Training in progress, epoch 13, checkpoint

checkpoint-1053/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df26ad71e6b9c578551fbade9cd231a514e617764710b9474de3c64dd997ae2f
+oid sha256:61ac4bd406da0e6701da1c3e5a63687ac74e6c6452c4399caa4763fe57957f1b
 size 592330980

checkpoint-1053/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0757064f821b5ebebf263ce2fc7cfa9511bd0f50b5dde8b3ee62cc9260c22c12
+oid sha256:e769f065859c4de2bde0aeeaaca7731b556384df8690576047dbe44e2395c177
 size 1014670074

checkpoint-1053/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:785282e90233e50783d511149f52a487e725aef0302182724c574e3af2083098
+oid sha256:1a92327c1c7423b14915d094fccec26dc0b7df7db995e9e1e5592edda9350390
 size 1064

checkpoint-1053/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": null,
-  "best_model_checkpoint": null,
+  "best_metric": 0.276384562253952,
+  "best_model_checkpoint": "longformer-sep_tok_full_labels/checkpoint-324",
   "epoch": 13.0,
   "eval_steps": 500,
   "global_step": 1053,
@@ -11,841 +11,841 @@
     {
       "epoch": 1.0,
       "eval_B-Claim": {
-        "f1-score": 0.37160751565762,
-        "precision": 0.42788461538461536,
-        "recall": 0.3284132841328413,
-        "support": 271.0
+        "f1-score": 0.21800947867298578,
+        "precision": 0.3333333333333333,
+        "recall": 0.1619718309859155,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.3578947368421052,
-        "precision": 0.6666666666666666,
-        "recall": 0.2446043165467626,
-        "support": 139.0
+        "f1-score": 0.0,
+        "precision": 0.0,
+        "recall": 0.0,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8640915593705293,
-        "precision": 0.7895424836601307,
-        "recall": 0.9541864139020537,
-        "support": 633.0
+        "f1-score": 0.8305489260143198,
+        "precision": 0.71900826446281,
+        "recall": 0.9830508474576272,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.5003402749421533,
-        "precision": 0.5493126120741183,
-        "recall": 0.4593851537115721,
-        "support": 4001.0
+        "f1-score": 0.5590492696211935,
+        "precision": 0.5643589102724319,
+        "recall": 0.5538386068187393,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.7718093699515347,
-        "precision": 0.6502211636611093,
-        "recall": 0.9493293591654247,
-        "support": 2013.0
+        "f1-score": 0.7399038461538461,
+        "precision": 0.7205056179775281,
+        "recall": 0.7603754940711462,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.875016720916752,
-        "precision": 0.8846812731043188,
-        "recall": 0.865561044460127,
-        "support": 11336.0
+        "f1-score": 0.8867229093291834,
+        "precision": 0.8883965205974068,
+        "recall": 0.8850555918901243,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.9992483530087988,
-        "precision": 0.9995577178239717,
-        "recall": 0.998939179632249,
-        "support": 11312.0
+        "f1-score": 0.9976548035383666,
+        "precision": 0.9961383616794018,
+        "recall": 0.9991758694577221,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.8614038040733883,
-      "eval_loss": 0.31713685393333435,
+      "eval_accuracy": 0.8699050632911393,
+      "eval_loss": 0.32671716809272766,
       "eval_macro avg": {
-        "f1-score": 0.6771440758127848,
-        "precision": 0.7096952189107044,
-        "recall": 0.685774107364433,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8338,
-      "eval_samples_per_second": 16.55,
-      "eval_steps_per_second": 2.069,
+        "f1-score": 0.6045556047614136,
+        "precision": 0.6031058583318446,
+        "recall": 0.6204954629544678,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.234,
+      "eval_samples_per_second": 12.833,
+      "eval_steps_per_second": 1.604,
       "eval_weighted avg": {
-        "f1-score": 0.8576207231627551,
-        "precision": 0.8601529227027923,
-        "recall": 0.8614038040733883,
-        "support": 29705.0
+        "f1-score": 0.8664142595402331,
+        "precision": 0.8644597558999653,
+        "recall": 0.8699050632911393,
+        "support": 31600.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_B-Claim": {
-        "f1-score": 0.4708624708624709,
-        "precision": 0.6392405063291139,
-        "recall": 0.3726937269372694,
-        "support": 271.0
+        "f1-score": 0.5352697095435685,
+        "precision": 0.6515151515151515,
+        "recall": 0.45422535211267606,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.796875,
-        "precision": 0.8717948717948718,
-        "recall": 0.7338129496402878,
-        "support": 139.0
+        "f1-score": 0.7766990291262136,
+        "precision": 0.7142857142857143,
+        "recall": 0.851063829787234,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8736616702355461,
-        "precision": 0.796875,
-        "recall": 0.966824644549763,
-        "support": 633.0
+        "f1-score": 0.8802721088435375,
+        "precision": 0.8490813648293963,
+        "recall": 0.9138418079096046,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.5100589925881107,
-        "precision": 0.6459770114942529,
-        "recall": 0.4213946513371657,
-        "support": 4001.0
+        "f1-score": 0.5747557221255521,
+        "precision": 0.6325869180907484,
+        "recall": 0.5266127054206524,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8401387776888176,
