806, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9995934041111363, + "precision": 0.9999096167751266, + "recall": 0.9992773913828923, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9103489546321576, - "precision": 0.9107236399012197, - "recall": 0.9099745775385075, - "support": 13374.0 + "f1-score": 0.9161992180860105, + "precision": 0.9055171023452725, + "recall": 0.9271363698912893, + "support": 14534.0 }, - "eval_accuracy": 0.9133781007573667, - "eval_loss": 0.5282083749771118, + "eval_accuracy": 0.9119361105115475, + "eval_loss": 0.5764271020889282, "eval_macro avg": { - "f1-score": 0.8736240895637626, - "precision": 0.8751202327904755, - "recall": 0.872203336865589, - "support": 32613.0 - }, - "eval_runtime": 4.89, - "eval_samples_per_second": 16.564, - "eval_steps_per_second": 2.249, + "f1-score": 0.8736823608334341, + "precision": 0.8728234806633418, + "recall": 0.8770959255453377, + "support": 32431.0 + }, + "eval_runtime": 4.8358, + "eval_samples_per_second": 16.543, + "eval_steps_per_second": 2.068, "eval_weighted avg": { - "f1-score": 0.9135835215352088, - "precision": 0.9138148255025323, - "recall": 0.9133781007573667, - "support": 32613.0 + "f1-score": 0.9098800911419986, + "precision": 0.909053016531805, + "recall": 0.9119361105115475, + "support": 32431.0 }, "step": 1053 }, { "epoch": 14.0, "eval_Claim": { - "f1-score": 0.6549993880797944, - "precision": 0.7010741419963322, - "recall": 0.6146072576940744, - "support": 4354.0 + "f1-score": 0.7047413793103449, + "precision": 0.6923565530383231, + "recall": 0.717577353522054, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9030425963488844, - "precision": 0.8624564122433166, - "recall": 0.9476372924648787, - "support": 2349.0 + "f1-score": 0.9112001687407719, + "precision": 0.8737864077669902, + "recall": 0.9519612163948876, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9992316384180792, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.998464456688646, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9098839358320396, - "precision": 0.8997076023391812, - "recall": 0.9202931060266188, - "support": 13374.0 + "f1-score": 0.9145424153781864, + "precision": 0.9258919757439007, + "recall": 0.903467730838035, + "support": 14534.0 }, - "eval_accuracy": 0.9120596081317266, - "eval_loss": 0.5426023602485657, + "eval_accuracy": 0.9131694983195091, + "eval_loss": 0.5468704700469971, "eval_macro avg": { - "f1-score": 0.866971508384752, - "precision": 0.8658095391447076, - "recall": 0.8706144714809814, - "support": 32613.0 + "f1-score": 0.8824289004618456, + "precision": 0.8730087341373035, + "recall": 0.8928676893609057, + "support": 32431.0 }, - "eval_runtime": 4.8976, - "eval_samples_per_second": 16.539, - "eval_steps_per_second": 2.246, + "eval_runtime": 4.8411, + "eval_samples_per_second": 16.525, + "eval_steps_per_second": 2.066, "eval_weighted avg": { - "f1-score": 0.9099868804586297, - "precision": 0.9090570140832118, - "recall": 0.9120596081317266, - "support": 32613.0 + "f1-score": 0.9137390947680665, + "precision": 0.9147298617643856, + "recall": 0.9131694983195091, + "support": 32431.0 }, "step": 1134 }, { "epoch": 15.0, "eval_Claim": { - "f1-score": 0.6663435909861886, - "precision": 0.7051282051282052, - "recall": 0.6316031235645384, - "support": 4354.0 + "f1-score": 0.6868638486408813, + "precision": 0.7559304164470216, + "recall": 0.6293614219881501, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.8880537974683544, - "precision": 0.8293313631326191, - "recall": 0.955725840783312, - "support": 2349.0 + "f1-score": 0.9058546000428909, + "precision": 0.8822055137844611, + "recall": 0.9308065226972234, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9999548349216385, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9116781523561184, - "precision": 0.9083958132284166, - "recall": 0.9149842978914311, - "support": 13374.0 + "f1-score": 0.9218029420675262, + "precision": 0.9023924075660713, + "recall": 0.9420668776661622, + "support": 14534.0 }, - "eval_accuracy": 0.9127341857541471, - "eval_loss": 0.6165347099304199, + "eval_accuracy": 0.9170855046097869, + "eval_loss": 0.6552095413208008, "eval_macro avg": { - "f1-score": 0.8665089135222377, - "precision": 0.8607138453723102, - "recall": 0.8755583729944088, - "support": 32613.0 - }, - "eval_runtime": 4.8787, - "eval_samples_per_second": 16.603, - "eval_steps_per_second": 2.255, + "f1-score": 0.8786190564182342, + "precision": 0.8851320844493885, + "recall": 0.8755361240685993, + "support": 32431.0 + }, + "eval_runtime": 4.8081, + "eval_samples_per_second": 16.639, + "eval_steps_per_second": 2.08, "eval_weighted avg": { - "f1-score": 0.9111575738225369, - "precision": 0.910775248619985, - "recall": 0.9127341857541471, - "support": 32613.0 + "f1-score": 0.9143537536055225, + "precision": 0.9137205288178378, + "recall": 0.9170855046097869, + "support": 32431.0 }, "step": 1215 }, { "epoch": 16.0, "eval_Claim": { - "f1-score": 0.674565560821485, - "precision": 0.6630434782608695, - "recall": 0.6864951768488746, - "support": 4354.0 + "f1-score": 0.7261954261954263, + "precision": 0.6899071696622556, + "recall": 0.7665130568356375, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9101030927835051, - "precision": 0.8824470211915234, - "recall": 0.9395487441464453, - "support": 2349.0 + "f1-score": 0.9104052573932091, + "precision": 0.9050522648083623, + "recall": 0.9158219479947113, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9999548349216385, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9041334190523012, - "precision": 0.9146836024179356, - "recall": 0.8938238372962465, - "support": 13374.0 + "f1-score": 0.9186291000841041, + "precision": 0.9360805599200114, + "recall": 0.9018164304389706, + "support": 14534.0 }, - "eval_accuracy": 0.9102198509796707, - "eval_loss": 0.6390220522880554, + "eval_accuracy": 0.9172705127809812, + "eval_loss": 0.604762613773346, "eval_macro avg": { - "f1-score": 0.8721905464838953, - "precision": 0.8650435254675821, - "recall": 0.8799469970074799, - "support": 32613.0 - }, - "eval_runtime": 4.9309, - "eval_samples_per_second": 16.427, - "eval_steps_per_second": 2.231, + "f1-score": 0.8887961546485946, + "precision": 0.8827599985976573, + "recall": 0.8960152772980452, + "support": 32431.0 + }, + "eval_runtime": 4.8523, + "eval_samples_per_second": 16.487, + "eval_steps_per_second": 2.061, "eval_weighted avg": { - "f1-score": 0.9107494233350908, - "precision": 0.9115609681986994, - "recall": 0.9102198509796707, - "support": 32613.0 + "f1-score": 0.918776399254959, + "precision": 0.92113920073012, + "recall": 0.9172705127809812, + "support": 32431.0 }, "step": 1296 }, { "epoch": 17.0, "eval_Claim": { - "f1-score": 0.6808510638297871, - "precision": 0.7280334728033473, - "recall": 0.6394120349104272, - "support": 4354.0 + "f1-score": 0.6726977499420088, + "precision": 0.7134071340713407, + "recall": 0.6363835856923414, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9149623250807319, - "precision": 0.9255226480836237, - "recall": 0.9046402724563644, - "support": 2349.0 + "f1-score": 0.898166556218246, + "precision": 0.9003542958370239, + "recall": 0.8959894226531512, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9999548349216385, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9178984340699547, - "precision": 0.8986960882647944, - "recall": 