{ "best_metric": null, "best_model_checkpoint": null, "epoch": 47.0, "eval_steps": 500, "global_step": 3807, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "eval_Claim": { "f1-score": 0.5724454649827784, "precision": 0.6002889477486154, "recall": 0.5470704410796576, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.7860447185813415, "precision": 0.698526892771497, "recall": 0.8986337593653592, "support": 2269.0 }, "eval_O": { "f1-score": 0.9987820830889982, "precision": 0.9975671292124707, "recall": 1.0, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.8965445389824622, "precision": 0.9051258677512096, "recall": 0.8881243979633962, "support": 14534.0 }, "eval_accuracy": 0.8791279948197712, "eval_loss": 0.3157936632633209, "eval_macro avg": { "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": 0.8781740995293698, "precision": 0.8793943851252293, "recall": 0.8791279948197712, "support": 32431.0 }, "step": 81 }, { "epoch": 2.0, "eval_Claim": { "f1-score": 0.6090644692195831, "precision": 0.680108254397835, "recall": 0.5514592933947773, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8585858585858586, "precision": 0.8960229995208433, "recall": 0.8241516086381666, "support": 2269.0 }, "eval_O": { "f1-score": 0.999458141425088, "precision": 0.9992776523702032, "recall": 0.9996386956914461, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9105885478942475, "precision": 0.8801849235905997, "recall": 0.9431677445988716, "support": 14534.0 }, "eval_accuracy": 0.8990780426135487, "eval_loss": 0.24451124668121338, "eval_macro avg": { "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": 0.8949194640367888, "precision": 0.8938342965082798, "recall": 0.8990780426135487, "support": 32431.0 }, "step": 162 }, { "epoch": 3.0, "eval_Claim": { "f1-score": 0.6542320859402667, "precision": 0.7079713847726111, "recall": 0.60807548825982, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8915187376725839, "precision": 0.8866608544027899, "recall": 0.8964301454385192, "support": 2269.0 }, "eval_O": { "f1-score": 0.9995936611133684, "precision": 0.9992778479870013, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9157990498332153, "precision": 0.8973258501155497, "recall": 0.935048850970139, "support": 14534.0 }, "eval_accuracy": 0.9085442940396534, "eval_loss": 0.22665414214134216, "eval_macro avg": { "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": 0.9059516032151627, "precision": 0.9047762338408377, "recall": 0.9085442940396534, "support": 32431.0 }, "step": 243 }, { "epoch": 4.0, "eval_Claim": { "f1-score": 0.6861642294713161, "precision": 0.6424741478360781, "recall": 0.7362299758613122, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8995515695067264, "precision": 0.9155636695572797, "recall": 0.884089907448215, "support": 2269.0 }, "eval_O": { "f1-score": 0.9976481230212574, "precision": 0.9990941208442794, "recall": 0.9962063047601842, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9062532879739067, "precision": 0.9242435081193219, "recall": 0.8889500481629283, "support": 14534.0 }, "eval_accuracy": 0.9037649162838025, "eval_loss": 0.2452327460050583, "eval_macro avg": { "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": 0.9060584182508294, "precision": 0.9095955973231696, "recall": 0.9037649162838025, "support": 32431.0 }, "step": 324 }, { "epoch": 5.0, "eval_Claim": { "f1-score": 0.6640249759846301, "precision": 0.7332272606735614, "recall": 0.6067588325652842, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8848007008322383, "precision": 0.8794079233783195, "recall": 0.8902600264433671, "support": 2269.0 }, "eval_O": { "f1-score": 0.999367774566474, "precision": 0.9992775219001174, "recall": 0.9994580435371692, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9201314377682404, "precision": 0.8973839110529758, "recall": 0.9440621989816981, "support": 14534.0 }, "eval_accuracy": 0.9118127717307515, "eval_loss": 0.28411003947257996, "eval_macro avg": { "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": 0.9087220731552942, "precision": 0.9078434958291443, "recall": 0.9118127717307515, "support": 32431.0 }, "step": 405 }, { "epoch": 6.0, "eval_Claim": { "f1-score": 0.6496087442553721, "precision": 0.7484258729250143, "recall": 0.5738424402018872, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8865800865800866, "precision": 0.8711186729051468, "recall": 0.9026002644336713, "support": 2269.0 }, "eval_O": { "f1-score": 0.9986892655367232, "precision": 0.9994572100597069, "recall": 0.9979225002258152, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9191778088205947, "precision": 0.889647180015452, "recall": 0.9507362047612495, "support": 14534.0 }, "eval_accuracy": 0.9105177145323918, "eval_loss": 0.3082066774368286, "eval_macro