diff --git "a/checkpoint-3969/trainer_state.json" "b/checkpoint-3969/trainer_state.json" --- "a/checkpoint-3969/trainer_state.json" +++ "b/checkpoint-3969/trainer_state.json" @@ -11,384 +11,384 @@ { "epoch": 1.0, "eval_B-Claim": { - "f1-score": 0.32421052631578945, - "precision": 0.42777777777777776, - "recall": 0.26101694915254237, - "support": 295.0 + "f1-score": 0.18764302059496568, + "precision": 0.3416666666666667, + "recall": 0.12933753943217666, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.192090395480226, - "precision": 0.8095238095238095, - "recall": 0.10897435897435898, - "support": 156.0 + "f1-score": 0.19540229885057472, + "precision": 0.8947368421052632, + "recall": 0.10967741935483871, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8566265060240965, - "precision": 0.7620578778135049, - "recall": 0.9779917469050894, - "support": 727.0 + "f1-score": 0.8476339053562143, + "precision": 0.740909090909091, + "recall": 0.9902794653705954, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.5001390820584144, - "precision": 0.5900886117492615, - "recall": 0.43398503499879315, - "support": 4143.0 + "f1-score": 0.4315998169894769, + "precision": 0.6394035246272028, + "recall": 0.3257366482504604, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.776470588235294, - "precision": 0.6649333766655834, - "recall": 0.93296853625171, - "support": 2193.0 + "f1-score": 0.7779799818016379, + "precision": 0.7493426818580193, + "recall": 0.8088930936613056, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.8861458374413376, - "precision": 0.8781460059402845, - "recall": 0.8942927644670858, - "support": 12563.0 + "f1-score": 0.8948683715038855, + "precision": 0.8376057421174058, + "recall": 0.9605350187403542, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9987647926046939, - "precision": 0.9977708781147998, - "recall": 0.9997606892150607, - "support": 12536.0 + "f1-score": 0.9980147987727848, + "precision": 0.9970251509961237, + "recall": 0.9990064131514769, + "support": 11071.0 }, - "eval_accuracy": 0.8713396498328887, - "eval_loss": 0.3092103600502014, + "eval_accuracy": 0.8673183065585397, + "eval_loss": 0.3193369209766388, "eval_macro avg": { - "f1-score": 0.647778246879979, - "precision": 0.732899762512146, - "recall": 0.6584271542806628, - "support": 32613.0 - }, - "eval_runtime": 4.965, - "eval_samples_per_second": 16.314, - "eval_steps_per_second": 2.215, + "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": 0.8639627599323451, - "precision": 0.8662078389010613, - "recall": 0.8713396498328887, - "support": 32613.0 + "f1-score": 0.8489527906249091, + "precision": 0.8526967051825685, + "recall": 0.8673183065585397, + "support": 32431.0 }, "step": 81 }, { "epoch": 2.0, "eval_B-Claim": { - "f1-score": 0.6186291739894552, - "precision": 0.6423357664233577, - "recall": 0.5966101694915255, - "support": 295.0 + "f1-score": 0.583941605839416, + "precision": 0.6926406926406926, + "recall": 0.5047318611987381, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8113879003558718, - "precision": 0.912, - "recall": 0.7307692307692307, - "support": 156.0 + "f1-score": 0.8117647058823529, + "precision": 0.745945945945946, + "recall": 0.8903225806451613, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8871181938911022, - "precision": 0.8575096277278562, - "recall": 0.9188445667125172, - "support": 727.0 + "f1-score": 0.8943661971830986, + "precision": 0.8649262202043133, + "recall": 0.9258809234507898, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.656253149249219, - "precision": 0.5633217993079584, - "recall": 0.7859039343470915, - "support": 4143.0 + "f1-score": 0.6197740770932805, + "precision": 0.6271506009898656, + "recall": 0.6125690607734806, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8368159203980099, - "precision": 0.9206349206349206, - "recall": 0.7669858641130871, - "support": 2193.0 + "f1-score": 0.8298737427776589, + "precision": 0.7577178585384916, + "recall": 0.9172185430463576, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.8831027253668763, - "precision": 0.933020288827855, - "recall": 0.8382551938231314, - "support": 12563.0 + "f1-score": 0.9013697613125952, + "precision": 0.9106660666066607, + "recall": 0.8922613360770192, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9998006141085457, - "precision": 0.9996013077107089, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9961487925331883, + "precision": 0.9993636363636363, + "recall": 0.9929545659831993, + "support": 11071.0 }, - "eval_accuracy": 0.8880814399165977, - "eval_loss": 0.2581149935722351, + "eval_accuracy": 0.887854213561099, + "eval_loss": 0.26407390832901, "eval_macro avg": { - "f1-score": 0.8133010967655829, - "precision": 0.8326319586618081, - "recall": 0.8053384227509405, - "support": 32613.0 - }, - "eval_runtime": 5.0296, - "eval_samples_per_second": 16.105, - "eval_steps_per_second": 2.187, + "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": 0.8933836566327042, - "precision": 0.9064023864086126, - "recall": 0.8880814399165977, - "support": 32613.0 + "f1-score": 0.8876368199002541, + "precision": 0.8889201304482091, + "recall": 0.887854213561099, + "support": 32431.0 }, "step": 162 }, { "epoch": 3.0, "eval_B-Claim": { - "f1-score": 0.6576271186440678, - "precision": 0.6576271186440678, - "recall": 0.6576271186440678, - "support": 295.0 + "f1-score": 0.5895522388059701, + "precision": 0.7214611872146118, + "recall": 0.49842271293375395, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8673139158576051, - "precision": 0.8758169934640523, - "recall": 0.8589743589743589, - "support": 156.0 + "f1-score": 0.8668941979522183, + "precision": 0.9202898550724637, + "recall": 0.8193548387096774, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8838487972508591, - "precision": 0.8832417582417582, - "recall": 0.8844566712517193, - "support": 727.0 + "f1-score": 0.8929178470254956, + "precision": 0.8365180467091295, + "recall": 0.9574726609963548, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6468460321049062, - "precision": 0.6083351052519669, - "recall": 0.6905623944001931, - "support": 4143.0 + "f1-score": 0.6548140728684112, + "precision": 0.7178149876475433, + "recall": 0.6019797421731123, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8695855771482088, - "precision": 0.8936477382098171, - "recall": 0.8467852257181943, - "support": 2193.0 + "f1-score": 0.8815622622368754, + "precision": 0.9502460360852925, + "recall": 0.8221381267738883, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.8926256077795786, - "precision": 0.9090533960551291, - "recall": 0.8767810236408501, - "support": 12563.0 + "f1-score": 0.9169943670953343, + "precision": 0.8852257181942544, + "recall": 0.9511280958330272, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9998803589232302, - "precision": 0.9997607464710104, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9984170774727511, + "precision": 0.9998188405797102, + "recall": 0.9970192394544305, + "support": 11071.0 }, - "eval_accuracy": 0.8965749854352559, - "eval_loss": 0.2399493306875229, + "eval_accuracy": 0.9067250470229101, + "eval_loss": 0.25323203206062317, "eval_macro avg": { - "f1-score": 0.8311039153869222, - "precision": 0.8324975509054003, - "recall": 0.8307409703756262, - "support": 32613.0 - }, - "eval_runtime": 5.0024, - "eval_samples_per_second": 16.192, - "eval_steps_per_second": 2.199, + "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": 0.8986385496019461, - "precision": 0.9016737059631544, - "recall": 0.8965749854352559, - "support": 32613.0 + "f1-score": 0.9033110548907601, + "precision": 0.9034896128542845, + "recall": 0.9067250470229101, + "support": 32431.0 }, "step": 243 }, { "epoch": 4.0, "eval_B-Claim": { - "f1-score": 0.6106346483704975, - "precision": 0.6180555555555556, - "recall": 0.6033898305084746, - "support": 295.0 + "f1-score": 0.6806136680613668, + "precision": 0.61, + "recall": 0.7697160883280757, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8498583569405098, - "precision": 0.7614213197969543, - "recall": 0.9615384615384616, - "support": 156.0 + "f1-score": 0.8861538461538462, + "precision": 0.8470588235294118, + "recall": 0.9290322580645162, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8746478873239436, - "precision": 0.8961038961038961, - "recall": 0.8541953232462174, - "support": 727.0 + "f1-score": 0.8687258687258688, + "precision": 0.9233926128590971, + "recall": 0.8201701093560145, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.5964912280701754, - "precision": 0.6109845608706657, - "recall": 0.5826695631185131, - "support": 4143.0 + "f1-score": 0.656155854589275, + "precision": 0.5819380121125757, + "recall": 0.7520718232044199, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8631961259079903, - "precision": 0.7741585233441911, - "recall": 0.9753761969904241, - "support": 2193.0 + "f1-score": 0.8818897637795275, + "precision": 0.8408408408408409, + "recall": 0.9271523178807947, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.8933694290678411, - "precision": 0.9071885770556376, - "recall": 0.8799649765183475, - "support": 12563.0 + "f1-score": 0.8829758349171871, + "precision": 0.9349433218334154, + "recall": 0.836481222899978, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999601132782897, - "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9971017118014671, + "precision": 0.9998183634547271, + "recall": 0.9943997832174148, + "support": 11071.0 }, - "eval_accuracy": 0.8920369177935179, - "eval_loss": 0.2815629839897156, + "eval_accuracy": 0.8843698930036077, + "eval_loss": 0.2791365385055542, "eval_macro avg": { - "f1-score": 0.8125939698513209, - "precision": 0.7954160618181286, - "recall": 0.8367220830941131, - "support": 32613.0 - }, - "eval_runtime": 4.9474, - "eval_samples_per_second": 16.372, - "eval_steps_per_second": 2.223, + "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": 0.8914155863760895, - "precision": 0.892730641712682, - "recall": 0.8920369177935179, - "support": 32613.0 + "f1-score": 0.889158243622726, + "precision": 0.8997827813567029, + "recall": 0.8843698930036077, + "support": 32431.0 }, "step": 324 }, { "epoch": 5.0, "eval_B-Claim": { - "f1-score": 0.6796116504854369, - "precision": 0.6501547987616099, - "recall": 0.711864406779661, - "support": 295.0 + "f1-score": 0.6078799249530958, + "precision": 0.75, + "recall": 0.5110410094637224, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8746177370030581, - "precision": 0.8362573099415205, - "recall": 0.9166666666666666, - "support": 156.0 + "f1-score": 0.8971962616822431, + "precision": 0.8674698795180723, + "recall": 0.9290322580645162, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8836879432624114, - "precision": 0.9121522693997072, - "recall": 0.8569463548830811, - "support": 727.0 + "f1-score": 0.893739230327398, + "precision": 0.8474945533769063, + "recall": 0.945321992709599, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6557152635181382, - "precision": 0.6216742375081116, - "recall": 0.6937002172338885, - "support": 4143.0 + "f1-score": 0.6119813135476778, + "precision": 0.7590320381731425, + "recall": 0.5126611418047882, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8859746610357857, - "precision": 0.8642671292281006, - "recall": 0.9088007295941632, - "support": 2193.0 + "f1-score": 0.8889393595275948, + "precision": 0.854958497160332, + "recall": 0.9257332071901608, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.8956745077255495, - "precision": 0.9180177168644492, - "recall": 0.874393058982727, - "support": 12563.0 + "f1-score": 0.9137743463765573, + "precision": 0.874462654486835, + "recall": 0.956786947894466, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9998006141085457, - "precision": 0.9996013077107089, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9976913675252366, + "precision": 1.0, + "recall": 0.9953933700659381, + "support": 11071.0 }, - "eval_accuracy": 0.9003771502161715, - "eval_loss": 0.2959797978401184, + "eval_accuracy": 0.9036724121982054, + "eval_loss": 0.3500836193561554, "eval_macro avg": { - "f1-score": 0.839297482448418, - "precision": 0.8288749670591724, - "recall": 0.8517673477343124, - "support": 32613.0 - }, - "eval_runtime": 4.9768, - "eval_samples_per_second": 16.275, - "eval_steps_per_second": 2.21, + "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": 0.9022413468876027, - "precision": 0.9051721478141229, - "recall": 0.9003771502161715, - "support": 32613.0 + "f1-score": 0.8968007164885599, + "precision": 0.8986502613295283, + "recall": 0.9036724121982054, + "support": 32431.0 }, "step": 405 }, { "epoch": 6.0, "eval_B-Claim": { - "f1-score": 0.6655518394648828, - "precision": 0.6567656765676567, - "recall": 0.6745762711864407, - "support": 295.0 + "f1-score": 0.672077922077922, + "precision": 0.6923076923076923, + "recall": 0.6529968454258676, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8807339449541285, - "precision": 0.8421052631578947, - "recall": 0.9230769230769231, - "support": 156.0 + "f1-score": 0.8910256410256411, + "precision": 0.8853503184713376, + "recall": 0.896774193548387, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8812849162011174, - "precision": 0.8950354609929078, - "recall": 0.8679504814305364, - "support": 727.0 + "f1-score": 0.8907563025210083, + "precision": 0.8801897983392646, + "recall": 0.9015795868772782, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.634203994799669, - "precision": 0.6213524779990737, - "recall": 0.647598358677287, - "support": 4143.0 + "f1-score": 0.6666666666666667, + "precision": 0.6998228296633764, + "recall": 0.6365101289134438, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8926454445664106, - "precision": 0.8607112616426758, - "recall": 0.9270405836753306, - "support": 2193.0 + "f1-score": 0.8853370396108408, + "precision": 0.867453472537449, + "recall": 0.9039735099337748, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.8911586596689544, - "precision": 0.9041533546325878, - "recall": 0.8785321977234737, - "support": 12563.0 + "f1-score": 0.9098497495826378, + "precision": 0.8987595898759589, + "recall": 0.9212170206511354, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9995614559661922, - "precision": 0.9991232964055152, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.998190209030857, + "precision": 1.0, + "recall": 0.9963869569144612, + "support": 11071.0 }, - "eval_accuracy": 0.8972802256768774, - "eval_loss": 0.33718833327293396, + "eval_accuracy": 0.9043816101877833, + "eval_loss": 0.30672115087509155, "eval_macro avg": { - "f1-score": 0.8350200365173363, - "precision": 0.8256066844854731, - "recall": 0.8455392593957132, - "support": 32613.0 - }, - "eval_runtime": 4.9878, - "eval_samples_per_second": 16.24, - "eval_steps_per_second": 2.205, + "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": 0.8979742337698349, - "precision": 0.8990738793396261, - "recall": 0.8972802256768774, - "support": 32613.0 + "f1-score": 0.9029367568413356, + "precision": 0.9020793621628027, + "recall": 0.9043816101877833, + "support": 32431.0 }, "step": 486 }, { "epoch": 6.17, - "grad_norm": 10.722956657409668, + "grad_norm": 0.43724608421325684, "learning_rate": 1.7530864197530865e-05, "loss": 0.2434, "step": 500 @@ -396,2751 +396,2751 @@ { "epoch": 7.0, "eval_B-Claim": { - "f1-score": 0.6608695652173914, - "precision": 0.6785714285714286, - "recall": 0.6440677966101694, - "support": 295.0 + "f1-score": 0.6763285024154589, + "precision": 0.6907894736842105, + "recall": 0.6624605678233438, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.875, - "precision": 