Training in progress, epoch 13, checkpoint

checkpoint-1053/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f6f11196cb52df2a9207b5132e8cee1efb0224893ec7dbcf6f3dec826221823
+oid sha256:640887274e215404287a624d422670f2d1e4301f7322280de47ddb6642fb1b76
 size 592324828

checkpoint-1053/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:260749ae0674b834271a7c3753d3e769d74ea97d61572890a6b89b9993a7dafb
+oid sha256:7e0044eb55df8da3359fcd42d00a5c539cabe677c6f4e0fde21d7075ae9c8972
 size 1014657786

checkpoint-1053/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a2d0b6fd3e0e222c69d0c818ba4ff34a17c0844523a335fc5dcaffe1d7766cfb
+oid sha256:2a8fed4ff2e0a1f5b060e9416aef90a5efb47a3301fd9fdfb8a444f0ad6d3aa3
 size 14244

checkpoint-1053/trainer_state.json

@@ -11,599 +11,599 @@
     {
       "epoch": 1.0,
       "eval_Claim": {
-        "f1-score": 0.5385059461955894,
-        "precision": 0.6267132491265789,
-        "recall": 0.4720647773279352,
-        "support": 4940.0
+        "f1-score": 0.5254768748366867,
+        "precision": 0.5768789443488239,
+        "recall": 0.4824856046065259,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.7814784727863526,
-        "precision": 0.9588039867109635,
-        "recall": 0.6595063985374772,
-        "support": 2188.0
+        "f1-score": 0.7984361424847959,
+        "precision": 0.7495921696574225,
+        "recall": 0.854089219330855,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9938608715302625,
-        "precision": 0.9996969237763298,
-        "recall": 0.9880925634688834,
-        "support": 13353.0
+        "f1-score": 0.9990277532260916,
+        "precision": 0.9988511841640155,
+        "recall": 0.9992043847241867,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.8968837690149166,
-        "precision": 0.8455112497215416,
-        "recall": 0.954902824076986,
-        "support": 15899.0
+        "f1-score": 0.8854183656825966,
+        "precision": 0.8719781543651113,
+        "recall": 0.8992793837488611,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.8837548103353491,
-      "eval_loss": 0.2860471308231354,
+      "eval_accuracy": 0.8755765022723447,
+      "eval_loss": 0.28303390741348267,
       "eval_macro avg": {
-        "f1-score": 0.8026822648817803,
-        "precision": 0.8576813523338533,
-        "recall": 0.7686416408528204,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9572,
-      "eval_samples_per_second": 16.34,
-      "eval_steps_per_second": 2.219,
+        "f1-score": 0.8020897840575427,
+        "precision": 0.7993251131338432,
+        "recall": 0.8087646481026072,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.8106,
+      "eval_samples_per_second": 16.63,
+      "eval_steps_per_second": 2.079,
       "eval_weighted avg": {
-        "f1-score": 0.8768739289107891,
-        "precision": 0.8792073379911217,
-        "recall": 0.8837548103353491,
-        "support": 36380.0
+        "f1-score": 0.8718761173649596,
+        "precision": 0.8700202202343714,
+        "recall": 0.8755765022723447,
+        "support": 29705.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_Claim": {
-        "f1-score": 0.6469459868753155,
-        "precision": 0.6453172205438067,
-        "recall": 0.648582995951417,
-        "support": 4940.0
+        "f1-score": 0.5578782336033447,
+        "precision": 0.6124497991967871,
+        "recall": 0.5122360844529751,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8435970110796187,
-        "precision": 0.9669226225634967,
-        "recall": 0.7481718464351006,
-        "support": 2188.0
+        "f1-score": 0.784421283598464,
+        "precision": 0.9571619812583668,
+        "recall": 0.6644981412639405,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.998877245508982,
-        "precision": 0.9983541557567143,
-        "recall": 0.9994008836965476,
-        "support": 13353.0
+        "f1-score": 0.9998674150351351,
+        "precision": 0.9997348652231551,