avg": { "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": 0.9061618943658084, "precision": 0.9059932736781678, "recall": 0.9105177145323918, "support": 32431.0 }, "step": 486 }, { "epoch": 6.17, "grad_norm": 0.8612700700759888, "learning_rate": 1.7530864197530865e-05, "loss": 0.2223, "step": 500 }, { "epoch": 7.0, "eval_Claim": { "f1-score": 0.6977381768334475, "precision": 0.629950495049505, "recall": 0.7818740399385561, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8892209178228387, "precision": 0.8621688741721855, "recall": 0.9180255619215514, "support": 2269.0 }, "eval_O": { "f1-score": 0.9988698521766648, "precision": 0.9998190045248869, "recall": 0.9979225002258152, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9034227633516503, "precision": 0.9449996243143738, "recall": 0.8653502132929682, "support": 14534.0 }, "eval_accuracy": 0.9025623631710401, "eval_loss": 0.3512111008167267, "eval_macro avg": { "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": 0.9061105026169647, "precision": 0.9136494810618672, "recall": 0.9025623631710401, "support": 32431.0 }, "step": 567 }, { "epoch": 8.0, "eval_Claim": { "f1-score": 0.6987032799389779, "precision": 0.6177710335525206, "recall": 0.80403774412991, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9015364639688379, "precision": 0.8856292517006803, "recall": 0.9180255619215514, "support": 2269.0 }, "eval_O": { "f1-score": 0.998915499322187, "precision": 0.9994574554661362, "recall": 0.9983741306115076, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.8976577489773016, "precision": 0.9472075788830315, "recall": 0.8530342644832806, "support": 14534.0 }, "eval_accuracy": 0.9003114304215103, "eval_loss": 0.3705728054046631, "eval_macro avg": { "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": 0.9045397397569994, "precision": 0.9144455555782818, "recall": 0.9003114304215103, "support": 32431.0 }, "step": 648 }, { "epoch": 9.0, "eval_Claim": { "f1-score": 0.6657139454567107, "precision": 0.7278645833333334, "recall": 0.6133421110379635, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8882521489971346, "precision": 0.8884479717813051, "recall": 0.8880564125165271, "support": 2269.0 }, "eval_O": { "f1-score": 0.9987339482727438, "precision": 0.9999094612947035, "recall": 0.9975611959172613, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9178854152690192, "precision": 0.8955360649299646, "recall": 0.9413788358332187, "support": 14534.0 }, "eval_accuracy": 0.9107335573987851, "eval_loss": 0.41826051473617554, "eval_macro avg": { "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": 0.9079779452489347, "precision": 0.9071100542045434, "recall": 0.9107335573987851, "support": 32431.0 }, "step": 729 }, { "epoch": 10.0, "eval_Claim": { "f1-score": 0.6738416222952305, "precision": 0.6964871194379391, "recall": 0.652622339258284, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8951019550136641, "precision": 0.8557073954983923, "recall": 0.9382988100484795, "support": 2269.0 }, "eval_O": { "f1-score": 0.9989605459393501, "precision": 0.9996382054992764, "recall": 0.998283804534369, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9109121470961545, "precision": 0.9083259218717931, "recall": 0.9135131415990092, "support": 14534.0 }, "eval_accuracy": 0.9075267490980852, "eval_loss": 0.5459077954292297, "eval_macro avg": { "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": 0.9065515080853328, "precision": 0.9060496255197702, "recall": 0.9075267490980852, "support": 32431.0 }, "step": 810 }, { "epoch": 11.0, "eval_Claim": { "f1-score": 0.6581417175035196, "precision": 0.7070834383665239, "recall": 0.6155365371955234, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.881979417560762, "precision": 0.876414273281114, "recall": 0.8876156897311591, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548390010388, "precision": 0.9999096820809249, "recall": 1.0, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9173079519373565, "precision": 0.9002915065588976, "recall": 0.9349800467868447, "support": 14534.0 }, "eval_accuracy": 0.9089759797724399, "eval_loss": 0.5726844668388367, "eval_macro avg": { "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": 0.9066330640790278, "precision": 0.9054793272287636, "recall": 0.9089759797724399, "support": 32431.0 }, "step": 891 }, { "epoch": 12.0, "eval_Claim": { "f1-score": 0.7079179557694512, "precision": 0.7407673860911271, "recall": 0.6778582400702217, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8947480331703169, "precision": 0.8644207066557108, "recall": 0.9272807404142794, "support": 2269.0 }, "eval_O": { "f1-score": 0.999186477447347, "precision": 0.9999095431931253, "recall": 0.998464456688646, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9243840851702722, "precision": 