0.9379392851802004, - "support": 13374.0 + "f1-score": 0.9172189909373732, + "precision": 0.9018486500864477, + "recall": 0.9331223338378973, + "support": 14534.0 }, - "eval_accuracy": 0.9195106245975532, - "eval_loss": 0.5812303423881531, + "eval_accuracy": 0.9116277635595572, + "eval_loss": 0.6666390299797058, "eval_macro avg": { - "f1-score": 0.8784179840646908, - "precision": 0.8880630522879414, - "recall": 0.8704779555713364, - "support": 32613.0 - }, - "eval_runtime": 4.8888, - "eval_samples_per_second": 16.568, - "eval_steps_per_second": 2.25, + "f1-score": 0.8720095330048165, + "precision": 0.8789025199987031, + "recall": 0.8663512540265629, + "support": 32431.0 + }, + "eval_runtime": 4.8204, + "eval_samples_per_second": 16.596, + "eval_steps_per_second": 2.075, "eval_weighted avg": { - "f1-score": 0.9175833462373209, - "precision": 0.91678385690944, - "recall": 0.9195106245975532, - "support": 32613.0 + "f1-score": 0.9097710324765249, + "precision": 0.9087714991080675, + "recall": 0.9116277635595572, + "support": 32431.0 }, "step": 1377 }, { "epoch": 18.0, "eval_Claim": { - "f1-score": 0.6756586411193748, - "precision": 0.7488125174629785, - "recall": 0.6155259531465319, - "support": 4354.0 + "f1-score": 0.7242054854157598, + "precision": 0.7184193478730296, + "recall": 0.7300855826201448, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9198384010206252, - "precision": 0.9188615123194562, - "recall": 0.9208173690932312, - "support": 2349.0 + "f1-score": 0.91550204257149, + "precision": 0.8937867338371117, + "recall": 0.9382988100484795, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999202297383536, - "precision": 0.9999202297383536, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9999548349216385, + "precision": 1.0, + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9191801729776873, - "precision": 0.8941600678733032, - "recall": 0.9456407955735008, - "support": 13374.0 + "f1-score": 0.9218198185721211, + "precision": 0.9277948146083078, + "recall": 0.9159212880143113, + "support": 14534.0 }, - "eval_accuracy": 0.9206451415079876, - "eval_loss": 0.5839426517486572, + "eval_accuracy": 0.9200456353488946, + "eval_loss": 0.5898464918136597, "eval_macro avg": { - "f1-score": 0.8786493612140102, - "precision": 0.8904385818485229, - "recall": 0.8704760868879043, - "support": 32613.0 + "f1-score": 0.8903705453702524, + "precision": 0.8850002240796123, + "recall": 0.8960538386514494, + "support": 32431.0 }, - "eval_runtime": 4.8961, - "eval_samples_per_second": 16.544, - "eval_steps_per_second": 2.247, + "eval_runtime": 4.8354, + "eval_samples_per_second": 16.545, + "eval_steps_per_second": 2.068, "eval_weighted avg": { - "f1-score": 0.9177516254510409, - "precision": 0.9171873835957982, - "recall": 0.9206451415079876, - "support": 32613.0 + "f1-score": 0.9202833015380038, + "precision": 0.9206441029524512, + "recall": 0.9200456353488946, + "support": 32431.0 }, "step": 1458 }, { "epoch": 18.52, - "grad_norm": 6.301952838897705, + "grad_norm": 0.2594422399997711, "learning_rate": 1.2592592592592593e-05, - "loss": 0.0099, + "loss": 0.0093, "step": 1500 }, { "epoch": 19.0, "eval_Claim": { - "f1-score": 0.6969172151021822, - "precision": 0.7007197585326214, - "recall": 0.6931557188791916, - "support": 4354.0 + "f1-score": 0.7198604752561586, + "precision": 0.7151830192765866, + "recall": 0.7245995172262454, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.93473143590841, - "precision": 0.9397590361445783, - "recall": 0.929757343550447, - "support": 2349.0 + "f1-score": 0.9025769956002515, + "precision": 0.860223642172524, + "recall": 0.9493168796826796, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9975098474215601, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.9950320657573841, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9126430781850042, - "precision": 0.9101658362460028, - "recall": 0.9151338417825632, - "support": 13374.0 + "f1-score": 0.9212571111419453, + "precision": 0.9289911851126347, + "recall": 0.9136507499655979, + "support": 14534.0 }, - "eval_accuracy": 0.9191426731671419, - "eval_loss": 0.6210956573486328, + "eval_accuracy": 0.9173630168665783, + "eval_loss": 0.6108261346817017, "eval_macro avg": { - "f1-score": 0.8860629606184716, - "precision": 0.8876611577308006, - "recall": 0.8844917834876388, - "support": 32613.0 - }, - "eval_runtime": 4.8804, - "eval_samples_per_second": 16.597, - "eval_steps_per_second": 2.254, + "f1-score": 0.8853011073549788, + "precision": 0.8760994616404363, + "recall": 0.8956498031579767, + "support": 32431.0 + }, + "eval_runtime": 4.8341, + "eval_samples_per_second": 16.549, + "eval_steps_per_second": 2.069, "eval_weighted avg": { - "f1-score": 0.9189970320181106, - "precision": 0.9188662710731514, - "recall": 0.9191426731671419, - "support": 32613.0 + "f1-score": 0.9176816552958715, + "precision": 0.9183773040411919, + "recall": 0.9173630168665783, + "support": 32431.0 }, "step": 1539 }, { "epoch": 20.0, "eval_Claim": { - "f1-score": 0.6770349931017183, - "precision": 0.7457861287648522, - "recall": 0.6198897565457051, - "support": 4354.0 + "f1-score": 0.7150392937883518, + "precision": 0.7018174133558749, + "recall": 0.7287689269256089, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.933418150975403, - "precision": 0.9298690325306295, - "recall": 0.9369944657300979, - "support": 2349.0 + "f1-score": 0.8972902097902098, + "precision": 0.8899003034243607, + "recall": 0.9048038783605112, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9992316384180792, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.998464456688646, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.916624189907522, - "precision": 0.8932727788816349, - "recall": 0.9412292507851054, - "support": 13374.0 + "f1-score": 0.9203380437794403, + "precision": 0.9266285395452644, + "recall": 0.9141323792486583, + "support": 14534.0 }, - "eval_accuracy": 0.9205838162695857, - "eval_loss": 0.6079531311988831, + "eval_accuracy": 0.9162221331442139, + "eval_loss": 0.6273905038833618, "eval_macro avg": { - "f1-score": 0.8817593618157333, - "precision": 0.8922319850442791, - "recall": 0.8745084256998155, - "support": 32613.0 - }, - "eval_runtime": 4.8699, - "eval_samples_per_second": 16.633, - "eval_steps_per_second": 2.259, + "f1-score": 0.8829747964440202, + "precision": 0.879586564081375, + "recall": 0.886542410305856, + "support": 32431.0 + }, + "eval_runtime": 4.8321, + "eval_samples_per_second": 16.556, + "eval_steps_per_second": 2.069, "eval_weighted avg": { - "f1-score": 0.9178806455243597, - "precision": 0.9172429800024408, - "recall": 0.9205838162695857, - "support": 32613.0 + "f1-score": 0.9168102169228345, + "precision": 0.9175167258142971, + "recall": 0.9162221331442139, + "support": 32431.0 }, "step": 1620 }, { "epoch": 21.0, "eval_Claim": { - "f1-score": 0.6940774487471526, - "precision": 0.6884319927699954, - "recall": 0.6998162609095085, - "support": 4354.0 + "f1-score": 0.7209400804573365, + "precision": 0.6964614440580896, + "recall": 0.7472021066491112, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9116146152275975, - "precision": 0.875686274509804, - "recall": 0.9506172839506173, - "support": 2349.0 + "f1-score": 0.9076372817168105, + "precision": 0.8683574879227053, + "recall": 0.9506390480387836, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.999960116459937, - "precision": 