Training in progress, epoch 12, checkpoint

checkpoint-492/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f68f549d3574ddc833a1bf0e5882e856c15052aa9fd527f955a2646cb3225ed
+oid sha256:6196e4ea61870f43126a6769d53a20551c226fff9c3b82c6c1b2f1e42d270778
 size 592324828

checkpoint-492/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3facc874d5ca280780eb6392396af2afb55fab003abdca8ae572d7e97cfc1329
+oid sha256:30e11cb045821a3d25c10f6c62443440986d45e50175ff61f4590568813b0e76
 size 1014657786

checkpoint-492/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b6e53e706c63ba0c275c9b6332ceadb76c337b40858f3ddadb0ab10622a334eb
+oid sha256:1c189466643703424c77db24cb1c70bb1b76b65752ddeeaa3d7bb424b6ea3dcb
 size 14244

checkpoint-492/trainer_state.json

@@ -11,540 +11,540 @@
     {
       "epoch": 1.0,
       "eval_Claim": {
-        "f1-score": 0.4643135219309629,
-        "precision": 0.5337112171837709,
-        "recall": 0.4108865411116215,
-        "support": 4354.0
+        "f1-score": 0.5675116962245675,
+        "precision": 0.5627967198964178,
+        "recall": 0.5723063418915953,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.7490950657268052,
-        "precision": 0.6779310344827586,
-        "recall": 0.8369518944231588,
-        "support": 2349.0
+        "f1-score": 0.7021748779405237,
+        "precision": 0.7071971390254805,
+        "recall": 0.6972234464521816,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9980494407069782,
-        "precision": 0.9961064759634486,
-        "recall": 1.0,
-        "support": 12536.0
+        "f1-score": 0.9884124087591241,
+        "precision": 0.9985252096967463,
+        "recall": 0.9785023936410442,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.8811786372007365,
-        "precision": 0.868321718931475,
-        "recall": 0.8944220128607746,
-        "support": 13374.0
+        "f1-score": 0.8946486578902376,
+        "precision": 0.8892665352457345,
+        "recall": 0.900096325856612,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.8663109802839358,
-      "eval_loss": 0.32735684514045715,
+      "eval_accuracy": 0.8666091085689618,
+      "eval_loss": 0.35281190276145935,
       "eval_macro avg": {
-        "f1-score": 0.7731591663913707,
-        "precision": 0.7690176116403633,
-        "recall": 0.7855651120988888,
-        "support": 32613.0
-      },
-      "eval_runtime": 1.4205,
-      "eval_samples_per_second": 57.022,
-      "eval_steps_per_second": 7.744,
+        "f1-score": 0.7881869102036132,
+        "precision": 0.7894464009660948,
+        "recall": 0.7870321269603583,
+        "support": 32431.0
+      },
+      "eval_runtime": 1.4161,
+      "eval_samples_per_second": 56.493,
+      "eval_steps_per_second": 7.062,
       "eval_weighted avg": {
-        "f1-score": 0.8609350954068932,
-        "precision": 0.8590551035257559,
-        "recall": 0.8663109802839358,
-        "support": 32613.0
+        "f1-score": 0.8672234272421873,
+        "precision": 0.8679524954775053,
+        "recall": 0.8666091085689618,
+        "support": 32431.0
       },
       "step": 41
     },
     {
       "epoch": 2.0,
       "eval_Claim": {
-        "f1-score": 0.5737245829782053,
-        "precision": 0.6105208603265094,
-        "recall": 0.5411116214974736,
-        "support": 4354.0
+        "f1-score": 0.5799535003874967,
+        "precision": 0.7048665620094191,
+        "recall": 0.4926486723721747,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8465994421797898,
-        "precision": 0.8533737024221453,
-        "recall": 0.8399318859088974,
-        "support": 2349.0
+        "f1-score": 0.8542402826855124,
+        "precision": 0.8561310314298363,
+        "recall": 0.8523578669017188,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9996808425756004,
-        "precision": 0.9999201915403033,
-        "recall": 0.9994416081684748,
-        "support": 12536.0
+        "f1-score": 0.9957381211461733,
+        "precision": 0.9996358670914884,
+        "recall": 0.9918706530575377,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.8958440225756799,
-        "precision": 0.8785221391604371,
-        "recall": 0.9138627187079408,
-        "support": 13374.0