0.9169374492282697, "recall": 0.9319526627218935, "support": 14534.0 }, "eval_accuracy": 0.9186272393697388, "eval_loss": 0.5660926699638367, "eval_macro avg": { "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": 0.917429533390557, "precision": 0.9168331226787283, "recall": 0.9186272393697388, "support": 32431.0 }, "step": 972 }, { "epoch": 12.35, "grad_norm": 2.9101791381835938, "learning_rate": 1.506172839506173e-05, "loss": 0.0294, "step": 1000 }, { "epoch": 13.0, "eval_Claim": { "f1-score": 0.6744648496900223, "precision": 0.7221943887775552, "recall": 0.6326530612244898, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9044719714465672, "precision": 0.863672814755413, "recall": 0.9493168796826796, "support": 2269.0 }, "eval_O": { "f1-score": 0.9995934041111363, "precision": 0.9999096167751266, "recall": 0.9992773913828923, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9161992180860105, "precision": 0.9055171023452725, "recall": 0.9271363698912893, "support": 14534.0 }, "eval_accuracy": 0.9119361105115475, "eval_loss": 0.5764271020889282, "eval_macro avg": { "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": 0.9098800911419986, "precision": 0.909053016531805, "recall": 0.9119361105115475, "support": 32431.0 }, "step": 1053 }, { "epoch": 14.0, "eval_Claim": { "f1-score": 0.7047413793103449, "precision": 0.6923565530383231, "recall": 0.717577353522054, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9112001687407719, "precision": 0.8737864077669902, "recall": 0.9519612163948876, "support": 2269.0 }, "eval_O": { "f1-score": 0.9992316384180792, "precision": 1.0, "recall": 0.998464456688646, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9145424153781864, "precision": 0.9258919757439007, "recall": 0.903467730838035, "support": 14534.0 }, "eval_accuracy": 0.9131694983195091, "eval_loss": 0.5468704700469971, "eval_macro avg": { "f1-score": 0.8824289004618456, "precision": 0.8730087341373035, "recall": 0.8928676893609057, "support": 32431.0 }, "eval_runtime": 4.8411, "eval_samples_per_second": 16.525, "eval_steps_per_second": 2.066, "eval_weighted avg": { "f1-score": 0.9137390947680665, "precision": 0.9147298617643856, "recall": 0.9131694983195091, "support": 32431.0 }, "step": 1134 }, { "epoch": 15.0, "eval_Claim": { "f1-score": 0.6868638486408813, "precision": 0.7559304164470216, "recall": 0.6293614219881501, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9058546000428909, "precision": 0.8822055137844611, "recall": 0.9308065226972234, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9218029420675262, "precision": 0.9023924075660713, "recall": 0.9420668776661622, "support": 14534.0 }, "eval_accuracy": 0.9170855046097869, "eval_loss": 0.6552095413208008, "eval_macro avg": { "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": 0.9143537536055225, "precision": 0.9137205288178378, "recall": 0.9170855046097869, "support": 32431.0 }, "step": 1215 }, { "epoch": 16.0, "eval_Claim": { "f1-score": 0.7261954261954263, "precision": 0.6899071696622556, "recall": 0.7665130568356375, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9104052573932091, "precision": 0.9050522648083623, "recall": 0.9158219479947113, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9186291000841041, "precision": 0.9360805599200114, "recall": 0.9018164304389706, "support": 14534.0 }, "eval_accuracy": 0.9172705127809812, "eval_loss": 0.604762613773346, "eval_macro avg": { "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": 0.918776399254959, "precision": 0.92113920073012, "recall": 0.9172705127809812, "support": 32431.0 }, "step": 1296 }, { "epoch": 17.0, "eval_Claim": { "f1-score": 0.6726977499420088, "precision": 0.7134071340713407, "recall": 0.6363835856923414, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.898166556218246, "precision": 0.9003542958370239, "recall": 0.8959894226531512, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9172189909373732, "precision": 0.9018486500864477, "recall": 0.9331223338378973, "support": 14534.0 }, "eval_accuracy": 0.9116277635595572, "eval_loss": 0.6666390299797058, "eval_macro avg": { "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": 0.9097710324765249, "precision": 0.9087714991080675, "recall": 0.9116277635595572, "support": 32431.0 }, "step": 1377 }, { "epoch": 18.0, "eval_Claim": { "f1-score": 0.7242054854157598, "precision": 0.7184193478730296, "recall": 0.7300855826201448, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.91550204257149, "precision": 0.8937867338371117, "recall": 0.9382988100484795, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9218198185721211, "precision": 0.9277948146083078, "recall": 0.9159212880143113, "support": 14534.0 }, "eval_accuracy": 0.9200456353488946, "eval_loss": 0.5898464918136597, "eval_macro