+        "recall": 1.0,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.9099026477336144,
-        "precision": 0.8972179761540813,
-        "recall": 0.922951129001824,
-        "support": 15899.0
+        "f1-score": 0.8977749872463995,
+        "precision": 0.8530201342281879,
+        "recall": 0.9474861260664292,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.903243540406817,
-      "eval_loss": 0.24885737895965576,
+      "eval_accuracy": 0.8859114627167144,
+      "eval_loss": 0.27363118529319763,
       "eval_macro avg": {
-        "f1-score": 0.8498307227993828,
-        "precision": 0.8769529937545246,
-        "recall": 0.8297767137712222,
-        "support": 36380.0
-      },
-      "eval_runtime": 5.0097,
-      "eval_samples_per_second": 16.169,
-      "eval_steps_per_second": 2.196,
+        "f1-score": 0.8099854798708358,
+        "precision": 0.8555916949766242,
+        "recall": 0.7810550879458362,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.7727,
+      "eval_samples_per_second": 16.762,
+      "eval_steps_per_second": 2.095,
       "eval_weighted avg": {
-        "f1-score": 0.9028656814459712,
-        "precision": 0.9043261520711511,
-        "recall": 0.903243540406817,
-        "support": 36380.0
+        "f1-score": 0.8807489883812781,
+        "precision": 0.8826802296805741,
+        "recall": 0.8859114627167144,
+        "support": 29705.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_Claim": {
-        "f1-score": 0.7040630102775911,
-        "precision": 0.6698958142935478,
-        "recall": 0.7419028340080972,
-        "support": 4940.0
+        "f1-score": 0.64760736196319,
+        "precision": 0.6627322953289804,
+        "recall": 0.6331573896353166,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8888411094388303,
-        "precision": 0.8392204628501827,
-        "recall": 0.9446983546617916,
-        "support": 2188.0
+        "f1-score": 0.9016018306636155,
+        "precision": 0.8881875563570785,
+        "recall": 0.9154275092936803,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.998989104796136,
-        "precision": 0.9988769092542678,
-        "recall": 0.9991013255448213,
-        "support": 13353.0
+        "f1-score": 0.9996904982977407,
+        "precision": 1.0,
+        "recall": 0.9993811881188119,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.914582593823615,
-        "precision": 0.9390987408880053,
-        "recall": 0.8913139191144097,
-        "support": 15899.0
+        "f1-score": 0.8982820417748115,
+        "precision": 0.8936065573770492,
+        "recall": 0.9030067091857865,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.9137987905442551,
-      "eval_loss": 0.24298855662345886,
+      "eval_accuracy": 0.9027436458508669,
+      "eval_loss": 0.22585685551166534,
       "eval_macro avg": {
-        "f1-score": 0.8766189545840432,
-        "precision": 0.8617729818215009,
-        "recall": 0.8942541083322799,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9876,
-      "eval_samples_per_second": 16.24,
-      "eval_steps_per_second": 2.205,
+        "f1-score": 0.8617954331748394,
+        "precision": 0.861131602265777,
+        "recall": 0.8627431990583988,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.8003,
+      "eval_samples_per_second": 16.666,
+      "eval_steps_per_second": 2.083,
       "eval_weighted avg": {
-        "f1-score": 0.9154289662937856,
-        "precision": 0.9184781183611038,
-        "recall": 0.9137987905442551,
-        "support": 36380.0
+        "f1-score": 0.9019670975035187,
+        "precision": 0.9013351218793044,
+        "recall": 0.9027436458508669,
+        "support": 29705.0
       },
       "step": 243
     },
     {
       "epoch": 4.0,
       "eval_Claim": {
-        "f1-score": 0.6981611893583725,
-        "precision": 0.6754352763058289,
-        "recall": 0.7224696356275304,
-        "support": 4940.0
+        "f1-score": 0.6477602966229081,