+        "f1-score": 0.9081739586062353,
+        "precision": 0.8665166854143233,
+        "recall": 0.954038805559378,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.8916689663631068,
-      "eval_loss": 0.24866953492164612,
+      "eval_accuracy": 0.8950078628472757,
+      "eval_loss": 0.266658216714859,
       "eval_macro avg": {
-        "f1-score": 0.8289622225773189,
-        "precision": 0.8355842233623487,
-        "recall": 0.8235869585706966,
-        "support": 32613.0
-      },
-      "eval_runtime": 1.4275,
-      "eval_samples_per_second": 56.744,
-      "eval_steps_per_second": 7.706,
+        "f1-score": 0.8345264657063545,
+        "precision": 0.8567875364862668,
+        "recall": 0.8227289994727023,
+        "support": 32431.0
+      },
+      "eval_runtime": 1.4081,
+      "eval_samples_per_second": 56.815,
+      "eval_steps_per_second": 7.102,
       "eval_weighted avg": {
-        "f1-score": 0.8892060198210008,
-        "precision": 0.8875950468565349,
-        "recall": 0.8916689663631068,
-        "support": 32613.0
+        "f1-score": 0.8881729319562012,
+        "precision": 0.8885190226564973,
+        "recall": 0.8950078628472757,
+        "support": 32431.0
       },
       "step": 82
     },
     {
       "epoch": 3.0,
       "eval_Claim": {
-        "f1-score": 0.6284023668639053,
-        "precision": 0.648193359375,
-        "recall": 0.6097841065686724,
-        "support": 4354.0
+        "f1-score": 0.6576058546785155,
+        "precision": 0.6279952076677316,
+        "recall": 0.6901470265525566,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8758085381630012,
-        "precision": 0.8872870249017037,
-        "recall": 0.8646232439335888,
-        "support": 2349.0
+        "f1-score": 0.836495761001211,
+        "precision": 0.7716945996275605,
+        "recall": 0.9131776112825033,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9998803589232302,
-        "precision": 0.9997607464710104,
-        "recall": 1.0,
-        "support": 12536.0
+        "f1-score": 0.997964077274578,
+        "precision": 0.9997280638143582,
+        "recall": 0.9962063047601842,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.9024128884454791,
-        "precision": 0.8920300971583023,
-        "recall": 0.9130402273067145,
-        "support": 13374.0
+        "f1-score": 0.9064447592067989,
+        "precision": 0.9338246023639282,
+        "recall": 0.8806247419843126,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.9024928709410358,
-      "eval_loss": 0.2409481257200241,
+      "eval_accuracy": 0.8955937220560575,
+      "eval_loss": 0.2511727809906006,
       "eval_macro avg": {
-        "f1-score": 0.8516260380989039,
-        "precision": 0.8568178069765041,
-        "recall": 0.8468618944522439,
-        "support": 32613.0
-      },
-      "eval_runtime": 1.419,
-      "eval_samples_per_second": 57.082,
-      "eval_steps_per_second": 7.752,
+        "f1-score": 0.8496276130402758,
+        "precision": 0.8333106183683946,
+        "recall": 0.8700389211448892,
+        "support": 32431.0
+      },
+      "eval_runtime": 1.4062,
+      "eval_samples_per_second": 56.892,
+      "eval_steps_per_second": 7.112,
       "eval_weighted avg": {
-        "f1-score": 0.9013800727011249,
-        "precision": 0.900545253284536,
-        "recall": 0.9024928709410358,
-        "support": 32613.0
+        "f1-score": 0.8978276091178259,
+        "precision": 0.9020056542549684,
+        "recall": 0.8955937220560575,
+        "support": 32431.0
       },
       "step": 123
     },
     {
       "epoch": 4.0,
       "eval_Claim": {
-        "f1-score": 0.6302973542495903,
-        "precision": 0.6427889207258835,
-        "recall": 0.6182820395039045,
-        "support": 4354.0
+        "f1-score": 0.6724791508718727,
+        "precision": 0.5918265221017515,
+        "recall": 0.7785824007022164,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8868001634654679,
-        "precision": 0.8526522593320236,
-        "recall": 0.9237973605789698,
-        "support": 2349.0
+        "f1-score": 0.9030544488711819,
+        "precision": 0.9070698088039129,