avg": { "f1-score": 0.8903705453702524, "precision": 0.8850002240796123, "recall": 0.8960538386514494, "support": 32431.0 }, "eval_runtime": 4.8354, "eval_samples_per_second": 16.545, "eval_steps_per_second": 2.068, "eval_weighted avg": { "f1-score": 0.9202833015380038, "precision": 0.9206441029524512, "recall": 0.9200456353488946, "support": 32431.0 }, "step": 1458 }, { "epoch": 18.52, "grad_norm": 0.2594422399997711, "learning_rate": 1.2592592592592593e-05, "loss": 0.0093, "step": 1500 }, { "epoch": 19.0, "eval_Claim": { "f1-score": 0.7198604752561586, "precision": 0.7151830192765866, "recall": 0.7245995172262454, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9025769956002515, "precision": 0.860223642172524, "recall": 0.9493168796826796, "support": 2269.0 }, "eval_O": { "f1-score": 0.9975098474215601, "precision": 1.0, "recall": 0.9950320657573841, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9212571111419453, "precision": 0.9289911851126347, "recall": 0.9136507499655979, "support": 14534.0 }, "eval_accuracy": 0.9173630168665783, "eval_loss": 0.6108261346817017, "eval_macro avg": { "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": 0.9176816552958715, "precision": 0.9183773040411919, "recall": 0.9173630168665783, "support": 32431.0 }, "step": 1539 }, { "epoch": 20.0, "eval_Claim": { "f1-score": 0.7150392937883518, "precision": 0.7018174133558749, "recall": 0.7287689269256089, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8972902097902098, "precision": 0.8899003034243607, "recall": 0.9048038783605112, "support": 2269.0 }, "eval_O": { "f1-score": 0.9992316384180792, "precision": 1.0, "recall": 0.998464456688646, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9203380437794403, "precision": 0.9266285395452644, "recall": 0.9141323792486583, "support": 14534.0 }, "eval_accuracy": 0.9162221331442139, "eval_loss": 0.6273905038833618, "eval_macro avg": { "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": 0.9168102169228345, "precision": 0.9175167258142971, "recall": 0.9162221331442139, "support": 32431.0 }, "step": 1620 }, { "epoch": 21.0, "eval_Claim": { "f1-score": 0.7209400804573365, "precision": 0.6964614440580896, "recall": 0.7472021066491112, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9076372817168105, "precision": 0.8683574879227053, "recall": 0.9506390480387836, "support": 2269.0 }, "eval_O": { "f1-score": 0.999186477447347, "precision": 0.9999095431931253, "recall": 0.998464456688646, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9181764025650909, "precision": 0.9355852317360566, "recall": 0.9014036053392046, "support": 14534.0 }, "eval_accuracy": 0.916314637229811, "eval_loss": 0.6812385320663452, "eval_macro avg": { "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": 0.917379120765938, "precision": 0.9192399941849548, "recall": 0.916314637229811, "support": 32431.0 }, "step": 1701 }, { "epoch": 22.0, "eval_Claim": { "f1-score": 0.7066195738862937, "precision": 0.7348341232227488, "recall": 0.6804915514592934, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9032682605834613, "precision": 0.8991266375545851, "recall": 0.9074482150727192, "support": 2269.0 }, "eval_O": { "f1-score": 0.9992316384180792, "precision": 1.0, "recall": 0.998464456688646, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9243903268596307, "precision": 0.914037801843008, "recall": 0.9349800467868447, "support": 14534.0 }, "eval_accuracy": 0.9189664210169283, "eval_loss": 0.6876958012580872, "eval_macro avg": { "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": 0.917861415342362, "precision": 0.9171589791287257, "recall": 0.9189664210169283, "support": 32431.0 }, "step": 1782 }, { "epoch": 23.0, "eval_Claim": { "f1-score": 0.7040384395378104, "precision": 0.7354206500956023, "recall": 0.6752249286811499, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.90536013400335, "precision": 0.8623853211009175, "recall": 0.9528426619656236, "support": 2269.0 }, "eval_O": { "f1-score": 0.9991412429378531, "precision": 0.9999095350099512, "recall": 0.9983741306115076, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9212183436002738, "precision": 0.9164510418085251, "recall": 0.9260355029585798, "support": 14534.0 }, "eval_accuracy": 0.9173630168665783, "eval_loss": 0.6735763549804688, "eval_macro avg": { "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": 0.9161926989447983, "precision": 0.9157214424409951, "recall": 0.9173630168665783, "support": 32431.0 }, "step": 1863 }, { "epoch": 24.0, "eval_Claim": { "f1-score": 0.7284623773173391, "precision": 0.724040754389768, "recall": 0.7329383366249725, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8918801563178463, "precision": 0.8789045785194695, "recall": 0.9052446011458792, "support": 2269.0 }, "eval_O": { "f1-score": 0.9992316384180792, "precision": 1.0, "recall": 0.998464456688646, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9255205276060909, "precision": 0.9289526582103002, "recall": 0.9221136645108022, "support": 14534.0 }, "eval_accuracy": 0.920415651691283, "eval_loss": 0.6322010159492493, "eval_macro avg": { "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": 0.9206403732624274, "precision": 0.9209116629411165, "recall": 0.920415651691283, "support": 32431.0 }, "step": 1944 }, { "epoch": 24.69, "grad_norm": 0.6325204968452454, "learning_rate": 1.0123456790123458e-05, "loss": 0.004, "step": 2000 }, { "epoch": 25.0, "eval_Claim": { "f1-score": 0.7145744441949248, "precision": 0.7316624511381927, "recall": 0.6982664033355278, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9116772823779193, "precision": 0.8795575583777141, "recall": 0.9462318201851035, "support": 2269.0 }, "eval_O": { "f1-score": 0.9992316384180792, "precision": 1.0, "recall": 0.998464456688646, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9222897565330862, "precision": 0.920614245561116, "recall": 0.9239713774597496, "support": 14534.0 }, "eval_accuracy": 0.9192439332737196, "eval_loss": 0.6634725332260132, "eval_macro avg": { "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": 0.9186261381483827, "precision": 0.9182914290271974, "recall": 0.9192439332737196, "support": 32431.0 }, "step": 2025 }, { "epoch": 26.0, "eval_Claim": { "f1-score": 0.714535519125683, "precision": 0.7117352492924015, "recall": 0.717357910906298, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9060118543607113, "precision": 0.8716904276985743, "recall": 0.9431467606875276, "support": 2269.0 }, "eval_O": { "f1-score": 0.9990506758283985, "precision": 1.0, "recall": 0.9981031523800922, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9199431877230055, "precision": 0.926393637061327, "recall": 0.9135819457823036, "support": 14534.0 }, "eval_accuracy": 0.9169313311337918, "eval_loss": 0.6303670406341553, "eval_macro avg": { "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": 0.9171108994684886, "precision": 0.9175310114557944, "recall": 0.9169313311337918, "support": 32431.0 }, "step": 2106 }, { "epoch": 27.0, "eval_Claim": { "f1-score": 0.7246061922868008, "precision": 0.717512908777969, "recall": 0.7318411235461927, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9032534246575342, "precision": 0.8780690803162713, "recall": 0.9299250771264874, "support": 2269.0 }, "eval_O": { "f1-score": 0.9991864039052614, "precision": 1.0, "recall": 0.9983741306115076, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9229063442015175, "precision": 0.9295735324911008, "recall": 0.9163341131140773, "support": 14534.0 }, "eval_accuracy": 0.9193672720545157, "eval_loss": 0.6543757319450378, "eval_macro avg": { "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": 0.9197072530313251, "precision": 0.9202142021449996, "recall": 0.9193672720545157, "support": 32431.0 }, "step": 2187 }, { "epoch": 28.0, "eval_Claim": { "f1-score": 0.7182029107783169, "precision": 0.6913705583756345, "recall": 0.7472021066491112, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9016903914590747, "precision": 0.9101930848675348, "recall": 0.8933450859409432, "support": 2269.0 }, "eval_O": { "f1-score": 0.9992316384180792, "precision": 1.0, "recall": 0.998464456688646, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9179735039465906, "precision": 0.9279437609841827, "recall": 0.9082152194853447, "support": 14534.0 }, "eval_accuracy": 0.9153587616786408, "eval_loss": 0.67218017578125, "eval_macro avg": { "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": 0.9165029304654106, "precision": 0.9180579805194509, "recall": 0.9153587616786408, "support": 32431.0 }, "step": 2268 }, { "epoch": 29.0, "eval_Claim": { "f1-score": 0.7078925956061839, "precision": 0.7525951557093425, "recall": 0.6682027649769585, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9066271437645564, "precision": 0.8724531377343113, "recall": 0.9435874834728956, "support": 2269.0 }, "eval_O": { "f1-score": 0.9992316384180792, "precision": 1.0, "recall": 0.998464456688646, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9239400224405835, "precision": 0.9132889695503126, "recall": 0.9348424384202559, "support": 14534.0 }, "eval_accuracy": 0.9197064537017051, "eval_loss": 0.7057467103004456, "eval_macro avg": { "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": 0.9180734884048335, "precision": 0.9174528746424986, "recall": 0.9197064537017051, "support": 32431.0 }, "step": 2349 }, { "epoch": 30.0, "eval_Claim": { "f1-score": 0.7153448663221748, "precision": 0.7327963176064441, "recall": 0.6987052885670397, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8877374784110534, "precision": 