-        "precision": 0.9077277970011534,
-        "recall": 0.7819175360158966,
-        "support": 2013.0
+        "f1-score": 0.7833260963960051,
+        "precision": 0.6986831913245546,
+        "recall": 0.8913043478260869,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8912891699864469,
-        "precision": 0.8338584492430646,
-        "recall": 0.9572159491884262,
-        "support": 11336.0
+        "f1-score": 0.8995515695067264,
+        "precision": 0.8971377459749553,
+        "recall": 0.9019784172661871,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.9996904982977407,
-        "precision": 1.0,
-        "recall": 0.9993811881188119,
-        "support": 11312.0
+        "f1-score": 0.9971231300345225,
+        "precision": 0.9945072962780783,
+        "recall": 0.9997527608373167,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.8830499915839084,
-      "eval_loss": 0.2966194748878479,
+      "eval_accuracy": 0.8864240506329114,
+      "eval_loss": 0.28716403245925903,
       "eval_macro avg": {
-        "f1-score": 0.7689395113798762,
-        "precision": 0.8136390908374939,
-        "recall": 0.7476058065410885,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8625,
-      "eval_samples_per_second": 16.452,
-      "eval_steps_per_second": 2.057,
+        "f1-score": 0.7781424807965893,
+        "precision": 0.7768281974712284,
+        "recall": 0.791254174451394,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.2816,
+      "eval_samples_per_second": 12.736,
+      "eval_steps_per_second": 1.592,
       "eval_weighted avg": {
-        "f1-score": 0.8731020208182413,
-        "precision": 0.874440834821272,
-        "recall": 0.8830499915839084,
-        "support": 29705.0
+        "f1-score": 0.8834146129726252,
+        "precision": 0.8835831101628018,
+        "recall": 0.8864240506329114,
+        "support": 31600.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_B-Claim": {
-        "f1-score": 0.6085192697768763,
-        "precision": 0.6756756756756757,
-        "recall": 0.5535055350553506,
-        "support": 271.0
+        "f1-score": 0.5580448065173116,
+        "precision": 0.6618357487922706,
+        "recall": 0.4823943661971831,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.8571428571428571,
-        "precision": 0.851063829787234,
-        "recall": 0.8633093525179856,
-        "support": 139.0
+        "f1-score": 0.8046875,
+        "precision": 0.8956521739130435,
+        "recall": 0.7304964539007093,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8834729626808834,
-        "precision": 0.8529411764705882,
-        "recall": 0.9162717219589257,
-        "support": 633.0
+        "f1-score": 0.8868421052631579,
+        "precision": 0.8300492610837439,
+        "recall": 0.9519774011299436,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.5764474423833614,
-        "precision": 0.6584269662921348,
-        "recall": 0.5126218445388653,
-        "support": 4001.0
+        "f1-score": 0.5488426262920367,
+        "precision": 0.6751432664756447,
+        "recall": 0.4623497669855286,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8581151832460733,
-        "precision": 0.9070282235749861,
-        "recall": 0.8142076502732241,
-        "support": 2013.0
+        "f1-score": 0.8110936682365254,
+        "precision": 0.8620689655172413,
+        "recall": 0.7658102766798419,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8959744247675935,
-        "precision": 0.8563158317922328,
-        "recall": 0.939484827099506,
-        "support": 11336.0
+        "f1-score": 0.9039491427470623,
+        "precision": 0.854842235662756,
+        "recall": 0.9590418574231524,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.9996020340481981,
-        "precision": 1.0,
-        "recall": 0.9992043847241867,
-        "support": 11312.0
+        "f1-score": 0.9979824597521307,
+        "precision": 0.9972023368715544,
+        "recall": 0.9987638041865832,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.8918700555462044,
-      "eval_loss": 0.2552729547023773,
+      "eval_accuracy": 0.892373417721519,
+      "eval_loss": 0.28271517157554626,
       "eval_macro avg": {
-        "f1-score": 0.811324882006549,
-        "precision": 0.8287788147989789,
-        "recall": 0.7998007594525776,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8422,
-      "eval_samples_per_second": 16.522,
-      "eval_steps_per_second": 2.065,
+        "f1-score": 0.7873489012583177,
+        "precision": 0.825256284045179,
+        "recall": 0.7644048466432775,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.2071,
+      "eval_samples_per_second": 12.888,
+      "eval_steps_per_second": 1.611,
       "eval_weighted avg": {
-        "f1-score": 0.8867633844066056,
-        "precision": 0.886070631898416,
-        "recall": 0.8918700555462044,
-        "support": 29705.0
+        "f1-score": 0.88435885840807,
+        "precision": 0.8846770016417851,
+        "recall": 0.892373417721519,
+        "support": 31600.0
       },
       "step": 243
     },
     {
       "epoch": 4.0,
       "eval_B-Claim": {
-        "f1-score": 0.6722408026755852,
-        "precision": 0.6146788990825688,
-        "recall": 0.7416974169741697,
-        "support": 271.0
+        "f1-score": 0.659016393442623,
+        "precision": 0.6165644171779141,
+        "recall": 0.7077464788732394,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.8664259927797834,