+        "precision": 0.5561865427637239,
+        "recall": 0.7754318618042226,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.901231802911534,
-        "precision": 0.8836187966622749,
-        "recall": 0.9195612431444241,
-        "support": 2188.0
+        "f1-score": 0.8984409799554566,
+        "precision": 0.8627031650983746,
+        "recall": 0.9372676579925651,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9989130842172332,
-        "precision": 0.9998499399759904,
-        "recall": 0.9979779824758481,
-        "support": 13353.0
+        "f1-score": 1.0,
+        "precision": 1.0,
+        "recall": 1.0,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.9146224912392481,
-        "precision": 0.9266670970240785,
-        "recall": 0.9028869740235235,
-        "support": 15899.0
+        "f1-score": 0.861495720750997,
+        "precision": 0.9384029675907849,
+        "recall": 0.7962395427814131,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.9142935678944475,
-      "eval_loss": 0.23980534076690674,
+      "eval_accuracy": 0.8811311227066151,
+      "eval_loss": 0.32341843843460083,
       "eval_macro avg": {
-        "f1-score": 0.8782321419315969,
-        "precision": 0.8713927774920431,
-        "recall": 0.8857239588178316,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9726,
-      "eval_samples_per_second": 16.289,
-      "eval_steps_per_second": 2.212,
+        "f1-score": 0.8519242493323405,
+        "precision": 0.8393231688632209,
+        "recall": 0.8772347656445502,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.7708,
+      "eval_samples_per_second": 16.769,
+      "eval_steps_per_second": 2.096,
       "eval_weighted avg": {
-        "f1-score": 0.9153623106642749,
-        "precision": 0.9168247558035478,
-        "recall": 0.9142935678944475,
-        "support": 36380.0
+        "f1-score": 0.8869263673393438,
+        "precision": 0.9027457246003854,
+        "recall": 0.8811311227066151,
+        "support": 29705.0
       },
       "step": 324
     },
     {
       "epoch": 5.0,
       "eval_Claim": {
-        "f1-score": 0.6650318783717509,
-        "precision": 0.6450999048525214,
-        "recall": 0.6862348178137652,
-        "support": 4940.0
+        "f1-score": 0.6188788236838649,
+        "precision": 0.6833864888373442,
+        "recall": 0.5654990403071017,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9172031076581575,
-        "precision": 0.8916702632714717,
-        "recall": 0.9442413162705667,
-        "support": 2188.0
+        "f1-score": 0.909733679000707,
+        "precision": 0.9230033476805356,
+        "recall": 0.8968401486988847,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9982383147794145,
-        "precision": 0.9992495872730002,
-        "recall": 0.9972290870965326,
-        "support": 13353.0
+        "f1-score": 0.9999558011049724,
+        "precision": 0.9999116061168567,
+        "recall": 1.0,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.8998438545616774,
-        "precision": 0.9119622787753521,
-        "recall": 0.8880432731618341,
-        "support": 15899.0
+        "f1-score": 0.8978936810431294,
+        "precision": 0.8706816059757236,
+        "recall": 0.9268615919821088,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.9040956569543706,
-      "eval_loss": 0.2932502031326294,
+      "eval_accuracy": 0.9018347079616226,
+      "eval_loss": 0.3072384297847748,
       "eval_macro avg": {
-        "f1-score": 0.87007928884275,
-        "precision": 0.8619955085430864,
-        "recall": 0.8789371235856747,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9595,
-      "eval_samples_per_second": 16.332,
-      "eval_steps_per_second": 2.218,
+        "f1-score": 0.8566154962081685,
+        "precision": 0.8692457621526151,
+        "recall": 0.8473001952470238,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.7922,
+      "eval_samples_per_second": 16.694,
