{ "best_metric": null, "best_model_checkpoint": null, "epoch": 36.0, "eval_steps": 500, "global_step": 2916, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "eval_B-Claim": { "f1-score": 0.18764302059496568, "precision": 0.3416666666666667, "recall": 0.12933753943217666, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.19540229885057472, "precision": 0.8947368421052632, "recall": 0.10967741935483871, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8476339053562143, "precision": 0.740909090909091, "recall": 0.9902794653705954, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.4315998169894769, "precision": 0.6394035246272028, "recall": 0.3257366482504604, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.7779799818016379, "precision": 0.7493426818580193, "recall": 0.8088930936613056, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.8948683715038855, "precision": 0.8376057421174058, "recall": 0.9605350187403542, "support": 13607.0 }, "eval_O": { "f1-score": 0.9980147987727848, "precision": 0.9970251509961237, "recall": 0.9990064131514769, "support": 11071.0 }, "eval_accuracy": 0.8673183065585397, "eval_loss": 0.3193369209766388, "eval_macro avg": { "f1-score": 0.6190203134099342, "precision": 0.7429556713256817, "recall": 0.6176379425658869, "support": 32431.0 }, "eval_runtime": 4.9083, "eval_samples_per_second": 16.299, "eval_steps_per_second": 2.037, "eval_weighted avg": { "f1-score": 0.8489527906249091, "precision": 0.8526967051825685, "recall": 0.8673183065585397, "support": 32431.0 }, "step": 81 }, { "epoch": 2.0, "eval_B-Claim": { "f1-score": 0.583941605839416, "precision": 0.6926406926406926, "recall": 0.5047318611987381, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.8117647058823529, "precision": 0.745945945945946, "recall": 0.8903225806451613, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8943661971830986, "precision": 0.8649262202043133, "recall": 0.9258809234507898, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6197740770932805, "precision": 0.6271506009898656, "recall": 0.6125690607734806, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.8298737427776589, "precision": 0.7577178585384916, "recall": 0.9172185430463576, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9013697613125952, "precision": 0.9106660666066607, "recall": 0.8922613360770192, "support": 13607.0 }, "eval_O": { "f1-score": 0.9961487925331883, "precision": 0.9993636363636363, "recall": 0.9929545659831993, "support": 11071.0 }, "eval_accuracy": 0.887854213561099, "eval_loss": 0.26407390832901, "eval_macro avg": { "f1-score": 0.8053198403745129, "precision": 0.7997730030413723, "recall": 0.8194198387392494, "support": 32431.0 }, "eval_runtime": 4.9329, "eval_samples_per_second": 16.218, "eval_steps_per_second": 2.027, "eval_weighted avg": { "f1-score": 0.8876368199002541, "precision": 0.8889201304482091, "recall": 0.887854213561099, "support": 32431.0 }, "step": 162 }, { "epoch": 3.0, "eval_B-Claim": { "f1-score": 0.5895522388059701, "precision": 0.7214611872146118, "recall": 0.49842271293375395, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.8668941979522183, "precision": 0.9202898550724637, "recall": 0.8193548387096774, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8929178470254956, "precision": 0.8365180467091295, "recall": 0.9574726609963548, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6548140728684112, "precision": 0.7178149876475433, "recall": 0.6019797421731123, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.8815622622368754, "precision": 0.9502460360852925, "recall": 0.8221381267738883, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9169943670953343, "precision": 0.8852257181942544, "recall": 0.9511280958330272, "support": 13607.0 }, "eval_O": { "f1-score": 0.9984170774727511, "precision": 0.9998188405797102, "recall": 0.9970192394544305, "support": 11071.0 }, "eval_accuracy": 0.9067250470229101, "eval_loss": 0.25323203206062317, "eval_macro avg": { "f1-score": 0.8287360090652937, "precision": 0.8616249530718579, "recall": 0.8067879166963207, "support": 32431.0 }, "eval_runtime": 4.9025, "eval_samples_per_second": 16.318, "eval_steps_per_second": 2.04, "eval_weighted avg": { "f1-score": 0.9033110548907601, "precision": 0.9034896128542845, "recall": 0.9067250470229101, "support": 32431.0 }, "step": 243 }, { "epoch": 4.0, "eval_B-Claim": { "f1-score": 0.6806136680613668, "precision": 0.61, "recall": 0.7697160883280757, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.8861538461538462, "precision": 0.8470588235294118, "recall": 0.9290322580645162, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8687258687258688, "precision": 0.9233926128590971, "recall": 0.8201701093560145, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.656155854589275, "precision": 0.5819380121125757, "recall": 0.7520718232044199, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.8818897637795275, "precision": 0.8408408408408409, "recall": 0.9271523178807947, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.8829758349171871, "precision": 0.9349433218334154, "recall": 0.836481222899978, "support": 13607.0 }, "eval_O": { "f1-score": 0.9971017118014671, "precision": 0.9998183634547271, "recall": 0.9943997832174148, "support": 11071.0 }, "eval_accuracy": 0.8843698930036077, "eval_loss": 0.2791365385055542, "eval_macro avg": { "f1-score": 0.8362309354326484, "precision": 0.8197131392328668, "recall": 0.8612890861358877, "support": 32431.0 }, "eval_runtime": 4.9234, "eval_samples_per_second": 16.249, "eval_steps_per_second": 2.031, "eval_weighted avg": { "f1-score": 0.889158243622726, "precision": 0.8997827813567029, "recall": 0.8843698930036077, "support": 32431.0 }, "step": 324 }, { "epoch": 5.0, "eval_B-Claim": { "f1-score": 0.6078799249530958, "precision": 0.75, "recall": 0.5110410094637224, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.8971962616822431, "precision": 0.8674698795180723, "recall": 0.9290322580645162, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.893739230327398, "precision": 0.8474945533769063, "recall": 0.945321992709599, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6119813135476778, "precision": 0.7590320381731425, "recall": 0.5126611418047882, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.8889393595275948, "precision": 0.854958497160332, "recall": 