0.8700804062632247, "recall": 0.9061260467166152, "support": 2269.0 }, "eval_O": { "f1-score": 0.9992316384180792, "precision": 1.0, "recall": 0.998464456688646, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9262746978050201, "precision": 0.9220124071170496, "recall": 0.9305765790560067, "support": 14534.0 }, "eval_accuracy": 0.9194597761401129, "eval_loss": 0.6960250735282898, "eval_macro avg": { "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": 0.9188453276546991, "precision": 0.918414282204774, "recall": 0.9194597761401129, "support": 32431.0 }, "step": 2430 }, { "epoch": 30.86, "grad_norm": 0.009368549101054668, "learning_rate": 7.654320987654322e-06, "loss": 0.002, "step": 2500 }, { "epoch": 31.0, "eval_Claim": { "f1-score": 0.723667100130039, "precision": 0.7148362235067437, "recall": 0.7327188940092166, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8953336240732664, "precision": 0.8860595597755718, "recall": 0.9048038783605112, "support": 2269.0 }, "eval_O": { "f1-score": 0.9991864039052614, "precision": 1.0, "recall": 0.9983741306115076, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9236620107868897, "precision": 0.928283530229326, "recall": 0.9190862804458511, "support": 14534.0 }, "eval_accuracy": 0.9189664210169283, "eval_loss": 0.6865902543067932, "eval_macro avg": { "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": 0.9193598504741339, "precision": 0.9198190200735107, "recall": 0.9189664210169283, "support": 32431.0 }, "step": 2511 }, { "epoch": 32.0, "eval_Claim": { "f1-score": 0.7149137639499492, "precision": 0.7350486787204451, "recall": 0.695852534562212, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8920269389528569, "precision": 0.8796058269065982, "recall": 0.9048038783605112, "support": 2269.0 }, "eval_O": { "f1-score": 0.999322095177837, "precision": 1.0, "recall": 0.9986451088429229, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9260107310071427, "precision": 0.919942961906702, "recall": 0.9321590752717765, "support": 14534.0 }, "eval_accuracy": 0.9197372883969042, "eval_loss": 0.6847501397132874, "eval_macro avg": { "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": 0.9189974415520827, "precision": 0.9184697807200564, "recall": 0.9197372883969042, "support": 32431.0 }, "step": 2592 }, { "epoch": 33.0, "eval_Claim": { "f1-score": 0.7068796869905445, "precision": 0.7002583979328165, "recall": 0.7136273864384464, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8840610884061089, "precision": 0.8634453781512605, "recall": 0.9056853239312472, "support": 2269.0 }, "eval_O": { "f1-score": 0.9987338337704622, "precision": 1.0, "recall": 0.9974708698401229, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9202020901100422, "precision": 0.9256474519632414, "recall": 0.9148204210816018, "support": 14534.0 }, "eval_accuracy": 0.9141253738706793, "eval_loss": 0.7124452590942383, "eval_macro avg": { "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": 0.9145072552354695, "precision": 0.9150070965816288, "recall": 0.9141253738706793, "support": 32431.0 }, "step": 2673 }, { "epoch": 34.0, "eval_Claim": { "f1-score": 0.709584487534626, "precision": 0.7166517457475381, "recall": 0.7026552556506473, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8878668418747262, "precision": 0.8824553765781454, "recall": 0.8933450859409432, "support": 2269.0 }, "eval_O": { "f1-score": 0.9992316384180792, "precision": 1.0, "recall": 0.998464456688646, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9230083030261442, "precision": 0.9205447577333699, "recall": 0.9254850694922251, "support": 14534.0 }, "eval_accuracy": 0.9168388270481946, "eval_loss": 0.6949135661125183, "eval_macro avg": { "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": 0.9165811266694699, "precision": 0.9163538208419149, "recall": 0.9168388270481946, "support": 32431.0 }, "step": 2754 }, { "epoch": 35.0, "eval_Claim": { "f1-score": 0.7161833075049812, "precision": 0.7225820862184499, "recall": 0.7098968619705947, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9034467994005567, "precision": 0.8784346378018318, "recall": 0.9299250771264874, "support": 2269.0 }, "eval_O": { "f1-score": 0.9995933673699905, "precision": 1.0, "recall": 0.9991870653057537, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9223401323042998, "precision": 0.9237405106970324, "recall": 0.9209439933947984, "support": 14534.0 }, "eval_accuracy": 0.9186272393697388, "eval_loss": 0.6991766691207886, "eval_macro avg": { "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": 0.9184224283310388, "precision": 0.9183379896562086, "recall": 0.9186272393697388, "support": 32431.0 }, "step": 2835 }, { "epoch": 36.0, "eval_Claim": { "f1-score": 0.7000227946204696, "precision": 