+      "eval_steps_per_second": 2.087,
       "eval_weighted avg": {
-        "f1-score": 0.9051179642561333,
-        "precision": 0.9065430476942037,
-        "recall": 0.9040956569543706,
-        "support": 36380.0
+        "f1-score": 0.8984684143294738,
+        "precision": 0.8974043833368577,
+        "recall": 0.9018347079616226,
+        "support": 29705.0
       },
       "step": 405
     },
     {
       "epoch": 6.0,
       "eval_Claim": {
-        "f1-score": 0.6622698072805139,
-        "precision": 0.5740163325909429,
-        "recall": 0.7825910931174089,
-        "support": 4940.0
+        "f1-score": 0.620837537840565,
+        "precision": 0.6545212765957447,
+        "recall": 0.5904510556621881,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9223704351629127,
-        "precision": 0.8842767295597485,
-        "recall": 0.9638939670932358,
-        "support": 2188.0
+        "f1-score": 0.8993613741466637,
+        "precision": 0.8547509418166597,
+        "recall": 0.9488847583643123,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9982004948639124,
-        "precision": 0.9993994444861497,
-        "recall": 0.9970044184827379,
-        "support": 13353.0
+        "f1-score": 0.9999558011049724,
+        "precision": 0.9999116061168567,
+        "recall": 1.0,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.8780749379985253,
-        "precision": 0.9398091685199799,
-        "recall": 0.8239511918988616,
-        "support": 15899.0
+        "f1-score": 0.8931567692054615,
+        "precision": 0.8869558114841134,
+        "recall": 0.8994450426571688,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.8902693787795491,
-      "eval_loss": 0.3604726195335388,
+      "eval_accuracy": 0.8979633058407676,
+      "eval_loss": 0.3634909987449646,
       "eval_macro avg": {
-        "f1-score": 0.865228918826466,
-        "precision": 0.8493754187892053,
-        "recall": 0.8918601676480611,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.991,
-      "eval_samples_per_second": 16.229,
-      "eval_steps_per_second": 2.204,
+        "f1-score": 0.8533278705744156,
+        "precision": 0.8490349090033436,
+        "recall": 0.8596952141709173,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.7765,
+      "eval_samples_per_second": 16.749,
+      "eval_steps_per_second": 2.094,
       "eval_weighted avg": {
-        "f1-score": 0.8955262233990811,
-        "precision": 0.9086708334194257,
-        "recall": 0.8902693787795491,
-        "support": 36380.0
+        "f1-score": 0.8960665959131486,
+        "precision": 0.8950239457358053,
+        "recall": 0.8979633058407676,
+        "support": 29705.0
       },
       "step": 486
     },
     {
       "epoch": 6.17,
-      "grad_norm": 17.04454231262207,
+      "grad_norm": 9.48759651184082,
       "learning_rate": 1.7530864197530865e-05,
-      "loss": 0.2104,
+      "loss": 0.214,
       "step": 500
     },
     {
       "epoch": 7.0,
       "eval_Claim": {
-        "f1-score": 0.7037529399734124,
-        "precision": 0.7110973341599504,
-        "recall": 0.6965587044534413,
-        "support": 4940.0
+        "f1-score": 0.6812872467222885,
+        "precision": 0.6769303647560397,
+        "recall": 0.685700575815739,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9196885428253615,
-        "precision": 0.8959687906371911,
-        "recall": 0.9446983546617916,
-        "support": 2188.0
+        "f1-score": 0.9276377217553688,
+        "precision": 0.9319887429643527,
+        "recall": 0.9233271375464684,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9987257327036954,
-        "precision": 0.9996248780853777,
-        "recall": 0.997828203399985,
-        "support": 13353.0
+        "f1-score": 0.9999115983026874,
+        "precision": 0.9999115983026874,
+        "recall": 0.9999115983026874,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.9188781285372909,
