Training in progress, epoch 6, checkpoint

checkpoint-486/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd6a0fbff8d9b865883a7415cc2909c725b4b159a2ff6760d916fd98b0755ebe
+oid sha256:41c521ac88c7d00ef6f9fa3dec79264347144248fe3dadb525665cfa50a96689
 size 592330980

checkpoint-486/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0bba62c91251b1393280269444a63d81b5f77c38c20135588e0d6e1f68fffebc
+oid sha256:304c38fe468cdd24f92cba6e49a1850deee3f6d8df6e2f03cded13a660b740e6
 size 1014670074

checkpoint-486/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76fa7a37b83088a34dfd87f85bd186720c2175bd38d06b25844c7882d2516d29
+oid sha256:0f891ea0618b7d3999f24558d90e9c033c85e76f5c1de754b03ef890648ce2bb
 size 14244

checkpoint-486/trainer_state.json

@@ -11,378 +11,378 @@
     {
       "epoch": 1.0,
       "eval_B-Claim": {
-        "f1-score": 0.17153996101364524,
-        "precision": 0.25287356321839083,
-        "recall": 0.12979351032448377,
-        "support": 339.0
+        "f1-score": 0.37160751565762,
+        "precision": 0.42788461538461536,
+        "recall": 0.3284132841328413,
+        "support": 271.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.08383233532934131,
-        "precision": 1.0,
-        "recall": 0.04375,
-        "support": 160.0
+        "f1-score": 0.3578947368421052,
+        "precision": 0.6666666666666666,
+        "recall": 0.2446043165467626,
+        "support": 139.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8469620831429876,
-        "precision": 0.7427884615384616,
-        "recall": 0.9851222104144527,
-        "support": 941.0
+        "f1-score": 0.8640915593705293,
+        "precision": 0.7895424836601307,
+        "recall": 0.9541864139020537,
+        "support": 633.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.5420081967213116,
-        "precision": 0.6803858520900321,
-        "recall": 0.4504044274159217,
-        "support": 4698.0
+        "f1-score": 0.5003402749421533,
+        "precision": 0.5493126120741183,
+        "recall": 0.4593851537115721,
+        "support": 4001.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8343351767251743,
-        "precision": 0.8141717503519474,
-        "recall": 0.8555226824457594,
-        "support": 2028.0
+        "f1-score": 0.7718093699515347,
+        "precision": 0.6502211636611093,
+        "recall": 0.9493293591654247,
+        "support": 2013.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.9004323458767015,
-        "precision": 0.8590809093131264,
-        "recall": 0.9459659511472983,
-        "support": 14861.0
+        "f1-score": 0.875016720916752,
+        "precision": 0.8846812731043188,
+        "recall": 0.865561044460127,
+        "support": 11336.0
       },
       "eval_O": {
-        "f1-score": 0.9988389078242631,
-        "precision": 0.9991008541885209,
-        "recall": 0.9985770987793006,
-        "support": 13353.0
+        "f1-score": 0.9992483530087988,
+        "precision": 0.9995577178239717,
+        "recall": 0.998939179632249,
+        "support": 11312.0
       },
-      "eval_accuracy": 0.8856789444749863,
-      "eval_loss": 0.28347474336624146,
+      "eval_accuracy": 0.8614038040733883,
+      "eval_loss": 0.31713685393333435,
       "eval_macro avg": {
-        "f1-score": 0.6254212866619178,
-        "precision": 0.76405734152864,
-        "recall": 0.629876554361031,
-        "support": 36380.0
-      },
-      "eval_runtime": 5.0448,
-      "eval_samples_per_second": 16.056,
-      "eval_steps_per_second": 2.18,
+        "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,
       "eval_weighted avg": {
-        "f1-score": 0.8748148382495958,
-        "precision": 0.8768575102351055,
-        "recall": 0.8856789444749863,
-        "support": 36380.0
+        "f1-score": 0.8576207231627551,
+        "precision": 0.8601529227027923,
+        "recall": 0.8614038040733883,
+        "support": 29705.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_B-Claim": {
-        "f1-score": 0.5137931034482759,
-        "precision": 0.6182572614107884,
-        "recall": 0.43952802359882004,
-        "support": 339.0
+        "f1-score": 0.4708624708624709,
+        "precision": 0.6392405063291139,
+        "recall": 0.3726937269372694,
+        "support": 271.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.7622641509433963,
-        "precision": 0.9619047619047619,
-        "recall": 0.63125,
-        "support": 160.0
+        "f1-score": 0.796875,
+        "precision": 0.8717948717948718,
+        "recall": 0.7338129496402878,
+        "support": 139.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8853910477127397,
-        "precision": 0.8241758241758241,
-        "recall": 0.9564293304994687,
-        "support": 941.0