0.9999202361011407, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.999186477447347, + "precision": 0.9999095431931253, + "recall": 0.998464456688646, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9093450177532674, - "precision": 0.9188549618320611, - "recall": 0.9000299087782264, - "support": 13374.0 + "f1-score": 0.9181764025650909, + "precision": 0.9355852317360566, + "recall": 0.9014036053392046, + "support": 14534.0 }, - "eval_accuracy": 0.9153711710054273, - "eval_loss": 0.641860842704773, + "eval_accuracy": 0.916314637229811, + "eval_loss": 0.6812385320663452, "eval_macro avg": { - "f1-score": 0.8787492995469885, - "precision": 0.8707233663032503, - "recall": 0.8876158634095881, - "support": 32613.0 - }, - "eval_runtime": 4.8748, - "eval_samples_per_second": 16.616, - "eval_steps_per_second": 2.257, + "f1-score": 0.8864850605466462, + "precision": 0.8750784267274943, + "recall": 0.8994273041789365, + "support": 32431.0 + }, + "eval_runtime": 4.8372, + "eval_samples_per_second": 16.539, + "eval_steps_per_second": 2.067, "eval_weighted avg": { - "f1-score": 0.9156004118108944, - "precision": 0.916143448767362, - "recall": 0.9153711710054273, - "support": 32613.0 + "f1-score": 0.917379120765938, + "precision": 0.9192399941849548, + "recall": 0.916314637229811, + "support": 32431.0 }, "step": 1701 }, { "epoch": 22.0, "eval_Claim": { - "f1-score": 0.6900875325289804, - "precision": 0.7114634146341463, - "recall": 0.6699586587046394, - "support": 4354.0 + "f1-score": 0.7066195738862937, + "precision": 0.7348341232227488, + "recall": 0.6804915514592934, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9227146221365875, - "precision": 0.9280792420327304, - "recall": 0.9174116645381013, - "support": 2349.0 + "f1-score": 0.9032682605834613, + "precision": 0.8991266375545851, + "recall": 0.9074482150727192, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.999960116459937, - "precision": 0.9999202361011407, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9992316384180792, + "precision": 1.0, + "recall": 0.998464456688646, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9152730501701939, - "precision": 0.9058883843562326, - "recall": 0.9248541947061463, - "support": 13374.0 + "f1-score": 0.9243903268596307, + "precision": 0.914037801843008, + "recall": 0.9349800467868447, + "support": 14534.0 }, - "eval_accuracy": 0.9191733357863429, - "eval_loss": 0.6659213304519653, + "eval_accuracy": 0.9189664210169283, + "eval_loss": 0.6876958012580872, "eval_macro avg": { - "f1-score": 0.8820088303239246, - "precision": 0.8863378192810625, - "recall": 0.8780561294872218, - "support": 32613.0 - }, - "eval_runtime": 4.8738, - "eval_samples_per_second": 16.62, - "eval_steps_per_second": 2.257, + "f1-score": 0.8833774499368663, + "precision": 0.8869996406550854, + "recall": 0.8803460675018758, + "support": 32431.0 + }, + "eval_runtime": 4.8487, + "eval_samples_per_second": 16.499, + "eval_steps_per_second": 2.062, "eval_weighted avg": { - "f1-score": 0.9182982110492126, - "precision": 0.91767458311091, - "recall": 0.9191733357863429, - "support": 32613.0 + "f1-score": 0.917861415342362, + "precision": 0.9171589791287257, + "recall": 0.9189664210169283, + "support": 32431.0 }, "step": 1782 }, { "epoch": 23.0, "eval_Claim": { - "f1-score": 0.6811174220470616, - "precision": 0.6875731336297684, - "recall": 0.6747818098300413, - "support": 4354.0 + "f1-score": 0.7040384395378104, + "precision": 0.7354206500956023, + "recall": 0.6752249286811499, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9136502398604448, - "precision": 0.9365221278497988, - "recall": 0.8918688803746275, - "support": 2349.0 + "f1-score": 0.90536013400335, + "precision": 0.8623853211009175, + "recall": 0.9528426619656236, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, - "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9991412429378531, + "precision": 0.9999095350099512, + "recall": 0.9983741306115076, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9115136218543538, - "precision": 0.9049970518867925, - "recall": 0.9181247196052041, - "support": 13374.0 + "f1-score": 0.9212183436002738, + "precision": 0.9164510418085251, + "recall": 0.9260355029585798, + "support": 14534.0 }, - "eval_accuracy": 0.9151871952902217, - "eval_loss": 0.6351564526557922, + "eval_accuracy": 0.9173630168665783, + "eval_loss": 0.6735763549804688, "eval_macro avg": { - "f1-score": 0.8765603492600376, - "precision": 0.88227307834159, - "recall": 0.8711739098870567, - "support": 32613.0 - }, - "eval_runtime": 4.8784, - "eval_samples_per_second": 16.604, - "eval_steps_per_second": 2.255, + "f1-score": 0.8824395400198218, + "precision": 0.878541637003749, + "recall": 0.8881193060542152, + "support": 32431.0 + }, + "eval_runtime": 4.8285, + "eval_samples_per_second": 16.568, + "eval_steps_per_second": 2.071, "eval_weighted avg": { - "f1-score": 0.9149061057787342, - "precision": 0.9147583624345246, - "recall": 0.9151871952902217, - "support": 32613.0 + "f1-score": 0.9161926989447983, + "precision": 0.9157214424409951, + "recall": 0.9173630168665783, + "support": 32431.0 }, "step": 1863 }, { "epoch": 24.0, "eval_Claim": { - "f1-score": 0.6994498708880655, - "precision": 0.6841642872831101, - "recall": 0.7154340836012861, - "support": 4354.0 + "f1-score": 0.7284623773173391, + "precision": 0.724040754389768, + "recall": 0.7329383366249725, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9161639276055753, - "precision": 0.8958502847843776, - "recall": 0.9374201787994891, - "support": 2349.0 + "f1-score": 0.8918801563178463, + "precision": 0.8789045785194695, + "recall": 0.9052446011458792, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9992316384180792, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.998464456688646, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9120683786543625, - "precision": 0.9227825820769878, - "recall": 0.9016001196351129, - "support": 13374.0 + "f1-score": 0.9255205276060909, + "precision": 0.9289526582103002, + "recall": 0.9221136645108022, + "support": 14534.0 }, - "eval_accuracy": 0.9171189402998804, - "eval_loss": 0.6426755785942078, + "eval_accuracy": 0.920415651691283, + "eval_loss": 0.6322010159492493, "eval_macro avg": { - "f1-score": 0.8819105726065732, - "precision": 0.8756992885361189, - "recall": 0.8885936529435605, - "support": 32613.0 - }, - "eval_runtime": 4.8958, - "eval_samples_per_second": 16.545, - "eval_steps_per_second": 2.247, + "f1-score": 0.8862736749148389, + "precision": 0.8829744977798845, + "recall": 0.8896902647425751, + "support": 32431.0 + }, + "eval_runtime": 4.8235, + "eval_samples_per_second": 16.586, + "eval_steps_per_second": 2.073, "eval_weighted avg": { - "f1-score": 0.9177621279849207, - "precision": 0.9186673375183761, - "recall": 0.9171189402998804, - "support": 32613.0 + "f1-score": 0.9206403732624274, + "precision": 0.9209116629411165, + "recall": 0.920415651691283, + "support": 32431.0 }, "step": 1944 }, { "epoch": 24.69, - "grad_norm": 0.09022147953510284, + "grad_norm": 0.6325204968452454, "learning_rate": 1.0123456790123458e-05, - "loss": 0.0049, + "loss": 0.004, "step": 2000 }, { "epoch": 25.0, "eval_Claim": { - "f1-score": 0.7043327556325824, - "precision": 0.708672401767031, - "recall": 0.7000459347726229, - "support": 4354.0 + "f1-score": 0.7145744441949248, + "precision": 