0.8536585365853658, - "recall": 0.8974358974358975, - "support": 156.0 + "f1-score": 0.8952380952380952, + "precision": 0.88125, + "recall": 0.9096774193548387, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8905608755129958, - "precision": 0.8857142857142857, - "recall": 0.8954607977991746, - "support": 727.0 + "f1-score": 0.8941034897713597, + "precision": 0.8855780691299165, + "recall": 0.9027946537059538, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6306948332909138, - "precision": 0.6670255720053836, - "recall": 0.5981173062997828, - "support": 4143.0 + "f1-score": 0.6960393719240684, + "precision": 0.7088305489260143, + "recall": 0.6837016574585635, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.890302764370338, - "precision": 0.8579281183932347, - "recall": 0.9252165982672138, - "support": 2193.0 + "f1-score": 0.9140037593984962, + "precision": 0.9080298786181139, + "recall": 0.9200567644276254, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9037468972853708, - "precision": 0.8947573724449992, - "recall": 0.9129188888004458, - "support": 12563.0 + "f1-score": 0.9164355568530141, + "precision": 0.9101253895774444, + "recall": 0.9228338355258323, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9967376529225193, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9934965224460302, + "support": 11071.0 }, - "eval_accuracy": 0.9043326280930917, - "eval_loss": 0.43479764461517334, + "eval_accuracy": 0.9116277635595572, + "eval_loss": 0.34577852487564087, "eval_macro avg": { - "f1-score": 0.8358821336681442, - "precision": 0.8339507591020997, - "recall": 0.8390310407446692, - "support": 32613.0 - }, - "eval_runtime": 4.9479, - "eval_samples_per_second": 16.371, - "eval_steps_per_second": 2.223, + "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": 0.9025256555797259, - "precision": 0.9014510706568353, - "recall": 0.9043326280930917, - "support": 32613.0 + "f1-score": 0.9111536914902465, + "precision": 0.910801887328957, + "recall": 0.9116277635595572, + "support": 32431.0 }, "step": 567 }, { "epoch": 8.0, "eval_B-Claim": { - "f1-score": 0.6611295681063122, - "precision": 0.6482084690553745, - "recall": 0.6745762711864407, - "support": 295.0 + "f1-score": 0.701923076923077, + "precision": 0.7133550488599348, + "recall": 0.6908517350157729, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8925081433224756, - "precision": 0.9072847682119205, - "recall": 0.8782051282051282, - "support": 156.0 + "f1-score": 0.9260450160771705, + "precision": 0.9230769230769231, + "recall": 0.9290322580645162, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8810511756569848, - "precision": 0.885952712100139, - "recall": 0.8762035763411279, - "support": 727.0 + "f1-score": 0.8985507246376812, + "precision": 0.8931572629051621, + "recall": 0.9040097205346294, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6376404494382022, - "precision": 0.6189502385821404, - "recall": 0.6574945691527878, - "support": 4143.0 + "f1-score": 0.6915371329879102, + "precision": 0.6917760884588804, + "recall": 0.6912983425414365, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8858131487889275, - "precision": 0.896358543417367, - "recall": 0.8755129958960328, - "support": 2193.0 + "f1-score": 0.9147286821705426, + "precision": 0.9085394307046197, + "recall": 0.9210028382213813, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.8946945982823662, - "precision": 0.9022258195062728, - "recall": 0.8872880681365916, - "support": 12563.0 + "f1-score": 0.912640635340834, + "precision": 0.9131778382753293, + "recall": 0.9121040640846623, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999202361011406, - "precision": 0.9998404849258254, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9993222177036736, + "precision": 0.9998191681735985, + "recall": 0.9988257609971999, + "support": 11071.0 }, - "eval_accuracy": 0.8984147425873118, - "eval_loss": 0.3937377631664276, + "eval_accuracy": 0.9104252104467947, + "eval_loss": 0.3896510601043701, "eval_macro avg": { - "f1-score": 0.8361081885280585, - "precision": 0.836974433685577, - "recall": 0.8356115155597299, - "support": 32613.0 - }, - "eval_runtime": 5.0071, - "eval_samples_per_second": 16.177, - "eval_steps_per_second": 2.197, + "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": 0.8994624730308206, - "precision": 0.9007308115646135, - "recall": 0.8984147425873118, - "support": 32613.0 + "f1-score": 0.9103982293589928, + "precision": 0.9103825180220693, + "recall": 0.9104252104467947, + "support": 32431.0 }, "step": 648 }, { "epoch": 9.0, "eval_B-Claim": { - "f1-score": 0.6821192052980132, - "precision": 0.6666666666666666, - "recall": 0.6983050847457627, - "support": 295.0 + "f1-score": 0.6862745098039215, + "precision": 0.6171284634760705, + "recall": 0.7728706624605678, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8510638297872339, - "precision": 0.9523809523809523, - "recall": 0.7692307692307693, - "support": 156.0 + "f1-score": 0.9119496855345912, + "precision": 0.8895705521472392, + "recall": 0.9354838709677419, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8959891230455472, - "precision": 0.885752688172043, - "recall": 0.90646492434663, - "support": 727.0 + "f1-score": 0.8690095846645367, + "precision": 0.9164420485175202, + "recall": 0.8262454434993924, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6547967670141736, - "precision": 0.636094674556213, - "recall": 0.6746319092445088, - "support": 4143.0 + "f1-score": 0.6841291120965247, + "precision": 0.619652531290865, + "recall": 0.7635819521178637, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8507386653082017, - "precision": 0.9636468551644547, - "recall": 0.7615139078887369, - "support": 2193.0 + "f1-score": 0.9025385312783318, + "precision": 0.8664055700609226, + "recall": 0.9418164616840113, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9060476867203537, - "precision": 0.8987391338397682, - "recall": 0.9134760805540078, - "support": 12563.0 + "f1-score": 0.8949834831374357, + "precision": 0.9374748531423513, + "recall": 0.8561769677371941, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999202361011406, - "precision": 0.9998404849258254, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9989151071331707, + "precision": 0.9998190209030857, + "recall": 0.9980128263029536, + "support": 11071.0 }, - "eval_accuracy": 0.9033820868978628, - "eval_loss": 0.4608707129955292, + "eval_accuracy": 0.8965804323024267, + "eval_loss": 0.4196818470954895, "eval_macro avg": { - "f1-score": 0.8343822161820948, - "precision": 0.8575887793865605, - "recall": 0.8176603822872022, - "support": 32613.0 - }, - "eval_runtime": 4.9809, - "eval_samples_per_second": 16.262, - "eval_steps_per_second": 2.208, + "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": 0.9039813964448579, - "precision": 0.9064685293306464, - "recall": 0.9033820868978628, - "support": 32613.0 + "f1-score": 0.9000940098435676, + "precision": 0.9076597589527613, + "recall": 0.8965804323024267, + "support": 32431.0 }, "step": 729 }, { "epoch": 10.0, "eval_B-Claim": { - "f1-score": 0.6863354037267081, - "precision": 0.6332378223495702, - "recall": 0.7491525423728813, - "support": 295.0 + "f1-score": 0.6905537459283387, + "precision": 0.7138047138047138, + "recall": 0.668769716088328, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8976897689768977, - "precision": 0.9251700680272109, - "recall": 0.8717948717948718, - "support": 156.0 + "f1-score": 0.9142857142857143, + "precision": 0.9, + "recall": 0.9290322580645162, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8780141843971632, - "precision": 0.9062957540263543, - "recall": 0.8514442916093535, - "support": 727.0 + "f1-score": 0.894484412470024, + "precision": 0.8828402366863906, + "recall": 0.9064398541919806, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6615143369175628, - "precision": 0.6171368861024034, - "recall": 0.7127685252232682, - "support": 4143.0 + "f1-score": 0.681682755333095, + "precision": 0.7066963182604399, + "recall": 0.6583793738489871, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9019424292066465, - "precision": 0.9264423076923077, - "recall": 0.8787049703602371, - "support": 2193.0 + "f1-score": 0.9083702494754021, + "precision": 0.895632183908046, + "recall": 0.9214758751182592, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.8934422897101273, - "precision": 0.9130796077779625, - "recall": 0.8746318554485394, - "support": 12563.0 + "f1-score": 0.9129089784183134, + "precision": 0.904253785147801, + "recall": 0.9217314617476299, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999601132782897, - "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9983716301791207, + "precision": 0.999909395669113, + "recall": 0.9968385873001535, + "support": 11071.0 }, - "eval_accuracy": 0.9008370895041854, - "eval_loss": 0.49426138401031494, + "eval_accuracy": 0.9092534920292313, + "eval_loss": 0.5311424136161804, "eval_macro avg": { - "f1-score": 0.8455569323161993, - "precision": 0.8459089208536871, - "recall": 0.8483453266496436, - "support": 32613.0 - }, - "eval_runtime": 4.962, - "eval_samples_per_second": 16.324, - "eval_steps_per_second": 2.217, + "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": 0.9032978864532892, - "precision": 0.9071693470023781, - "recall": 0.9008370895041854, - "support": 32613.0 + "f1-score": 0.9081813659228948, + "precision": 0.9074585127295745, + "recall": 0.9092534920292313, + "support": 32431.0 }, "step": 810 }, { "epoch": 11.0, "eval_B-Claim": { - "f1-score": 0.6654991243432573, - "precision": 0.6884057971014492, - "recall": 0.6440677966101694, - "support": 295.0 + "f1-score": 0.680921052631579, + "precision": 0.711340206185567, + "recall": 0.6529968454258676, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8888888888888888, - "precision": 0.9361702127659575, - "recall": 0.8461538461538461, - "support": 156.0 + "f1-score": 0.9009584664536742, + "precision": 0.8924050632911392, + "recall": 0.9096774193548387, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8911290322580646, - "precision": 0.871222076215506, - "recall": 0.9119669876203577, - "support": 727.0 + "f1-score": 0.8942020322773461, + "precision": 0.88, + "recall": 0.9088699878493317, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6376628304034401, - "precision": 0.6697662061636557, - "recall": 0.6084962587496983, - "support": 4143.0 + "f1-score": 0.6838570400575678, + "precision": 0.713820731096645, + "recall": 0.6563075506445673, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8884068450228971, - "precision": 0.9422290388548057, - "recall": 0.8404012767897857, - "support": 2193.0 + "f1-score": 0.902502979737783, + "precision": 0.9096588178760211, + "recall": 0.8954588457899716, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.904972804972805, - "precision": 0.8838885937618578, - "recall": 0.9270874791053092, - "support": 12563.0 + "f1-score": 0.9134434303925829, + "precision": 0.9004641199571581, + "recall": 0.9268023811273609, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999202361011406, - "precision": 0.9998404849258254, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9988699543461556, + "precision": 0.999728555917481, + "recall": 0.9980128263029536, + "support": 11071.0 }, - "eval_accuracy": 0.9055284702419281, - "eval_loss": 0.5303362607955933, + "eval_accuracy": 0.9096235083716198, + "eval_loss": 0.49213308095932007, "eval_macro avg": { - "f1-score": 0.8394971088557848, - "precision": 0.8559317728270083, - "recall": 0.8254533778613096, - "support": 32613.0 - }, - "eval_runtime": 4.9749, - "eval_samples_per_second": 16.282, - "eval_steps_per_second": 2.211, + "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": 0.9038459038093231, - "precision": 0.9033803424876604, - "recall": 0.9055284702419281, - "support": 32613.0 + "f1-score": 0.9083194817194259, + "precision": 0.9075428987655072, + "recall": 0.9096235083716198, + "support": 32431.0 }, "step": 891 }, { "epoch": 12.0, "eval_B-Claim": { - "f1-score": 0.6374045801526718, - "precision": 0.7292576419213974, - "recall": 0.5661016949152542, - "support": 295.0 + "f1-score": 0.6577181208053691, + "precision": 0.7025089605734767, + "recall": 0.6182965299684543, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9114754098360656, - "precision": 0.9328859060402684, - "recall": 0.8910256410256411, - "support": 156.0 + "f1-score": 0.8978328173374612, + "precision": 0.8630952380952381, + "recall": 0.9354838709677419, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8913612565445026, - "precision": 0.850187265917603, - "recall": 0.936726272352132, - "support": 727.0 + "f1-score": 0.8902147971360382, + "precision": 0.8745603751465416, + "recall": 0.9064398541919806, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.5983549421441516, - "precision": 0.7082508250825083, - "recall": 0.5179821385469466, - "support": 4143.0 + "f1-score": 0.6518804243008679, + "precision": 0.6842105263157895, + "recall": 0.6224677716390423, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9206057824690226, - "precision": 0.9265588914549654, - "recall": 0.9147286821705426, - "support": 2193.0 + "f1-score": 0.8919829328542555, + "precision": 0.8490808037622916, + "recall": 0.9394512771996215, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9017204628501827, - "precision": 0.8640309285870595, - "recall": 0.9428480458489215, - "support": 12563.0 + "f1-score": 0.9075452422650322, + "precision": 0.9011665821317296, + "recall": 0.9140148453002132, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9997606319317004, + "f1-score": 0.9985526910900046, "precision": 1.0, - "recall": 0.9995213784301212, - "support": 12536.0 + "recall": 0.9971095655315689, + "support": 11071.0 }, - "eval_accuracy": 0.9049765430963113, - "eval_loss": 0.5691992044448853, + "eval_accuracy": 0.9020073386574574, + "eval_loss": 0.5465030670166016, "eval_macro avg": { - "f1-score": 0.8372404379897568, - "precision": 0.8587387798576859, - "recall": 0.8241334076127942, - "support": 32613.0 - }, - "eval_runtime": 4.9632, - "eval_samples_per_second": 16.32, - "eval_steps_per_second": 2.216, + "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": 0.8995624631750451, - "precision": 0.8995123024912762, - "recall": 0.9049765430963113, - "support": 32613.0 + "f1-score": 0.9004246846165758, + "precision": 0.8996509302731299, + "recall": 0.9020073386574574, + "support": 32431.0 }, "step": 972 }, { "epoch": 12.35, - "grad_norm": 8.504526138305664, + "grad_norm": 0.1318705528974533, "learning_rate": 1.506172839506173e-05, - "loss": 0.0319, + "loss": 0.0258, "step": 1000 }, { "epoch": 13.0, "eval_B-Claim": { - "f1-score": 0.6779661016949152, - "precision": 0.6779661016949152, - "recall": 0.6779661016949152, - "support": 295.0 + "f1-score": 0.6857142857142857, + "precision": 0.6551724137931034, + "recall": 0.7192429022082019, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9096774193548387, - "precision": 0.9155844155844156, - "recall": 0.9038461538461539, - "support": 156.0 + "f1-score": 0.9022082018927444, + "precision": 0.8827160493827161, + "recall": 0.9225806451612903, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8873626373626373, - "precision": 0.8861454046639232, - "recall": 0.8885832187070152, - "support": 727.0 + "f1-score": 0.8858560794044665, + "precision": 0.9049429657794676, + "recall": 0.8675577156743621, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.663892288861689, - "precision": 0.6734541842562701, - "recall": 0.6545981173062998, - "support": 4143.0 + "f1-score": 0.6909411894516165, + "precision": 0.6634880271243908, + "recall": 0.7207642725598526, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9122568973315242, - "precision": 0.9048900852400179, - "recall": 0.9197446420428637, - "support": 2193.0 + "f1-score": 0.9048498845265589, + "precision": 0.8840252707581228, + "recall": 0.9266792809839167, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9055930186433956, - "precision": 0.9025855934213648, - "recall": 0.9086205524158243, - "support": 12563.0 + "f1-score": 0.906777130915062, + "precision": 0.9211464098870271, + "recall": 0.8928492687587272, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9998404340194671, + "f1-score": 0.9971466099008107, "precision": 1.0, - "recall": 0.9996809189534142, - "support": 12536.0 + "recall": 0.9943094571402764, + "support": 11071.0 }, - "eval_accuracy": 0.9095452733572502, - "eval_loss": 0.5995727777481079, + "eval_accuracy": 0.9044432795781814, + "eval_loss": 0.5739177465438843, "eval_macro avg": { - "f1-score": 0.8509412567526382, - "precision": 0.8515179692658439, - "recall": 0.8504342435666409, - "support": 32613.0 - }, - "eval_runtime": 4.967, - "eval_samples_per_second": 16.308, - "eval_steps_per_second": 2.215, + "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": 0.9091182110651892, - "precision": 0.908741493590885, - "recall": 0.9095452733572502, - "support": 32613.0 + "f1-score": 0.9058771229581195, + "precision": 0.9079380628166581, + "recall": 0.9044432795781814, + "support": 32431.0 }, "step": 1053 }, { "epoch": 14.0, "eval_B-Claim": { - "f1-score": 0.6543438077634012, - "precision": 0.7195121951219512, - "recall": 0.6, - "support": 295.0 + "f1-score": 0.6666666666666666, + "precision": 0.7276119402985075, + "recall": 0.6151419558359621, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8918032786885245, - "precision": 0.912751677852349, - "recall": 0.8717948717948718, - "support": 156.0 + "f1-score": 0.9003215434083601, + "precision": 0.8974358974358975, + "recall": 0.9032258064516129, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.895364238410596, - "precision": 0.8633461047254151, - "recall": 0.9298486932599724, - "support": 727.0 + "f1-score": 0.8976470588235295, + "precision": 0.8700114025085519, + "recall": 0.9270959902794653, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.612272977435333, - "precision": 0.712, - "recall": 0.5370504465363264, - "support": 4143.0 + "f1-score": 0.6725553620303559, + "precision": 0.7317271250676773, + "recall": 0.6222375690607734, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9033142857142857, - "precision": 0.9055912007332723, - "recall": 0.9010487916096671, - "support": 2193.0 + "f1-score": 0.8921406655652585, + "precision": 0.8902496467263307, + "recall": 0.8940397350993378, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9070541005543873, - "precision": 0.8727192466156563, - "recall": 0.9442012258218578, - "support": 12563.0 + "f1-score": 0.9186321600114704, + "precision": 0.8966482401511441, + "recall": 0.9417211729257, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9977821029285294, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.995574022220215, + "support": 11071.0 }, - "eval_accuracy": 0.9072455769171802, - "eval_loss": 0.6337578892707825, + "eval_accuracy": 0.9104560451419937, + "eval_loss": 0.5471230745315552, "eval_macro avg": { - "f1-score": 0.8377360983666469, - "precision": 0.8551314892926634, - "recall": 0.8262777184318136, - "support": 32613.0 - }, - "eval_runtime": 4.9902, - "eval_samples_per_second": 16.232, - "eval_steps_per_second": 2.204, + "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": 0.9024627717609818, - "precision": 0.902034384321647, - "recall": 0.9072455769171802, - "support": 32613.0 + "f1-score": 0.9078809171606194, + "precision": 0.9070974660556812, + "recall": 0.9104560451419937, + "support": 32431.0 }, "step": 1134 }, { "epoch": 15.0, "eval_B-Claim": { - "f1-score": 0.6715867158671587, - "precision": 0.7368421052631579, - "recall": 0.6169491525423729, - "support": 295.0 + "f1-score": 0.6688524590163935, + "precision": 0.6962457337883959, + "recall": 0.6435331230283912, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8902077151335311, - "precision": 0.8287292817679558, - "recall": 0.9615384615384616, - "support": 156.0 + "f1-score": 0.9061488673139159, + "precision": 0.9090909090909091, + "recall": 0.9032258064516129, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.896551724137931, - "precision": 0.8816489361702128, - "recall": 0.9119669876203577, - "support": 727.0 + "f1-score": 0.8902147971360382, + "precision": 0.8745603751465416, + "recall": 0.9064398541919806, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6393111798735368, - "precision": 0.7221884498480243, - "recall": 0.5734974656046343, - "support": 4143.0 + "f1-score": 0.6658545498842734, + "precision": 0.7071151358344114, + "recall": 0.6291436464088398, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.898728813559322, - "precision": 0.8393351800554016, - "recall": 0.9671682626538988, - "support": 2193.0 + "f1-score": 0.8983730252298987, + "precision": 0.8956276445698167, + "recall": 0.9011352885525071, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9101119213820535, - "precision": 0.892125382262997, - "recall": 0.9288386531879328, - "support": 12563.0 + "f1-score": 0.9134990605578841, + "precision": 0.8985002487739001, + "recall": 0.9290071286837657, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, - "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9998645047649157, + "precision": 0.9999096657633243, + "recall": 0.999819347845723, + "support": 11071.0 }, - "eval_accuracy": 0.9105878024100819, - "eval_loss": 0.6874175667762756, + "eval_accuracy": 0.9077117572692794, + "eval_loss": 0.6182886362075806, "eval_macro avg": { - "f1-score": 0.8437854385647904, - "precision": 0.8429813336239642, - "recall": 0.8514227118782368, - "support": 32613.0 - }, - "eval_runtime": 4.9796, - "eval_samples_per_second": 16.266, - "eval_steps_per_second": 2.209, + "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": 0.9069419606556991, - "precision": 0.9065117879543761, - "recall": 0.9105878024100819, - "support": 32613.0 + "f1-score": 0.9058074210256117, + "precision": 0.9047621091758847, + "recall": 0.9077117572692794, + "support": 32431.0 }, "step": 1215 }, { "epoch": 16.0, "eval_B-Claim": { - "f1-score": 0.676923076923077, - "precision": 0.6827586206896552, - "recall": 0.6711864406779661, - "support": 295.0 + "f1-score": 0.6643835616438356, + "precision": 0.7265917602996255, + "recall": 0.61198738170347, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8980891719745223, - "precision": 0.8924050632911392, - "recall": 0.9038461538461539, - "support": 156.0 + "f1-score": 0.9113924050632912, + "precision": 0.8944099378881988, + "recall": 0.9290322580645162, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8902606310013718, - "precision": 0.8878248974008208, - "recall": 0.8927097661623109, - "support": 727.0 + "f1-score": 0.8954518606024808, + "precision": 0.871264367816092, + "recall": 0.9210206561360875, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6403442166540116, - "precision": 0.6730513434424049, - "recall": 0.6106685976345644, - "support": 4143.0 + "f1-score": 0.6607098535616778, + "precision": 0.7167474421109317, + "recall": 0.6127992633517495, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9115737631520036, - "precision": 0.8953386103781882, - "recall": 0.9284085727314182, - "support": 2193.0 + "f1-score": 0.9103066635923449, + "precision": 0.8879892037786775, + "recall": 0.9337748344370861, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9030989460437312, - "precision": 0.8924990283715507, - "recall": 0.9139536734856324, - "support": 12563.0 + "f1-score": 0.9118409680207432, + "precision": 0.8940046606878045, + "recall": 0.9304034688028221, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9983714828553334, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9967482612230151, + "support": 11071.0 }, - "eval_accuracy": 0.9067549750099654, - "eval_loss": 0.6374069452285767, + "eval_accuracy": 0.90737257562209, + "eval_loss": 0.6626638174057007, "eval_macro avg": { - "f1-score": 0.8457556865355311, - "precision": 0.8462682233676798, - "recall": 0.8458247435054351, - "support": 32613.0 - }, - "eval_runtime": 5.004, - "eval_samples_per_second": 16.187, - "eval_steps_per_second": 2.198, + "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": 0.905181158084066, - "precision": 0.9041324706485238, - "recall": 0.9067549750099654, - "support": 32613.0 + "f1-score": 0.9048051223322606, + "precision": 0.9038418545958269, + "recall": 0.90737257562209, + "support": 32431.0 }, "step": 1296 }, { "epoch": 17.0, "eval_B-Claim": { - "f1-score": 0.7001620745542951, - "precision": 0.6708074534161491, - "recall": 0.7322033898305085, - "support": 295.0 + "f1-score": 0.6856240126382307, + "precision": 0.6867088607594937, + "recall": 0.6845425867507886, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.910828025477707, - "precision": 0.9050632911392406, - "recall": 0.9166666666666666, - "support": 156.0 + "f1-score": 0.9131832797427654, + "precision": 0.9102564102564102, + "recall": 0.9161290322580645, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8870175438596491, - "precision": 0.9054441260744985, - "recall": 0.8693259972489684, - "support": 727.0 + "f1-score": 0.8919902912621359, + "precision": 0.8909090909090909, + "recall": 0.8930741190765492, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6722241519045965, - "precision": 0.6459724076546507, - "recall": 0.7006999758629012, - "support": 4143.0 + "f1-score": 0.6720430107526882, + "precision": 0.682573599240266, + "recall": 0.6618324125230203, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9159550561797752, - "precision": 0.9029685423128047, - "recall": 0.9293205654354765, - "support": 2193.0 + "f1-score": 0.9010318949343339, + "precision": 0.8934883720930232, + "recall": 0.9087038789025544, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.8987500505642975, - "precision": 0.9137193617371279, - "recall": 0.884263312902969, - "support": 12563.0 + "f1-score": 0.9102217019097096, + "precision": 0.9062431703941138, + "recall": 0.9142353200558536, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.999600989545926, + "f1-score": 0.9988243805389763, "precision": 1.0, - "recall": 0.9992022973835354, - "support": 12536.0 + "recall": 0.9976515219943998, + "support": 11071.0 }, - "eval_accuracy": 0.9066016619139607, - "eval_loss": 0.5605483055114746, + "eval_accuracy": 0.90576917147174, + "eval_loss": 0.6958470940589905, "eval_macro avg": { - "f1-score": 0.8549339845837495, - "precision": 0.8491393117620675, - "recall": 0.8616688864758607, - "support": 32613.0 - }, - "eval_runtime": 5.0022, - "eval_samples_per_second": 16.193, - "eval_steps_per_second": 2.199, + "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": 0.9078958614075368, - "precision": 0.9097251400220888, - "recall": 0.9066016619139607, - "support": 32613.0 + "f1-score": 0.9053221272704657, + "precision": 0.9049421722279843, + "recall": 0.90576917147174, + "support": 32431.0 }, "step": 1377 }, { "epoch": 18.0, "eval_B-Claim": { - "f1-score": 0.7140495867768595, - "precision": 0.6967741935483871, - "recall": 0.7322033898305085, - "support": 295.0 + "f1-score": 0.688, + "precision": 0.698051948051948, + "recall": 0.6782334384858044, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9066666666666667, - "precision": 0.9444444444444444, - "recall": 0.8717948717948718, - "support": 156.0 + "f1-score": 0.9180327868852459, + "precision": 0.9333333333333333, + "recall": 0.9032258064516129, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8988300068823124, - "precision": 0.8994490358126722, - "recall": 0.8982118294360385, - "support": 727.0 + "f1-score": 0.8942307692307692, + "precision": 0.8846611177170036, + "recall": 0.9040097205346294, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6946223564954682, - "precision": 0.6955469506292352, - "recall": 0.6937002172338885, - "support": 4143.0 + "f1-score": 0.6798171758479673, + "precision": 0.7118387909319899, + "recall": 0.6505524861878453, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9217432052483598, - "precision": 0.9479518072289157, - "recall": 0.8969448244414044, - "support": 2193.0 + "f1-score": 0.9276937618147448, + "precision": 0.9268177525967894, + "recall": 0.9285714285714286, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9124099152609487, - "precision": 0.9078086833188874, - "recall": 0.9170580275411924, - "support": 12563.0 + "f1-score": 0.9138506163886876, + "precision": 0.9018822013883919, + "recall": 0.9261409568604395, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999601132782897, - "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9999096739228616, + "precision": 0.9999096739228616, + "recall": 0.9999096739228616, + "support": 11071.0 }, - "eval_accuracy": 0.916873639346273, - "eval_loss": 0.636456310749054, + "eval_accuracy": 0.911473590083562, + "eval_loss": 0.6612581014633179, "eval_macro avg": { - "f1-score": 0.8640402643727008, - "precision": 0.8702821592832203, - "recall": 0.8585476271451797, - "support": 32613.0 - }, - "eval_runtime": 4.9811, - "eval_samples_per_second": 16.261, - "eval_steps_per_second": 2.208, + "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": 0.9168994897940371, - "precision": 0.917060404983014, - "recall": 0.916873639346273, - "support": 32613.0 + "f1-score": 0.9100977079524204, + "precision": 0.9092367421655284, + "recall": 0.911473590083562, + "support": 32431.0 }, "step": 1458 }, { "epoch": 18.52, - "grad_norm": 2.2237088680267334, + "grad_norm": 0.010771242901682854, "learning_rate": 1.2592592592592593e-05, - "loss": 0.0112, + "loss": 0.0083, "step": 1500 }, { "epoch": 19.0, "eval_B-Claim": { - "f1-score": 0.6907216494845362, - "precision": 0.7003484320557491, - "recall": 0.6813559322033899, - "support": 295.0 + "f1-score": 0.6950819672131147, + "precision": 0.7235494880546075, + "recall": 0.668769716088328, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9067524115755627, - "precision": 0.9096774193548387, - "recall": 0.9038461538461539, - "support": 156.0 + "f1-score": 0.9049180327868852, + "precision": 0.92, + "recall": 0.8903225806451613, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8940533151059467, - "precision": 0.8885869565217391, - "recall": 0.8995873452544704, - "support": 727.0 + "f1-score": 0.9021479713603818, + "precision": 0.8862837045720985, + "recall": 0.9185905224787363, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6695981433980701, - "precision": 0.6777942631058358, - "recall": 0.6615978759353126, - "support": 4143.0 + "f1-score": 0.6773359596208298, + "precision": 0.7279703625297698, + "recall": 0.6332872928176796, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9135184342284934, - "precision": 0.9118582462517038, - "recall": 0.9151846785225718, - "support": 2193.0 + "f1-score": 0.8959923664122137, + "precision": 0.9037536092396535, + "recall": 0.8883632923368022, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9059761272157671, - "precision": 0.9027185079816659, - "recall": 0.9092573429913238, - "support": 12563.0 + "f1-score": 0.9165171200918814, + "precision": 0.8956857243072606, + "recall": 0.9383405600058793, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9978274644699919, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9956643482973534, + "support": 11071.0 }, - "eval_accuracy": 0.9107717781252874, - "eval_loss": 