0.7282428266540194, "recall": 0.673908272986614, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9057333047025982, "precision": 0.8831658291457286, "recall": 0.9294843543411194, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9210652099692729, "precision": 0.9141366223908919, "recall": 0.9280996284574102, "support": 14534.0 }, "eval_accuracy": 0.9169930005241899, "eval_loss": 0.7325878739356995, "eval_macro avg": { "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": 0.9158636638576222, "precision": 0.9151604174161526, "recall": 0.9169930005241899, "support": 32431.0 }, "step": 2916 }, { "epoch": 37.0, "eval_Claim": { "f1-score": 0.7043226381461676, "precision": 0.7153202081918986, "recall": 0.6936581084046521, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8989154013015185, "precision": 0.8850918410935498, "recall": 0.9131776112825033, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9208855328642527, "precision": 0.9187726868022739, "recall": 0.9230081188936288, "support": 14534.0 }, "eval_accuracy": 0.91634547192501, "eval_loss": 0.7374063730239868, "eval_macro avg": { "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": 0.9159103209784385, "precision": 0.9155570227916499, "recall": 0.91634547192501, "support": 32431.0 }, "step": 2997 }, { "epoch": 37.04, "grad_norm": 0.13183258473873138, "learning_rate": 5.185185185185185e-06, "loss": 0.0012, "step": 3000 }, { "epoch": 38.0, "eval_Claim": { "f1-score": 0.6968245125348189, "precision": 0.7077863286555002, "recall": 0.6861970594689488, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8926233085988651, "precision": 0.8841331603977518, "recall": 0.9012780960775673, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9194212042243862, "precision": 0.9164046479835953, "recall": 0.922457685427274, "support": 14534.0 }, "eval_accuracy": 0.9142178779562764, "eval_loss": 0.7381414175033569, "eval_macro avg": { "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": 0.9137602710507441, "precision": 0.9133700963405134, "recall": 0.9142178779562764, "support": 32431.0 }, "step": 3078 }, { "epoch": 39.0, "eval_Claim": { "f1-score": 0.6997294860234445, "precision": 0.719351100811124, "recall": 0.6811498793065613, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9017094017094017, "precision": 0.8751555371215264, "recall": 0.9299250771264874, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9202920909184409, "precision": 0.9171165015374103, "recall": 0.9234897481766892, "support": 14534.0 }, "eval_accuracy": 0.9159754555826216, "eval_loss": 0.7733151316642761, "eval_macro avg": { "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": 0.9151944475074327, "precision": 0.9146860137975935, "recall": 0.9159754555826216, "support": 32431.0 }, "step": 3159 }, { "epoch": 40.0, "eval_Claim": { "f1-score": 0.6766934825696631, "precision": 0.7218905472636816, "recall": 0.6368224709238535, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8926333981421473, "precision": 0.8754237288135593, "recall": 0.9105332745702953, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999096657633243, "precision": 1.0, "recall": 0.999819347845723, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9188236888467272, "precision": 0.9050861033239888, "recall": 0.9329847254713086, "support": 14534.0 }, "eval_accuracy": 0.9126144738059264, "eval_loss": 0.8124102354049683, "eval_macro avg": { "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": 0.9106490945027469, "precision": 0.9096701640488858, "recall": 0.9126144738059264, "support": 32431.0 }, "step": 3240 }, { "epoch": 41.0, "eval_Claim": { "f1-score": 0.6830284788145403, "precision": 0.7228620436167607, "recall": 0.6473557164801405, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9045826876245294, "precision": 0.9088078291814946, "recall": 0.9003966505068312, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9190962592166678, "precision": 0.9038717402873869, "recall": 0.9348424384202559, "support": 14534.0 }, "eval_accuracy": 0.9142487126514754, "eval_loss": 0.8041703104972839, "eval_macro avg": { "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": 0.9125128397228092, "precision": 0.911598136675134, "recall": 0.9142487126514754, "support": 32431.0 }, "step": 3321 }, { "epoch": 42.0, "eval_Claim": { "f1-score": 0.6939400568803326, "precision": 0.6918211559432933, "recall": 0.696071977177968, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8969750109601051, "precision": 0.8922808547754034, "recall": 0.9017188188629353, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9167729262156668, "precision": 0.9183870745011393, "recall": 0.9151644419980735, "support": 14534.0 }, "eval_accuracy": 0.9123677962443342, "eval_loss": 0.7728341221809387, "eval_macro