+        "recall": 0.8990744821507272,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.999680893498205,
-        "precision": 0.9997606510292005,
-        "recall": 0.9996011486917677,
-        "support": 12536.0
+        "f1-score": 0.9991864039052614,
+        "precision": 1.0,
+        "recall": 0.9983741306115076,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.8995508982035928,
-        "precision": 0.9004945301963135,
-        "recall": 0.8986092418124719,
-        "support": 13374.0
+        "f1-score": 0.8886077779384126,
+        "precision": 0.9359677173747526,
+        "recall": 0.8458098252373745,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.9018182933186153,
-      "eval_loss": 0.23009340465068817,
+      "eval_accuracy": 0.8921710708889643,
+      "eval_loss": 0.24931207299232483,
       "eval_macro avg": {
-        "f1-score": 0.854082327354214,
-        "precision": 0.8489240903208553,
-        "recall": 0.8600724476467785,
-        "support": 32613.0
-      },
-      "eval_runtime": 1.4216,
-      "eval_samples_per_second": 56.979,
-      "eval_steps_per_second": 7.738,
+        "f1-score": 0.8658319453966822,
+        "precision": 0.8587160120701043,
+        "recall": 0.8804602096754564,
+        "support": 32431.0
+      },
+      "eval_runtime": 1.4073,
+      "eval_samples_per_second": 56.847,
+      "eval_steps_per_second": 7.106,
       "eval_weighted avg": {
-        "f1-score": 0.9011744291494633,
-        "precision": 0.9008001866175751,
-        "recall": 0.9018182933186153,
-        "support": 32613.0
+        "f1-score": 0.8969978155839744,
+        "precision": 0.907448110194518,
+        "recall": 0.8921710708889643,
+        "support": 32431.0
       },
       "step": 164
     },
     {
       "epoch": 5.0,
       "eval_Claim": {
-        "f1-score": 0.6072607260726073,
-        "precision": 0.6787741203178207,
-        "recall": 0.5493798805695912,
-        "support": 4354.0
+        "f1-score": 0.6831918505942275,
+        "precision": 0.7054698457223001,
+        "recall": 0.6622778143515471,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.875,
-        "precision": 0.9056947608200455,
-        "recall": 0.8463175819497658,
-        "support": 2349.0
+        "f1-score": 0.9057110862262038,
+        "precision": 0.9207650273224044,
+        "recall": 0.8911414720141031,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9998404722022812,
-        "precision": 0.9997607273887382,
-        "recall": 0.9999202297383536,
-        "support": 12536.0
+        "f1-score": 0.9989149109322725,
+        "precision": 1.0,
+        "recall": 0.9978321741486768,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.9058781103498017,
-        "precision": 0.8748955140707718,
-        "recall": 0.9391356363092568,
-        "support": 13374.0
+        "f1-score": 0.9169270479554409,
+        "precision": 0.9053655264922871,
+        "recall": 0.9287876702903537,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.903780700947475,
-      "eval_loss": 0.263094037771225,
+      "eval_accuracy": 0.912275292158737,
+      "eval_loss": 0.22767245769500732,
       "eval_macro avg": {
-        "f1-score": 0.8469948271561726,
-        "precision": 0.864781280649344,
-        "recall": 0.8336883321417418,
-        "support": 32613.0
-      },
-      "eval_runtime": 1.4217,
-      "eval_samples_per_second": 56.974,
-      "eval_steps_per_second": 7.737,
+        "f1-score": 0.8761862239270362,
+        "precision": 0.8829000998842478,
+        "recall": 0.8700097827011701,
+        "support": 32431.0
+      },
+      "eval_runtime": 1.4166,
+      "eval_samples_per_second": 56.472,
+      "eval_steps_per_second": 7.059,
       "eval_weighted avg": {
-        "f1-score": 0.899905013603967,
-        "precision": 0.8989271945775551,
-        "recall": 0.903780700947475,
-        "support": 32613.0
+        "f1-score": 0.9112876078974042,
+        "precision": 0.9106603094566913,
+        "recall": 0.912275292158737,
+        "support": 32431.0
       },
       "step": 205
     },
     {
       "epoch": 6.0,
       "eval_Claim": {
-        "f1-score": 0.57247396566282,
-        "precision": 0.7390988372093024,
-        "recall": 0.467156637574644,