0.6661651134490967, + "eval_accuracy": 0.9104252104467947, + "eval_loss": 0.6561155319213867, "eval_macro avg": { - "f1-score": 0.8543742972869109, - "precision": 0.8558548321816474, - "recall": 0.8529756183933175, - "support": 32613.0 - }, - "eval_runtime": 4.9645, - "eval_samples_per_second": 16.316, - "eval_steps_per_second": 2.216, + "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": 0.9103873346045664, - "precision": 0.9100412278575488, - "recall": 0.9107717781252874, - "support": 32613.0 + "f1-score": 0.9083143486955053, + "precision": 0.9075517471550982, + "recall": 0.9104252104467947, + "support": 32431.0 }, "step": 1539 }, { "epoch": 20.0, "eval_B-Claim": { - "f1-score": 0.697278911564626, - "precision": 0.6996587030716723, - "recall": 0.6949152542372882, - "support": 295.0 + "f1-score": 0.6833333333333333, + "precision": 0.7243816254416962, + "recall": 0.6466876971608833, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9072847682119205, - "precision": 0.9383561643835616, - "recall": 0.8782051282051282, - "support": 156.0 + "f1-score": 0.9050632911392404, + "precision": 0.8881987577639752, + "recall": 0.9225806451612903, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8958475153165418, - "precision": 0.8867924528301887, - "recall": 0.9050894085281981, - "support": 727.0 + "f1-score": 0.8997613365155133, + "precision": 0.8839390386869871, + "recall": 0.9161603888213852, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6639919759277834, - "precision": 0.6908426819723454, - "recall": 0.6391503741250302, - "support": 4143.0 + "f1-score": 0.6925646418409007, + "precision": 0.7485958812516715, + "recall": 0.6443370165745856, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9188432835820896, - "precision": 0.9403341288782816, - "recall": 0.898312813497492, - "support": 2193.0 + "f1-score": 0.9094717244589248, + "precision": 0.8950984883188273, + "recall": 0.924314096499527, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9079586401378662, - "precision": 0.8937466265710541, - "recall": 0.9226299450768128, - "support": 12563.0 + "f1-score": 0.9203195854027206, + "precision": 0.9017561182029762, + "recall": 0.9396634085397222, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9982808541440463, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.9965676090687381, + "support": 11071.0 }, - "eval_accuracy": 0.9120289455125257, - "eval_loss": 0.6076620817184448, + "eval_accuracy": 0.9149887453362523, + "eval_loss": 0.6437746286392212, "eval_macro avg": { - "f1-score": 0.8558807440027308, - "precision": 0.8642472511010147, - "recall": 0.8483175933440433, - "support": 32613.0 - }, - "eval_runtime": 4.9557, - "eval_samples_per_second": 16.345, - "eval_steps_per_second": 2.22, + "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": 0.910883429910181, - "precision": 0.9102484848959365, - "recall": 0.9120289455125257, - "support": 32613.0 + "f1-score": 0.9128082705193213, + "precision": 0.9120938553978903, + "recall": 0.9149887453362523, + "support": 32431.0 }, "step": 1620 }, { "epoch": 21.0, "eval_B-Claim": { - "f1-score": 0.6752767527675276, - "precision": 0.7408906882591093, - "recall": 0.6203389830508474, - "support": 295.0 + "f1-score": 0.6677631578947368, + "precision": 0.697594501718213, + "recall": 0.6403785488958991, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9120521172638437, - "precision": 0.9271523178807947, - "recall": 0.8974358974358975, - "support": 156.0 + "f1-score": 0.8903654485049834, + "precision": 0.9178082191780822, + "recall": 0.864516129032258, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8992042440318302, - "precision": 0.8681177976952625, - "recall": 0.9325997248968363, - "support": 727.0 + "f1-score": 0.8954869358669834, + "precision": 0.875725900116144, + "recall": 0.9161603888213852, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6303986710963455, - "precision": 0.73904576436222, - "recall": 0.5496017378711079, - "support": 4143.0 + "f1-score": 0.6630760023937762, + "precision": 0.6906008476689105, + "recall": 0.6376611418047882, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9182505152278452, - "precision": 0.922263109475621, - "recall": 0.9142726858185134, - "support": 2193.0 + "f1-score": 0.8958130477117819, + "precision": 0.9227683049147443, + "recall": 0.8703878902554399, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.909555793008701, - "precision": 0.8736163037900447, - "recall": 0.9485791610284168, - "support": 12563.0 + "f1-score": 0.9106195648251723, + "precision": 0.8945129732028924, + "recall": 0.9273168222238554, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999202361011406, - "precision": 0.9998404849258254, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9977821029285294, + "precision": 1.0, + "recall": 0.995574022220215, + "support": 11071.0 }, - "eval_accuracy": 0.9117836445589182, - "eval_loss": 0.685625433921814, + "eval_accuracy": 0.9047207918349727, + "eval_loss": 0.7204756736755371, "eval_macro avg": { - "f1-score": 0.8492369042138906, - "precision": 0.8672752094841254, - "recall": 0.8375468843002313, - "support": 32613.0 - }, - "eval_runtime": 4.9777, - "eval_samples_per_second": 16.273, - "eval_steps_per_second": 2.21, + "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": 0.9070745665639394, - "precision": 0.9072440958491244, - "recall": 0.9117836445589182, - "support": 32613.0 + "f1-score": 0.9033970685809433, + "precision": 0.9027615692407923, + "recall": 0.9047207918349727, + "support": 32431.0 }, "step": 1701 }, { "epoch": 22.0, "eval_B-Claim": { - "f1-score": 0.7060653188180404, - "precision": 0.6522988505747126, - "recall": 0.7694915254237288, - "support": 295.0 + "f1-score": 0.6976, + "precision": 0.7077922077922078, + "recall": 0.6876971608832808, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9032258064516129, - "precision": 0.9090909090909091, - "recall": 0.8974358974358975, - "support": 156.0 + "f1-score": 0.9155844155844156, + "precision": 0.9215686274509803, + "recall": 0.9096774193548387, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8852459016393444, - "precision": 0.9186390532544378, - "recall": 0.8541953232462174, - "support": 727.0 + "f1-score": 0.8986731001206273, + "precision": 0.8922155688622755, + "recall": 0.905224787363305, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6783193277310924, - "precision": 0.6329987452948557, - "recall": 0.7306299782766111, - "support": 4143.0 + "f1-score": 0.695459320649002, + "precision": 0.7054226852948141, + "recall": 0.6857734806629834, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9076923076923077, - "precision": 0.9007633587786259, - "recall": 0.9147286821705426, - "support": 2193.0 + "f1-score": 0.9150853889943075, + "precision": 0.917697431018078, + "recall": 0.9124881740775781, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.8980493191019507, - "precision": 0.9234650967199327, - "recall": 0.8739950648730399, - "support": 12563.0 + "f1-score": 0.9140016812251014, + "precision": 0.9091173476806748, + "recall": 0.9189387815095171, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.999322095177837, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9986451088429229, + "support": 11071.0 }, - "eval_accuracy": 0.9056817833379327, - "eval_loss": 0.714400589466095, + "eval_accuracy": 0.9118436064259505, + "eval_loss": 0.7220445871353149, "eval_macro avg": { - "f1-score": 0.8540854259191927, - "precision": 0.8481794305304963, - "recall": 0.8629252102037197, - "support": 32613.0 - }, - "eval_runtime": 5.1631, - "eval_samples_per_second": 15.688, - "eval_steps_per_second": 2.13, + "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": 0.9079755332777275, - "precision": 0.911828926872852, - "recall": 0.9056817833379327, - "support": 32613.0 + "f1-score": 0.9114287052262304, + "precision": 0.9110799834813146, + "recall": 0.9118436064259505, + "support": 32431.0 }, "step": 1782 }, { "epoch": 23.0, "eval_B-Claim": { - "f1-score": 0.6942148760330579, - "precision": 0.6774193548387096, - "recall": 0.711864406779661, - "support": 295.0 + "f1-score": 0.6832504145936982, + "precision": 0.7202797202797203, + "recall": 0.6498422712933754, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.88659793814433, - "precision": 0.9555555555555556, - "recall": 0.8269230769230769, - "support": 156.0 + "f1-score": 0.910828025477707, + "precision": 0.89937106918239, + "recall": 0.9225806451612903, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8945205479452054, - "precision": 0.8908594815825375, - "recall": 0.8982118294360385, - "support": 727.0 + "f1-score": 0.8956469886702446, + "precision": 0.8793911007025761, + "recall": 0.9125151883353585, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6700834326579261, - "precision": 0.6618789733929833, - "recall": 0.6784938450398262, - "support": 4143.0 + "f1-score": 0.676206645646374, + "precision": 0.7145053818554588, + "recall": 0.641804788213628, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8958790538892953, - "precision": 0.9627882599580713, - "recall": 0.8376652986776105, - "support": 2193.0 + "f1-score": 0.9027875380651206, + "precision": 0.894199535962877, + "recall": 0.9115421002838221, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9067846840795036, - "precision": 0.900345260514752, - "recall": 0.9133168829101329, - "support": 12563.0 + "f1-score": 0.913811834576664, + "precision": 0.9007639037624046, + "recall": 0.9272433306386418, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9998644925245043, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9997290217685846, + "support": 11071.0 }, - "eval_accuracy": 0.909146659307638, - "eval_loss": 0.6947330236434937, + "eval_accuracy": 0.9096235083716198, + "eval_loss": 0.7609456181526184, "eval_macro avg": { - "f1-score": 0.8497257903927597, - "precision": 0.8641209836918013, - "recall": 0.8380679056809066, - "support": 32613.0 - }, - "eval_runtime": 4.9764, - "eval_samples_per_second": 16.277, - "eval_steps_per_second": 2.21, + "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": 0.9095199312515945, - "precision": 0.9105926154648943, - "recall": 0.909146659307638, - "support": 32613.0 + "f1-score": 0.907913975857415, + "precision": 0.9069505627094704, + "recall": 0.9096235083716198, + "support": 32431.0 }, "step": 1863 }, { "epoch": 24.0, "eval_B-Claim": { - "f1-score": 0.7096774193548386, - "precision": 0.676923076923077, - "recall": 0.7457627118644068, - "support": 295.0 + "f1-score": 0.6968749999999999, + "precision": 0.6904024767801857, + "recall": 0.7034700315457413, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9130434782608695, - "precision": 0.8855421686746988, - "recall": 0.9423076923076923, - "support": 156.0 + "f1-score": 0.9131832797427654, + "precision": 0.9102564102564102, + "recall": 0.9161290322580645, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8896746817538898, - "precision": 0.9155749636098981, - "recall": 0.8651994497936726, - "support": 727.0 + "f1-score": 0.8922702373706634, + "precision": 0.8939024390243903, + "recall": 0.8906439854191981, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6777306967984934, - "precision": 0.6613829542844015, - "recall": 0.6949070721699252, - "support": 4143.0 + "f1-score": 0.6863417982155113, + "precision": 0.6821282401091405, + "recall": 0.6906077348066298, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9096057503811806, - "precision": 0.8707256046705588, - "recall": 0.9521203830369357, - "support": 2193.0 + "f1-score": 0.903548463919981, + "precision": 0.9098321342925659, + "recall": 0.8973509933774835, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9036461494880823, - "precision": 0.9190813302601252, - "recall": 0.8887208469314655, - "support": 12563.0 + "f1-score": 0.9113179092948837, + "precision": 0.9122238586156112, + "recall": 0.910413757624752, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9999096657633243, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.999819347845723, + "support": 11071.0 }, - "eval_accuracy": 0.9095759359764511, - "eval_loss": 0.7040016055107117, + "eval_accuracy": 0.908143443002066, + "eval_loss": 0.7767113447189331, "eval_macro avg": { - "f1-score": 0.8576254537196221, - "precision": 0.8470328712032513, - "recall": 0.8698597365862997, - "support": 32613.0 - }, - "eval_runtime": 4.987, - "eval_samples_per_second": 16.242, - "eval_steps_per_second": 2.206, + "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": 0.9103637622541415, - "precision": 0.9117679902876374, - "recall": 0.9095759359764511, - "support": 32613.0 + "f1-score": 0.9083489289362786, + "precision": 0.9085692481624786, + "recall": 0.908143443002066, + "support": 32431.0 }, "step": 1944 }, { "epoch": 24.69, - "grad_norm": 0.0041221510618925095, + "grad_norm": 0.007521071936935186, "learning_rate": 1.0123456790123458e-05, - "loss": 0.0061, + "loss": 0.0037, "step": 2000 }, { "epoch": 25.0, "eval_B-Claim": { - "f1-score": 0.6870748299319729, - "precision": 0.689419795221843, - "recall": 0.6847457627118644, - "support": 295.0 + "f1-score": 0.6971080669710807, + "precision": 0.6735294117647059, + "recall": 0.722397476340694, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.898360655737705, - "precision": 0.9194630872483222, - "recall": 0.8782051282051282, - "support": 156.0 + "f1-score": 0.9196141479099678, + "precision": 0.9166666666666666, + "recall": 0.9225806451612903, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8940533151059467, - "precision": 0.8885869565217391, - "recall": 0.8995873452544704, - "support": 727.0 + "f1-score": 0.8868388683886839, + "precision": 0.8978829389788294, + "recall": 0.8760631834750912, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6645924082140635, - "precision": 0.6860226104830421, - "recall": 0.6444605358435916, - "support": 4143.0 + "f1-score": 0.6842635052007605, + "precision": 0.6654339786817489, + "recall": 0.7041896869244936, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9111880046136103, - "precision": 0.9220354808590103, - "recall": 0.9005927952576379, - "support": 2193.0 + "f1-score": 0.9163140573577808, + "precision": 0.9107476635514019, + "recall": 0.9219489120151372, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9084473808400189, - "precision": 0.8977846871356393, - "recall": 0.919366393377378, - "support": 12563.0 + "f1-score": 0.9042813682570305, + "precision": 0.9130890836892186, + "recall": 0.8956419489968399, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9989601699896017, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.9979225002258152, + "support": 11071.0 }, - "eval_accuracy": 0.911415693128507, - "eval_loss": 0.6578116416931152, + "eval_accuracy": 0.9045666183589776, + "eval_loss": 0.7725499272346497, "eval_macro avg": { - "f1-score": 0.8519595134919024, - "precision": 0.8576160882099424, - "recall": 0.8467082800928672, - "support": 32613.0 - }, - "eval_runtime": 4.9806, - "eval_samples_per_second": 16.263, - "eval_steps_per_second": 2.209, + "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": 0.9104735591343546, - "precision": 0.9098182414323299, - "recall": 0.911415693128507, - "support": 32613.0 + "f1-score": 0.9055214160067376, + "precision": 0.906722541443147, + "recall": 0.9045666183589776, + "support": 32431.0 }, "step": 2025 }, { "epoch": 26.0, "eval_B-Claim": { - "f1-score": 0.7008264462809918, - "precision": 0.6838709677419355, - "recall": 0.7186440677966102, - "support": 295.0 + "f1-score": 0.6904376012965964, + "precision": 0.71, + "recall": 0.6719242902208202, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8947368421052632, - "precision": 