0.9257332071901608, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9137743463765573, "precision": 0.874462654486835, "recall": 0.956786947894466, "support": 13607.0 }, "eval_O": { "f1-score": 0.9976913675252366, "precision": 1.0, "recall": 0.9953933700659381, "support": 11071.0 }, "eval_accuracy": 0.9036724121982054, "eval_loss": 0.3500836193561554, "eval_macro avg": { "f1-score": 0.8301716862771149, "precision": 0.8504882318164697, "recall": 0.8251385610275985, "support": 32431.0 }, "eval_runtime": 4.9239, "eval_samples_per_second": 16.247, "eval_steps_per_second": 2.031, "eval_weighted avg": { "f1-score": 0.8968007164885599, "precision": 0.8986502613295283, "recall": 0.9036724121982054, "support": 32431.0 }, "step": 405 }, { "epoch": 6.0, "eval_B-Claim": { "f1-score": 0.672077922077922, "precision": 0.6923076923076923, "recall": 0.6529968454258676, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.8910256410256411, "precision": 0.8853503184713376, "recall": 0.896774193548387, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8907563025210083, "precision": 0.8801897983392646, "recall": 0.9015795868772782, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6666666666666667, "precision": 0.6998228296633764, "recall": 0.6365101289134438, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.8853370396108408, "precision": 0.867453472537449, "recall": 0.9039735099337748, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9098497495826378, "precision": 0.8987595898759589, "recall": 0.9212170206511354, "support": 13607.0 }, "eval_O": { "f1-score": 0.998190209030857, "precision": 1.0, "recall": 0.9963869569144612, "support": 11071.0 }, "eval_accuracy": 0.9043816101877833, "eval_loss": 0.30672115087509155, "eval_macro avg": { "f1-score": 0.8448433615022248, "precision": 0.8462691001707254, "recall": 0.8442054631806213, "support": 32431.0 }, "eval_runtime": 4.9018, "eval_samples_per_second": 16.321, "eval_steps_per_second": 2.04, "eval_weighted avg": { "f1-score": 0.9029367568413356, "precision": 0.9020793621628027, "recall": 0.9043816101877833, "support": 32431.0 }, "step": 486 }, { "epoch": 6.17, "grad_norm": 0.43724608421325684, "learning_rate": 1.7530864197530865e-05, "loss": 0.2434, "step": 500 }, { "epoch": 7.0, "eval_B-Claim": { "f1-score": 0.6763285024154589, "precision": 0.6907894736842105, "recall": 0.6624605678233438, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.8952380952380952, "precision": 0.88125, "recall": 0.9096774193548387, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8941034897713597, "precision": 0.8855780691299165, "recall": 0.9027946537059538, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6960393719240684, "precision": 0.7088305489260143, "recall": 0.6837016574585635, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9140037593984962, "precision": 0.9080298786181139, "recall": 0.9200567644276254, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9164355568530141, "precision": 0.9101253895774444, "recall": 0.9228338355258323, "support": 13607.0 }, "eval_O": { "f1-score": 0.9967376529225193, "precision": 1.0, "recall": 0.9934965224460302, "support": 11071.0 }, "eval_accuracy": 0.9116277635595572, "eval_loss": 0.34577852487564087, "eval_macro avg": { "f1-score": 0.855555204074716, "precision": 0.8549433371336713, "recall": 0.8564316315345982, "support": 32431.0 }, "eval_runtime": 4.8989, "eval_samples_per_second": 16.33, "eval_steps_per_second": 2.041, "eval_weighted avg": { "f1-score": 0.9111536914902465, "precision": 0.910801887328957, "recall": 0.9116277635595572, "support": 32431.0 }, "step": 567 }, { "epoch": 8.0, "eval_B-Claim": { "f1-score": 0.701923076923077, "precision": 0.7133550488599348, "recall": 0.6908517350157729, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9260450160771705, "precision": 0.9230769230769231, "recall": 0.9290322580645162, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8985507246376812, "precision": 0.8931572629051621, "recall": 0.9040097205346294, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6915371329879102, "precision": 0.6917760884588804, "recall": 0.6912983425414365, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9147286821705426, "precision": 0.9085394307046197, "recall": 0.9210028382213813, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.912640635340834, "precision": 0.9131778382753293, "recall": 0.9121040640846623, "support": 13607.0 }, "eval_O": { "f1-score": 0.9993222177036736, "precision": 0.9998191681735985, "recall": 0.9988257609971999, "support": 11071.0 }, "eval_accuracy": 0.9104252104467947, "eval_loss": 0.3896510601043701, "eval_macro avg": { "f1-score": 0.8635353551201269, "precision": 0.863271680064921, "recall": 0.8638749599227997, "support": 32431.0 }, "eval_runtime": 4.9786, "eval_samples_per_second": 16.069, "eval_steps_per_second": 2.009, "eval_weighted avg": { "f1-score": 0.9103982293589928, "precision": 0.9103825180220693, "recall": 0.9104252104467947, "support": 32431.0 }, "step": 648 }, { "epoch": 9.0, "eval_B-Claim": { "f1-score": 0.6862745098039215, "precision": 0.6171284634760705, "recall": 0.7728706624605678, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9119496855345912, "precision": 0.8895705521472392, "recall": 0.9354838709677419, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8690095846645367, "precision": 0.9164420485175202, "recall": 0.8262454434993924, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6841291120965247, "precision": 0.619652531290865, "recall": 0.7635819521178637, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9025385312783318, "precision": 0.8664055700609226, "recall": 0.9418164616840113, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.8949834831374357, "precision": 0.9374748531423513, "recall": 0.8561769677371941, "support": 13607.0 }, "eval_O": { "f1-score": 0.9989151071331707, "precision": 0.9998190209030857, "recall": 0.9980128263029536, "support": 11071.0 }, "eval_accuracy": 0.8965804323024267, "eval_loss": 0.4196818470954895, "eval_macro avg": { "f1-score": 0.8496857162355017, "precision": 0.8352132913625792, "recall": 0.8705983121099606, "support": 32431.0 }, "eval_runtime": 4.9285, "eval_samples_per_second": 16.232, "eval_steps_per_second": 2.029, "eval_weighted avg": { "f1-score": 0.9000940098435676, "precision": 0.9076597589527613, "recall": 