0.7316624511381927, + "recall": 0.6982664033355278, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9363867684478372, - "precision": 0.9328263624841572, - "recall": 0.9399744572158365, - "support": 2349.0 + "f1-score": 0.9116772823779193, + "precision": 0.8795575583777141, + "recall": 0.9462318201851035, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9992316384180792, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.998464456688646, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9144606653474219, - "precision": 0.9132672085912447, - "recall": 0.9156572454015254, - "support": 13374.0 + "f1-score": 0.9222897565330862, + "precision": 0.920614245561116, + "recall": 0.9239713774597496, + "support": 14534.0 }, - "eval_accuracy": 0.9210437555575998, - "eval_loss": 0.6521892547607422, + "eval_accuracy": 0.9192439332737196, + "eval_loss": 0.6634725332260132, "eval_macro avg": { - "f1-score": 0.8887950473569604, - "precision": 0.8886914932106081, - "recall": 0.8889194093474961, - "support": 32613.0 - }, - "eval_runtime": 4.8719, - "eval_samples_per_second": 16.626, - "eval_steps_per_second": 2.258, + "f1-score": 0.8869432803810023, + "precision": 0.8829585637692556, + "recall": 0.8917335144172567, + "support": 32431.0 + }, + "eval_runtime": 4.8751, + "eval_samples_per_second": 16.41, + "eval_steps_per_second": 2.051, "eval_weighted avg": { - "f1-score": 0.9208669633417549, - "precision": 0.9207004694590575, - "recall": 0.9210437555575998, - "support": 32613.0 + "f1-score": 0.9186261381483827, + "precision": 0.9182914290271974, + "recall": 0.9192439332737196, + "support": 32431.0 }, "step": 2025 }, { "epoch": 26.0, "eval_Claim": { - "f1-score": 0.70013357079252, - "precision": 0.6792656587473002, - "recall": 0.7223242994947175, - "support": 4354.0 + "f1-score": 0.714535519125683, + "precision": 0.7117352492924015, + "recall": 0.717357910906298, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9349714043634824, - "precision": 0.9304384485666105, - "recall": 0.9395487441464453, - "support": 2349.0 + "f1-score": 0.9060118543607113, + "precision": 0.8716904276985743, + "recall": 0.9431467606875276, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9990506758283985, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9981031523800922, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9075579416991192, - "precision": 0.9179349904397706, - "recall": 0.8974128906834156, - "support": 13374.0 + "f1-score": 0.9199431877230055, + "precision": 0.926393637061327, + "recall": 0.9135819457823036, + "support": 14534.0 }, - "eval_accuracy": 0.9165056879158617, - "eval_loss": 0.6482347846031189, + "eval_accuracy": 0.9169313311337918, + "eval_loss": 0.6303670406341553, "eval_macro avg": { - "f1-score": 0.8856657292137804, - "precision": 0.8819097744384202, - "recall": 0.8898214835811445, - "support": 32613.0 - }, - "eval_runtime": 4.9087, - "eval_samples_per_second": 16.501, - "eval_steps_per_second": 2.241, + "f1-score": 0.8848853092594496, + "precision": 0.8774548285130757, + "recall": 0.8930474424390553, + "support": 32431.0 + }, + "eval_runtime": 4.8348, + "eval_samples_per_second": 16.547, + "eval_steps_per_second": 2.068, "eval_weighted avg": { - "f1-score": 0.9173737254580834, - "precision": 0.9185167005798364, - "recall": 0.9165056879158617, - "support": 32613.0 + "f1-score": 0.9171108994684886, + "precision": 0.9175310114557944, + "recall": 0.9169313311337918, + "support": 32431.0 }, "step": 2106 }, { "epoch": 27.0, "eval_Claim": { - "f1-score": 0.6918882072256305, - "precision": 0.6845773381294964, - "recall": 0.6993569131832797, - "support": 4354.0 + "f1-score": 0.7246061922868008, + "precision": 0.717512908777969, + "recall": 0.7318411235461927, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9247223364870424, - "precision": 0.8945483485873458, - "recall": 0.9570029799914858, - "support": 2349.0 + "f1-score": 0.9032534246575342, + "precision": 0.8780690803162713, + "recall": 0.9299250771264874, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9991864039052614, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9983741306115076, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9091732729331824, - "precision": 0.9181152790484904, - "recall": 0.9004037685060565, - "support": 13374.0 + "f1-score": 0.9229063442015175, + "precision": 0.9295735324911008, + "recall": 0.9163341131140773, + "support": 14534.0 }, - "eval_accuracy": 0.9159230981510441, - "eval_loss": 0.6850813627243042, + "eval_accuracy": 0.9193672720545157, + "eval_loss": 0.6543757319450378, "eval_macro avg": { - "f1-score": 0.8814459541614638, - "precision": 0.8743102414413331, - "recall": 0.8891909154202055, - "support": 32613.0 - }, - "eval_runtime": 4.8749, - "eval_samples_per_second": 16.616, - "eval_steps_per_second": 2.256, + "f1-score": 0.8874880912627785, + "precision": 0.8812888803963352, + "recall": 0.8941186110995663, + "support": 32431.0 + }, + "eval_runtime": 4.8562, + "eval_samples_per_second": 16.474, + "eval_steps_per_second": 2.059, "eval_weighted avg": { - "f1-score": 0.9161971414735486, - "precision": 0.9167147316420451, - "recall": 0.9159230981510441, - "support": 32613.0 + "f1-score": 0.9197072530313251, + "precision": 0.9202142021449996, + "recall": 0.9193672720545157, + "support": 32431.0 }, "step": 2187 }, { "epoch": 28.0, "eval_Claim": { - "f1-score": 0.6900420757363254, - "precision": 0.7025226082817706, - "recall": 0.6779972439136426, - "support": 4354.0 + "f1-score": 0.7182029107783169, + "precision": 0.6913705583756345, + "recall": 0.7472021066491112, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9335321687260333, - "precision": 0.9343283582089552, - "recall": 0.9327373350361856, - "support": 2349.0 + "f1-score": 0.9016903914590747, + "precision": 0.9101930848675348, + "recall": 0.8933450859409432, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9992316384180792, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.998464456688646, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9118346714243236, - "precision": 0.9065779748706578, - "recall": 0.9171526843128458, - "support": 13374.0 + "f1-score": 0.9179735039465906, + "precision": 0.9279437609841827, + "recall": 0.9082152194853447, + "support": 14534.0 }, - "eval_accuracy": 0.9181921319719131, - "eval_loss": 0.6860224604606628, + "eval_accuracy": 0.9153587616786408, + "eval_loss": 0.67218017578125, "eval_macro avg": { - "f1-score": 0.8838522289716705, - "precision": 0.8858572353403459, - "recall": 0.8819718158156685, - "support": 32613.0 - }, - "eval_runtime": 4.8824, - "eval_samples_per_second": 16.59, - "eval_steps_per_second": 2.253, + "f1-score": 0.8842746111505153, + "precision": 0.882376851056838, + "recall": 0.8868067171910112, + "support": 32431.0 + }, + "eval_runtime": 4.8266, + "eval_samples_per_second": 16.575, + "eval_steps_per_second": 2.072, "eval_weighted avg": { - "f1-score": 0.9176766061914672, - "precision": 0.9172444910254145, - "recall": 0.9181921319719131, - "support": 32613.0 + "f1-score": 0.9165029304654106, + "precision": 0.9180579805194509, + "recall": 0.9153587616786408, + "support": 32431.0 }, "step": 2268 }, { "epoch": 29.0, "eval_Claim": { - "f1-score": 0.702871276481342, - "precision": 0.6608695652173913, - "recall": 0.7505741846577859, - "support": 4354.0 + "f1-score": 0.7078925956061839, + "precision": 0.7525951557093425, + "recall": 