-        "precision": 0.8695652173913043,
-        "recall": 0.8633093525179856,
-        "support": 139.0
+        "f1-score": 0.8675496688741722,
+        "precision": 0.8136645962732919,
+        "recall": 0.9290780141843972,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8687035507844755,
-        "precision": 0.9100346020761245,
-        "recall": 0.8309636650868878,
-        "support": 633.0
+        "f1-score": 0.8698224852071005,
+        "precision": 0.9130434782608695,
+        "recall": 0.8305084745762712,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6483151400094921,
-        "precision": 0.6171222046532641,
-        "recall": 0.6828292926768308,
-        "support": 4001.0
+        "f1-score": 0.6472172351885099,
+        "precision": 0.5964839710444674,
+        "recall": 0.70738287956831,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8696993060909791,
-        "precision": 0.9009584664536742,
-        "recall": 0.8405365126676602,
-        "support": 2013.0
+        "f1-score": 0.8549332083430982,
+        "precision": 0.8131966116807846,
+        "recall": 0.9011857707509882,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8906159274643798,
-        "precision": 0.9020175517958925,
-        "recall": 0.879498941425547,
-        "support": 11336.0
+        "f1-score": 0.8876102800153433,
+        "precision": 0.9271658801531475,
+        "recall": 0.8512916939175932,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.999557835160948,
-        "precision": 0.9999115357395613,
-        "recall": 0.9992043847241867,
-        "support": 11312.0
+        "f1-score": 0.9981065283609122,
+        "precision": 0.9970394736842105,
+        "recall": 0.9991758694577221,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.8935869382258879,
-      "eval_loss": 0.29465603828430176,
+      "eval_accuracy": 0.8912974683544304,
+      "eval_loss": 0.276384562253952,
       "eval_macro avg": {
-        "f1-score": 0.8307940792808061,
-        "precision": 0.8306126395989127,
-        "recall": 0.8340056522961811,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8494,
-      "eval_samples_per_second": 16.497,
-      "eval_steps_per_second": 2.062,
+        "f1-score": 0.8263222570616798,
+        "precision": 0.8110226326106693,
+        "recall": 0.8466241687612174,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.2529,
+      "eval_samples_per_second": 12.794,
+      "eval_steps_per_second": 1.599,
       "eval_weighted avg": {
-        "f1-score": 0.8954766464091573,
-        "precision": 0.8982496227307208,
-        "recall": 0.8935869382258879,
-        "support": 29705.0
+        "f1-score": 0.8943886873545748,
+        "precision": 0.9004180663566287,
+        "recall": 0.8912974683544304,
+        "support": 31600.0
       },
       "step": 324
     },
     {
       "epoch": 5.0,
       "eval_B-Claim": {
-        "f1-score": 0.6438095238095239,
-        "precision": 0.6653543307086615,
-        "recall": 0.6236162361623616,
-        "support": 271.0
+        "f1-score": 0.665474060822898,
+        "precision": 0.6763636363636364,
+        "recall": 0.6549295774647887,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.8785714285714286,
-        "precision": 0.8723404255319149,
-        "recall": 0.8848920863309353,
-        "support": 139.0
+        "f1-score": 0.8896551724137931,
+        "precision": 0.8657718120805369,
+        "recall": 0.9148936170212766,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8783151326053042,
-        "precision": 0.8674884437596302,
-        "recall": 0.8894154818325435,
-        "support": 633.0
+        "f1-score": 0.8843441466854725,
+        "precision": 0.8830985915492958,
+        "recall": 0.885593220338983,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6283729628640129,
-        "precision": 0.6748923959827834,
-        "recall": 0.5878530367408148,
-        "support": 4001.0
+        "f1-score": 0.6698067632850242,
+        "precision": 0.6597668332143707,
+        "recall": 0.6801569781702232,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.885450461692039,
-        "precision": 0.8896690070210632,
-        "recall": 0.8812717337307501,
-        "support": 2013.0
+        "f1-score": 0.8809815950920246,
+        "precision": 0.8751828376401756,
+        "recall": 0.8868577075098815,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8989315871101154,
-        "precision": 0.878494442573257,
-        "recall": 0.9203422724064926,
-        "support": 11336.0
+        "f1-score": 0.9024631353488756,
+        "precision": 0.9094305163539764,
+        "recall": 0.8956017004578156,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.9997347245556637,
-        "precision": 1.0,
-        "recall": 0.9994695898161244,
-        "support": 11312.0
+        "f1-score": 0.9980246913580246,
+        "precision": 0.9967121486108828,
+        "recall": 0.9993406955661777,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.8995118666891095,
-      "eval_loss": 0.3176642060279846,
+      "eval_accuracy": 0.9047784810126582,
+      "eval_loss": 0.2893199920654297,
       "eval_macro avg": {
-        "f1-score": 0.830455117315441,
-        "precision": 0.8354627207967587,
-        "recall": 0.8266943481457174,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8682,
-      "eval_samples_per_second": 16.433,
-      "eval_steps_per_second": 2.054,
+        "f1-score": 0.8415356521437304,