0.8965804323024267, "support": 32431.0 }, "step": 729 }, { "epoch": 10.0, "eval_B-Claim": { "f1-score": 0.6905537459283387, "precision": 0.7138047138047138, "recall": 0.668769716088328, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9142857142857143, "precision": 0.9, "recall": 0.9290322580645162, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.894484412470024, "precision": 0.8828402366863906, "recall": 0.9064398541919806, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.681682755333095, "precision": 0.7066963182604399, "recall": 0.6583793738489871, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9083702494754021, "precision": 0.895632183908046, "recall": 0.9214758751182592, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9129089784183134, "precision": 0.904253785147801, "recall": 0.9217314617476299, "support": 13607.0 }, "eval_O": { "f1-score": 0.9983716301791207, "precision": 0.999909395669113, "recall": 0.9968385873001535, "support": 11071.0 }, "eval_accuracy": 0.9092534920292313, "eval_loss": 0.5311424136161804, "eval_macro avg": { "f1-score": 0.8572367837271441, "precision": 0.8575909476395005, "recall": 0.857523875194265, "support": 32431.0 }, "eval_runtime": 4.8937, "eval_samples_per_second": 16.348, "eval_steps_per_second": 2.043, "eval_weighted avg": { "f1-score": 0.9081813659228948, "precision": 0.9074585127295745, "recall": 0.9092534920292313, "support": 32431.0 }, "step": 810 }, { "epoch": 11.0, "eval_B-Claim": { "f1-score": 0.680921052631579, "precision": 0.711340206185567, "recall": 0.6529968454258676, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9009584664536742, "precision": 0.8924050632911392, "recall": 0.9096774193548387, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8942020322773461, "precision": 0.88, "recall": 0.9088699878493317, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6838570400575678, "precision": 0.713820731096645, "recall": 0.6563075506445673, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.902502979737783, "precision": 0.9096588178760211, "recall": 0.8954588457899716, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9134434303925829, "precision": 0.9004641199571581, "recall": 0.9268023811273609, "support": 13607.0 }, "eval_O": { "f1-score": 0.9988699543461556, "precision": 0.999728555917481, "recall": 0.9980128263029536, "support": 11071.0 }, "eval_accuracy": 0.9096235083716198, "eval_loss": 0.49213308095932007, "eval_macro avg": { "f1-score": 0.8535364222709555, "precision": 0.8582024991891446, "recall": 0.8497322652135558, "support": 32431.0 }, "eval_runtime": 4.9244, "eval_samples_per_second": 16.246, "eval_steps_per_second": 2.031, "eval_weighted avg": { "f1-score": 0.9083194817194259, "precision": 0.9075428987655072, "recall": 0.9096235083716198, "support": 32431.0 }, "step": 891 }, { "epoch": 12.0, "eval_B-Claim": { "f1-score": 0.6577181208053691, "precision": 0.7025089605734767, "recall": 0.6182965299684543, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.8978328173374612, "precision": 0.8630952380952381, "recall": 0.9354838709677419, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8902147971360382, "precision": 0.8745603751465416, "recall": 0.9064398541919806, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6518804243008679, "precision": 0.6842105263157895, "recall": 0.6224677716390423, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.8919829328542555, "precision": 0.8490808037622916, "recall": 0.9394512771996215, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9075452422650322, "precision": 0.9011665821317296, "recall": 0.9140148453002132, "support": 13607.0 }, "eval_O": { "f1-score": 0.9985526910900046, "precision": 1.0, "recall": 0.9971095655315689, "support": 11071.0 }, "eval_accuracy": 0.9020073386574574, "eval_loss": 0.5465030670166016, "eval_macro avg": { "f1-score": 0.8422467179698613, "precision": 0.8392317837178667, "recall": 0.8476091021140889, "support": 32431.0 }, "eval_runtime": 4.8988, "eval_samples_per_second": 16.331, "eval_steps_per_second": 2.041, "eval_weighted avg": { "f1-score": 0.9004246846165758, "precision": 0.8996509302731299, "recall": 0.9020073386574574, "support": 32431.0 }, "step": 972 }, { "epoch": 12.35, "grad_norm": 0.1318705528974533, "learning_rate": 1.506172839506173e-05, "loss": 0.0258, "step": 1000 }, { "epoch": 13.0, "eval_B-Claim": { "f1-score": 0.6857142857142857, "precision": 0.6551724137931034, "recall": 0.7192429022082019, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9022082018927444, "precision": 0.8827160493827161, "recall": 0.9225806451612903, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8858560794044665, "precision": 0.9049429657794676, "recall": 0.8675577156743621, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6909411894516165, "precision": 0.6634880271243908, "recall": 0.7207642725598526, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9048498845265589, "precision": 0.8840252707581228, "recall": 0.9266792809839167, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.906777130915062, "precision": 0.9211464098870271, "recall": 0.8928492687587272, "support": 13607.0 }, "eval_O": { "f1-score": 0.9971466099008107, "precision": 1.0, "recall": 0.9943094571402764, "support": 11071.0 }, "eval_accuracy": 0.9044432795781814, "eval_loss": 0.5739177465438843, "eval_macro avg": { "f1-score": 0.8533561974007922, "precision": 0.8444987338178326, "recall": 0.8634262203552325, "support": 32431.0 }, "eval_runtime": 4.929, "eval_samples_per_second": 16.231, "eval_steps_per_second": 2.029, "eval_weighted avg": { "f1-score": 0.9058771229581195, "precision": 0.9079380628166581, "recall": 0.9044432795781814, "support": 32431.0 }, "step": 1053 }, { "epoch": 14.0, "eval_B-Claim": { "f1-score": 0.6666666666666666, "precision": 0.7276119402985075, "recall": 0.6151419558359621, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9003215434083601, "precision": 0.8974358974358975, "recall": 0.9032258064516129, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8976470588235295, "precision": 0.8700114025085519, "recall": 0.9270959902794653, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6725553620303559, "precision": 0.7317271250676773, "recall": 0.6222375690607734, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.8921406655652585, "precision": 0.8902496467263307, "recall": 