-        "support": 4354.0
+        "f1-score": 0.7047057585008949,
+        "precision": 0.6772561715904493,
+        "recall": 0.7344744349352644,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8933023452355799,
-        "precision": 0.886744966442953,
-        "recall": 0.8999574286930608,
-        "support": 2349.0
+        "f1-score": 0.9053571428571427,
+        "precision": 0.9172320217096337,
+        "recall": 0.8937858087263112,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9998005982053838,
-        "precision": 0.9996809952946806,
-        "recall": 0.9999202297383536,
-        "support": 12536.0
+        "f1-score": 0.9990960043391791,
+        "precision": 0.9999095268252963,
+        "recall": 0.998283804534369,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.9065413958745407,
-        "precision": 0.8590842147543178,
-        "recall": 0.9595483774487812,
-        "support": 13374.0
+        "f1-score": 0.9127577454014394,
+        "precision": 0.9226713532513181,
+        "recall": 0.9030549057382689,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.9050378683347131,
-      "eval_loss": 0.28428906202316284,
+      "eval_accuracy": 0.9112269125219697,
+      "eval_loss": 0.2558422088623047,
       "eval_macro avg": {
-        "f1-score": 0.8430295762445812,
-        "precision": 0.8711522534253133,
-        "recall": 0.8316456683637099,
-        "support": 32613.0
-      },
-      "eval_runtime": 1.4177,
-      "eval_samples_per_second": 57.135,
-      "eval_steps_per_second": 7.759,
+        "f1-score": 0.880479162774664,
+        "precision": 0.8792672683441743,
+        "recall": 0.8823997384835534,
+        "support": 32431.0
+      },
+      "eval_runtime": 1.4147,
+      "eval_samples_per_second": 56.547,
+      "eval_steps_per_second": 7.068,
       "eval_weighted avg": {
-        "f1-score": 0.896835733694634,
-        "precision": 0.8991013862116997,
-        "recall": 0.9050378683347131,
-        "support": 32613.0
+        "f1-score": 0.9124791845559805,
+        "precision": 0.9141734652287734,
+        "recall": 0.9112269125219697,
+        "support": 32431.0
       },
       "step": 246
     },
     {
       "epoch": 7.0,
       "eval_Claim": {
-        "f1-score": 0.6294631710362047,
-        "precision": 0.6895514223194749,
-        "recall": 0.5790078089113458,
-        "support": 4354.0
+        "f1-score": 0.671239246686817,
+        "precision": 0.7137206427688504,
+        "recall": 0.6335308316875137,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8833069620253163,
-        "precision": 0.8248984115256742,
-        "recall": 0.9506172839506173,
-        "support": 2349.0
+        "f1-score": 0.9053934571175951,
+        "precision": 0.9082039911308204,
+        "recall": 0.9026002644336713,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9995215311004785,
-        "precision": 0.9992028061224489,
-        "recall": 0.9998404594767071,
-        "support": 12536.0
+        "f1-score": 0.999186477447347,
+        "precision": 0.9999095431931253,
+        "recall": 0.998464456688646,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.9074593796159526,
-        "precision": 0.8964686998394864,
-        "recall": 0.9187228951697323,
-        "support": 13374.0
+        "f1-score": 0.9147585275244849,
+        "precision": 0.8983152029716105,
+        "recall": 0.9318150543553048,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.9068469628675682,
-      "eval_loss": 0.2983298599720001,
+      "eval_accuracy": 0.910610218617989,
+      "eval_loss": 0.2902248203754425,
       "eval_macro avg": {
-        "f1-score": 0.854937760944488,
-        "precision": 0.8525303349517711,
-        "recall": 0.8620471118771007,
-        "support": 32613.0
-      },
-      "eval_runtime": 1.4145,
-      "eval_samples_per_second": 57.265,
-      "eval_steps_per_second": 7.777,
+        "f1-score": 0.872644427194061,
+        "precision": 0.8800373450161016,
+        "recall": 0.8666026517912839,
+        "support": 32431.0
+      },
+      "eval_runtime": 1.4074,
+      "eval_samples_per_second": 56.842,
+      "eval_steps_per_second": 7.105,
       "eval_weighted avg": {