+        "precision": 0.838046625116125,
+        "recall": 0.8453390709327352,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.2466,
+      "eval_samples_per_second": 12.807,
+      "eval_steps_per_second": 1.601,
       "eval_weighted avg": {
-        "f1-score": 0.8971010593476484,
-        "precision": 0.8958911872123141,
-        "recall": 0.8995118666891095,
-        "support": 29705.0
+        "f1-score": 0.9051715590931133,
+        "precision": 0.9056611908459661,
+        "recall": 0.9047784810126582,
+        "support": 31600.0
       },
       "step": 405
     },
     {
       "epoch": 6.0,
       "eval_B-Claim": {
-        "f1-score": 0.6150712830957231,
-        "precision": 0.6863636363636364,
-        "recall": 0.5571955719557196,
-        "support": 271.0
+        "f1-score": 0.6263345195729537,
+        "precision": 0.6330935251798561,
+        "recall": 0.6197183098591549,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.8686131386861314,
-        "precision": 0.8814814814814815,
-        "recall": 0.8561151079136691,
-        "support": 139.0
+        "f1-score": 0.8372093023255814,
+        "precision": 0.9230769230769231,
+        "recall": 0.7659574468085106,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8804841149773072,
-        "precision": 0.8447024673439768,
-        "recall": 0.919431279620853,
-        "support": 633.0
+        "f1-score": 0.8784530386740332,
+        "precision": 0.8594594594594595,
+        "recall": 0.8983050847457628,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.5758052970651396,
-        "precision": 0.6739276139410187,
-        "recall": 0.5026243439140214,
-        "support": 4001.0
+        "f1-score": 0.6109127995920448,
+        "precision": 0.6360499070878683,
+        "recall": 0.5876870247731175,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8627552339105711,
-        "precision": 0.8992456896551724,
-        "recall": 0.829110779930452,
-        "support": 2013.0
+        "f1-score": 0.8409272581934453,
+        "precision": 0.9126662810873337,
+        "recall": 0.7796442687747036,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.894107779408681,
-        "precision": 0.8521864257734432,
-        "recall": 0.9403669724770642,
-        "support": 11336.0
+        "f1-score": 0.8964938902643559,
+        "precision": 0.8762685402029664,
+        "recall": 0.9176749509483323,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 1.0,
-        "precision": 1.0,
-        "recall": 1.0,
-        "support": 11312.0
+        "f1-score": 0.9987239122380934,
+        "precision": 0.9976971790443293,
+        "recall": 0.9997527608373167,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.8922403635751557,
-      "eval_loss": 0.41071853041648865,
+      "eval_accuracy": 0.8939873417721519,
+      "eval_loss": 0.33532533049583435,
       "eval_macro avg": {
-        "f1-score": 0.8138338353062219,
-        "precision": 0.8339867592226755,
-        "recall": 0.8006920079731114,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8571,
-      "eval_samples_per_second": 16.471,
-      "eval_steps_per_second": 2.059,
+        "f1-score": 0.8127221029800725,
+        "precision": 0.8340445450198194,
+        "recall": 0.7955342638209855,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.2278,
+      "eval_samples_per_second": 12.846,
+      "eval_steps_per_second": 1.606,
       "eval_weighted avg": {
-        "f1-score": 0.8864802913843919,
-        "precision": 0.8861194550557427,
-        "recall": 0.8922403635751557,
-        "support": 29705.0
+        "f1-score": 0.8922477132246446,
+        "precision": 0.891880888702747,
+        "recall": 0.8939873417721519,
+        "support": 31600.0
       },
       "step": 486
     },
     {
       "epoch": 6.17,
-      "grad_norm": 5.239528656005859,
-      "learning_rate": 1.7530864197530865e-05,
-      "loss": 0.2363,
+      "grad_norm": 4.557807445526123,
+      "learning_rate": 1.3827160493827162e-05,
+      "loss": 0.2608,
       "step": 500
     },
     {
       "epoch": 7.0,
       "eval_B-Claim": {
-        "f1-score": 0.6954954954954955,
-        "precision": 0.6795774647887324,
-        "recall": 0.7121771217712177,
-        "support": 271.0
+        "f1-score": 0.5553235908141962,
+        "precision": 0.6820512820512821,
+        "recall": 0.46830985915492956,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.8920863309352518,
-        "precision": 0.8920863309352518,
-        "recall": 0.8920863309352518,
-        "support": 139.0
+        "f1-score": 0.8412698412698413,
+        "precision": 0.954954954954955,
+        "recall": 0.75177304964539,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8876494023904382,
-        "precision": 0.8954983922829582,
-        "recall": 0.8799368088467614,
-        "support": 633.0
+        "f1-score": 0.8846905537459283,
+        "precision": 0.8210399032648126,
+        "recall": 0.9590395480225988,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6831545741324923,
-        "precision": 0.6898572884811417,
-        "recall": 0.6765808547863035,
-        "support": 4001.0
+        "f1-score": 0.5484584747012834,
+        "precision": 0.6880088823094005,
+        "recall": 0.45597252882021094,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8984707946853848,
-        "precision": 0.9068825910931174,
-        "recall": 0.8902136115250869,
-        "support": 2013.0
+        "f1-score": 0.8530857454942655,