0.8940397350993378, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9186321600114704, "precision": 0.8966482401511441, "recall": 0.9417211729257, "support": 13607.0 }, "eval_O": { "f1-score": 0.9977821029285294, "precision": 1.0, "recall": 0.995574022220215, "support": 11071.0 }, "eval_accuracy": 0.9104560451419937, "eval_loss": 0.5471230745315552, "eval_macro avg": { "f1-score": 0.8493922227763101, "precision": 0.8590977503125871, "recall": 0.842719464553295, "support": 32431.0 }, "eval_runtime": 4.9223, "eval_samples_per_second": 16.253, "eval_steps_per_second": 2.032, "eval_weighted avg": { "f1-score": 0.9078809171606194, "precision": 0.9070974660556812, "recall": 0.9104560451419937, "support": 32431.0 }, "step": 1134 }, { "epoch": 15.0, "eval_B-Claim": { "f1-score": 0.6688524590163935, "precision": 0.6962457337883959, "recall": 0.6435331230283912, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9061488673139159, "precision": 0.9090909090909091, "recall": 0.9032258064516129, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8902147971360382, "precision": 0.8745603751465416, "recall": 0.9064398541919806, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6658545498842734, "precision": 0.7071151358344114, "recall": 0.6291436464088398, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.8983730252298987, "precision": 0.8956276445698167, "recall": 0.9011352885525071, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9134990605578841, "precision": 0.8985002487739001, "recall": 0.9290071286837657, "support": 13607.0 }, "eval_O": { "f1-score": 0.9998645047649157, "precision": 0.9999096657633243, "recall": 0.999819347845723, "support": 11071.0 }, "eval_accuracy": 0.9077117572692794, "eval_loss": 0.6182886362075806, "eval_macro avg": { "f1-score": 0.8489724662719027, "precision": 0.8544356732810428, "recall": 0.8446148850232601, "support": 32431.0 }, "eval_runtime": 4.9224, "eval_samples_per_second": 16.252, "eval_steps_per_second": 2.032, "eval_weighted avg": { "f1-score": 0.9058074210256117, "precision": 0.9047621091758847, "recall": 0.9077117572692794, "support": 32431.0 }, "step": 1215 }, { "epoch": 16.0, "eval_B-Claim": { "f1-score": 0.6643835616438356, "precision": 0.7265917602996255, "recall": 0.61198738170347, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9113924050632912, "precision": 0.8944099378881988, "recall": 0.9290322580645162, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8954518606024808, "precision": 0.871264367816092, "recall": 0.9210206561360875, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6607098535616778, "precision": 0.7167474421109317, "recall": 0.6127992633517495, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9103066635923449, "precision": 0.8879892037786775, "recall": 0.9337748344370861, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9118409680207432, "precision": 0.8940046606878045, "recall": 0.9304034688028221, "support": 13607.0 }, "eval_O": { "f1-score": 0.9983714828553334, "precision": 1.0, "recall": 0.9967482612230151, "support": 11071.0 }, "eval_accuracy": 0.90737257562209, "eval_loss": 0.6626638174057007, "eval_macro avg": { "f1-score": 0.8503509707628153, "precision": 0.8558581960830471, "recall": 0.8479665891026781, "support": 32431.0 }, "eval_runtime": 4.9177, "eval_samples_per_second": 16.268, "eval_steps_per_second": 2.033, "eval_weighted avg": { "f1-score": 0.9048051223322606, "precision": 0.9038418545958269, "recall": 0.90737257562209, "support": 32431.0 }, "step": 1296 }, { "epoch": 17.0, "eval_B-Claim": { "f1-score": 0.6856240126382307, "precision": 0.6867088607594937, "recall": 0.6845425867507886, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9131832797427654, "precision": 0.9102564102564102, "recall": 0.9161290322580645, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8919902912621359, "precision": 0.8909090909090909, "recall": 0.8930741190765492, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6720430107526882, "precision": 0.682573599240266, "recall": 0.6618324125230203, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9010318949343339, "precision": 0.8934883720930232, "recall": 0.9087038789025544, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9102217019097096, "precision": 0.9062431703941138, "recall": 0.9142353200558536, "support": 13607.0 }, "eval_O": { "f1-score": 0.9988243805389763, "precision": 1.0, "recall": 0.9976515219943998, "support": 11071.0 }, "eval_accuracy": 0.90576917147174, "eval_loss": 0.6958470940589905, "eval_macro avg": { "f1-score": 0.8532740816826916, "precision": 0.8528827862360568, "recall": 0.853738410223033, "support": 32431.0 }, "eval_runtime": 4.9433, "eval_samples_per_second": 16.183, "eval_steps_per_second": 2.023, "eval_weighted avg": { "f1-score": 0.9053221272704657, "precision": 0.9049421722279843, "recall": 0.90576917147174, "support": 32431.0 }, "step": 1377 }, { "epoch": 18.0, "eval_B-Claim": { "f1-score": 0.688, "precision": 0.698051948051948, "recall": 0.6782334384858044, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9180327868852459, "precision": 0.9333333333333333, "recall": 0.9032258064516129, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8942307692307692, "precision": 0.8846611177170036, "recall": 0.9040097205346294, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6798171758479673, "precision": 0.7118387909319899, "recall": 0.6505524861878453, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9276937618147448, "precision": 0.9268177525967894, "recall": 0.9285714285714286, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9138506163886876, "precision": 0.9018822013883919, "recall": 0.9261409568604395, "support": 13607.0 }, "eval_O": { "f1-score": 0.9999096739228616, "precision": 0.9999096739228616, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_accuracy": 0.911473590083562, "eval_loss": 0.6612581014633179, "eval_macro avg": { "f1-score": 0.8602192548700395, "precision": 0.865213545420331, "recall": 0.8558062158592316, "support": 32431.0 }, "eval_runtime": 4.9167, "eval_samples_per_second": 16.271, "eval_steps_per_second": 2.034, "eval_weighted avg": { "f1-score": 0.9100977079524204, "precision": 0.9092367421655284, "recall": 0.911473590083562, "support": 32431.0 }, "step": 1458 }, { "epoch": 18.52, "grad_norm": 0.010771242901682854, "learning_rate": 1.2592592592592593e-05, "loss": 