-        "precision": 0.9187048098082364,
-        "recall": 0.9190515126737531,
-        "support": 15899.0
+        "f1-score": 0.9017086927670869,
+        "precision": 0.9029819752471135,
+        "recall": 0.9004389961070156,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.9192963166575041,
-      "eval_loss": 0.36414769291877747,
+      "eval_accuracy": 0.9098468271334792,
+      "eval_loss": 0.3468942642211914,
       "eval_macro avg": {
-        "f1-score": 0.8852613360099401,
-        "precision": 0.8813489531726889,
-        "recall": 0.8895341937972427,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9835,
-      "eval_samples_per_second": 16.254,
-      "eval_steps_per_second": 2.207,
+        "f1-score": 0.8776363148868579,
+        "precision": 0.8779531703175483,
+        "recall": 0.8773445769429775,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.7913,
+      "eval_samples_per_second": 16.697,
+      "eval_steps_per_second": 2.087,
       "eval_weighted avg": {
-        "f1-score": 0.9190227083446231,
-        "precision": 0.9188476996393494,
-        "recall": 0.9192963166575041,
-        "support": 36380.0
+        "f1-score": 0.9100559053806798,
+        "precision": 0.910277290769933,
+        "recall": 0.9098468271334792,
+        "support": 29705.0
       },
       "step": 567
     },
     {
       "epoch": 8.0,
       "eval_Claim": {
-        "f1-score": 0.7025399811853246,
-        "precision": 0.6562390158172232,
-        "recall": 0.7558704453441295,
-        "support": 4940.0
+        "f1-score": 0.6928905033731187,
+        "precision": 0.6105725260654838,
+        "recall": 0.8008637236084453,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9105835367206567,
-        "precision": 0.8848641655886158,
-        "recall": 0.9378427787934186,
-        "support": 2188.0
+        "f1-score": 0.9302433371958285,
+        "precision": 0.9278779472954231,
+        "recall": 0.9326208178438662,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9986513823331086,
-        "precision": 0.9991005172026085,
-        "recall": 0.9982026510896428,
-        "support": 13353.0
+        "f1-score": 0.9999115904871364,
+        "precision": 1.0,
+        "recall": 0.9998231966053748,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.9083384525849526,
-        "precision": 0.9345974717232203,
-        "recall": 0.8835146864582678,
-        "support": 15899.0
+        "f1-score": 0.8820387074174622,
+        "precision": 0.9356247097073851,
+        "recall": 0.8342582622380519,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.911544804837823,
-      "eval_loss": 0.35222452878952026,
+      "eval_accuracy": 0.8997475172529877,
+      "eval_loss": 0.40174800157546997,
       "eval_macro avg": {
-        "f1-score": 0.8800283382060107,
-        "precision": 0.8687002925829169,
-        "recall": 0.8938576404213647,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9741,
-      "eval_samples_per_second": 16.285,
-      "eval_steps_per_second": 2.211,
+        "f1-score": 0.8762710346183864,
+        "precision": 0.868518795767073,
+        "recall": 0.8918915000739345,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.7722,
+      "eval_samples_per_second": 16.764,
+      "eval_steps_per_second": 2.095,
       "eval_weighted avg": {
-        "f1-score": 0.9136769997620249,
-        "precision": 0.9174837257168467,
-        "recall": 0.911544804837823,
-        "support": 36380.0
+        "f1-score": 0.9038782866839282,
+        "precision": 0.9139692560686062,
+        "recall": 0.8997475172529877,
+        "support": 29705.0
       },
       "step": 648
     },
     {
       "epoch": 9.0,
       "eval_Claim": {
-        "f1-score": 0.6566933693658913,
-        "precision": 0.6275594908688434,
-        "recall": 0.6886639676113361,
-        "support": 4940.0
+        "f1-score": 0.6602241979126401,
+        "precision": 0.7130531589201224,