0.918918918918919, - "recall": 0.8717948717948718, - "support": 156.0 + "f1-score": 0.9022082018927444, + "precision": 0.8827160493827161, + "recall": 0.9225806451612903, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8964088397790055, - "precision": 0.9001386962552012, - "recall": 0.8927097661623109, - "support": 727.0 + "f1-score": 0.8958458759783263, + "precision": 0.8878281622911695, + "recall": 0.9040097205346294, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6849380362249762, - "precision": 0.6763944457519416, - "recall": 0.6937002172338885, - "support": 4143.0 + "f1-score": 0.6721173500060118, + "precision": 0.7034986156556758, + "recall": 0.6434162062615101, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9123211813567144, - "precision": 0.9234002802428771, - "recall": 0.9015047879616963, - "support": 2193.0 + "f1-score": 0.9087940284581292, + "precision": 0.8964565117349287, + "recall": 0.9214758751182592, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9089821314613912, - "precision": 0.910944120233432, - "recall": 0.9070285759770755, - "support": 12563.0 + "f1-score": 0.911746539258075, + "precision": 0.9016240298936476, + "recall": 0.9220989196736974, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999601132782897, - "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9998193315266486, + "precision": 0.9999096576023128, + "recall": 0.9997290217685846, + "support": 11071.0 }, - "eval_accuracy": 0.9130714745653574, - "eval_loss": 0.7021818161010742, + "eval_accuracy": 0.9083284511732601, + "eval_loss": 0.7629147171974182, "eval_macro avg": { - "f1-score": 0.8568819414980903, - "precision": 0.8590953470206152, - "recall": 0.8550432166664008, - "support": 32613.0 - }, - "eval_runtime": 4.965, - "eval_samples_per_second": 16.314, - "eval_steps_per_second": 2.215, + "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": 0.913484595343639, - "precision": 0.9139608163869709, - "recall": 0.9130714745653574, - "support": 32613.0 + "f1-score": 0.9069099155264195, + "precision": 0.9059874383002041, + "recall": 0.9083284511732601, + "support": 32431.0 }, "step": 2106 }, { "epoch": 27.0, "eval_B-Claim": { - "f1-score": 0.692967409948542, - "precision": 0.7013888888888888, - "recall": 0.6847457627118644, - "support": 295.0 + "f1-score": 0.701639344262295, + "precision": 0.7303754266211604, + "recall": 0.6750788643533123, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9073482428115015, - "precision": 0.9044585987261147, - "recall": 0.9102564102564102, - "support": 156.0 + "f1-score": 0.9166666666666666, + "precision": 0.910828025477707, + "recall": 0.9225806451612903, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8958904109589041, - "precision": 0.8922237380627558, - "recall": 0.8995873452544704, - "support": 727.0 + "f1-score": 0.9017964071856287, + "precision": 0.8890200708382526, + "recall": 0.9149453219927096, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6764050507551375, - "precision": 0.69428208386277, - "recall": 0.6594255370504465, - "support": 4143.0 + "f1-score": 0.6878357407186343, + "precision": 0.7143565583932556, + "recall": 0.6632136279926335, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9191374663072776, - "precision": 0.9057104913678619, - "recall": 0.93296853625171, - "support": 2193.0 + "f1-score": 0.9198392054859305, + "precision": 0.9196217494089834, + "recall": 0.9200567644276254, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9100839005857211, - "precision": 0.9049980322707596, - "recall": 0.9152272546366313, - "support": 12563.0 + "f1-score": 0.9161731041749935, + "precision": 0.906065832973983, + "recall": 0.9265084147865069, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, - "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9997741950051934, + "precision": 0.9997290462427746, + "recall": 0.999819347845723, + "support": 11071.0 }, - "eval_accuracy": 0.9140526783797872, - "eval_loss": 0.7241345643997192, + "eval_accuracy": 0.913076994233912, + "eval_loss": 0.7461331486701965, "eval_macro avg": { - "f1-score": 0.8574046401952976, - "precision": 0.8575802618827357, - "recall": 0.8574586923087905, - "support": 32613.0 - }, - "eval_runtime": 4.9786, - "eval_samples_per_second": 16.27, - "eval_steps_per_second": 2.209, + "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": 0.9132763828083692, - "precision": 0.9126659915256446, - "recall": 0.9140526783797872, - "support": 32613.0 + "f1-score": 0.9119067413298663, + "precision": 0.9111175788791426, + "recall": 0.913076994233912, + "support": 32431.0 }, "step": 2187 }, { "epoch": 28.0, "eval_B-Claim": { - "f1-score": 0.702791461412151, - "precision": 0.6815286624203821, - "recall": 0.7254237288135593, - "support": 295.0 + "f1-score": 0.69375, + "precision": 0.6873065015479877, + "recall": 0.7003154574132492, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8970099667774087, - "precision": 0.9310344827586207, - "recall": 0.8653846153846154, - "support": 156.0 + "f1-score": 0.9096989966555183, + "precision": 0.9444444444444444, + "recall": 0.8774193548387097, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8956461644782308, - "precision": 0.9, - "recall": 0.8913342503438789, - "support": 727.0 + "f1-score": 0.8948004836759371, + "precision": 0.8904933814681107, + "recall": 0.8991494532199271, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6791507633587787, - "precision": 0.6713039377505305, - "recall": 0.6871832005792904, - "support": 4143.0 + "f1-score": 0.6768418636676612, + "precision": 0.6748283752860412, + "recall": 0.6788674033149171, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9076923076923077, - "precision": 0.9284692417739628, - "recall": 0.8878248974008208, - "support": 2193.0 + "f1-score": 0.9071392910634049, + "precision": 0.9603594080338267, + "recall": 0.8595080416272469, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9079402985074626, - "precision": 0.9079764368731094, - "recall": 0.9079041630183873, - "support": 12563.0 + "f1-score": 0.9102119741690686, + "precision": 0.9037820605709318, + "recall": 0.9167340339531124, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999202233745512, - "precision": 1.0, - "recall": 0.9998404594767071, - "support": 12536.0 + "f1-score": 0.9998193315266486, + "precision": 0.9999096576023128, + "recall": 0.9997290217685846, + "support": 11071.0 }, - "eval_accuracy": 0.9116303314629136, - "eval_loss": 0.7758038640022278, + "eval_accuracy": 0.9067250470229101, + "eval_loss": 0.7586968541145325, "eval_macro avg": { - "f1-score": 0.85573588365727, - "precision": 0.8600446802252293, - "recall": 0.8521279021453229, - "support": 32613.0 - }, - "eval_runtime": 4.9782, - "eval_samples_per_second": 16.271, - "eval_steps_per_second": 2.21, + "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": 0.9120331319181557, - "precision": 0.9125454748864557, - "recall": 0.9116303314629136, - "support": 32613.0 + "f1-score": 0.9068326774221362, + "precision": 0.9073589365470631, + "recall": 0.9067250470229101, + "support": 32431.0 }, "step": 2268 }, { "epoch": 29.0, "eval_B-Claim": { - "f1-score": 0.7140495867768595, - "precision": 0.6967741935483871, - "recall": 0.7322033898305085, - "support": 295.0 + "f1-score": 0.6825396825396827, + "precision": 0.6869009584664537, + "recall": 0.6782334384858044, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9155844155844156, - "precision": 0.9276315789473685, - "recall": 0.9038461538461539, - "support": 156.0 + "f1-score": 0.9055374592833876, + "precision": 0.9144736842105263, + "recall": 0.896774193548387, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8975069252077562, - "precision": 0.9037656903765691, - "recall": 0.8913342503438789, - "support": 727.0 + "f1-score": 0.893719806763285, + "precision": 0.8883553421368547, + "recall": 0.8991494532199271, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6932471264367815, - "precision": 0.6878118317890235, - "recall": 0.6987690079652426, - "support": 4143.0 + "f1-score": 0.6749443598453789, + "precision": 0.6870975435249225, + "recall": 0.6632136279926335, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9248, - "precision": 0.9271310724106324, - "recall": 0.9224806201550387, - "support": 2193.0 + "f1-score": 0.9090909090909092, + "precision": 0.9196515004840271, + "recall": 0.8987701040681173, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9102149894300187, - "precision": 0.912216181643748, - "recall": 0.9082225583061371, - "support": 12563.0 + "f1-score": 0.9108261693137113, + "precision": 0.9029974720115566, + "recall": 0.9187917983390902, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999601132782897, + "f1-score": 0.9980995475113122, "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "recall": 0.9962063047601842, + "support": 11071.0 }, - "eval_accuracy": 0.9158311102934412, - "eval_loss": 0.7231832146644592, + "eval_accuracy": 0.9067250470229101, + "eval_loss": 0.796188473701477, "eval_macro avg": { - "f1-score": 0.8650518795305887, - "precision": 0.865047221245104, - "recall": 0.8652537443121877, - "support": 32613.0 - }, - "eval_runtime": 4.968, - "eval_samples_per_second": 16.304, - "eval_steps_per_second": 2.214, + "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": 0.9160979814383814, - "precision": 0.9163913568578534, - "recall": 0.9158311102934412, - "support": 32613.0 + "f1-score": 0.9062194518780211, + "precision": 0.9058490047862192, + "recall": 0.9067250470229101, + "support": 32431.0 }, "step": 2349 }, { "epoch": 30.0, "eval_B-Claim": { - "f1-score": 0.6937394247038917, - "precision": 0.6925675675675675, - "recall": 0.6949152542372882, - "support": 295.0 + "f1-score": 0.6710097719869706, + "precision": 0.6936026936026936, + "recall": 0.6498422712933754, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9061488673139159, - "precision": 0.9150326797385621, - "recall": 0.8974358974358975, - "support": 156.0 + "f1-score": 0.89375, + "precision": 0.8666666666666667, + "recall": 0.9225806451612903, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8942307692307693, - "precision": 0.8930041152263375, - "recall": 0.8954607977991746, - "support": 727.0 + "f1-score": 0.893848009650181, + "precision": 0.8874251497005988, + "recall": 0.9003645200486027, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6808769792935445, - "precision": 0.6872387509220556, - "recall": 0.6746319092445088, - "support": 4143.0 + "f1-score": 0.6677052127022169, + "precision": 0.69632591852037, + "recall": 0.6413443830570903, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9230070406540993, - "precision": 0.9194570135746606, - "recall": 0.9265845873233014, - "support": 2193.0 + "f1-score": 0.8996555683122847, + "precision": 0.8741633199464525, + "recall": 0.9266792809839167, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.908715505261068, - "precision": 0.9065916653462209, - "recall": 0.9108493194300724, - "support": 12563.0 + "f1-score": 0.9106609341980617, + "precision": 0.903027675410073, + "recall": 0.9184243404130227, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, - "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9988248056409328, + "precision": 0.9996381073011852, + "recall": 0.9980128263029536, + "support": 11071.0 }, - "eval_accuracy": 0.9138073774261798, - "eval_loss": 0.7184767127037048, + "eval_accuracy": 0.9059541796429342, + "eval_loss": 0.8091421127319336, "eval_macro avg": { - "f1-score": 0.8581026552081842, - "precision": 0.8591273989107719, - "recall": 0.8571253950671774, - "support": 32613.0 - }, - "eval_runtime": 4.9612, - "eval_samples_per_second": 16.327, - "eval_steps_per_second": 2.217, + "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": 0.9135418017594757, - "precision": 0.9132976723952129, - "recall": 0.9138073774261798, - "support": 32613.0 + "f1-score": 0.9046472304257149, + "precision": 0.9038225277994681, + "recall": 0.9059541796429342, + "support": 32431.0 }, "step": 2430 }, { "epoch": 30.86, - "grad_norm": 0.002683656057342887, + "grad_norm": 0.0011433030012995005, "learning_rate": 7.654320987654322e-06, - "loss": 0.0024, + "loss": 0.002, "step": 2500 }, { "epoch": 31.0, "eval_B-Claim": { - "f1-score": 0.6986754966887417, - "precision": 0.6828478964401294, - "recall": 0.7152542372881356, - "support": 295.0 + "f1-score": 0.6802507836990596, + "precision": 0.67601246105919, + "recall": 0.6845425867507886, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9206349206349206, - "precision": 0.9119496855345912, - "recall": 0.9294871794871795, - "support": 156.0 + "f1-score": 0.9096774193548387, + "precision": 0.9096774193548387, + "recall": 0.9096774193548387, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.890125173852573, - "precision": 0.90014064697609, - "recall": 0.8803301237964236, - "support": 727.0 + "f1-score": 0.8876745598057073, + "precision": 0.8871359223300971, + "recall": 0.8882138517618469, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6821539194662855, - "precision": 0.6734885909197835, - "recall": 0.6910451363746077, - "support": 4143.0 + "f1-score": 0.6758359023002134, + "precision": 0.6968215158924206, + "recall": 0.6560773480662984, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9271943751417554, - "precision": 0.9223826714801444, - "recall": 0.9320565435476517, - "support": 2193.0 + "f1-score": 0.9129821260583255, + "precision": 0.9078578110383536, + "recall": 0.9181646168401135, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.905381076215243, - "precision": 0.9101512226512226, - "recall": 0.9006606702220807, - "support": 12563.0 + "f1-score": 0.9112844437964718, + "precision": 0.9039699182876564, + "recall": 0.9187183067538767, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999601132782897, - "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9996839015579136, + "precision": 0.999548491963157, + "recall": 0.999819347845723, + "support": 11071.0 }, - "eval_accuracy": 0.912304909085334, - "eval_loss": 0.73140949010849, + "eval_accuracy": 0.9080817736116679, + "eval_loss": 0.8016564249992371, "eval_macro avg": { - "f1-score": 0.8605892964682583, - "precision": 0.8572801020002803, - "recall": 0.8641077314934904, - "support": 32613.0 - }, - "eval_runtime": 4.9803, - "eval_samples_per_second": 16.264, - "eval_steps_per_second": 2.209, + "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": 0.9127082057497462, - "precision": 0.913175260529506, - "recall": 0.912304909085334, - "support": 32613.0 + "f1-score": 0.9071696505232147, + "precision": 0.9064763012077777, + "recall": 0.9080817736116679, + "support": 32431.0 }, "step": 2511 }, { "epoch": 32.0, "eval_B-Claim": { - "f1-score": 0.6913996627318718, - "precision": 0.6879194630872483, - "recall": 0.6949152542372882, - "support": 295.0 + "f1-score": 0.699367088607595, + "precision": 0.7015873015873015, + "recall": 0.6971608832807571, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9240506329113924, - "precision": 0.9125, - "recall": 0.9358974358974359, - "support": 156.0 + "f1-score": 0.9073482428115016, + "precision": 0.8987341772151899, + "recall": 0.9161290322580645, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8895027624309392, - "precision": 0.8932038834951457, - "recall": 0.8858321870701513, - "support": 727.0 + "f1-score": 0.8998178506375226, + "precision": 0.8992718446601942, + "recall": 0.9003645200486027, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6711132670385641, - "precision": 0.6743845966366073, - "recall": 0.6678735216027033, - "support": 4143.0 + "f1-score": 0.674114114114114, + "precision": 0.7048480281336348, + "recall": 