0.0083, "step": 1500 }, { "epoch": 19.0, "eval_B-Claim": { "f1-score": 0.6950819672131147, "precision": 0.7235494880546075, "recall": 0.668769716088328, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9049180327868852, "precision": 0.92, "recall": 0.8903225806451613, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.9021479713603818, "precision": 0.8862837045720985, "recall": 0.9185905224787363, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6773359596208298, "precision": 0.7279703625297698, "recall": 0.6332872928176796, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.8959923664122137, "precision": 0.9037536092396535, "recall": 0.8883632923368022, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9165171200918814, "precision": 0.8956857243072606, "recall": 0.9383405600058793, "support": 13607.0 }, "eval_O": { "f1-score": 0.9978274644699919, "precision": 1.0, "recall": 0.9956643482973534, "support": 11071.0 }, "eval_accuracy": 0.9104252104467947, "eval_loss": 0.6561155319213867, "eval_macro avg": { "f1-score": 0.8556886974221856, "precision": 0.865320412671913, "recall": 0.8476197589528487, "support": 32431.0 }, "eval_runtime": 4.9567, "eval_samples_per_second": 16.14, "eval_steps_per_second": 2.017, "eval_weighted avg": { "f1-score": 0.9083143486955053, "precision": 0.9075517471550982, "recall": 0.9104252104467947, "support": 32431.0 }, "step": 1539 }, { "epoch": 20.0, "eval_B-Claim": { "f1-score": 0.6833333333333333, "precision": 0.7243816254416962, "recall": 0.6466876971608833, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9050632911392404, "precision": 0.8881987577639752, "recall": 0.9225806451612903, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8997613365155133, "precision": 0.8839390386869871, "recall": 0.9161603888213852, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6925646418409007, "precision": 0.7485958812516715, "recall": 0.6443370165745856, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9094717244589248, "precision": 0.8950984883188273, "recall": 0.924314096499527, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9203195854027206, "precision": 0.9017561182029762, "recall": 0.9396634085397222, "support": 13607.0 }, "eval_O": { "f1-score": 0.9982808541440463, "precision": 1.0, "recall": 0.9965676090687381, "support": 11071.0 }, "eval_accuracy": 0.9149887453362523, "eval_loss": 0.6437746286392212, "eval_macro avg": { "f1-score": 0.8583992524049543, "precision": 0.8631385585237333, "recall": 0.8557586945465904, "support": 32431.0 }, "eval_runtime": 4.9174, "eval_samples_per_second": 16.269, "eval_steps_per_second": 2.034, "eval_weighted avg": { "f1-score": 0.9128082705193213, "precision": 0.9120938553978903, "recall": 0.9149887453362523, "support": 32431.0 }, "step": 1620 }, { "epoch": 21.0, "eval_B-Claim": { "f1-score": 0.6677631578947368, "precision": 0.697594501718213, "recall": 0.6403785488958991, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.8903654485049834, "precision": 0.9178082191780822, "recall": 0.864516129032258, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8954869358669834, "precision": 0.875725900116144, "recall": 0.9161603888213852, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6630760023937762, "precision": 0.6906008476689105, "recall": 0.6376611418047882, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.8958130477117819, "precision": 0.9227683049147443, "recall": 0.8703878902554399, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9106195648251723, "precision": 0.8945129732028924, "recall": 0.9273168222238554, "support": 13607.0 }, "eval_O": { "f1-score": 0.9977821029285294, "precision": 1.0, "recall": 0.995574022220215, "support": 11071.0 }, "eval_accuracy": 0.9047207918349727, "eval_loss": 0.7204756736755371, "eval_macro avg": { "f1-score": 0.8458437514465661, "precision": 0.8570015352569981, "recall": 0.8359992776076915, "support": 32431.0 }, "eval_runtime": 4.9231, "eval_samples_per_second": 16.25, "eval_steps_per_second": 2.031, "eval_weighted avg": { "f1-score": 0.9033970685809433, "precision": 0.9027615692407923, "recall": 0.9047207918349727, "support": 32431.0 }, "step": 1701 }, { "epoch": 22.0, "eval_B-Claim": { "f1-score": 0.6976, "precision": 0.7077922077922078, "recall": 0.6876971608832808, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9155844155844156, "precision": 0.9215686274509803, "recall": 0.9096774193548387, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8986731001206273, "precision": 0.8922155688622755, "recall": 0.905224787363305, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.695459320649002, "precision": 0.7054226852948141, "recall": 0.6857734806629834, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9150853889943075, "precision": 0.917697431018078, "recall": 0.9124881740775781, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9140016812251014, "precision": 0.9091173476806748, "recall": 0.9189387815095171, "support": 13607.0 }, "eval_O": { "f1-score": 0.999322095177837, "precision": 1.0, "recall": 0.9986451088429229, "support": 11071.0 }, "eval_accuracy": 0.9118436064259505, "eval_loss": 0.7220445871353149, "eval_macro avg": { "f1-score": 0.8622465716787558, "precision": 0.8648305525855757, "recall": 0.8597778446706323, "support": 32431.0 }, "eval_runtime": 4.8992, "eval_samples_per_second": 16.329, "eval_steps_per_second": 2.041, "eval_weighted avg": { "f1-score": 0.9114287052262304, "precision": 0.9110799834813146, "recall": 0.9118436064259505, "support": 32431.0 }, "step": 1782 }, { "epoch": 23.0, "eval_B-Claim": { "f1-score": 0.6832504145936982, "precision": 0.7202797202797203, "recall": 0.6498422712933754, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.910828025477707, "precision": 0.89937106918239, "recall": 0.9225806451612903, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8956469886702446, "precision": 0.8793911007025761, "recall": 0.9125151883353585, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.676206645646374, "precision": 0.7145053818554588, "recall": 0.641804788213628, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9027875380651206, "precision": 0.894199535962877, "recall": 0.9115421002838221, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.913811834576664, "precision": 0.9007639037624046, "recall": 0.9272433306386418, "support": 13607.0 }, "eval_O": { "f1-score": 