avg": { "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": 0.9124725980114123, "precision": 0.9125852427590434, "recall": 0.9123677962443342, "support": 32431.0 }, "step": 3402 }, { "epoch": 43.0, "eval_Claim": { "f1-score": 0.7087594486438418, "precision": 0.7181797702185176, "recall": 0.6995830590300637, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9011231006386258, "precision": 0.9005281690140845, "recall": 0.9017188188629353, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9225603070175438, "precision": 0.9189078498293516, "recall": 0.9262419155084629, "support": 14534.0 }, "eval_accuracy": 0.9178255372945638, "eval_loss": 0.7432543039321899, "eval_macro avg": { "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": 0.9174387654537162, "precision": 0.9170993900218476, "recall": 0.9178255372945638, "support": 32431.0 }, "step": 3483 }, { "epoch": 43.21, "grad_norm": 0.0030191524419933558, "learning_rate": 2.7160493827160496e-06, "loss": 0.0009, "step": 3500 }, { "epoch": 44.0, "eval_Claim": { "f1-score": 0.7071469411092054, "precision": 0.7382999044890163, "recall": 0.6785165679174896, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.9027962716378164, "precision": 0.9092534644613322, "recall": 0.8964301454385192, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9254835425856802, "precision": 0.9130289234065345, "recall": 0.9382826475849732, "support": 14534.0 }, "eval_accuracy": 0.9198914618728994, "eval_loss": 0.7702658772468567, "eval_macro avg": { "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": 0.9186392999395205, "precision": 0.9179023510964751, "recall": 0.9198914618728994, "support": 32431.0 }, "step": 3564 }, { "epoch": 45.0, "eval_Claim": { "f1-score": 0.7032942490228922, "precision": 0.7160072760345612, "recall": 0.6910247970155804, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8986205386468141, "precision": 0.8929503916449086, "recall": 0.9043631555751432, "support": 2269.0 }, "eval_O": { "f1-score": 0.9999548349216385, "precision": 1.0, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9215384088496181, "precision": 0.9174224343675418, "recall": 0.9256914820421082, "support": 14534.0 }, "eval_accuracy": 0.9165613147914032, "eval_loss": 0.7518438696861267, "eval_macro avg": { "f1-score": 0.8808520078602406, "precision": 0.8815950255117528, "recall": 0.8802472771389234, "support": 32431.0 }, "eval_runtime": 4.8413, "eval_samples_per_second": 16.524, "eval_steps_per_second": 2.066, "eval_weighted avg": { "f1-score": 0.9160377757893605, "precision": 0.9155982626693486, "recall": 0.9165613147914032, "support": 32431.0 }, "step": 3645 }, { "epoch": 46.0, "eval_Claim": { "f1-score": 0.7006939780613387, "precision": 0.7151016678090016, "recall": 0.6868553873162168, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8954833078769365, "precision": 0.8867761452031115, "recall": 0.9043631555751432, "support": 2269.0 }, "eval_O": { "f1-score": 0.9997289483194796, "precision": 1.0, "recall": 0.9994580435371692, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.9213598548392617, "precision": 0.9169335604770017, "recall": 0.9258290904086969, "support": 14534.0 }, "eval_accuracy": 0.9158829514970245, "eval_loss": 0.750240683555603, "eval_macro avg": { "f1-score": 0.8793165222742542, "precision": 0.8797028433722787, "recall": 0.8791264192093065, "support": 32431.0 }, "eval_runtime": 4.8402, "eval_samples_per_second": 16.528, "eval_steps_per_second": 2.066, "eval_weighted avg": { "f1-score": 0.9152957787511048, "precision": 0.9148199482484113, "recall": 0.9158829514970245, "support": 32431.0 }, "step": 3726 }, { "epoch": 47.0, "eval_Claim": { "f1-score": 0.7070640176600441, "precision": 0.7113035753941817, "recall": 0.7028746982664034, "support": 4557.0 }, "eval_MajorClaim": { "f1-score": 0.8924216027874564, "precision": 0.8820490744726647, "recall": 0.9030409872190392, "support": 2269.0 }, "eval_O": { "f1-score": 0.9995029596493606, "precision": 1.0, "recall": 0.9990064131514769, "support": 11071.0 }, "eval_Premise": { "f1-score": 0.921558513016266, "precision": 0.9212100378136817, "recall": 0.9219072519609193, "support": 14534.0 }, "eval_accuracy": 0.9161296290586167, "eval_loss": 0.7462829351425171, "eval_macro avg": { "f1-score": 0.8801367732782818, "precision": 0.8786406719201321, "recall": 0.8817073376494596, "support": 32431.0 }, "eval_runtime": 4.8323, "eval_samples_per_second": 16.555, "eval_steps_per_second": 2.069, "eval_weighted avg": { "f1-score": 0.9159885307162297, "precision": 0.9158720493550557, "recall": 0.9161296290586167, "support": 32431.0 }, "step": 3807 } ], "logging_steps": 500, "max_steps": 4050, "num_input_tokens_seen": 0, "num_train_epochs": 50, "save_steps": 500, "total_flos": 6758070438159600.0, "train_batch_size": 4, "trial_name": null, "trial_params": null }