0.6459484346224678, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.928603302097278, - "precision": 0.909130624726955, - "recall": 0.9489284085727314, - "support": 2193.0 + "f1-score": 0.9130636913063692, + "precision": 0.8976234003656307, + "recall": 0.9290444654683065, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9038691663342641, - "precision": 0.9058927000879508, - "recall": 0.9018546525511423, - "support": 12563.0 + "f1-score": 0.913888989858602, + "precision": 0.9041283084004603, + "recall": 0.9238627177188212, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999601132782897, - "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9991866979938551, + "precision": 0.9996383690443902, + "recall": 0.9987354349200614, + "support": 11071.0 }, - "eval_accuracy": 0.9109250912212922, - "eval_loss": 0.7511080503463745, + "eval_accuracy": 0.9096851777620178, + "eval_loss": 0.8070256114006042, "eval_macro avg": { - "f1-score": 0.8583569866889427, - "precision": 0.8547187525762724, - "recall": 0.8621745270956865, - "support": 32613.0 - }, - "eval_runtime": 5.0707, - "eval_samples_per_second": 15.974, - "eval_steps_per_second": 2.169, + "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": 0.9107546356079316, - "precision": 0.9106514058813452, - "recall": 0.9109250912212922, - "support": 32613.0 + "f1-score": 0.9083512753053686, + "precision": 0.90748708577928, + "recall": 0.9096851777620178, + "support": 32431.0 }, "step": 2592 }, { "epoch": 33.0, "eval_B-Claim": { - "f1-score": 0.7094594594594594, - "precision": 0.7070707070707071, - "recall": 0.711864406779661, - "support": 295.0 + "f1-score": 0.6867088607594937, + "precision": 0.6888888888888889, + "recall": 0.6845425867507886, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9185667752442997, - "precision": 0.9337748344370861, - "recall": 0.9038461538461539, - "support": 156.0 + "f1-score": 0.9161290322580645, + "precision": 0.9161290322580645, + "recall": 0.9161290322580645, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8984910836762688, - "precision": 0.896032831737346, - "recall": 0.9009628610729024, - "support": 727.0 + "f1-score": 0.8917120387174833, + "precision": 0.8879518072289156, + "recall": 0.8955042527339003, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6867204695524578, - "precision": 0.695910780669145, - "recall": 0.6777697320782042, - "support": 4143.0 + "f1-score": 0.6777020447906524, + "precision": 0.71900826446281, + "recall": 0.6408839779005525, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9270425776754891, - "precision": 0.9358736059479554, - "recall": 0.9183766529867761, - "support": 2193.0 + "f1-score": 0.9200660221645838, + "precision": 0.9172543488481429, + "recall": 0.9228949858088931, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9104648862512364, - "precision": 0.9051290119572057, - "recall": 0.9158640452121308, - "support": 12563.0 + "f1-score": 0.9157149053057683, + "precision": 0.9009316549320816, + "recall": 0.93099140148453, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999601132782897, - "precision": 1.0, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9996386956914461, + "precision": 0.9996386956914461, + "recall": 0.9996386956914461, + "support": 11071.0 }, - "eval_accuracy": 0.9158617729126421, - "eval_loss": 0.7425341010093689, + "eval_accuracy": 0.9116585982547563, + "eval_loss": 0.823725700378418, "eval_macro avg": { - "f1-score": 0.8643864807339287, - "precision": 0.8676845388313493, - "recall": 0.8612291545305973, - "support": 32613.0 - }, - "eval_runtime": 4.9731, - "eval_samples_per_second": 16.288, - "eval_steps_per_second": 2.212, + "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": 0.9155106123184129, - "precision": 0.9152281470826237, - "recall": 0.9158617729126421, - "support": 32613.0 + "f1-score": 0.9099212385263286, + "precision": 0.9089940716622809, + "recall": 0.9116585982547563, + "support": 32431.0 }, "step": 2673 }, { "epoch": 34.0, "eval_B-Claim": { - "f1-score": 0.6989247311827957, - "precision": 0.7414448669201521, - "recall": 0.6610169491525424, - "support": 295.0 + "f1-score": 0.6927899686520376, + "precision": 0.6884735202492211, + "recall": 0.6971608832807571, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9022082018927445, - "precision": 0.8881987577639752, - "recall": 0.9166666666666666, - "support": 156.0 + "f1-score": 0.9196141479099678, + "precision": 0.9166666666666666, + "recall": 0.9225806451612903, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.9055330634278002, - "precision": 0.8887417218543047, - "recall": 0.922971114167813, - "support": 727.0 + "f1-score": 0.8929440389294404, + "precision": 0.8940316686967114, + "recall": 0.8918590522478737, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6595328200443691, - "precision": 0.7178977272727273, - "recall": 0.6099444846729423, - "support": 4143.0 + "f1-score": 0.6815160349854228, + "precision": 0.6906168754431576, + "recall": 0.6726519337016574, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.8977876106194691, - "precision": 0.8719381177481736, - "recall": 0.9252165982672138, - "support": 2193.0 + "f1-score": 0.9175862878610003, + "precision": 0.9109557109557109, + "recall": 0.924314096499527, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9145199063231851, - "precision": 0.8972198820556023, - "recall": 0.9325001989970548, - "support": 12563.0 + "f1-score": 0.9118692513466964, + "precision": 0.9093699751498319, + "recall": 0.9143823032262806, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9997606319317004, - "precision": 1.0, - "recall": 0.9995213784301212, - "support": 12536.0 + "f1-score": 0.999729070712544, + "precision": 0.999548532731377, + "recall": 0.9999096739228616, + "support": 11071.0 }, - "eval_accuracy": 0.9140526783797872, - "eval_loss": 0.7677819728851318, + "eval_accuracy": 0.9091918226388332, + "eval_loss": 0.8422325849533081, "eval_macro avg": { - "f1-score": 0.8540381379174375, - "precision": 0.8579201533735622, - "recall": 0.8525481986220506, - "support": 32613.0 - }, - "eval_runtime": 5.0132, - "eval_samples_per_second": 16.157, - "eval_steps_per_second": 2.194, + "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": 0.9115584392994356, - "precision": 0.9106057225130435, - "recall": 0.9140526783797872, - "support": 32613.0 + "f1-score": 0.9087951672928857, + "precision": 0.9084430526412389, + "recall": 0.9091918226388332, + "support": 32431.0 }, "step": 2754 }, { "epoch": 35.0, "eval_B-Claim": { - "f1-score": 0.7009966777408638, - "precision": 0.6872964169381107, - "recall": 0.7152542372881356, - "support": 295.0 + "f1-score": 0.68, + "precision": 0.7208480565371025, + "recall": 0.6435331230283912, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9042904290429044, - "precision": 0.9319727891156463, - "recall": 0.8782051282051282, - "support": 156.0 + "f1-score": 0.9206349206349206, + "precision": 0.90625, + "recall": 0.9354838709677419, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8945554789800138, - "precision": 0.8964088397790055, - "recall": 0.8927097661623109, - "support": 727.0 + "f1-score": 0.897910447761194, + "precision": 0.8826291079812206, + "recall": 0.913730255164034, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.690017931858936, - "precision": 0.6835622927522501, - "recall": 0.6965966690803765, - "support": 4143.0 + "f1-score": 0.6727050183598531, + "precision": 0.7182435964453737, + "recall": 0.6325966850828729, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9171501508470644, - "precision": 0.9338374291115312, - "recall": 0.9010487916096671, - "support": 2193.0 + "f1-score": 0.9136773894931729, + "precision": 0.894426823742637, + "recall": 0.9337748344370861, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.910284986467123, - "precision": 0.910357455616591, - "recall": 0.9102125288545729, - "support": 12563.0 + "f1-score": 0.9146681135829264, + "precision": 0.9006268699244907, + "recall": 0.9291541118541927, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, - "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9992771955186122, + "precision": 0.9995481247175779, + "recall": 0.9990064131514769, + "support": 11071.0 }, - "eval_accuracy": 0.9146659307638059, - "eval_loss": 0.808364987373352, + "eval_accuracy": 0.9104252104467947, + "eval_loss": 0.8378845453262329, "eval_macro avg": { - "f1-score": 0.8596136649909865, - "precision": 0.8633478890447337, - "recall": 0.8562895887428844, - "support": 32613.0 - }, - "eval_runtime": 4.9841, - "eval_samples_per_second": 16.252, - "eval_steps_per_second": 2.207, + "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": 0.9149777701397716, - "precision": 0.9153575016233867, - "recall": 0.9146659307638059, - "support": 32613.0 + "f1-score": 0.9083861171551627, + "precision": 0.9073749479187287, + "recall": 0.9104252104467947, + "support": 32431.0 }, "step": 2835 }, { "epoch": 36.0, "eval_B-Claim": { - "f1-score": 0.6881720430107526, - "precision": 0.7300380228136882, - "recall": 0.6508474576271186, - "support": 295.0 + "f1-score": 0.674496644295302, + "precision": 0.7204301075268817, + "recall": 0.6340694006309149, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9050632911392406, - "precision": 0.89375, - "recall": 0.9166666666666666, - "support": 156.0 + "f1-score": 0.9108910891089109, + "precision": 0.9324324324324325, + "recall": 0.8903225806451613, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.9001349527665317, - "precision": 0.8834437086092716, - "recall": 0.9174690508940853, - "support": 727.0 + "f1-score": 0.9000591366055587, + "precision": 0.8767281105990783, + "recall": 0.9246658566221142, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6585080067699518, - "precision": 0.7148106274731487, - "recall": 0.6104272266473569, - "support": 4143.0 + "f1-score": 0.6687843616371411, + "precision": 0.7125748502994012, + "recall": 0.6300644567219152, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9073909171861086, - "precision": 0.8864723792953458, - "recall": 0.9293205654354765, - "support": 2193.0 + "f1-score": 0.9068203650336216, + "precision": 0.9209756097560976, + "recall": 0.8930936613055819, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9130146341463415, - "precision": 0.8955749502373297, - "recall": 0.9311470190241184, - "support": 12563.0 + "f1-score": 0.9150956972226219, + "precision": 0.8963281415180774, + "recall": 0.9346659807452047, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, - "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9989605459393501, + "precision": 0.9996382054992764, + "recall": 0.998283804534369, + "support": 11071.0 }, - "eval_accuracy": 0.9138380400453807, - "eval_loss": 0.8094782829284668, + "eval_accuracy": 0.9094693348956245, + "eval_loss": 0.8216572999954224, "eval_macro avg": { - "f1-score": 0.8531834064312752, - "precision": 0.8577270983469691, - "recall": 0.8508397123278317, - "support": 32613.0 - }, - "eval_runtime": 4.948, - "eval_samples_per_second": 16.37, - "eval_steps_per_second": 2.223, + "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": 0.9113822260750222, - "precision": 0.9103625011540336, - "recall": 0.9138380400453807, - "support": 32613.0 + "f1-score": 0.9074394591383382, + "precision": 0.9065446575677608, + "recall": 0.9094693348956245, + "support": 32431.0 }, "step": 2916 }, { "epoch": 37.0, "eval_B-Claim": { - "f1-score": 0.6933333333333332, - "precision": 0.6819672131147541, - "recall": 0.7050847457627119, - "support": 295.0 + "f1-score": 0.7020280811232449, + "precision": 0.6944444444444444, + "recall": 0.7097791798107256, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9135802469135802, - "precision": 0.8809523809523809, - "recall": 0.9487179487179487, - "support": 156.0 + "f1-score": 0.9166666666666666, + "precision": 0.910828025477707, + "recall": 0.9225806451612903, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8896648044692737, - "precision": 0.9035460992907801, - "recall": 0.8762035763411279, - "support": 727.0 + "f1-score": 0.896888346552776, + "precision": 0.9007352941176471, + "recall": 0.8930741190765492, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6752997601918465, - "precision": 0.6709554443650226, - "recall": 0.6797006999758629, - "support": 4143.0 + "f1-score": 0.6857803468208092, + "precision": 0.6888063167673014, + "recall": 0.682780847145488, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9153129737925059, - "precision": 0.8716996699669967, - "recall": 0.9635202918376653, - "support": 2193.0 + "f1-score": 0.908329455560726, + "precision": 0.8937728937728938, + "recall": 0.923368022705771, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9059946052578606, - "precision": 0.9165852069077876, - "recall": 0.8956459444400223, - "support": 12563.0 + "f1-score": 0.9133458439121316, + "precision": 0.9145910095799558, + "recall": 0.9121040640846623, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999202361011406, - "precision": 0.9998404849258254, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9997742153985099, + "precision": 0.9996387935705255, + "recall": 0.9999096739228616, + "support": 11071.0 }, - "eval_accuracy": 0.910986416459694, - "eval_loss": 0.8418905735015869, + "eval_accuracy": 0.9096851777620178, + "eval_loss": 0.8227846026420593, "eval_macro avg": { - "f1-score": 0.8561579942942201, - "precision": 0.8465066427890783, - "recall": 0.8669818867250484, - "support": 32613.0 - }, - "eval_runtime": 4.9617, - "eval_samples_per_second": 16.325, - "eval_steps_per_second": 2.217, + "f1-score": 0.8604018508621235, + "precision": 0.8575452539614965, + "recall": 0.863370935986764, + "support": 32431.0 + }, + "eval_runtime": 4.9103, + "eval_samples_per_second": 16.292, + "eval_steps_per_second": 2.037, "eval_weighted avg": { - "f1-score": 0.9111672255947384, - "precision": 0.9117822024694574, - "recall": 0.910986416459694, - "support": 32613.0 + "f1-score": 0.9095741998919733, + "precision": 0.9095024472986821, + "recall": 0.9096851777620178, + "support": 32431.0 }, "step": 2997 }, { "epoch": 37.04, - "grad_norm": 0.953349769115448, + "grad_norm": 0.10209230333566666, "learning_rate": 5.185185185185185e-06, - "loss": 0.0013, + "loss": 0.0007, "step": 3000 }, { "epoch": 38.0, "eval_B-Claim": { - "f1-score": 0.6727272727272727, - "precision": 0.7254901960784313, - "recall": 0.6271186440677966, - "support": 295.0 + "f1-score": 0.6929133858267716, + "precision": 0.6918238993710691, + "recall": 0.694006309148265, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9142857142857143, - "precision": 0.9056603773584906, - "recall": 0.9230769230769231, - "support": 156.0 + "f1-score": 0.9120521172638436, + "precision": 0.9210526315789473, + "recall": 0.9032258064516129, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8960429242119383, - "precision": 0.8743455497382199, - "recall": 0.9188445667125172, - "support": 727.0 + "f1-score": 0.8957575757575758, + "precision": 0.8935912938331319, + "recall": 0.8979343863912516, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6539833531510107, - "precision": 0.7224168126094571, - "recall": 0.5973931933381608, - "support": 4143.0 + "f1-score": 0.668, + "precision": 0.6831087584215592, + "recall": 0.6535451197053407, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9178635547576303, - "precision": 0.9036676977463544, - "recall": 0.9325125398996807, - "support": 2193.0 + "f1-score": 0.9133821099834789, + "precision": 0.9114460668864814, + "recall": 0.9153263954588458, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.91184230948316, - "precision": 0.889545007192066, - "recall": 0.9352861577648651, - "support": 