0.9998644925245043, "precision": 1.0, "recall": 0.9997290217685846, "support": 11071.0 }, "eval_accuracy": 0.9096235083716198, "eval_loss": 0.7609456181526184, "eval_macro avg": { "f1-score": 0.8546279913649018, "precision": 0.8583586731064896, "recall": 0.8521796208135287, "support": 32431.0 }, "eval_runtime": 4.9475, "eval_samples_per_second": 16.17, "eval_steps_per_second": 2.021, "eval_weighted avg": { "f1-score": 0.907913975857415, "precision": 0.9069505627094704, "recall": 0.9096235083716198, "support": 32431.0 }, "step": 1863 }, { "epoch": 24.0, "eval_B-Claim": { "f1-score": 0.6968749999999999, "precision": 0.6904024767801857, "recall": 0.7034700315457413, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9131832797427654, "precision": 0.9102564102564102, "recall": 0.9161290322580645, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8922702373706634, "precision": 0.8939024390243903, "recall": 0.8906439854191981, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6863417982155113, "precision": 0.6821282401091405, "recall": 0.6906077348066298, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.903548463919981, "precision": 0.9098321342925659, "recall": 0.8973509933774835, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9113179092948837, "precision": 0.9122238586156112, "recall": 0.910413757624752, "support": 13607.0 }, "eval_O": { "f1-score": 0.9999096657633243, "precision": 1.0, "recall": 0.999819347845723, "support": 11071.0 }, "eval_accuracy": 0.908143443002066, "eval_loss": 0.7767113447189331, "eval_macro avg": { "f1-score": 0.857635193472447, "precision": 0.8569636512969006, "recall": 0.8583478404110846, "support": 32431.0 }, "eval_runtime": 4.9392, "eval_samples_per_second": 16.197, "eval_steps_per_second": 2.025, "eval_weighted avg": { "f1-score": 0.9083489289362786, "precision": 0.9085692481624786, "recall": 0.908143443002066, "support": 32431.0 }, "step": 1944 }, { "epoch": 24.69, "grad_norm": 0.007521071936935186, "learning_rate": 1.0123456790123458e-05, "loss": 0.0037, "step": 2000 }, { "epoch": 25.0, "eval_B-Claim": { "f1-score": 0.6971080669710807, "precision": 0.6735294117647059, "recall": 0.722397476340694, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9196141479099678, "precision": 0.9166666666666666, "recall": 0.9225806451612903, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8868388683886839, "precision": 0.8978829389788294, "recall": 0.8760631834750912, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6842635052007605, "precision": 0.6654339786817489, "recall": 0.7041896869244936, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9163140573577808, "precision": 0.9107476635514019, "recall": 0.9219489120151372, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9042813682570305, "precision": 0.9130890836892186, "recall": 0.8956419489968399, "support": 13607.0 }, "eval_O": { "f1-score": 0.9989601699896017, "precision": 1.0, "recall": 0.9979225002258152, "support": 11071.0 }, "eval_accuracy": 0.9045666183589776, "eval_loss": 0.7725499272346497, "eval_macro avg": { "f1-score": 0.8581971691535578, "precision": 0.8539071061903674, "recall": 0.8629634790199088, "support": 32431.0 }, "eval_runtime": 4.957, "eval_samples_per_second": 16.139, "eval_steps_per_second": 2.017, "eval_weighted avg": { "f1-score": 0.9055214160067376, "precision": 0.906722541443147, "recall": 0.9045666183589776, "support": 32431.0 }, "step": 2025 }, { "epoch": 26.0, "eval_B-Claim": { "f1-score": 0.6904376012965964, "precision": 0.71, "recall": 0.6719242902208202, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9022082018927444, "precision": 0.8827160493827161, "recall": 0.9225806451612903, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8958458759783263, "precision": 0.8878281622911695, "recall": 0.9040097205346294, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6721173500060118, "precision": 0.7034986156556758, "recall": 0.6434162062615101, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9087940284581292, "precision": 0.8964565117349287, "recall": 0.9214758751182592, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.911746539258075, "precision": 0.9016240298936476, "recall": 0.9220989196736974, "support": 13607.0 }, "eval_O": { "f1-score": 0.9998193315266486, "precision": 0.9999096576023128, "recall": 0.9997290217685846, "support": 11071.0 }, "eval_accuracy": 0.9083284511732601, "eval_loss": 0.7629147171974182, "eval_macro avg": { "f1-score": 0.8544241326309331, "precision": 0.8545761466514928, "recall": 0.8550335255341129, "support": 32431.0 }, "eval_runtime": 4.9235, "eval_samples_per_second": 16.249, "eval_steps_per_second": 2.031, "eval_weighted avg": { "f1-score": 0.9069099155264195, "precision": 0.9059874383002041, "recall": 0.9083284511732601, "support": 32431.0 }, "step": 2106 }, { "epoch": 27.0, "eval_B-Claim": { "f1-score": 0.701639344262295, "precision": 0.7303754266211604, "recall": 0.6750788643533123, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9166666666666666, "precision": 0.910828025477707, "recall": 0.9225806451612903, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.9017964071856287, "precision": 0.8890200708382526, "recall": 0.9149453219927096, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6878357407186343, "precision": 0.7143565583932556, "recall": 0.6632136279926335, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9198392054859305, "precision": 0.9196217494089834, "recall": 0.9200567644276254, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9161731041749935, "precision": 0.906065832973983, "recall": 0.9265084147865069, "support": 13607.0 }, "eval_O": { "f1-score": 0.9997741950051934, "precision": 0.9997290462427746, "recall": 0.999819347845723, "support": 11071.0 }, "eval_accuracy": 0.913076994233912, "eval_loss": 0.7461331486701965, "eval_macro avg": { "f1-score": 0.8633892376427632, "precision": 0.8671423871365881, "recall": 0.8603147123656859, "support": 32431.0 }, "eval_runtime": 4.9043, "eval_samples_per_second": 16.312, "eval_steps_per_second": 2.039, "eval_weighted avg": { "f1-score": 0.9119067413298663, "precision": 0.9111175788791426, "recall": 0.913076994233912, "support": 32431.0 }, "step": 2187 }, { "epoch": 28.0, "eval_B-Claim": { "f1-score": 0.69375, "precision": 0.6873065015479877, "recall": 