-        "f1-score": 0.9039933265062536,
-        "precision": 0.9031788559978264,
-        "recall": 0.9068469628675682,
-        "support": 32613.0
+        "f1-score": 0.9087067599584377,
+        "precision": 0.9077503480513693,
+        "recall": 0.910610218617989,
+        "support": 32431.0
       },
       "step": 287
     },
     {
       "epoch": 8.0,
       "eval_Claim": {
-        "f1-score": 0.6635588402501421,
-        "precision": 0.6570592208961945,
-        "recall": 0.6701883325677538,
-        "support": 4354.0
+        "f1-score": 0.6637626559924652,
+        "precision": 0.7160274320548641,
+        "recall": 0.618608733816107,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9037098791162985,
-        "precision": 0.8852592895059208,
-        "recall": 0.9229459344401874,
-        "support": 2349.0
+        "f1-score": 0.9002159827213824,
+        "precision": 0.8826768318509106,
+        "recall": 0.9184662847069194,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9997208374875374,
-        "precision": 0.9996012441183507,
-        "recall": 0.9998404594767071,
-        "support": 12536.0
+        "f1-score": 0.999186477447347,
+        "precision": 0.9999095431931253,
+        "recall": 0.998464456688646,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.9033812787107464,
-        "precision": 0.9098907766990292,
-        "recall": 0.8969642590100194,
-        "support": 13374.0
+        "f1-score": 0.914696744563015,
+        "precision": 0.8981960472211169,
+        "recall": 0.9318150543553048,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.9081041302548064,
-      "eval_loss": 0.3042994737625122,
+      "eval_accuracy": 0.9096235083716198,
+      "eval_loss": 0.3177148699760437,
       "eval_macro avg": {
-        "f1-score": 0.8675927088911811,
-        "precision": 0.8629526328048738,
-        "recall": 0.8724847463736669,
-        "support": 32613.0
-      },
-      "eval_runtime": 1.4198,
-      "eval_samples_per_second": 57.05,
-      "eval_steps_per_second": 7.747,
+        "f1-score": 0.8694654651810524,
+        "precision": 0.8742024635800042,
+        "recall": 0.8668386323917443,
+        "support": 32431.0
+      },
+      "eval_runtime": 1.4095,
+      "eval_samples_per_second": 56.758,
+      "eval_steps_per_second": 7.095,
       "eval_weighted avg": {
-        "f1-score": 0.9084190763411705,
-        "precision": 0.9088458701337473,
-        "recall": 0.9081041302548064,
-        "support": 32613.0
+        "f1-score": 0.9072662719450809,
+        "precision": 0.906235103522757,
+        "recall": 0.9096235083716198,
+        "support": 32431.0
       },
       "step": 328
     },
     {
       "epoch": 9.0,
       "eval_Claim": {
-        "f1-score": 0.6706725213773977,
-        "precision": 0.6748837209302325,
-        "recall": 0.6665135507579237,
-        "support": 4354.0
+        "f1-score": 0.7096774193548386,
+        "precision": 0.6808467741935483,
+        "recall": 0.7410577134079438,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9129511677282377,
-        "precision": 0.9106310885218127,
-        "recall": 0.9152830991911451,
-        "support": 2349.0
+        "f1-score": 0.8967198581560283,
+        "precision": 0.9019170753455193,
+        "recall": 0.8915821947994711,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9996809444045626,
-        "precision": 0.9996012123145638,
-        "recall": 0.9997606892150607,
-        "support": 12536.0
+        "f1-score": 0.9995030045633218,
+        "precision": 0.999909600433918,
+        "recall": 0.9990967392286153,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.9052560848140958,
-        "precision": 0.9039063664827792,
-        "recall": 0.9066098399880365,
-        "support": 13374.0
+        "f1-score": 0.9160975609756098,
+        "precision": 0.927996611605252,
+        "recall": 0.9044997935874501,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.910986416459694,
-      "eval_loss": 0.3104316294193268,
+      "eval_accuracy": 0.9129228207579168,
+      "eval_loss": 0.31878241896629333,
       "eval_macro avg": {
-        "f1-score": 0.8721401795810735,
-        "precision": 0.8722555970623471,