+        "precision": 0.9536019536019537,
+        "recall": 0.7717391304347826,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.9068164859763859,
-        "precision": 0.9024198479951079,
-        "recall": 0.9112561750176429,
-        "support": 11336.0
+        "f1-score": 0.900171657448026,
+        "precision": 0.8438103411285132,
+        "recall": 0.9646010464355788,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.9999558011049724,
-        "precision": 0.9999116061168567,
-        "recall": 1.0,
-        "support": 11312.0
+        "f1-score": 0.9986819342614713,
+        "precision": 0.9982707509881423,
+        "recall": 0.9990934564034943,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.9094428547382596,
-      "eval_loss": 0.3520519435405731,
+      "eval_accuracy": 0.8943354430379746,
+      "eval_loss": 0.45137375593185425,
       "eval_macro avg": {
-        "f1-score": 0.8519469835314888,
-        "precision": 0.8523190745275951,
-        "recall": 0.8517501289831806,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8905,
-      "eval_samples_per_second": 16.358,
-      "eval_steps_per_second": 2.045,
+        "f1-score": 0.7973831139621446,
+        "precision": 0.8488197240427228,
+        "recall": 0.7672183741309979,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.1951,
+      "eval_samples_per_second": 12.913,
+      "eval_steps_per_second": 1.614,
       "eval_weighted avg": {
-        "f1-score": 0.9091888981291354,
-        "precision": 0.908989097041669,
-        "recall": 0.9094428547382596,
-        "support": 29705.0
+        "f1-score": 0.8858958517067363,
+        "precision": 0.8885840321741321,
+        "recall": 0.8943354430379746,
+        "support": 31600.0
       },
       "step": 567
     },
     {
       "epoch": 8.0,
       "eval_B-Claim": {
-        "f1-score": 0.6666666666666667,
-        "precision": 0.658273381294964,
-        "recall": 0.6752767527675276,
-        "support": 271.0
+        "f1-score": 0.6642599277978338,
+        "precision": 0.6814814814814815,
+        "recall": 0.647887323943662,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.8819444444444444,
-        "precision": 0.8523489932885906,
-        "recall": 0.9136690647482014,
-        "support": 139.0
+        "f1-score": 0.9015151515151516,
+        "precision": 0.967479674796748,
+        "recall": 0.8439716312056738,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8759007205764612,
-        "precision": 0.887987012987013,
-        "recall": 0.8641390205371248,
-        "support": 633.0
+        "f1-score": 0.8873531444367657,
+        "precision": 0.8687415426251691,
+        "recall": 0.9067796610169492,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.642274412855377,
-        "precision": 0.6353631694790902,
-        "recall": 0.6493376655836041,
-        "support": 4001.0
+        "f1-score": 0.6521516393442622,
+        "precision": 0.6823907799517556,
+        "recall": 0.6244787834191807,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.87683284457478,
-        "precision": 0.862914862914863,
-        "recall": 0.8912071535022354,
-        "support": 2013.0
+        "f1-score": 0.8962213943587014,
+        "precision": 0.9711649365628604,
+        "recall": 0.8320158102766798,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8911585772014743,
-        "precision": 0.8972547616918537,
-        "recall": 0.8851446718419196,
-        "support": 11336.0
+        "f1-score": 0.9046460618145563,
+        "precision": 0.8831269952503309,
+        "recall": 0.9272400261608895,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.9998673915926268,
-        "precision": 0.9999115904871364,
-        "recall": 0.9998231966053748,
-        "support": 11312.0
+        "f1-score": 0.9979418786531653,
+        "precision": 0.996875,
+        "recall": 0.9990110433492665,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.8952364921730348,
-      "eval_loss": 0.4631531238555908,
+      "eval_accuracy": 0.9062974683544304,
+      "eval_loss": 0.40965867042541504,
       "eval_macro avg": {
-        "f1-score": 0.8335207225588329,
-        "precision": 0.827721967449073,
-        "recall": 0.8397996465122839,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8517,
-      "eval_samples_per_second": 16.489,
-      "eval_steps_per_second": 2.061,
+        "f1-score": 0.8434413139886338,
+        "precision": 0.8644657729526208,
+        "recall": 0.8259120399103289,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.2033,
+      "eval_samples_per_second": 12.896,
+      "eval_steps_per_second": 1.612,
       "eval_weighted avg": {
-        "f1-score": 0.8956465277692122,
-        "precision": 0.8961582200263131,
-        "recall": 0.8952364921730348,
-        "support": 29705.0
+        "f1-score": 0.9047924430886448,
+        "precision": 0.9047867115327305,
+        "recall": 0.9062974683544304,
+        "support": 31600.0
       },
       "step": 648
     },
     {
       "epoch": 9.0,
       "eval_B-Claim": {
-        "f1-score": 0.6604127579737337,
-        "precision": 0.6717557251908397,
-        "recall": 0.6494464944649446,
-        "support": 271.0
+        "f1-score": 0.6900175131348512,
+        "precision": 0.686411149825784,
+        "recall": 0.6936619718309859,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.8661417322834645,
-        "precision": 0.9565217391304348,
-        "recall": 0.7913669064748201,