+        "recall": 0.6146833013435701,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8702745722244328,
-        "precision": 0.7706131078224101,
-        "recall": 0.9995429616087751,
-        "support": 2188.0
+        "f1-score": 0.9052154195011338,
+        "precision": 0.8839681133746679,
+        "recall": 0.9275092936802974,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9987648313807688,
-        "precision": 0.9983537862915295,
-        "recall": 0.9991762150827529,
-        "support": 13353.0
+        "f1-score": 0.9999558011049724,
+        "precision": 0.9999116061168567,
+        "recall": 1.0,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.9001826722338205,
-        "precision": 0.9350138917124077,
-        "recall": 0.8678533241084345,
-        "support": 15899.0
+        "f1-score": 0.9072885349800925,
+        "precision": 0.8903596204449405,
+        "recall": 0.9248736850824153,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.8996426608026388,
-      "eval_loss": 0.4436376690864563,
+      "eval_accuracy": 0.910149806429894,
+      "eval_loss": 0.4713464379310608,
       "eval_macro avg": {
-        "f1-score": 0.8564788613012284,
-        "precision": 0.8328850691737977,
-        "recall": 0.8888091171028246,
-        "support": 36380.0
-      },
-      "eval_runtime": 5.0402,
-      "eval_samples_per_second": 16.071,
-      "eval_steps_per_second": 2.182,
+        "f1-score": 0.8681709883747097,
+        "precision": 0.8718231247141469,
+        "recall": 0.8667665700265708,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.7731,
+      "eval_samples_per_second": 16.761,
+      "eval_steps_per_second": 2.095,
       "eval_weighted avg": {
-        "f1-score": 0.9015045934020747,
-        "precision": 0.90662587513727,
-        "recall": 0.8996426608026388,
-        "support": 36380.0
+        "f1-score": 0.9077607320175198,
+        "precision": 0.9067368029754924,
+        "recall": 0.910149806429894,
+        "support": 29705.0
       },
       "step": 729
     },
     {
       "epoch": 10.0,
       "eval_Claim": {
-        "f1-score": 0.7371103705155853,
-        "precision": 0.7145572025845686,
-        "recall": 0.7611336032388664,
-        "support": 4940.0
+        "f1-score": 0.6952499153785399,
+        "precision": 0.6562300319488817,
+        "recall": 0.7392034548944337,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9365328549002018,
-        "precision": 0.9194187582562747,
-        "recall": 0.9542961608775137,
-        "support": 2188.0
+        "f1-score": 0.9103908484270734,
+        "precision": 0.9344422700587084,
+        "recall": 0.887546468401487,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9991386090408599,
-        "precision": 0.9993257416841475,
-        "recall": 0.9989515464689583,
-        "support": 13353.0
+        "f1-score": 1.0,
+        "precision": 1.0,
+        "recall": 1.0,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.9225428371233836,
-        "precision": 0.9344473837021743,
-        "recall": 0.9109377948298635,
-        "support": 15899.0
+        "f1-score": 0.9012517385257303,
+        "precision": 0.9174532349407929,
+        "recall": 0.8856125238134681,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.9255085211654756,
-      "eval_loss": 0.44094258546829224,
+      "eval_accuracy": 0.9087695674128935,
+      "eval_loss": 0.4968256950378418,
       "eval_macro avg": {
-        "f1-score": 0.8988311678950076,
-        "precision": 0.8919372715567913,
-        "recall": 0.9063297763538004,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9458,
-      "eval_samples_per_second": 16.378,
-      "eval_steps_per_second": 2.224,
+        "f1-score": 0.8767231255828358,
+        "precision": 0.8770313842370958,
+        "recall": 0.8780906117773473,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.7941,
+      "eval_samples_per_second": 16.687,
+      "eval_steps_per_second": 2.086,
       "eval_weighted avg": {