-        "recall": 0.8720417947880416,
-        "support": 32613.0
-      },
-      "eval_runtime": 1.4235,
-      "eval_samples_per_second": 56.901,
-      "eval_steps_per_second": 7.727,
+        "f1-score": 0.8804994607624497,
+        "precision": 0.8776675153945594,
+        "recall": 0.8840591102558701,
+        "support": 32431.0
+      },
+      "eval_runtime": 1.4158,
+      "eval_samples_per_second": 56.504,
+      "eval_steps_per_second": 7.063,
       "eval_weighted avg": {
-        "f1-score": 0.9107878958829343,
-        "precision": 0.9105988621342419,
-        "recall": 0.910986416459694,
-        "support": 32613.0
+        "f1-score": 0.9142091539852634,
+        "precision": 0.9159930478071482,
+        "recall": 0.9129228207579168,
+        "support": 32431.0
       },
       "step": 369
     },
     {
       "epoch": 10.0,
       "eval_Claim": {
-        "f1-score": 0.6484411354118194,
-        "precision": 0.657001414427157,
-        "recall": 0.6401010564997703,
-        "support": 4354.0
+        "f1-score": 0.6989775051124745,
+        "precision": 0.6544131725062224,
+        "recall": 0.750054860653939,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8997963340122199,
-        "precision": 0.8625536899648575,
-        "recall": 0.9404001702852277,
-        "support": 2349.0
+        "f1-score": 0.8996763754045308,
+        "precision": 0.8812341504649197,
+        "recall": 0.9189070074922874,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9997607083034219,
-        "precision": 0.999680969851651,
-        "recall": 0.9998404594767071,
-        "support": 12536.0
+        "f1-score": 0.9993221564462921,
+        "precision": 0.9999095677337674,
+        "recall": 0.9987354349200614,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.9021992043833972,
-        "precision": 0.905666063893912,
-        "recall": 0.898758785703604,
-        "support": 13374.0
+        "f1-score": 0.9087506179814958,
+        "precision": 0.933473592571097,
+        "recall": 0.885303426448328,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.9060803973875449,
-      "eval_loss": 0.3316749632358551,
+      "eval_accuracy": 0.90737257562209,
+      "eval_loss": 0.3552992045879364,
       "eval_macro avg": {
-        "f1-score": 0.8625493455277146,
-        "precision": 0.8562255345343944,
-        "recall": 0.8697751179913272,
-        "support": 32613.0
-      },
-      "eval_runtime": 1.4233,
-      "eval_samples_per_second": 56.909,
-      "eval_steps_per_second": 7.728,
+        "f1-score": 0.8766816637361983,
+        "precision": 0.8672576208190016,
+        "recall": 0.888250182378654,
+        "support": 32431.0
+      },
+      "eval_runtime": 1.4067,
+      "eval_samples_per_second": 56.872,
+      "eval_steps_per_second": 7.109,
       "eval_weighted avg": {
-        "f1-score": 0.9056494861218845,
-        "precision": 0.905500915362609,
-        "recall": 0.9060803973875449,
-        "support": 32613.0
+        "f1-score": 0.9095582394113776,
+        "precision": 0.9132862117518615,
+        "recall": 0.90737257562209,
+        "support": 32431.0
       },
       "step": 410
     },
     {
       "epoch": 11.0,
       "eval_Claim": {
-        "f1-score": 0.6478906714200832,
-        "precision": 0.6712632356562424,
-        "recall": 0.6260909508497933,
-        "support": 4354.0
+        "f1-score": 0.7129222520107238,
+        "precision": 0.6971476510067114,
+        "recall": 0.7294272547728768,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8952419848887073,
-        "precision": 0.8602825745682888,
-        "recall": 0.9331630481055768,
-        "support": 2349.0
+        "f1-score": 0.8950740004417939,
+        "precision": 0.8972542072630647,
+        "recall": 0.8929043631555752,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9997607083034219,
-        "precision": 0.999680969851651,
-        "recall": 0.9998404594767071,
-        "support": 12536.0
+        "f1-score": 0.9984622342831299,
+        "precision": 0.999909412084428,
+        "recall": 0.9970192394544305,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.9058867362146051,
-        "precision": 0.9027922174365067,