-        "support": 139.0
+        "f1-score": 0.9064748201438848,
+        "precision": 0.9197080291970803,
+        "recall": 0.8936170212765957,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8852963818321786,
-        "precision": 0.8633633633633634,
-        "recall": 0.9083728278041074,
-        "support": 633.0
+        "f1-score": 0.8892025405786873,
+        "precision": 0.8885754583921015,
+        "recall": 0.8898305084745762,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6148568412719356,
-        "precision": 0.6512017887087759,
-        "recall": 0.5823544113971507,
-        "support": 4001.0
+        "f1-score": 0.6670815183571872,
+        "precision": 0.6771096513390601,
+        "recall": 0.657346087809664,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8500414250207126,
-        "precision": 0.957089552238806,
-        "recall": 0.7645305514157973,
-        "support": 2013.0
+        "f1-score": 0.8905813658288906,
+        "precision": 0.9159268929503916,
+        "recall": 0.866600790513834,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8985408601693112,
-        "precision": 0.8677183468901487,
-        "recall": 0.9316337332392378,
-        "support": 11336.0
+        "f1-score": 0.9051692681937971,
+        "precision": 0.8978524893428779,
+        "recall": 0.9126062786134729,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.9995136401821638,
-        "precision": 0.9998230871295887,
-        "recall": 0.9992043847241867,
-        "support": 11312.0
+        "f1-score": 0.9987240172875077,
+        "precision": 0.9976153276868679,
+        "recall": 0.9998351738915444,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.8952701565393032,
-      "eval_loss": 0.5578464865684509,
+      "eval_accuracy": 0.9076582278481012,
+      "eval_loss": 0.44169458746910095,
       "eval_macro avg": {
-        "f1-score": 0.8249719483905,
-        "precision": 0.8524962289502797,
-        "recall": 0.8038441870743206,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8325,
-      "eval_samples_per_second": 16.555,
-      "eval_steps_per_second": 2.069,
+        "f1-score": 0.849607291932115,
+        "precision": 0.8547427141048803,
+        "recall": 0.8447854046300962,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.2166,
+      "eval_samples_per_second": 12.869,
+      "eval_steps_per_second": 1.609,
       "eval_weighted avg": {
-        "f1-score": 0.8928897918536849,
-        "precision": 0.8934537922744324,
-        "recall": 0.8952701565393032,
-        "support": 29705.0
+        "f1-score": 0.9071553503542205,
+        "precision": 0.9068271879381139,
+        "recall": 0.9076582278481012,
+        "support": 31600.0
       },
       "step": 729
     },
     {
       "epoch": 10.0,
       "eval_B-Claim": {
-        "f1-score": 0.6775956284153005,
-        "precision": 0.6690647482014388,
-        "recall": 0.6863468634686347,
-        "support": 271.0
+        "f1-score": 0.6761565836298933,
+        "precision": 0.6834532374100719,
+        "recall": 0.6690140845070423,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.8832116788321168,
-        "precision": 0.8962962962962963,
-        "recall": 0.8705035971223022,
-        "support": 139.0
+        "f1-score": 0.8872727272727273,
+        "precision": 0.9104477611940298,
+        "recall": 0.8652482269503546,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8820269200316706,
-        "precision": 0.8841269841269841,
-        "recall": 0.8799368088467614,
-        "support": 633.0
+        "f1-score": 0.8901329601119664,
+        "precision": 0.8821081830790569,
+        "recall": 0.8983050847457628,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6348430262480701,
-        "precision": 0.6542031291434632,
-        "recall": 0.6165958510372407,
-        "support": 4001.0
+        "f1-score": 0.6512742871561948,
+        "precision": 0.6705637828007275,
+        "recall": 0.6330635271032622,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8864321608040201,
-        "precision": 0.896797153024911,
-        "recall": 0.8763040238450075,
-        "support": 2013.0
+        "f1-score": 0.8682926829268293,
+        "precision": 0.9037947621592731,
+        "recall": 0.8354743083003953,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8934983440822729,
-        "precision": 0.8828797795384086,
-        "recall": 0.9043754410726887,
-        "support": 11336.0
+        "f1-score": 0.905623866156017,
+        "precision": 0.8933428775948461,
+        "recall": 0.9182472204054938,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 1.0,
-        "precision": 1.0,
-        "recall": 1.0,
-        "support": 11312.0
+        "f1-score": 0.9981898963304262,
+        "precision": 0.9965500246426812,
+        "recall": 0.9998351738915444,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.897458340346743,
-      "eval_loss": 0.6301500201225281,
+      "eval_accuracy": 0.9045569620253164,
+      "eval_loss": 0.45926380157470703,
       "eval_macro avg": {
-        "f1-score": 0.8368011083447787,
-        "precision": 0.8404811557616432,
-        "recall": 0.8334375121989479,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8426,
-      "eval_samples_per_second": 16.52,
-      "eval_steps_per_second": 2.065,
+        "f1-score": 0.8395632862262934,
+        "precision": 0.8486086612686695,
+        "recall": 0.8313125179862652,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.2031,