0.6682027649769585, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9320307628351693, - "precision": 0.9106417546709992, - "recall": 0.9544487015751384, - "support": 2349.0 + "f1-score": 0.9066271437645564, + "precision": 0.8724531377343113, + "recall": 0.9435874834728956, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9992316384180792, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.998464456688646, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9048953733922058, - "precision": 0.9299976323889195, - "recall": 0.8811126065500224, - "support": 13374.0 + "f1-score": 0.9239400224405835, + "precision": 0.9132889695503126, + "recall": 0.9348424384202559, + "support": 14534.0 }, - "eval_accuracy": 0.9146352681446049, - "eval_loss": 0.7058882713317871, + "eval_accuracy": 0.9197064537017051, + "eval_loss": 0.7057467103004456, "eval_macro avg": { - "f1-score": 0.8849393814967517, - "precision": 0.8753772380693275, - "recall": 0.8965139306303251, - "support": 32613.0 - }, - "eval_runtime": 4.8777, - "eval_samples_per_second": 16.606, - "eval_steps_per_second": 2.255, + "f1-score": 0.8844228500573508, + "precision": 0.8845843157484916, + "recall": 0.8862742858896889, + "support": 32431.0 + }, + "eval_runtime": 4.8302, + "eval_samples_per_second": 16.562, + "eval_steps_per_second": 2.07, "eval_weighted avg": { - "f1-score": 0.9164202159722681, - "precision": 0.9195815136371419, - "recall": 0.9146352681446049, - "support": 32613.0 + "f1-score": 0.9180734884048335, + "precision": 0.9174528746424986, + "recall": 0.9197064537017051, + "support": 32431.0 }, "step": 2349 }, { "epoch": 30.0, "eval_Claim": { - "f1-score": 0.6935711954748178, - "precision": 0.658813804505063, - "recall": 0.7322002756086358, - "support": 4354.0 + "f1-score": 0.7153448663221748, + "precision": 0.7327963176064441, + "recall": 0.6987052885670397, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9355112431056427, - "precision": 0.9323467230443975, - "recall": 0.9386973180076629, - "support": 2349.0 + "f1-score": 0.8877374784110534, + "precision": 0.8700804062632247, + "recall": 0.9061260467166152, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9992316384180792, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.998464456688646, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9033831149039927, - "precision": 0.9209258971570607, - "recall": 0.8864961866307761, - "support": 13374.0 + "f1-score": 0.9262746978050201, + "precision": 0.9220124071170496, + "recall": 0.9305765790560067, + "support": 14534.0 }, - "eval_accuracy": 0.913255450280563, - "eval_loss": 0.6731629967689514, + "eval_accuracy": 0.9194597761401129, + "eval_loss": 0.6960250735282898, "eval_macro avg": { - "f1-score": 0.8831064166906858, - "precision": 0.8780216061766303, - "recall": 0.8893285024963571, - "support": 32613.0 - }, - "eval_runtime": 4.9031, - "eval_samples_per_second": 16.52, - "eval_steps_per_second": 2.243, + "f1-score": 0.882147170239082, + "precision": 0.8812222827466796, + "recall": 0.8834680927570769, + "support": 32431.0 + }, + "eval_runtime": 4.8194, + "eval_samples_per_second": 16.6, + "eval_steps_per_second": 2.075, "eval_weighted avg": { - "f1-score": 0.9148091452468388, - "precision": 0.9171502378139044, - "recall": 0.913255450280563, - "support": 32613.0 + "f1-score": 0.9188453276546991, + "precision": 0.918414282204774, + "recall": 0.9194597761401129, + "support": 32431.0 }, "step": 2430 }, { "epoch": 30.86, - "grad_norm": 0.004342585802078247, + "grad_norm": 0.009368549101054668, "learning_rate": 7.654320987654322e-06, - "loss": 0.0026, + "loss": 0.002, "step": 2500 }, { "epoch": 31.0, "eval_Claim": { - "f1-score": 0.7007920557985577, - "precision": 0.7220462850182704, - "recall": 0.6807533302710151, - "support": 4354.0 + "f1-score": 0.723667100130039, + "precision": 0.7148362235067437, + "recall": 0.7327188940092166, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9436951402269322, - "precision": 0.9491817398794143, - "recall": 0.9382716049382716, - "support": 2349.0 + "f1-score": 0.8953336240732664, + "precision": 0.8860595597755718, + "recall": 0.9048038783605112, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9991864039052614, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.9983741306115076, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9151526364477336, - "precision": 0.9058677019998534, - "recall": 0.9246298788694481, - "support": 13374.0 + "f1-score": 0.9236620107868897, + "precision": 0.928283530229326, + "recall": 0.9190862804458511, + "support": 14534.0 }, - "eval_accuracy": 0.9219942967528286, - "eval_loss": 0.6748707294464111, + "eval_accuracy": 0.9189664210169283, + "eval_loss": 0.6865902543067932, "eval_macro avg": { - "f1-score": 0.8898999864378783, - "precision": 0.8942739317243846, - "recall": 0.8858937609542722, - "support": 32613.0 - }, - "eval_runtime": 4.8881, - "eval_samples_per_second": 16.571, - "eval_steps_per_second": 2.25, + "f1-score": 0.8854622847238642, + "precision": 0.8822948283779104, + "recall": 0.8887457958567716, + "support": 32431.0 + }, + "eval_runtime": 4.8664, + "eval_samples_per_second": 16.439, + "eval_steps_per_second": 2.055, "eval_weighted avg": { - "f1-score": 0.9211890913209031, - "precision": 0.9206295673042141, - "recall": 0.9219942967528286, - "support": 32613.0 + "f1-score": 0.9193598504741339, + "precision": 0.9198190200735107, + "recall": 0.9189664210169283, + "support": 32431.0 }, "step": 2511 }, { "epoch": 32.0, "eval_Claim": { - "f1-score": 0.7010334054313868, - "precision": 0.7351310483870968, - "recall": 0.6699586587046394, - "support": 4354.0 + "f1-score": 0.7149137639499492, + "precision": 0.7350486787204451, + "recall": 0.695852534562212, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9412010008340285, - "precision": 0.9223539027380466, - "recall": 0.9608343976160069, - "support": 2349.0 + "f1-score": 0.8920269389528569, + "precision": 0.8796058269065982, + "recall": 0.9048038783605112, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.999322095177837, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.9986451088429229, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9177053667196804, - "precision": 0.9079997072385274, - "recall": 0.9276207566920891, - "support": 13374.0 + "f1-score": 0.9260107310071427, + "precision": 0.919942961906702, + "recall": 0.9321590752717765, + "support": 14534.0 }, - "eval_accuracy": 0.9234047772360715, - "eval_loss": 0.7106999754905701, + "eval_accuracy": 0.9197372883969042, + "eval_loss": 0.6847501397132874, "eval_macro avg": { - "f1-score": 0.8899749715658463, - "precision": 0.8913711645909177, - "recall": 0.8895835106877722, - "support": 32613.0 - }, - "eval_runtime": 4.8986, - "eval_samples_per_second": 16.535, - "eval_steps_per_second": 2.246, + "f1-score": 0.8830683822719464, + "precision": 0.8836493668834363, + "recall": 0.8828651492593558, + "support": 32431.0 + }, + "eval_runtime": 4.8433, + "eval_samples_per_second": 16.518, + "eval_steps_per_second": 2.065, "eval_weighted avg": { - "f1-score": 0.9220884970034353, - "precision": 0.9213184308961812, - "recall": 0.9234047772360715, - "support": 32613.0 + "f1-score": 0.9189974415520827, + "precision": 0.9184697807200564, + "recall": 0.9197372883969042, + "support": 32431.0 }, "step": 2592 }, { "epoch": 33.0, "eval_Claim": { - "f1-score": 0.711044912923923, - "precision": 0.709419295839049, - "recall": 