12563.0 + "f1-score": 0.9093298291721419, + "precision": 0.9033287403002394, + "recall": 0.9154111854192695, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.999880349379811, - "precision": 0.9998404722022812, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9994127478881509, + "precision": 0.9996385324417133, + "recall": 0.9991870653057537, + "support": 11071.0 }, - "eval_accuracy": 0.9138073774261798, - "eval_loss": 0.8167690634727478, + "eval_accuracy": 0.9062625265949246, + "eval_loss": 0.8360013961791992, "eval_macro avg": { - "f1-score": 0.8523750682852196, - "precision": 0.8601380161321857, - "recall": 0.8477360363711853, - "support": 32613.0 - }, - "eval_runtime": 5.0064, - "eval_samples_per_second": 16.179, - "eval_steps_per_second": 2.197, + "f1-score": 0.8558353951274232, + "precision": 0.8577128461190201, + "recall": 0.854090895411477, + "support": 32431.0 + }, + "eval_runtime": 4.9237, + "eval_samples_per_second": 16.248, + "eval_steps_per_second": 2.031, "eval_weighted avg": { - "f1-score": 0.9108272506158761, - "precision": 0.909913900654684, - "recall": 0.9138073774261798, - "support": 32613.0 + "f1-score": 0.9055737228038481, + "precision": 0.905007886365788, + "recall": 0.9062625265949246, + "support": 32431.0 }, "step": 3078 }, { "epoch": 39.0, "eval_B-Claim": { - "f1-score": 0.7077922077922078, - "precision": 0.6791277258566978, - "recall": 0.7389830508474576, - "support": 295.0 + "f1-score": 0.6883720930232559, + "precision": 0.676829268292683, + "recall": 0.7003154574132492, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9126213592233009, + "f1-score": 0.9155844155844156, "precision": 0.9215686274509803, - "recall": 0.9038461538461539, - "support": 156.0 + "recall": 0.9096774193548387, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8916841369671559, - "precision": 0.90625, - "recall": 0.8775790921595599, - "support": 727.0 + "f1-score": 0.8902439024390243, + "precision": 0.8935128518971848, + "recall": 0.8869987849331713, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6838785046728972, - "precision": 0.662666968530677, - "recall": 0.7064928795558774, - "support": 4143.0 + "f1-score": 0.6723443434227184, + "precision": 0.6722669735327963, + "recall": 0.6724217311233885, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9182648401826483, - "precision": 0.9195244627343393, - "recall": 0.9170086639306886, - "support": 2193.0 + "f1-score": 0.9199905593580363, + "precision": 0.9180405087140838, + "recall": 0.9219489120151372, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9046661303298471, - "precision": 0.9144506790274051, - "recall": 0.8950887526864603, - "support": 12563.0 + "f1-score": 0.9068117487041871, + "precision": 0.9071785819358635, + "recall": 0.9064452120232234, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9998404467491026, - "precision": 0.9999202170097335, - "recall": 0.9997606892150607, - "support": 12536.0 + "f1-score": 0.9995483288166215, + "precision": 0.999638630409251, + "recall": 0.9994580435371692, + "support": 11071.0 }, - "eval_accuracy": 0.9110784043172968, - "eval_loss": 0.8231661915779114, + "eval_accuracy": 0.9053683204341525, + "eval_loss": 0.8432661294937134, "eval_macro avg": { - "f1-score": 0.8598210894167371, - "precision": 0.857644097229976, - "recall": 0.8626798974630369, - "support": 32613.0 - }, - "eval_runtime": 4.9876, - "eval_samples_per_second": 16.24, - "eval_steps_per_second": 2.205, + "f1-score": 0.8561279130497513, + "precision": 0.8555764917475491, + "recall": 0.8567522229143112, + "support": 32431.0 + }, + "eval_runtime": 4.9317, + "eval_samples_per_second": 16.222, + "eval_steps_per_second": 2.028, "eval_weighted avg": { - "f1-score": 0.9120843184028957, - "precision": 0.9133824319585561, - "recall": 0.9110784043172968, - "support": 32613.0 + "f1-score": 0.9054087494332319, + "precision": 0.9054547409122511, + "recall": 0.9053683204341525, + "support": 32431.0 }, "step": 3159 }, { "epoch": 40.0, "eval_B-Claim": { - "f1-score": 0.7035175879396984, - "precision": 0.695364238410596, - "recall": 0.711864406779661, - "support": 295.0 + "f1-score": 0.6732026143790849, + "precision": 0.6983050847457627, + "recall": 0.6498422712933754, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9067524115755627, - "precision": 0.9096774193548387, - "recall": 0.9038461538461539, - "support": 156.0 + "f1-score": 0.9090909090909091, + "precision": 0.9507042253521126, + "recall": 0.8709677419354839, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8964088397790055, - "precision": 0.9001386962552012, - "recall": 0.8927097661623109, - "support": 727.0 + "f1-score": 0.8942992874109265, + "precision": 0.8745644599303136, + "recall": 0.9149453219927096, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6803130644190246, - "precision": 0.6787602114368092, - "recall": 0.6818730388607289, - "support": 4143.0 + "f1-score": 0.6805373915111164, + "precision": 0.7037128104253749, + "recall": 0.6588397790055248, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9128065395095367, - "precision": 0.9090909090909091, - "recall": 0.9165526675786594, - "support": 2193.0 + "f1-score": 0.9203758654797229, + "precision": 0.9642487046632124, + "recall": 0.880321665089877, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9086126499541668, - "precision": 0.9098818646232439, - "recall": 0.9073469712648253, - "support": 12563.0 + "f1-score": 0.913268070657082, + "precision": 0.8980534242682581, + "recall": 0.9290071286837657, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9998404467491026, - "precision": 0.9999202170097335, - "recall": 0.9997606892150607, - "support": 12536.0 + "f1-score": 0.9991414757579864, + "precision": 0.9996383363471971, + "recall": 0.9986451088429229, + "support": 11071.0 }, - "eval_accuracy": 0.9127341857541471, - "eval_loss": 0.8246187567710876, + "eval_accuracy": 0.9100551941044063, + "eval_loss": 0.8238250017166138, "eval_macro avg": { - "f1-score": 0.858321648560871, - "precision": 0.8575476508830473, - "recall": 0.8591362419581999, - "support": 32613.0 - }, - "eval_runtime": 5.0718, - "eval_samples_per_second": 15.971, - "eval_steps_per_second": 2.169, + "f1-score": 0.855702230612404, + "precision": 0.8698895779617474, + "recall": 0.84322414526338, + "support": 32431.0 + }, + "eval_runtime": 4.89, + "eval_samples_per_second": 16.36, + "eval_steps_per_second": 2.045, "eval_weighted avg": { - "f1-score": 0.9128231876865274, - "precision": 0.9129190379677707, - "recall": 0.9127341857541471, - "support": 32613.0 + "f1-score": 0.9090248699424149, + "precision": 0.9087184358539823, + "recall": 0.9100551941044063, + "support": 32431.0 }, "step": 3240 }, { "epoch": 41.0, "eval_B-Claim": { - "f1-score": 0.679646017699115, - "precision": 0.7111111111111111, - "recall": 0.6508474576271186, - "support": 295.0 + "f1-score": 0.6917057902973396, + "precision": 0.6863354037267081, + "recall": 0.6971608832807571, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.8990228013029317, - "precision": 0.9139072847682119, - "recall": 0.8846153846153846, - "support": 156.0 + "f1-score": 0.9108910891089109, + "precision": 0.9324324324324325, + "recall": 0.8903225806451613, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8975741239892182, - "precision": 0.8797886393659181, - "recall": 0.9160935350756534, - "support": 727.0 + "f1-score": 0.895631067961165, + "precision": 0.8945454545454545, + "recall": 0.8967193195625759, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6559774243201641, - "precision": 0.6999726252395292, - "recall": 0.6171856142891624, - "support": 4143.0 + "f1-score": 0.6696025031869277, + "precision": 0.6742123687281214, + "recall": 0.6650552486187845, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9044762344254731, - "precision": 0.9154600653900047, - "recall": 0.8937528499772002, - "support": 2193.0 + "f1-score": 0.9039163220627584, + "precision": 0.9303955933900852, + "recall": 0.8789025543992431, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9108617734143681, - "precision": 0.8920259442960702, - "recall": 0.930510228448619, - "support": 12563.0 + "f1-score": 0.9091307203853735, + "precision": 0.9029358463211309, + "recall": 0.9154111854192695, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9999202297383536, - "precision": 0.9999202297383536, - "recall": 0.9999202297383536, - "support": 12536.0 + "f1-score": 0.9990962494351558, + "precision": 0.9996383036440908, + "recall": 0.9985547827657845, + "support": 11071.0 }, - "eval_accuracy": 0.9118449697973201, - "eval_loss": 0.8509364128112793, + "eval_accuracy": 0.9051524775677593, + "eval_loss": 0.8386729955673218, "eval_macro avg": { - "f1-score": 0.8496398006985176, - "precision": 0.8588836999870283, - "recall": 0.8418464713959273, - "support": 32613.0 - }, - "eval_runtime": 4.9776, - "eval_samples_per_second": 16.273, - "eval_steps_per_second": 2.21, + "f1-score": 0.854281963205376, + "precision": 0.8600707718268605, + "recall": 0.8488752220987965, + "support": 32431.0 + }, + "eval_runtime": 4.9226, + "eval_samples_per_second": 16.251, + "eval_steps_per_second": 2.031, "eval_weighted avg": { - "f1-score": 0.9098416837096517, - "precision": 0.9088727226253279, - "recall": 0.9118449697973201, - "support": 32613.0 + "f1-score": 0.9049592023598794, + "precision": 0.904911498067166, + "recall": 0.9051524775677593, + "support": 32431.0 }, "step": 3321 }, { "epoch": 42.0, "eval_B-Claim": { - "f1-score": 0.7035175879396984, - "precision": 0.695364238410596, - "recall": 0.711864406779661, - "support": 295.0 + "f1-score": 0.6782884310618067, + "precision": 0.6815286624203821, + "recall": 0.6750788643533123, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.910828025477707, - "precision": 0.9050632911392406, - "recall": 0.9166666666666666, - "support": 156.0 + "f1-score": 0.9144736842105262, + "precision": 0.9328859060402684, + "recall": 0.896774193548387, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8955017301038062, - "precision": 0.9011142061281338, - "recall": 0.889958734525447, - "support": 727.0 + "f1-score": 0.8913043478260869, + "precision": 0.885954381752701, + "recall": 0.8967193195625759, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6800193050193051, - "precision": 0.6798552472858866, - "recall": 0.6801834419502776, - "support": 4143.0 + "f1-score": 0.6689879294336118, + "precision": 0.674625468164794, + "recall": 0.6634438305709024, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9161636485580147, - "precision": 0.8986842105263158, - "recall": 0.9343365253077975, - "support": 2193.0 + "f1-score": 0.9094869312681512, + "precision": 0.9311199207135779, + "recall": 0.8888363292336803, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9082920594344145, - "precision": 0.9115689890162751, - "recall": 0.9050386054286397, - "support": 12563.0 + "f1-score": 0.9086592892881926, + "precision": 0.903150863946566, + "recall": 0.9142353200558536, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.999960116459937, - "precision": 0.9999202361011407, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9996386956914461, + "precision": 0.9996386956914461, + "recall": 0.9996386956914461, + "support": 11071.0 }, - "eval_accuracy": 0.9129181614693527, - "eval_loss": 0.8349179029464722, + "eval_accuracy": 0.9052758163485554, + "eval_loss": 0.8507019877433777, "eval_macro avg": { - "f1-score": 0.8591832104275545, - "precision": 0.8559386312296555, - "recall": 0.86257834009407, - "support": 32613.0 + "f1-score": 0.852977044111403, + "precision": 0.8584148426756765, + "recall": 0.8478180790023082, + "support": 32431.0 }, - "eval_runtime": 4.9839, - "eval_samples_per_second": 16.252, - "eval_steps_per_second": 2.207, + "eval_runtime": 4.9177, + "eval_samples_per_second": 16.268, + "eval_steps_per_second": 2.033, "eval_weighted avg": { - "f1-score": 0.9129333898384456, - "precision": 0.9130079539928556, - "recall": 0.9129181614693527, - "support": 32613.0 + "f1-score": 0.9050035527570847, + "precision": 0.9048415623762223, + "recall": 0.9052758163485554, + "support": 32431.0 }, "step": 3402 }, { "epoch": 43.0, "eval_B-Claim": { - "f1-score": 0.6853146853146852, - "precision": 0.7075812274368231, - "recall": 0.6644067796610169, - "support": 295.0 + "f1-score": 0.6795491143317229, + "precision": 0.694078947368421, + "recall": 0.6656151419558359, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9073482428115015, - "precision": 0.9044585987261147, - "recall": 0.9102564102564102, - "support": 156.0 + "f1-score": 0.9196141479099678, + "precision": 0.9166666666666666, + "recall": 0.9225806451612903, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8959891230455472, - "precision": 0.885752688172043, - "recall": 0.90646492434663, - "support": 727.0 + "f1-score": 0.8933092224231465, + "precision": 0.8863636363636364, + "recall": 0.9003645200486027, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6696002011566508, - "precision": 0.6987667278929415, - "recall": 0.6427709389331402, - "support": 4143.0 + "f1-score": 0.6700023826542769, + "precision": 0.694320987654321, + "recall": 0.647329650092081, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.909255487666893, - "precision": 0.9025157232704403, - "recall": 0.9160966712266302, - "support": 2193.0 + "f1-score": 0.9232214134773422, + "precision": 0.9165501165501165, + "recall": 0.9299905392620624, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9115148655025954, - "precision": 0.900816167897396, - "recall": 0.922470747432938, - "support": 12563.0 + "f1-score": 0.9102820746132848, + "precision": 0.9017161811364292, + "recall": 0.9190122730947307, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9998005504806733, - "precision": 0.9999202106439001, - "recall": 0.9996809189534142, - "support": 12536.0 + "f1-score": 0.9996838730072709, + "precision": 0.9996387283236994, + "recall": 0.9997290217685846, + "support": 11071.0 }, - "eval_accuracy": 0.9134394259957686, - "eval_loss": 0.8265806436538696, + "eval_accuracy": 0.9079584348308717, + "eval_loss": 0.8612508773803711, "eval_macro avg": { - "f1-score": 0.854117593711221, - "precision": 0.8571159062913799, - "recall": 0.8517353415443115, - "support": 32613.0 - }, - "eval_runtime": 4.9504, - "eval_samples_per_second": 16.362, - "eval_steps_per_second": 2.222, + "f1-score": 0.8565231754881447, + "precision": 0.8584764662947556, + "recall": 0.8549459701975982, + "support": 32431.0 + }, + "eval_runtime": 4.9087, + "eval_samples_per_second": 16.298, + "eval_steps_per_second": 2.037, "eval_weighted avg": { - "f1-score": 0.912154972694351, - "precision": 0.9112910502105193, - "recall": 0.9134394259957686, - "support": 32613.0 + "f1-score": 0.9068187696306371, + "precision": 0.9059835770444568, + "recall": 0.9079584348308717, + "support": 32431.0 }, "step": 3483 }, { "epoch": 43.21, - "grad_norm": 0.005421491339802742, + "grad_norm": 0.0010790067026391625, "learning_rate": 2.7160493827160496e-06, - "loss": 0.0008, + "loss": 0.0007, "step": 3500 }, { "epoch": 44.0, "eval_B-Claim": { - "f1-score": 0.697594501718213, - "precision": 0.7073170731707317, - "recall": 0.688135593220339, - "support": 295.0 + "f1-score": 0.6791277258566978, + "precision": 0.6707692307692308, + "recall": 0.6876971608832808, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.910828025477707, - "precision": 0.9050632911392406, - "recall": 0.9166666666666666, - "support": 156.0 + "f1-score": 0.922077922077922, + "precision": 0.9281045751633987, + "recall": 0.9161290322580645, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8972602739726028, - "precision": 0.8935879945429741, - "recall": 