0.7003154574132492, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9096989966555183, "precision": 0.9444444444444444, "recall": 0.8774193548387097, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8948004836759371, "precision": 0.8904933814681107, "recall": 0.8991494532199271, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6768418636676612, "precision": 0.6748283752860412, "recall": 0.6788674033149171, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9071392910634049, "precision": 0.9603594080338267, "recall": 0.8595080416272469, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9102119741690686, "precision": 0.9037820605709318, "recall": 0.9167340339531124, "support": 13607.0 }, "eval_O": { "f1-score": 0.9998193315266486, "precision": 0.9999096576023128, "recall": 0.9997290217685846, "support": 11071.0 }, "eval_accuracy": 0.9067250470229101, "eval_loss": 0.7586968541145325, "eval_macro avg": { "f1-score": 0.8560374201083196, "precision": 0.8658748327076651, "recall": 0.847388966590821, "support": 32431.0 }, "eval_runtime": 5.0076, "eval_samples_per_second": 15.976, "eval_steps_per_second": 1.997, "eval_weighted avg": { "f1-score": 0.9068326774221362, "precision": 0.9073589365470631, "recall": 0.9067250470229101, "support": 32431.0 }, "step": 2268 }, { "epoch": 29.0, "eval_B-Claim": { "f1-score": 0.6825396825396827, "precision": 0.6869009584664537, "recall": 0.6782334384858044, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9055374592833876, "precision": 0.9144736842105263, "recall": 0.896774193548387, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.893719806763285, "precision": 0.8883553421368547, "recall": 0.8991494532199271, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6749443598453789, "precision": 0.6870975435249225, "recall": 0.6632136279926335, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9090909090909092, "precision": 0.9196515004840271, "recall": 0.8987701040681173, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9108261693137113, "precision": 0.9029974720115566, "recall": 0.9187917983390902, "support": 13607.0 }, "eval_O": { "f1-score": 0.9980995475113122, "precision": 1.0, "recall": 0.9962063047601842, "support": 11071.0 }, "eval_accuracy": 0.9067250470229101, "eval_loss": 0.796188473701477, "eval_macro avg": { "f1-score": 0.8535368477639526, "precision": 0.857068071547763, "recall": 0.8501627029163064, "support": 32431.0 }, "eval_runtime": 4.9618, "eval_samples_per_second": 16.123, "eval_steps_per_second": 2.015, "eval_weighted avg": { "f1-score": 0.9062194518780211, "precision": 0.9058490047862192, "recall": 0.9067250470229101, "support": 32431.0 }, "step": 2349 }, { "epoch": 30.0, "eval_B-Claim": { "f1-score": 0.6710097719869706, "precision": 0.6936026936026936, "recall": 0.6498422712933754, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.89375, "precision": 0.8666666666666667, "recall": 0.9225806451612903, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.893848009650181, "precision": 0.8874251497005988, "recall": 0.9003645200486027, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6677052127022169, "precision": 0.69632591852037, "recall": 0.6413443830570903, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.8996555683122847, "precision": 0.8741633199464525, "recall": 0.9266792809839167, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9106609341980617, "precision": 0.903027675410073, "recall": 0.9184243404130227, "support": 13607.0 }, "eval_O": { "f1-score": 0.9988248056409328, "precision": 0.9996381073011852, "recall": 0.9980128263029536, "support": 11071.0 }, "eval_accuracy": 0.9059541796429342, "eval_loss": 0.8091421127319336, "eval_macro avg": { "f1-score": 0.8479220432129496, "precision": 0.8458356473068627, "recall": 0.8510354667514646, "support": 32431.0 }, "eval_runtime": 5.049, "eval_samples_per_second": 15.845, "eval_steps_per_second": 1.981, "eval_weighted avg": { "f1-score": 0.9046472304257149, "precision": 0.9038225277994681, "recall": 0.9059541796429342, "support": 32431.0 }, "step": 2430 }, { "epoch": 30.86, "grad_norm": 0.0011433030012995005, "learning_rate": 7.654320987654322e-06, "loss": 0.002, "step": 2500 }, { "epoch": 31.0, "eval_B-Claim": { "f1-score": 0.6802507836990596, "precision": 0.67601246105919, "recall": 0.6845425867507886, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9096774193548387, "precision": 0.9096774193548387, "recall": 0.9096774193548387, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8876745598057073, "precision": 0.8871359223300971, "recall": 0.8882138517618469, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6758359023002134, "precision": 0.6968215158924206, "recall": 0.6560773480662984, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9129821260583255, "precision": 0.9078578110383536, "recall": 0.9181646168401135, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9112844437964718, "precision": 0.9039699182876564, "recall": 0.9187183067538767, "support": 13607.0 }, "eval_O": { "f1-score": 0.9996839015579136, "precision": 0.999548491963157, "recall": 0.999819347845723, "support": 11071.0 }, "eval_accuracy": 0.9080817736116679, "eval_loss": 0.8016564249992371, "eval_macro avg": { "f1-score": 0.8539127337960757, "precision": 0.8544319342751019, "recall": 0.8536019253390694, "support": 32431.0 }, "eval_runtime": 4.9624, "eval_samples_per_second": 16.121, "eval_steps_per_second": 2.015, "eval_weighted avg": { "f1-score": 0.9071696505232147, "precision": 0.9064763012077777, "recall": 0.9080817736116679, "support": 32431.0 }, "step": 2511 }, { "epoch": 32.0, "eval_B-Claim": { "f1-score": 0.699367088607595, "precision": 0.7015873015873015, "recall": 0.6971608832807571, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9073482428115016, "precision": 0.8987341772151899, "recall": 0.9161290322580645, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8998178506375226, "precision": 0.8992718446601942, "recall": 0.9003645200486027, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.674114114114114, "precision": 0.7048480281336348, "recall": 0.6459484346224678, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9130636913063692, "precision": 0.8976234003656307, "recall": 0.9290444654683065, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.913888989858602, "precision": 0.9041283084004603, "recall": 0.9238627177188212, "support": 13607.0 }, "eval_O": { "f1-score": 