-        "recall": 0.9090025422461493,
-        "support": 13374.0
+        "f1-score": 0.9182698961937715,
+        "precision": 0.9236391479883057,
+        "recall": 0.9129627081326545,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.9078894919203998,
-      "eval_loss": 0.3783251941204071,
+      "eval_accuracy": 0.9144645555178688,
+      "eval_loss": 0.3296458423137665,
       "eval_macro avg": {
-        "f1-score": 0.8621950252067043,
-        "precision": 0.8585047493781722,
-        "recall": 0.8670242501695566,
-        "support": 32613.0
-      },
-      "eval_runtime": 1.4193,
-      "eval_samples_per_second": 57.069,
-      "eval_steps_per_second": 7.75,
+        "f1-score": 0.8811820957323547,
+        "precision": 0.8794876045856275,
+        "recall": 0.8830783913788842,
+        "support": 32431.0
+      },
+      "eval_runtime": 1.4103,
+      "eval_samples_per_second": 56.725,
+      "eval_steps_per_second": 7.091,
       "eval_weighted avg": {
-        "f1-score": 0.9067601525554976,
-        "precision": 0.9060628476302188,
-        "recall": 0.9078894919203998,
-        "support": 32613.0
+        "f1-score": 0.9151681932855632,
+        "precision": 0.9160044438952304,
+        "recall": 0.9144645555178688,
+        "support": 32431.0
       },
       "step": 451
     },
     {
       "epoch": 12.0,
       "eval_Claim": {
-        "f1-score": 0.6313421644588852,
-        "precision": 0.6924342105263158,
-        "recall": 0.5801561782269178,
-        "support": 4354.0
+        "f1-score": 0.7077807250221043,
+        "precision": 0.7129815185927411,
+        "recall": 0.7026552556506473,
+        "support": 4557.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.9006842214389384,
-        "precision": 0.8779304769603881,
-        "recall": 0.9246487867177522,
-        "support": 2349.0
+        "f1-score": 0.8944790739091719,
+        "precision": 0.9037336932073774,
+        "recall": 0.8854120758043191,
+        "support": 2269.0
       },
       "eval_O": {
-        "f1-score": 0.9998005982053838,
-        "precision": 0.9996809952946806,
-        "recall": 0.9999202297383536,
-        "support": 12536.0
+        "f1-score": 0.9991415533366468,
+        "precision": 0.9995480021695896,
+        "recall": 0.9987354349200614,
+        "support": 11071.0
       },
       "eval_Premise": {
-        "f1-score": 0.9077069457659372,
-        "precision": 0.8889048165137615,
-        "recall": 0.927321668909825,
-        "support": 13374.0
+        "f1-score": 0.9209633766144781,
+        "precision": 0.9171613783691572,
+        "recall": 0.9247970276592817,
+        "support": 14534.0
       },
-      "eval_accuracy": 0.908686720019624,
-      "eval_loss": 0.4068906903266907,
+      "eval_accuracy": 0.9160679596682186,
+      "eval_loss": 0.3575716018676758,
       "eval_macro avg": {
-        "f1-score": 0.8598834824672862,
-        "precision": 0.8647376248237866,
-        "recall": 0.8580117158982121,
-        "support": 32613.0
-      },
-      "eval_runtime": 1.418,
-      "eval_samples_per_second": 57.122,
-      "eval_steps_per_second": 7.757,
+        "f1-score": 0.8805911822206004,
+        "precision": 0.8833561480847163,
+        "recall": 0.8778999485085774,
+        "support": 32431.0
+      },
+      "eval_runtime": 1.4095,
+      "eval_samples_per_second": 56.758,
+      "eval_steps_per_second": 7.095,
       "eval_weighted avg": {
-        "f1-score": 0.905704596694275,
-        "precision": 0.9044654345224509,
-        "recall": 0.908686720019624,
-        "support": 32613.0
+        "f1-score": 0.9158431018263538,
+        "precision": 0.9156562528245048,
+        "recall": 0.9160679596682186,
+        "support": 32431.0
       },
       "step": 492
     }
@@ -554,7 +554,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 16,
   "save_steps": 500,
-  "total_flos": 1720106206408800.0,
+  "total_flos": 1725464792721600.0,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null

checkpoint-492/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1fa0ab387f66efa2185b605999d4a743ce5870243a43ebfb649f6fb9a993302b
+oid sha256:a875ae3b17c481d24bf195a549c4a53e60bc990aca9336c453f8458ad25ff8f3
 size 4664