+      "eval_samples_per_second": 12.897,
+      "eval_steps_per_second": 1.612,
       "eval_weighted avg": {
-        "f1-score": 0.8964756773362035,
-        "precision": 0.895761611933671,
-        "recall": 0.897458340346743,
-        "support": 29705.0
+        "f1-score": 0.9034697801243392,
+        "precision": 0.9028380907030437,
+        "recall": 0.9045569620253164,
+        "support": 31600.0
       },
       "step": 810
     },
     {
       "epoch": 11.0,
       "eval_B-Claim": {
-        "f1-score": 0.6483300589390962,
-        "precision": 0.6932773109243697,
-        "recall": 0.6088560885608856,
-        "support": 271.0
+        "f1-score": 0.6407766990291263,
+        "precision": 0.7142857142857143,
+        "recall": 0.5809859154929577,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.8805970149253732,
-        "precision": 0.9147286821705426,
-        "recall": 0.8489208633093526,
-        "support": 139.0
+        "f1-score": 0.9032258064516129,
+        "precision": 0.9130434782608695,
+        "recall": 0.8936170212765957,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8848207475209764,
-        "precision": 0.855457227138643,
-        "recall": 0.9162717219589257,
-        "support": 633.0
+        "f1-score": 0.8872282608695653,
+        "precision": 0.8547120418848168,
+        "recall": 0.922316384180791,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6216666666666667,
-        "precision": 0.6995936230071897,
-        "recall": 0.55936015996001,
-        "support": 4001.0
+        "f1-score": 0.5974395448079659,
+        "precision": 0.7111412123264477,
+        "recall": 0.515084621044886,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8927083333333333,
-        "precision": 0.9381499726327313,
-        "recall": 0.851465474416294,
-        "support": 2013.0
+        "f1-score": 0.8862901219208758,
+        "precision": 0.8927318295739348,
+        "recall": 0.8799407114624506,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.9015977681968045,
-        "precision": 0.8655250770978737,
-        "recall": 0.9408080451658434,
-        "support": 11336.0
+        "f1-score": 0.9007579901539423,
+        "precision": 0.862670258943272,
+        "recall": 0.9423642903858731,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 1.0,
-        "precision": 1.0,
-        "recall": 1.0,
-        "support": 11312.0
+        "f1-score": 0.9989708122349842,
+        "precision": 0.9980258287406433,
+        "recall": 0.9999175869457723,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.9019357010604275,
-      "eval_loss": 0.5967662334442139,
+      "eval_accuracy": 0.9014240506329114,
+      "eval_loss": 0.5345993638038635,
       "eval_macro avg": {
-        "f1-score": 0.8328172270831785,
-        "precision": 0.8523902704244785,
-        "recall": 0.8179546219101873,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8502,
-      "eval_samples_per_second": 16.494,
-      "eval_steps_per_second": 2.062,
+        "f1-score": 0.8306698907811532,
+        "precision": 0.8495157662879569,
+        "recall": 0.8191752186841895,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.2101,
+      "eval_samples_per_second": 12.882,
+      "eval_steps_per_second": 1.61,
       "eval_weighted avg": {
-        "f1-score": 0.8979974574260295,
-        "precision": 0.8977509549706758,
-        "recall": 0.9019357010604275,
-        "support": 29705.0
+        "f1-score": 0.8957812921551218,
+        "precision": 0.8957333024681017,
+        "recall": 0.9014240506329114,
+        "support": 31600.0
       },
       "step": 891
     },
     {
       "epoch": 12.0,
       "eval_B-Claim": {
-        "f1-score": 0.6921898928024502,
-        "precision": 0.5916230366492147,
-        "recall": 0.8339483394833949,
-        "support": 271.0
+        "f1-score": 0.6859083191850596,
+        "precision": 0.6622950819672131,
+        "recall": 0.7112676056338029,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.8978102189781022,
-        "precision": 0.9111111111111111,
-        "recall": 0.8848920863309353,
-        "support": 139.0
+        "f1-score": 0.8921933085501859,
+        "precision": 0.9375,
+        "recall": 0.851063829787234,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8482758620689655,
-        "precision": 0.9335863377609108,
-        "recall": 0.7772511848341233,
-        "support": 633.0
+        "f1-score": 0.8863636363636365,
+        "precision": 0.8914285714285715,
+        "recall": 0.8813559322033898,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6750635055038103,
-        "precision": 0.5854598861758766,
-        "recall": 0.7970507373156711,
-        "support": 4001.0
+        "f1-score": 0.6563715953307394,
+        "precision": 0.6508319266939957,
+        "recall": 0.6620063772381654,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8948170731707318,
-        "precision": 0.9157566302652106,
-        "recall": 0.8748137108792846,
-        "support": 2013.0
+        "f1-score": 0.8721605916534602,
+        "precision": 0.9370034052213394,
+        "recall": 0.8157114624505929,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8742200328407225,
-        "precision": 0.9336606874436316,
-        "recall": 0.821894848270995,
-        "support": 11336.0
+        "f1-score": 0.9015022330491272,
+        "precision": 0.8954670108081949,
+        "recall": 0.907619359058208,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 1.0,