0.9009628610729024, - "support": 727.0 + "f1-score": 0.8867235079171742, + "precision": 0.8888888888888888, + "recall": 0.8845686512758202, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6770255271920089, - "precision": 0.6921331316187594, - "recall": 0.662563359884142, - "support": 4143.0 + "f1-score": 0.671923924388264, + "precision": 0.6770273428371115, + "recall": 0.6668968692449355, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9156572454015253, - "precision": 0.9011037527593819, - "recall": 0.9306885544915641, - "support": 2193.0 + "f1-score": 0.9232954545454545, + "precision": 0.9241706161137441, + "recall": 0.9224219489120151, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9102726696258719, - "precision": 0.9064645986265688, - "recall": 0.9141128711295072, - "support": 12563.0 + "f1-score": 0.9072981195703971, + "precision": 0.9050753254351324, + "recall": 0.9095318586021901, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9998803398348689, - "precision": 0.9999202233745512, - "recall": 0.9998404594767071, - "support": 12536.0 + "f1-score": 0.9996386956914461, + "precision": 0.9996386956914461, + "recall": 0.9996386956914461, + "support": 11071.0 }, - "eval_accuracy": 0.9138993652837826, - "eval_loss": 0.8316358327865601, + "eval_accuracy": 0.9058616755573371, + "eval_loss": 0.860885739326477, "eval_macro avg": { - "f1-score": 0.8583597976032568, - "precision": 0.8579414378903154, - "recall": 0.8589957665631184, - "support": 32613.0 - }, - "eval_runtime": 4.951, - "eval_samples_per_second": 16.36, - "eval_steps_per_second": 2.222, + "f1-score": 0.8557264785781936, + "precision": 0.8562392392712789, + "recall": 0.8552691738382502, + "support": 32431.0 + }, + "eval_runtime": 4.9389, + "eval_samples_per_second": 16.198, + "eval_steps_per_second": 2.025, "eval_weighted avg": { - "f1-score": 0.9132369409794028, - "precision": 0.9127044322479574, - "recall": 0.9138993652837826, - "support": 32613.0 + "f1-score": 0.9056541162924544, + "precision": 0.905464186235529, + "recall": 0.9058616755573371, + "support": 32431.0 }, "step": 3564 }, { "epoch": 45.0, "eval_B-Claim": { - "f1-score": 0.6840277777777778, - "precision": 0.701067615658363, - "recall": 0.6677966101694915, - "support": 295.0 + "f1-score": 0.685337726523888, + "precision": 0.7172413793103448, + "recall": 0.6561514195583596, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9038461538461539, - "precision": 0.9038461538461539, - "recall": 0.9038461538461539, - "support": 156.0 + "f1-score": 0.9260450160771705, + "precision": 0.9230769230769231, + "recall": 0.9290322580645162, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8950953678474116, - "precision": 0.8866396761133604, - "recall": 0.9037138927097662, - "support": 727.0 + "f1-score": 0.8977884040645547, + "precision": 0.8835294117647059, + "recall": 0.9125151883353585, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6764705882352942, - "precision": 0.6993043030146869, - "recall": 0.6550808592807145, - "support": 4143.0 + "f1-score": 0.6761074800290486, + "precision": 0.7128637059724349, + "recall": 0.6429558011049724, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9135073779795687, - "precision": 0.9095840867992767, - "recall": 0.9174646602827178, - "support": 2193.0 + "f1-score": 0.923368022705771, + "precision": 0.923368022705771, + "recall": 0.923368022705771, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9118354155992591, - "precision": 0.9030444964871195, - "recall": 0.9207991721722518, - "support": 12563.0 + "f1-score": 0.913971492656103, + "precision": 0.9000356252226577, + "recall": 0.9283457044168443, + "support": 13607.0 }, "eval_O": { - "f1-score": 0.9998404340194671, - "precision": 1.0, - "recall": 0.9996809189534142, - "support": 12536.0 + "f1-score": 0.9995031392565156, + "precision": 0.9996385977593061, + "recall": 0.9993677174600307, + "support": 11071.0 }, - "eval_accuracy": 0.9143899671909974, - "eval_loss": 0.8290512561798096, + "eval_accuracy": 0.9109802349603774, + "eval_loss": 0.8607417345046997, "eval_macro avg": { - "f1-score": 0.8549461593292761, - "precision": 0.8576409045598515, - "recall": 0.8526260382020728, - "support": 32613.0 - }, - "eval_runtime": 4.9524, - "eval_samples_per_second": 16.356, - "eval_steps_per_second": 2.221, + "f1-score": 0.8603030401875786, + "precision": 0.8656790951160204, + "recall": 0.8559623016636931, + "support": 32431.0 + }, + "eval_runtime": 4.916, + "eval_samples_per_second": 16.273, + "eval_steps_per_second": 2.034, "eval_weighted avg": { - "f1-score": 0.9134041597178751, - "precision": 0.9126816186877005, - "recall": 0.9143899671909974, - "support": 32613.0 + "f1-score": 0.9093333274296762, + "precision": 0.9083916855213499, + "recall": 0.9109802349603774, + "support": 32431.0 }, "step": 3645 }, { "epoch": 46.0, "eval_B-Claim": { - "f1-score": 0.702970297029703, - "precision": 0.684887459807074, - "recall": 0.7220338983050848, - "support": 295.0 + "f1-score": 0.6800618238021638, + "precision": 0.6666666666666666, + "recall": 0.694006309148265, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.910828025477707, - "precision": 0.9050632911392406, - "recall": 0.9166666666666666, - "support": 156.0 + "f1-score": 0.922077922077922, + "precision": 0.9281045751633987, + "recall": 0.9161290322580645, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8927576601671309, - "precision": 0.9040902679830748, - "recall": 0.8817056396148556, - "support": 727.0 + "f1-score": 0.8868501529051988, + "precision": 0.8928571428571429, + "recall": 0.8809234507897934, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.682891911238368, - "precision": 0.6751592356687898, - "recall": 0.6908037653874004, - "support": 4143.0 + "f1-score": 0.6812327506899724, + "precision": 0.6806066176470589, + "recall": 0.6818600368324125, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.916331096196868, - "precision": 0.8994290733421169, - "recall": 0.9338805289557683, - "support": 2193.0 + "f1-score": 0.9240596167494677, + "precision": 0.9242782773308093, + "recall": 0.9238410596026491, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.907720676661589, - "precision": 0.9143180166357102, - "recall": 0.9012178619756428, - "support": 12563.0 + "f1-score": 0.9089239134428156, + "precision": 0.9087569791360565, + "recall": 0.9090909090909091, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, - "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9993673746046091, + "precision": 0.9999095759110227, + "recall": 0.9988257609971999, + "support": 11071.0 }, - "eval_accuracy": 0.9126728605157453, - "eval_loss": 0.8511553406715393, + "eval_accuracy": 0.9074650797076871, + "eval_loss": 0.839061439037323, "eval_macro avg": { - "f1-score": 0.859071380967338, - "precision": 0.8547067635108581, - "recall": 0.8637583372722026, - "support": 32613.0 - }, - "eval_runtime": 4.9644, - "eval_samples_per_second": 16.316, - "eval_steps_per_second": 2.216, + "f1-score": 0.8575105077531642, + "precision": 0.8573114049588793, + "recall": 0.857810936959899, + "support": 32431.0 + }, + "eval_runtime": 4.9104, + "eval_samples_per_second": 16.292, + "eval_steps_per_second": 2.036, "eval_weighted avg": { - "f1-score": 0.9130386769588699, - "precision": 0.913522681540919, - "recall": 0.9126728605157453, - "support": 32613.0 + "f1-score": 0.90755266088551, + "precision": 0.9076484082022293, + "recall": 0.9074650797076871, + "support": 32431.0 }, "step": 3726 }, { "epoch": 47.0, "eval_B-Claim": { - "f1-score": 0.7048903878583475, - "precision": 0.7013422818791947, - "recall": 0.7084745762711865, - "support": 295.0 + "f1-score": 0.682615629984051, + "precision": 0.6903225806451613, + "recall": 0.6750788643533123, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9038461538461539, - "precision": 0.9038461538461539, - "recall": 0.9038461538461539, - "support": 156.0 + "f1-score": 0.9114754098360656, + "precision": 0.9266666666666666, + "recall": 0.896774193548387, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.8986905582356995, - "precision": 0.9005524861878453, - "recall": 0.8968363136176066, - "support": 727.0 + "f1-score": 0.8950542822677926, + "precision": 0.888622754491018, + "recall": 0.9015795868772782, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.676923076923077, - "precision": 0.6906077348066298, - "recall": 0.6637702148201786, - "support": 4143.0 + "f1-score": 0.6786858023959199, + "precision": 0.7000244678248104, + "recall": 0.658609576427256, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9125252979536767, - "precision": 0.9001774622892635, - "recall": 0.9252165982672138, - "support": 2193.0 + "f1-score": 0.9192015209125475, + "precision": 0.9235912129894938, + "recall": 0.9148533585619678, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9106477443907081, - "precision": 0.9070520413803996, - "recall": 0.9142720687733822, - "support": 12563.0 + "f1-score": 0.9126799476515922, + "precision": 0.9030285590964678, + "recall": 0.9225398691849783, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, - "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "f1-score": 0.9991412429378531, + "precision": 0.9999095350099512, + "recall": 0.9983741306115076, + "support": 11071.0 }, - "eval_accuracy": 0.9138380400453807, - "eval_loss": 0.8366554379463196, + "eval_accuracy": 0.9095001695908236, + "eval_loss": 0.8389514088630676, "eval_macro avg": { - "f1-score": 0.8582176027439518, - "precision": 0.8576540229127839, - "recall": 0.8589165607993888, - "support": 32613.0 - }, - "eval_runtime": 4.9791, - "eval_samples_per_second": 16.268, - "eval_steps_per_second": 2.209, + "f1-score": 0.856979119426546, + "precision": 0.8617379681033671, + "recall": 0.8525442256520982, + "support": 32431.0 + }, + "eval_runtime": 4.9029, + "eval_samples_per_second": 16.317, + "eval_steps_per_second": 2.04, "eval_weighted avg": { - "f1-score": 0.9132682856266137, - "precision": 0.9128007066684791, - "recall": 0.9138380400453807, - "support": 32613.0 + "f1-score": 0.9085760354788047, + "precision": 0.90791798885497, + "recall": 0.9095001695908236, + "support": 32431.0 }, "step": 3807 }, { "epoch": 48.0, "eval_B-Claim": { - "f1-score": 0.7046979865771812, - "precision": 0.6976744186046512, - "recall": 0.711864406779661, - "support": 295.0 + "f1-score": 0.6810631229235881, + "precision": 0.7192982456140351, + "recall": 0.6466876971608833, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.910828025477707, - "precision": 0.9050632911392406, - "recall": 0.9166666666666666, - "support": 156.0 + "f1-score": 0.9114754098360656, + "precision": 0.9266666666666666, + "recall": 0.896774193548387, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.896265560165975, - "precision": 0.9012517385257302, - "recall": 0.8913342503438789, - "support": 727.0 + "f1-score": 0.9007724301841948, + "precision": 0.8813953488372093, + "recall": 0.9210206561360875, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.674466788543571, - "precision": 0.6811915312653866, - "recall": 0.6678735216027033, - "support": 4143.0 + "f1-score": 0.6752427184466019, + "precision": 0.7140657084188912, + "recall": 0.6404235727440147, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9142091152815013, - "precision": 0.8961892247043364, - "recall": 0.93296853625171, - "support": 2193.0 + "f1-score": 0.909614001438504, + "precision": 0.9222168206125425, + "recall": 0.8973509933774835, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9082294860054942, - "precision": 0.9085550422176198, - "recall": 0.9079041630183873, - "support": 12563.0 + "f1-score": 0.9159276083351359, + "precision": 0.898945580638313, + "recall": 0.9335636069670022, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9991411653030782, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.998283804534369, + "support": 11071.0 }, - "eval_accuracy": 0.9123968969429369, - "eval_loss": 0.841332733631134, + "eval_accuracy": 0.9107335573987851, + "eval_loss": 0.8414669036865234, "eval_macro avg": { - "f1-score": 0.8583852802930613, - "precision": 0.8557036066367092, - "recall": 0.8612302206661439, - "support": 32613.0 - }, - "eval_runtime": 4.9999, - "eval_samples_per_second": 16.2, - "eval_steps_per_second": 2.2, + "f1-score": 0.8561766366381668, + "precision": 0.8660840529696653, + "recall": 0.8477292177811753, + "support": 32431.0 + }, + "eval_runtime": 4.9177, + "eval_samples_per_second": 16.268, + "eval_steps_per_second": 2.033, "eval_weighted avg": { - "f1-score": 0.9121155265596795, - "precision": 0.9119035452121351, - "recall": 0.9123968969429369, - "support": 32613.0 + "f1-score": 0.9089824114024411, + "precision": 0.9081267787357191, + "recall": 0.9107335573987851, + "support": 32431.0 }, "step": 3888 }, { "epoch": 49.0, "eval_B-Claim": { - "f1-score": 0.699490662139219, - "precision": 0.7006802721088435, - "recall": 0.6983050847457627, - "support": 295.0 + "f1-score": 0.6820428336079077, + "precision": 0.7137931034482758, + "recall": 0.6529968454258676, + "support": 317.0 }, "eval_B-MajorClaim": { - "f1-score": 0.9038461538461539, - "precision": 0.9038461538461539, - "recall": 0.9038461538461539, - "support": 156.0 + "f1-score": 0.9114754098360656, + "precision": 0.9266666666666666, + "recall": 0.896774193548387, + "support": 155.0 }, "eval_B-Premise": { - "f1-score": 0.897594501718213, - "precision": 0.896978021978022, - "recall": 0.8982118294360385, - "support": 727.0 + "f1-score": 0.899880810488677, + "precision": 0.8830409356725146, + "recall": 0.9173754556500607, + "support": 823.0 }, "eval_I-Claim": { - "f1-score": 0.6754786905497222, - "precision": 0.6918016194331984, - "recall": 0.6599082790248613, - "support": 4143.0 + "f1-score": 0.6787878787878788, + "precision": 0.7168458781362007, + "recall": 0.6445672191528545, + "support": 4344.0 }, "eval_I-MajorClaim": { - "f1-score": 0.9153225806451613, - "precision": 0.8996036988110965, - "recall": 0.9316005471956225, - "support": 2193.0 + "f1-score": 0.9103977000479158, + "precision": 0.9223300970873787, + "recall": 0.8987701040681173, + "support": 2114.0 }, "eval_I-Premise": { - "f1-score": 0.9101786917072785, - "precision": 0.9061218049858, - "recall": 0.9142720687733822, - "support": 12563.0 + "f1-score": 0.9167177637511272, + "precision": 0.9001274968125796, + "recall": 0.9339310648930698, + "support": 13607.0 }, "eval_O": { - "f1-score": 1.0, + "f1-score": 0.9991411653030782, "precision": 1.0, - "recall": 1.0, - "support": 12536.0 + "recall": 0.998283804534369, + "support": 11071.0 }, - "eval_accuracy": 0.9137153895685769, - "eval_loss": 0.8413227200508118, + "eval_accuracy": 0.911504424778761, + "eval_loss": 0.8389942049980164, "eval_macro avg": { - "f1-score": 0.8574158972293925, - "precision": 0.8570045101661591, - "recall": 0.8580205661459743, - "support": 32613.0 - }, - "eval_runtime": 4.9903, - "eval_samples_per_second": 16.232, - "eval_steps_per_second": 2.204, + "f1-score": 0.8569205088318073, + "precision": 0.866114882546231, + "recall": 0.8489569553246751, + "support": 32431.0 + }, + "eval_runtime": 4.9206, + "eval_samples_per_second": 16.258, + "eval_steps_per_second": 2.032, "eval_weighted avg": { - "f1-score": 0.913018933747312, - "precision": 0.9124697807344069, - "recall": 0.9137153895685769, - "support": 32613.0 + "f1-score": 0.9098268295642529, + "precision": 0.9089903982664754, + "recall": 0.911504424778761, + "support": 32431.0 }, "step": 3969 } @@ -3150,7 +3150,7 @@ "num_input_tokens_seen": 0, "num_train_epochs": 50, "save_steps": 500, - "total_flos": 7023868612551000.0, + "total_flos": 7045749823182000.0, "train_batch_size": 4, "trial_name": null, "trial_params": null