-        "f1-score": 0.9263184587909816,
-        "precision": 0.927497977075915,
-        "recall": 0.9255085211654756,
-        "support": 36380.0
+        "f1-score": 0.9106135328171674,
+        "precision": 0.9134657277821738,
+        "recall": 0.9087695674128935,
+        "support": 29705.0
       },
       "step": 810
     },
     {
       "epoch": 11.0,
       "eval_Claim": {
-        "f1-score": 0.6750768578394996,
-        "precision": 0.7086579123080348,
-        "recall": 0.6445344129554655,
-        "support": 4940.0
+        "f1-score": 0.6796779702366431,
+        "precision": 0.6913151364764268,
+        "recall": 0.668426103646833,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9008264462809917,
-        "precision": 0.8220211161387632,
-        "recall": 0.9963436928702011,
-        "support": 2188.0
+        "f1-score": 0.9301903898458749,
+        "precision": 0.9079646017699115,
+        "recall": 0.9535315985130112,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9990631440884392,
-        "precision": 0.9998499849984999,
-        "recall": 0.9982775406275743,
-        "support": 13353.0
+        "f1-score": 0.9999558011049724,
+        "precision": 0.9999116061168567,
+        "recall": 1.0,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.9168605748066159,
-        "precision": 0.9167452681883921,
-        "recall": 0.9169759104346186,
-        "support": 15899.0
+        "f1-score": 0.901923474663909,
+        "precision": 0.9008428358948934,
+        "recall": 0.9030067091857865,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.9145959318306762,
-      "eval_loss": 0.4954843521118164,
+      "eval_accuracy": 0.9106884362901868,
+      "eval_loss": 0.5864725708961487,
       "eval_macro avg": {
-        "f1-score": 0.8729567557538865,
-        "precision": 0.8618185704084225,
-        "recall": 0.8890328892219649,
-        "support": 36380.0
-      },
-      "eval_runtime": 5.0503,
-      "eval_samples_per_second": 16.039,
-      "eval_steps_per_second": 2.178,
+        "f1-score": 0.8779369089628499,
+        "precision": 0.8750085450645222,
+        "recall": 0.8812411028364077,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.8031,
+      "eval_samples_per_second": 16.656,
+      "eval_steps_per_second": 2.082,
       "eval_weighted avg": {
-        "f1-score": 0.9132365141301059,
-        "precision": 0.9132952764575458,
-        "recall": 0.9145959318306762,
-        "support": 36380.0
+        "f1-score": 0.9101191594213591,
+        "precision": 0.9096858090555641,
+        "recall": 0.9106884362901868,
+        "support": 29705.0
       },
       "step": 891
     },
     {
       "epoch": 12.0,
       "eval_Claim": {
-        "f1-score": 0.6857200770244248,
-        "precision": 0.6866247209255125,
-        "recall": 0.6848178137651821,
-        "support": 4940.0
+        "f1-score": 0.697112745636184,
+        "precision": 0.6425824731835661,
+        "recall": 0.7617562380038387,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9214571488734471,
-        "precision": 0.8543537680593518,
-        "recall": 1.0,
-        "support": 2188.0
+        "f1-score": 0.9365296803652968,
+        "precision": 0.920556552962298,
+        "recall": 0.9530669144981413,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9987652935234033,
-        "precision": 0.9979811574697174,
-        "recall": 0.9995506627724107,
-        "support": 13353.0
+        "f1-score": 1.0,
+        "precision": 1.0,
+        "recall": 1.0,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.9127542413343095,
-        "precision": 0.9239592731022039,
-        "recall": 0.9018177243851815,
-        "support": 15899.0
+        "f1-score": 0.8923037300940034,
+        "precision": 0.9260513186029936,
+        "recall": 0.8609293464756067,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.91412864211105,
-      "eval_loss": 0.5210011005401611,
+      "eval_accuracy": 0.9066487123379903,