-        "precision": 1.0,
-        "recall": 1.0,
-        "support": 11312.0
+        "f1-score": 0.9988474520457726,
+        "precision": 0.9977796052631579,
+        "recall": 0.9999175869457723,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.8894125568086181,
-      "eval_loss": 0.6401852965354919,
+      "eval_accuracy": 0.902879746835443,
+      "eval_loss": 0.6067038774490356,
       "eval_macro avg": {
-        "f1-score": 0.8403395121949689,
-        "precision": 0.8387425270579937,
-        "recall": 0.8556929867306292,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8727,
-      "eval_samples_per_second": 16.418,
-      "eval_steps_per_second": 2.052,
+        "f1-score": 0.8419067337397116,
+        "precision": 0.8531865144832105,
+        "recall": 0.8327060219024521,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.2295,
+      "eval_samples_per_second": 12.842,
+      "eval_steps_per_second": 1.605,
       "eval_weighted avg": {
-        "f1-score": 0.8945864954779644,
-        "precision": 0.9075830699664568,
-        "recall": 0.8894125568086181,
-        "support": 29705.0
+        "f1-score": 0.9030573735174033,
+        "precision": 0.9038530884689406,
+        "recall": 0.902879746835443,
+        "support": 31600.0
       },
       "step": 972
     },
     {
       "epoch": 12.35,
-      "grad_norm": 23.30899429321289,
-      "learning_rate": 1.506172839506173e-05,
-      "loss": 0.0272,
+      "grad_norm": 0.581019937992096,
+      "learning_rate": 7.654320987654322e-06,
+      "loss": 0.0322,
       "step": 1000
     },
     {
       "epoch": 13.0,
       "eval_B-Claim": {
-        "f1-score": 0.706766917293233,
-        "precision": 0.7203065134099617,
-        "recall": 0.6937269372693727,
-        "support": 271.0
+        "f1-score": 0.6728624535315986,
+        "precision": 0.7125984251968503,
+        "recall": 0.6373239436619719,
+        "support": 284.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.8958333333333333,
-        "precision": 0.8657718120805369,
-        "recall": 0.9280575539568345,
-        "support": 139.0
+        "f1-score": 0.9154929577464789,
+        "precision": 0.9090909090909091,
+        "recall": 0.9219858156028369,
+        "support": 141.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8957345971563981,
-        "precision": 0.8957345971563981,
-        "recall": 0.8957345971563981,
-        "support": 633.0
+        "f1-score": 0.8905817174515235,
+        "precision": 0.873641304347826,
+        "recall": 0.9081920903954802,
+        "support": 708.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6850271528316524,
-        "precision": 0.7096169300830432,
-        "recall": 0.6620844788802799,
-        "support": 4001.0
+        "f1-score": 0.6460571129159025,
+        "precision": 0.6932808546527973,
+        "recall": 0.6048565121412803,
+        "support": 4077.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8991678903573177,
-        "precision": 0.8861553304389773,
-        "recall": 0.912568306010929,
-        "support": 2013.0
+        "f1-score": 0.9011916583912611,
+        "precision": 0.905688622754491,
+        "recall": 0.8967391304347826,
+        "support": 2024.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.9089637690660374,
-        "precision": 0.9007362494586401,
-        "recall": 0.9173429781227946,
-        "support": 11336.0
+        "f1-score": 0.9046914370275567,
+        "precision": 0.8856001879257693,
+        "recall": 0.9246239372138653,
+        "support": 12232.0
       },
       "eval_O": {
-        "f1-score": 0.9999557971975424,
-        "precision": 1.0,
-        "recall": 0.9999115983026874,
-        "support": 11312.0
+        "f1-score": 0.9991346985866744,
+        "precision": 0.9990935311083642,
+        "recall": 0.9991758694577221,
+        "support": 12134.0
       },
-      "eval_accuracy": 0.9116310385456994,
-      "eval_loss": 0.6060317754745483,
+      "eval_accuracy": 0.907246835443038,
+      "eval_loss": 0.5913504362106323,
       "eval_macro avg": {
-        "f1-score": 0.855921351033645,
-        "precision": 0.8540459189467938,
-        "recall": 0.8584894928141853,
-        "support": 29705.0
-      },
-      "eval_runtime": 4.8344,
-      "eval_samples_per_second": 16.548,
-      "eval_steps_per_second": 2.068,
+        "f1-score": 0.8471445765215709,
+        "precision": 0.8541419764395725,
+        "recall": 0.8418424712725627,
+        "support": 31600.0
+      },
+      "eval_runtime": 6.2496,
+      "eval_samples_per_second": 12.801,
+      "eval_steps_per_second": 1.6,
       "eval_weighted avg": {
-        "f1-score": 0.9106004556040976,
-        "precision": 0.9098905735839876,
-        "recall": 0.9116310385456994,
-        "support": 29705.0
+        "f1-score": 0.9050120935479347,
+        "precision": 0.9039360771033997,
+        "recall": 0.907246835443038,
+        "support": 31600.0
       },
       "step": 1053
     }
   ],
   "logging_steps": 500,
-  "max_steps": 4050,
+  "max_steps": 1620,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 50,
+  "num_train_epochs": 20,
   "save_steps": 500,
   "total_flos": 1869280565334000.0,
   "train_batch_size": 4,

checkpoint-1053/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3aa7d1dcb1ecb96052506e9849f1d3fa19e688784f8ea37f784dfaca9546785
+oid sha256:4a6eed06a4e3317966121cd80a8275e0b82aa59ebaa0185c3b4676ff92bfed33
 size 4920