0.7126779972439137, - "support": 4354.0 + "f1-score": 0.7068796869905445, + "precision": 0.7002583979328165, + "recall": 0.7136273864384464, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9253022452504318, - "precision": 0.938677179150241, - "recall": 0.9123031077054066, - "support": 2349.0 + "f1-score": 0.8840610884061089, + "precision": 0.8634453781512605, + "recall": 0.9056853239312472, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9987338337704622, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.9974708698401229, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9179324500839708, - "precision": 0.916325162059459, - "recall": 0.9195453865709585, - "support": 13374.0 + "f1-score": 0.9202020901100422, + "precision": 0.9256474519632414, + "recall": 0.9148204210816018, + "support": 14534.0 }, - "eval_accuracy": 0.9223009229448379, - "eval_loss": 0.7237989902496338, + "eval_accuracy": 0.9141253738706793, + "eval_loss": 0.7124452590942383, "eval_macro avg": { - "f1-score": 0.8885599303841538, - "precision": 0.8911054092621873, - "recall": 0.8861116803146581, - "support": 32613.0 - }, - "eval_runtime": 4.8767, - "eval_samples_per_second": 16.609, - "eval_steps_per_second": 2.256, + "f1-score": 0.8774691748192894, + "precision": 0.8723378070118297, + "recall": 0.8829010003228546, + "support": 32431.0 + }, + "eval_runtime": 4.8097, + "eval_samples_per_second": 16.633, + "eval_steps_per_second": 2.079, "eval_weighted avg": { - "f1-score": 0.9223730749223835, - "precision": 0.9224756086618937, - "recall": 0.9223009229448379, - "support": 32613.0 + "f1-score": 0.9145072552354695, + "precision": 0.9150070965816288, + "recall": 0.9141253738706793, + "support": 32431.0 }, "step": 2673 }, { "epoch": 34.0, "eval_Claim": { - "f1-score": 0.7148519468407553, - "precision": 0.7258522727272727, - "recall": 0.7041800643086816, - "support": 4354.0 + "f1-score": 0.709584487534626, + "precision": 0.7166517457475381, + "recall": 0.7026552556506473, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9396058486967578, - "precision": 0.9354430379746835, - "recall": 0.9438058748403576, - "support": 2349.0 + "f1-score": 0.8878668418747262, + "precision": 0.8824553765781454, + "recall": 0.8933450859409432, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9992316384180792, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.998464456688646, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9186834462729914, - "precision": 0.914936220706022, - "recall": 0.9224614924480335, - "support": 13374.0 + "f1-score": 0.9230083030261442, + "precision": 0.9205447577333699, + "recall": 0.9254850694922251, + "support": 14534.0 }, - "eval_accuracy": 0.9246312820041088, - "eval_loss": 0.6951220035552979, + "eval_accuracy": 0.9168388270481946, + "eval_loss": 0.6949135661125183, "eval_macro avg": { - "f1-score": 0.8932753387721986, - "precision": 0.8940578828519945, - "recall": 0.8925919153338566, - "support": 32613.0 - }, - "eval_runtime": 4.9327, - "eval_samples_per_second": 16.421, - "eval_steps_per_second": 2.23, + "f1-score": 0.879922817713394, + "precision": 0.8799129700147633, + "recall": 0.8799874669431155, + "support": 32431.0 + }, + "eval_runtime": 4.8274, + "eval_samples_per_second": 16.572, + "eval_steps_per_second": 2.072, "eval_weighted avg": { - "f1-score": 0.9242195414603059, - "precision": 0.9238669704528689, - "recall": 0.9246312820041088, - "support": 32613.0 + "f1-score": 0.9165811266694699, + "precision": 0.9163538208419149, + "recall": 0.9168388270481946, + "support": 32431.0 }, "step": 2754 }, { "epoch": 35.0, "eval_Claim": { - "f1-score": 0.7058540110816671, - "precision": 0.7421479229989868, - "recall": 0.6729444189251264, - "support": 4354.0 + "f1-score": 0.7161833075049812, + "precision": 0.7225820862184499, + "recall": 0.7098968619705947, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.939966414777498, - "precision": 0.9271221532091097, - "recall": 0.9531715623669647, - "support": 2349.0 + "f1-score": 0.9034467994005567, + "precision": 0.8784346378018318, + "recall": 0.9299250771264874, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9995933673699905, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.9991870653057537, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9197091070176086, - "precision": 0.9082756106452788, - "recall": 0.9314341259159563, - "support": 13374.0 + "f1-score": 0.9223401323042998, + "precision": 0.9237405106970324, + "recall": 0.9209439933947984, + "support": 14534.0 }, - "eval_accuracy": 0.9248152577193144, - "eval_loss": 0.7126197814941406, + "eval_accuracy": 0.9186272393697388, + "eval_loss": 0.6991766691207886, "eval_macro avg": { - "f1-score": 0.8913724115387659, - "precision": 0.8943864217133439, - "recall": 0.8893675842366002, - "support": 32613.0 - }, - "eval_runtime": 4.9139, - "eval_samples_per_second": 16.484, - "eval_steps_per_second": 2.239, + "f1-score": 0.885390901644957, + "precision": 0.8811893086793285, + "recall": 0.8899882494494086, + "support": 32431.0 + }, + "eval_runtime": 4.8369, + "eval_samples_per_second": 16.539, + "eval_steps_per_second": 2.067, "eval_weighted avg": { - "f1-score": 0.923464846836294, - "precision": 0.9227118023915538, - "recall": 0.9248152577193144, - "support": 32613.0 + "f1-score": 0.9184224283310388, + "precision": 0.9183379896562086, + "recall": 0.9186272393697388, + "support": 32431.0 }, "step": 2835 }, { "epoch": 36.0, "eval_Claim": { - "f1-score": 0.7094183778822989, - "precision": 0.7086866834746733, - "recall": 0.7101515847496555, - "support": 4354.0 + "f1-score": 0.7000227946204696, + "precision": 0.7282428266540194, + "recall": 0.673908272986614, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9417721518987342, - "precision": 0.9335006273525721, - "recall": 0.9501915708812261, - "support": 2349.0 + "f1-score": 0.9057333047025982, + "precision": 0.8831658291457286, + "recall": 0.9294843543411194, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9999548349216385, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9147501685519513, - "precision": 0.9164665265685981, - "recall": 0.9130402273067145, - "support": 13374.0 + "f1-score": 0.9210652099692729, + "precision": 0.9141366223908919, + "recall": 0.9280996284574102, + "support": 14534.0 }, - "eval_accuracy": 0.9220249593720296, - "eval_loss": 0.7317129373550415, + "eval_accuracy": 0.9169930005241899, + "eval_loss": 0.7325878739356995, "eval_macro avg": { - "f1-score": 0.8914752029028185, - "precision": 0.889663459348961, - "recall": 0.8933259031689874, - "support": 32613.0 - }, - "eval_runtime": 4.9011, - "eval_samples_per_second": 16.527, - "eval_steps_per_second": 2.244, + "f1-score": 0.8816940360534948, + "precision": 0.8813863195476599, + "recall": 0.8828504824270013, + "support": 32431.0 + }, + "eval_runtime": 4.8521, + "eval_samples_per_second": 16.488, + "eval_steps_per_second": 2.061, "eval_weighted avg": { - "f1-score": 0.9220371979388616, - "precision": 0.9220629233688514, - "recall": 0.9220249593720296, - "support": 32613.0 + "f1-score": 0.9158636638576222, + "precision": 0.9151604174161526, + "recall": 0.9169930005241899, + "support": 32431.0 }, "step": 2916 }, { "epoch": 37.0, "eval_Claim": { - "f1-score": 0.7016791711325474, - "precision": 0.7286668315607222, - "recall": 0.6766192007349564, - "support": 4354.0 + "f1-score": 0.7043226381461676, + "precision": 0.7153202081918986, + "recall": 