+      "eval_loss": 0.5556066632270813,
       "eval_macro avg": {
-        "f1-score": 0.8796741901888963,
-        "precision": 0.8657297298891964,
-        "recall": 0.8965465502306935,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9497,
-      "eval_samples_per_second": 16.365,
-      "eval_steps_per_second": 2.222,
+        "f1-score": 0.8814865390238711,
+        "precision": 0.8722975861872144,
+        "recall": 0.8939381247443967,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.7999,
+      "eval_samples_per_second": 16.667,
+      "eval_steps_per_second": 2.083,
       "eval_weighted avg": {
-        "f1-score": 0.9140186385274314,
-        "precision": 0.9147147620844138,
-        "recall": 0.91412864211105,
-        "support": 36380.0
+        "f1-score": 0.9091318205481447,
+        "precision": 0.9140393543072853,
+        "recall": 0.9066487123379903,
+        "support": 29705.0
       },
       "step": 972
     },
     {
       "epoch": 12.35,
-      "grad_norm": 3.2667109966278076,
+      "grad_norm": 3.987734794616699,
       "learning_rate": 1.506172839506173e-05,
-      "loss": 0.0286,
+      "loss": 0.0252,
       "step": 1000
     },
     {
       "epoch": 13.0,
       "eval_Claim": {
-        "f1-score": 0.7024380291747424,
-        "precision": 0.7079991774624718,
-        "recall": 0.6969635627530364,
-        "support": 4940.0
+        "f1-score": 0.6895428929242329,
+        "precision": 0.7212470526591565,
+        "recall": 0.6605086372360844,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9075119409465915,
-        "precision": 0.8643507030603805,
-        "recall": 0.9552102376599635,
-        "support": 2188.0
+        "f1-score": 0.9265742214139577,
+        "precision": 0.9069870939029817,
+        "recall": 0.9470260223048327,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9986895802912875,
-        "precision": 0.9985774183887391,
-        "recall": 0.9988017673930952,
-        "support": 13353.0
+        "f1-score": 0.9999558011049724,
+        "precision": 0.9999116061168567,
+        "recall": 1.0,
+        "support": 11312.0
       },
       "eval_Premise": {
-        "f1-score": 0.9200429808482398,
-        "precision": 0.9246014101505431,
-        "recall": 0.9155292785709793,
-        "support": 15899.0
+        "f1-score": 0.9083234293676489,
+        "precision": 0.8989292667099286,
+        "recall": 0.917916010933488,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.9188015393073117,
-      "eval_loss": 0.5909315347671509,
+      "eval_accuracy": 0.9151657970038714,
+      "eval_loss": 0.5547633767127991,
       "eval_macro avg": {
-        "f1-score": 0.8821706328152152,
-        "precision": 0.8738821772655336,
-        "recall": 0.8916262115942686,
-        "support": 36380.0
-      },
-      "eval_runtime": 4.9699,
-      "eval_samples_per_second": 16.298,
-      "eval_steps_per_second": 2.213,
+        "f1-score": 0.8810990862027029,
+        "precision": 0.8817687548472308,
+        "recall": 0.8813626676186013,
+        "support": 29705.0
+      },
+      "eval_runtime": 4.8002,
+      "eval_samples_per_second": 16.666,
+      "eval_steps_per_second": 2.083,
       "eval_weighted avg": {
-        "f1-score": 0.9186076225687217,
-        "precision": 0.918717904416961,
-        "recall": 0.9188015393073117,
-        "support": 36380.0
+        "f1-score": 0.9138424940934561,
+        "precision": 0.9130371003853032,
+        "recall": 0.9151657970038714,
+        "support": 29705.0
       },
       "step": 1053
     }
@@ -613,7 +613,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 1863448390276200.0,
+  "total_flos": 1869253525448400.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

checkpoint-1053/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:95e8ce1619d475b315f2cb3494747dc2424fb9856a35d6759fcf75e79995f1f1
+oid sha256:6c4152f1ae85467509d6d1bf2d3b859abf08e786a1c118f9c84f899edc2df5d1
 size 4856