0.9991866979938551, "precision": 0.9996383690443902, "recall": 0.9987354349200614, "support": 11071.0 }, "eval_accuracy": 0.9096851777620178, "eval_loss": 0.8070256114006042, "eval_macro avg": { "f1-score": 0.858112382189937, "precision": 0.8579759184866861, "recall": 0.8587493554738688, "support": 32431.0 }, "eval_runtime": 4.9602, "eval_samples_per_second": 16.128, "eval_steps_per_second": 2.016, "eval_weighted avg": { "f1-score": 0.9083512753053686, "precision": 0.90748708577928, "recall": 0.9096851777620178, "support": 32431.0 }, "step": 2592 }, { "epoch": 33.0, "eval_B-Claim": { "f1-score": 0.6867088607594937, "precision": 0.6888888888888889, "recall": 0.6845425867507886, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9161290322580645, "precision": 0.9161290322580645, "recall": 0.9161290322580645, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8917120387174833, "precision": 0.8879518072289156, "recall": 0.8955042527339003, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6777020447906524, "precision": 0.71900826446281, "recall": 0.6408839779005525, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9200660221645838, "precision": 0.9172543488481429, "recall": 0.9228949858088931, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9157149053057683, "precision": 0.9009316549320816, "recall": 0.93099140148453, "support": 13607.0 }, "eval_O": { "f1-score": 0.9996386956914461, "precision": 0.9996386956914461, "recall": 0.9996386956914461, "support": 11071.0 }, "eval_accuracy": 0.9116585982547563, "eval_loss": 0.823725700378418, "eval_macro avg": { "f1-score": 0.858238799955356, "precision": 0.8614003846157642, "recall": 0.8557978475183107, "support": 32431.0 }, "eval_runtime": 4.9155, "eval_samples_per_second": 16.275, "eval_steps_per_second": 2.034, "eval_weighted avg": { "f1-score": 0.9099212385263286, "precision": 0.9089940716622809, "recall": 0.9116585982547563, "support": 32431.0 }, "step": 2673 }, { "epoch": 34.0, "eval_B-Claim": { "f1-score": 0.6927899686520376, "precision": 0.6884735202492211, "recall": 0.6971608832807571, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9196141479099678, "precision": 0.9166666666666666, "recall": 0.9225806451612903, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.8929440389294404, "precision": 0.8940316686967114, "recall": 0.8918590522478737, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6815160349854228, "precision": 0.6906168754431576, "recall": 0.6726519337016574, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9175862878610003, "precision": 0.9109557109557109, "recall": 0.924314096499527, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9118692513466964, "precision": 0.9093699751498319, "recall": 0.9143823032262806, "support": 13607.0 }, "eval_O": { "f1-score": 0.999729070712544, "precision": 0.999548532731377, "recall": 0.9999096739228616, "support": 11071.0 }, "eval_accuracy": 0.9091918226388332, "eval_loss": 0.8422325849533081, "eval_macro avg": { "f1-score": 0.8594355429138727, "precision": 0.8585232785560966, "recall": 0.8604083697200353, "support": 32431.0 }, "eval_runtime": 4.9469, "eval_samples_per_second": 16.172, "eval_steps_per_second": 2.021, "eval_weighted avg": { "f1-score": 0.9087951672928857, "precision": 0.9084430526412389, "recall": 0.9091918226388332, "support": 32431.0 }, "step": 2754 }, { "epoch": 35.0, "eval_B-Claim": { "f1-score": 0.68, "precision": 0.7208480565371025, "recall": 0.6435331230283912, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9206349206349206, "precision": 0.90625, "recall": 0.9354838709677419, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.897910447761194, "precision": 0.8826291079812206, "recall": 0.913730255164034, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6727050183598531, "precision": 0.7182435964453737, "recall": 0.6325966850828729, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9136773894931729, "precision": 0.894426823742637, "recall": 0.9337748344370861, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9146681135829264, "precision": 0.9006268699244907, "recall": 0.9291541118541927, "support": 13607.0 }, "eval_O": { "f1-score": 0.9992771955186122, "precision": 0.9995481247175779, "recall": 0.9990064131514769, "support": 11071.0 }, "eval_accuracy": 0.9104252104467947, "eval_loss": 0.8378845453262329, "eval_macro avg": { "f1-score": 0.8569818693358113, "precision": 0.8603675113354861, "recall": 0.8553256133836852, "support": 32431.0 }, "eval_runtime": 4.9164, "eval_samples_per_second": 16.272, "eval_steps_per_second": 2.034, "eval_weighted avg": { "f1-score": 0.9083861171551627, "precision": 0.9073749479187287, "recall": 0.9104252104467947, "support": 32431.0 }, "step": 2835 }, { "epoch": 36.0, "eval_B-Claim": { "f1-score": 0.674496644295302, "precision": 0.7204301075268817, "recall": 0.6340694006309149, "support": 317.0 }, "eval_B-MajorClaim": { "f1-score": 0.9108910891089109, "precision": 0.9324324324324325, "recall": 0.8903225806451613, "support": 155.0 }, "eval_B-Premise": { "f1-score": 0.9000591366055587, "precision": 0.8767281105990783, "recall": 0.9246658566221142, "support": 823.0 }, "eval_I-Claim": { "f1-score": 0.6687843616371411, "precision": 0.7125748502994012, "recall": 0.6300644567219152, "support": 4344.0 }, "eval_I-MajorClaim": { "f1-score": 0.9068203650336216, "precision": 0.9209756097560976, "recall": 0.8930936613055819, "support": 2114.0 }, "eval_I-Premise": { "f1-score": 0.9150956972226219, "precision": 0.8963281415180774, "recall": 0.9346659807452047, "support": 13607.0 }, "eval_O": { "f1-score": 0.9989605459393501, "precision": 0.9996382054992764, "recall": 0.998283804534369, "support": 11071.0 }, "eval_accuracy": 0.9094693348956245, "eval_loss": 0.8216572999954224, "eval_macro avg": { "f1-score": 0.8535868342632151, "precision": 0.8655867796616066, "recall": 0.8435951058864658, "support": 32431.0 }, "eval_runtime": 4.9406, "eval_samples_per_second": 16.192, "eval_steps_per_second": 2.024, "eval_weighted avg": { "f1-score": 0.9074394591383382, "precision": 0.9065446575677608, "recall": 0.9094693348956245, "support": 32431.0 }, "step": 2916 } ], "logging_steps": 500, "max_steps": 4050, "num_input_tokens_seen": 0, "num_train_epochs": 50, "save_steps": 500, "total_flos": 5176469257848000.0, "train_batch_size": 4, "trial_name": null, "trial_params": null }