0.6936581084046521, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9388185654008439, - "precision": 0.9305729820158929, - "recall": 0.9472115793954874, - "support": 2349.0 + "f1-score": 0.8989154013015185, + "precision": 0.8850918410935498, + "recall": 0.9131776112825033, + "support": 2269.0 }, "eval_O": { - "f1-score": 0.999960116459937, - "precision": 0.9999202361011407, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9999548349216385, + "precision": 1.0, + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9166419899318923, - "precision": 0.9076381762204955, - "recall": 0.9258262299985046, - "support": 13374.0 + "f1-score": 0.9208855328642527, + "precision": 0.9187726868022739, + "recall": 0.9230081188936288, + "support": 14534.0 }, - "eval_accuracy": 0.9226075491368473, - "eval_loss": 0.741832971572876, + "eval_accuracy": 0.91634547192501, + "eval_loss": 0.7374063730239868, "eval_macro avg": { - "f1-score": 0.8892749607313051, - "precision": 0.8916995564745628, - "recall": 0.887414252532237, - "support": 32613.0 - }, - "eval_runtime": 4.8986, - "eval_samples_per_second": 16.535, - "eval_steps_per_second": 2.246, + "f1-score": 0.8810196018083942, + "precision": 0.8797961840219305, + "recall": 0.8824383781259115, + "support": 32431.0 + }, + "eval_runtime": 4.8356, + "eval_samples_per_second": 16.544, + "eval_steps_per_second": 2.068, "eval_weighted avg": { - "f1-score": 0.9215670411961056, - "precision": 0.9208684993072555, - "recall": 0.9226075491368473, - "support": 32613.0 + "f1-score": 0.9159103209784385, + "precision": 0.9155570227916499, + "recall": 0.91634547192501, + "support": 32431.0 }, "step": 2997 }, { "epoch": 37.04, - "grad_norm": 0.02783563733100891, + "grad_norm": 0.13183258473873138, "learning_rate": 5.185185185185185e-06, - "loss": 0.001, + "loss": 0.0012, "step": 3000 }, { "epoch": 38.0, "eval_Claim": { - "f1-score": 0.7192414431082331, - "precision": 0.7242664182580345, - "recall": 0.7142857142857143, - "support": 4354.0 + "f1-score": 0.6968245125348189, + "precision": 0.7077863286555002, + "recall": 0.6861970594689488, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.940041928721174, - "precision": 0.9260636100784799, - "recall": 0.9544487015751384, - "support": 2349.0 + "f1-score": 0.8926233085988651, + "precision": 0.8841331603977518, + "recall": 0.9012780960775673, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9999548349216385, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9176391382405745, - "precision": 0.9180511899416255, - "recall": 0.9172274562584118, - "support": 13374.0 + "f1-score": 0.9194212042243862, + "precision": 0.9164046479835953, + "recall": 0.922457685427274, + "support": 14534.0 }, - "eval_accuracy": 0.9246312820041088, - "eval_loss": 0.7159741520881653, + "eval_accuracy": 0.9142178779562764, + "eval_loss": 0.7381414175033569, "eval_macro avg": { - "f1-score": 0.8942306275174954, - "precision": 0.892095304569535, - "recall": 0.8964904680298161, - "support": 32613.0 - }, - "eval_runtime": 4.8897, - "eval_samples_per_second": 16.565, - "eval_steps_per_second": 2.25, + "f1-score": 0.8772059650699272, + "precision": 0.8770810342592118, + "recall": 0.877460628724163, + "support": 32431.0 + }, + "eval_runtime": 4.8281, + "eval_samples_per_second": 16.57, + "eval_steps_per_second": 2.071, "eval_weighted avg": { - "f1-score": 0.9244240507984156, - "precision": 0.9242570759957419, - "recall": 0.9246312820041088, - "support": 32613.0 + "f1-score": 0.9137602710507441, + "precision": 0.9133700963405134, + "recall": 0.9142178779562764, + "support": 32431.0 }, "step": 3078 }, { "epoch": 39.0, "eval_Claim": { - "f1-score": 0.6989553656220323, - "precision": 0.7233415233415234, - "recall": 0.6761598530087276, - "support": 4354.0 + "f1-score": 0.6997294860234445, + "precision": 0.719351100811124, + "recall": 0.6811498793065613, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.937261551504875, - "precision": 0.9333051920641621, - "recall": 0.9412515964240102, - "support": 2349.0 + "f1-score": 0.9017094017094017, + "precision": 0.8751555371215264, + "recall": 0.9299250771264874, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9999548349216385, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9162594402487784, - "precision": 0.907391113066432, - "recall": 0.9253028263795424, - "support": 13374.0 + "f1-score": 0.9202920909184409, + "precision": 0.9171165015374103, + "recall": 0.9234897481766892, + "support": 14534.0 }, - "eval_accuracy": 0.9219023088952258, - "eval_loss": 0.7222980260848999, + "eval_accuracy": 0.9159754555826216, + "eval_loss": 0.7733151316642761, "eval_macro avg": { - "f1-score": 0.8881190893439215, - "precision": 0.8910094571180294, - "recall": 0.8856785689530701, - "support": 32613.0 - }, - "eval_runtime": 4.8729, - "eval_samples_per_second": 16.623, - "eval_steps_per_second": 2.257, + "f1-score": 0.8804214533932313, + "precision": 0.8779057848675151, + "recall": 0.8836185946331498, + "support": 32431.0 + }, + "eval_runtime": 4.9105, + "eval_samples_per_second": 16.292, + "eval_steps_per_second": 2.036, "eval_weighted avg": { - "f1-score": 0.9209497071808923, - "precision": 0.920283679359095, - "recall": 0.9219023088952258, - "support": 32613.0 + "f1-score": 0.9151944475074327, + "precision": 0.9146860137975935, + "recall": 0.9159754555826216, + "support": 32431.0 }, "step": 3159 }, { "epoch": 40.0, "eval_Claim": { - "f1-score": 0.680875980794004, - "precision": 0.6946236559139785, - "recall": 0.6676619200734957, - "support": 4354.0 + "f1-score": 0.6766934825696631, + "precision": 0.7218905472636816, + "recall": 0.6368224709238535, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9202291758483913, - "precision": 0.9538602101416171, - "recall": 0.8888888888888888, - "support": 2349.0 + "f1-score": 0.8926333981421473, + "precision": 0.8754237288135593, + "recall": 0.9105332745702953, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9999096657633243, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.999819347845723, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9115485467370831, - "precision": 0.9006057067795373, - "recall": 0.9227605802302976, - "support": 13374.0 + "f1-score": 0.9188236888467272, + "precision": 0.9050861033239888, + "recall": 0.9329847254713086, + "support": 14534.0 }, - "eval_accuracy": 0.915953760770245, - "eval_loss": 0.7755006551742554, + "eval_accuracy": 0.9126144738059264, + "eval_loss": 0.8124102354049683, "eval_macro avg": { - "f1-score": 0.8781634258448696, - "precision": 0.8872723932087833, - "recall": 0.8698278472981706, - "support": 32613.0 - }, - "eval_runtime": 4.9172, - "eval_samples_per_second": 16.473, - "eval_steps_per_second": 2.237, + "f1-score": 0.8720150588304654, + "precision": 0.8756000948503074, + "recall": 0.870039954702795, + "support": 32431.0 + }, + "eval_runtime": 4.8273, + "eval_samples_per_second": 16.573, + "eval_steps_per_second": 2.072, "eval_weighted avg": { - "f1-score": 0.9153773838195417, - "precision": 0.915147632966659, - "recall": 0.915953760770245, - "support": 32613.0 + "f1-score": 0.9106490945027469, + "precision": 0.9096701640488858, + "recall": 0.9126144738059264, + "support": 32431.0 }, "step": 3240 }, { "epoch": 41.0, "eval_Claim": { - "f1-score": 0.7111719730685837, - "precision": 0.7067362213653889, - "recall": 0.7156637574644006, - "support": 4354.0 + "f1-score": 0.6830284788145403, + "precision": 0.7228620436167607, + "recall": 0.6473557164801405, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9299660441426145, - "precision": 0.9272111722386797, - "recall": 0.9327373350361856, - "support": 2349.0 + "f1-score": 0.9045826876245294, + "precision": 0.9088078291814946, + "recall": 0.9003966505068312, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9999548349216385, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9154765920761648, - "precision": 0.9178504321683577, - "recall": 0.9131149992522806, - "support": 13374.0 + "f1-score": 0.9190962592166678, + "precision": 0.9038717402873869, + "recall": 0.9348424384202559, + "support": 14534.0 }, - "eval_accuracy": 0.9215650200840155, - "eval_loss": 0.7457175254821777, + "eval_accuracy": 0.9142487126514754, + "eval_loss": 0.8041703104972839, "eval_macro avg": { - "f1-score": 0.8891536523218408, - "precision": 0.8879494564431065, - "recall": 0.8903790229382167, - "support": 32613.0 - }, - "eval_runtime": 4.9009, - "eval_samples_per_second": 16.527, - "eval_steps_per_second": 2.244, + "f1-score": 0.876665565144344, + "precision": 0.8838854032714106, + "recall": 0.8706261198325222, + "support": 32431.0 + }, + "eval_runtime": 4.8224, + "eval_samples_per_second": 16.589, + "eval_steps_per_second": 2.074, "eval_weighted avg": { - "f1-score": 0.9217341842473323, - "precision": 0.9219170340426572, - "recall": 0.9215650200840155, - "support": 32613.0 + "f1-score": 0.9125128397228092, + "precision": 0.911598136675134, + "recall": 0.9142487126514754, + "support": 32431.0 }, "step": 3321 }, { "epoch": 42.0, "eval_Claim": { - "f1-score": 0.7124631992149166, - "precision": 0.6782229603487648, - "recall": 0.7503445107946716, - "support": 4354.0 + "f1-score": 0.6939400568803326, + "precision": 0.6918211559432933, + "recall": 0.696071977177968, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9302426343154246, - "precision": 0.9470666078517865, - "recall": 0.9140059599829715, - "support": 2349.0 + "f1-score": 0.8969750109601051, + "precision": 0.8922808547754034, + "recall": 0.9017188188629353, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9999548349216385, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9112906284370614, - "precision": 0.9246517355499115, - "recall": 0.8983101540302079, - "support": 13374.0 + "f1-score": 0.9167729262156668, + "precision": 0.9183870745011393, + "recall": 0.9151644419980735, + "support": 14534.0 }, - "eval_accuracy": 0.9187747217367308, - "eval_loss": 0.7305595278739929, + "eval_accuracy": 0.9123677962443342, + "eval_loss": 0.7728341221809387, "eval_macro avg": { - "f1-score": 0.8884991154918507, - "precision": 0.8874853259376158, - "recall": 0.8906651562019627, - "support": 32613.0 - }, - "eval_runtime": 4.8913, - "eval_samples_per_second": 16.56, - "eval_steps_per_second": 2.249, + "f1-score": 0.8769107072444358, + "precision": 0.8756222713049591, + "recall": 0.8782162279904596, + "support": 32431.0 + }, + "eval_runtime": 4.8152, + "eval_samples_per_second": 16.614, + "eval_steps_per_second": 2.077, "eval_weighted avg": { - "f1-score": 0.9202099034773231, - "precision": 0.9223295784640139, - "recall": 0.9187747217367308, - "support": 32613.0 + "f1-score": 0.9124725980114123, + "precision": 0.9125852427590434, + "recall": 0.9123677962443342, + "support": 32431.0 }, "step": 3402 }, { "epoch": 43.0, "eval_Claim": { - "f1-score": 0.7247613700168444, - "precision": 0.7090749285871237, - "recall": 0.7411575562700965, - "support": 4354.0 + "f1-score": 0.7087594486438418, + "precision": 0.7181797702185176, + "recall": 0.6995830590300637, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.933246921581335, - "precision": 0.9473684210526315, - "recall": 0.9195402298850575, - "support": 2349.0 + "f1-score": 0.9011231006386258, + "precision": 0.9005281690140845, + "recall": 0.9017188188629353, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9999548349216385, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9186326070623592, - "precision": 0.923071115808546, - "recall": 0.9142365784357709, - "support": 13374.0 + "f1-score": 0.9225603070175438, + "precision": 0.9189078498293516, + "recall": 0.9262419155084629, + "support": 14534.0 }, - "eval_accuracy": 0.9244779689081041, - "eval_loss": 0.713043749332428, + "eval_accuracy": 0.9178255372945638, + "eval_loss": 0.7432543039321899, "eval_macro avg": { - "f1-score": 0.8941602246651347, - "precision": 0.8948786163620753, - "recall": 0.8937335911477313, - "support": 32613.0 - }, - "eval_runtime": 4.8816, - "eval_samples_per_second": 16.593, - "eval_steps_per_second": 2.253, + "f1-score": 0.8830994228054124, + "precision": 0.8844039472654884, + "recall": 0.8818633668310809, + "support": 32431.0 + }, + "eval_runtime": 4.8581, + "eval_samples_per_second": 16.467, + "eval_steps_per_second": 2.058, "eval_weighted avg": { - "f1-score": 0.9250789719038385, - "precision": 0.9258220268894141, - "recall": 0.9244779689081041, - "support": 32613.0 + "f1-score": 0.9174387654537162, + "precision": 0.9170993900218476, + "recall": 0.9178255372945638, + "support": 32431.0 }, "step": 3483 }, { "epoch": 43.21, - "grad_norm": 0.0021550117526203394, + "grad_norm": 0.0030191524419933558, "learning_rate": 2.7160493827160496e-06, - "loss": 0.001, + "loss": 0.0009, "step": 3500 }, { "epoch": 44.0, "eval_Claim": { - "f1-score": 0.7143999086445129, - "precision": 0.7104247104247104, - "recall": 0.718419843821773, - "support": 4354.0 + "f1-score": 0.7071469411092054, + "precision": 0.7382999044890163, + "recall": 0.6785165679174896, + "support": 4557.0 }, "eval_MajorClaim": { - "f1-score": 0.9345754816853695, - "precision": 0.9296545914069082, - "recall": 0.9395487441464453, - "support": 2349.0 + "f1-score": 0.9027962716378164, + "precision": 0.9092534644613322, + "recall": 0.8964301454385192, + "support": 2269.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9999548349216385, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9999096739228616, + "support": 11071.0 }, "eval_Premise": { - "f1-score": 0.9163230111719278, - "precision": 0.9188721804511278, - "recall": 0.9137879467623747, - "support": 13374.0 + "f1-score": 0.9254835425856802, + "precision": 0.9130289234065345, + "recall": 0.9382826475849732, + "support": 14534.0 }, - "eval_accuracy": 0.9226995369944501, - "eval_loss": 0.7186141014099121, + "eval_accuracy": 0.9198914618728994, + "eval_loss": 0.7702658772468567, "eval_macro avg": { - "f1-score": 0.8913246003754525, - "precision": 0.8897378705706865, - "recall": 0.8929391336826483, - "support": 32613.0 - }, - "eval_runtime": 4.8784, - "eval_samples_per_second": 16.604, - "eval_steps_per_second": 2.255, + "f1-score": 0.8838453975635852, + "precision": 0.8901455730892207, + "recall": 0.8782847587159609, + "support": 32431.0 + }, + "eval_runtime": 4.8623, + "eval_samples_per_second": 16.453, + "eval_steps_per_second": 2.057, "eval_weighted avg": { - "f1-score": 0.922844232671956, - "precision": 0.923004457294864, - "recall": 0.9226995369944501, - "support": 32613.0 + "f1-score": 0.9186392999395205, + "precision": 0.9179023510964751, + "recall": 0.9198914618728994, + "support": 32431.0 }, "step": 3564 } @@ -2043,7 +2043,7 @@ "num_input_tokens_seen": 0, "num_train_epochs": 50, "save_steps": 500, - "total_flos": 6307056090165600.0, + "total_flos": 6326704239979200.0, "train_batch_size": 4, "trial_name": null, "trial_params": null