+        "f1-score": 0.8736616702355461,
+        "precision": 0.796875,
+        "recall": 0.966824644549763,
+        "support": 633.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.5413588470388762,
-        "precision": 0.6279853891542568,
-        "recall": 0.47573435504469985,
-        "support": 4698.0
+        "f1-score": 0.5100589925881107,
+        "precision": 0.6459770114942529,
+        "recall": 0.4213946513371657,
+        "support": 4001.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.7949260042283298,
-        "precision": 0.856492027334852,
-        "recall": 0.7416173570019724,
-        "support": 2028.0
+        "f1-score": 0.8401387776888176,
+        "precision": 0.9077277970011534,
+        "recall": 0.7819175360158966,
+        "support": 2013.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.9007064868336545,
-        "precision": 0.8614779777627618,
-        "recall": 0.9436780835744566,
-        "support": 14861.0
+        "f1-score": 0.8912891699864469,
+        "precision": 0.8338584492430646,
+        "recall": 0.9572159491884262,
+        "support": 11336.0
       },
       "eval_O": {
-        "f1-score": 0.9988389947942025,
-        "precision": 0.9990260713215463,
-        "recall": 0.9986519883172321,
-        "support": 13353.0
+        "f1-score": 0.9996904982977407,
+        "precision": 1.0,
+        "recall": 0.9993811881188119,
+        "support": 11312.0
       },
-      "eval_accuracy": 0.8864211105002748,
-      "eval_loss": 0.2809496521949768,
+      "eval_accuracy": 0.8830499915839084,
+      "eval_loss": 0.2966194748878479,
       "eval_macro avg": {
-        "f1-score": 0.7710398049999251,
-        "precision": 0.8213313304378275,
-        "recall": 0.7409841625766643,
-        "support": 36380.0
-      },
-      "eval_runtime": 5.0579,
-      "eval_samples_per_second": 16.014,
-      "eval_steps_per_second": 2.175,
+        "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,
       "eval_weighted avg": {
-        "f1-score": 0.8798131143596267,
-        "precision": 0.8787439244541853,
-        "recall": 0.8864211105002748,
-        "support": 36380.0
+        "f1-score": 0.8731020208182413,
+        "precision": 0.874440834821272,
+        "recall": 0.8830499915839084,
+        "support": 29705.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_B-Claim": {
-        "f1-score": 0.6401137980085349,
-        "precision": 0.6181318681318682,
-        "recall": 0.6637168141592921,
-        "support": 339.0
+        "f1-score": 0.6085192697768763,
+        "precision": 0.6756756756756757,
+        "recall": 0.5535055350553506,
+        "support": 271.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.7918781725888324,
-        "precision": 0.6666666666666666,
-        "recall": 0.975,
-        "support": 160.0
+        "f1-score": 0.8571428571428571,
+        "precision": 0.851063829787234,
+        "recall": 0.8633093525179856,
+        "support": 139.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8866442199775534,
-        "precision": 0.93935790725327,
-        "recall": 0.8395324123273114,
-        "support": 941.0
+        "f1-score": 0.8834729626808834,
+        "precision": 0.8529411764705882,
+        "recall": 0.9162717219589257,
+        "support": 633.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6254390633315594,
-        "precision": 0.6255056418990845,
-        "recall": 0.6253724989357173,
-        "support": 4698.0
+        "f1-score": 0.5764474423833614,
+        "precision": 0.6584269662921348,
+        "recall": 0.5126218445388653,
+        "support": 4001.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.76480605487228,
-        "precision": 0.6205096714768191,
-        "recall": 0.9965483234714004,
-        "support": 2028.0
+        "f1-score": 0.8581151832460733,
+        "precision": 0.9070282235749861,
+        "recall": 0.8142076502732241,
+        "support": 2013.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.9049205625504156,
-        "precision": 0.9449897450922942,
-        "recall": 0.868111163447951,
-        "support": 14861.0
+        "f1-score": 0.8959744247675935,
+        "precision": 0.8563158317922328,
+        "recall": 0.939484827099506,
+        "support": 11336.0
       },
       "eval_O": {
-        "f1-score": 0.9988758992805756,
-        "precision": 0.9995500562429697,
-        "recall": 0.9982026510896428,
-        "support": 13353.0
+        "f1-score": 0.9996020340481981,
+        "precision": 1.0,
+        "recall": 0.9992043847241867,
+        "support": 11312.0
       },
-      "eval_accuracy": 0.8894997251236944,
-      "eval_loss": 0.318760484457016,
+      "eval_accuracy": 0.8918700555462044,
+      "eval_loss": 0.2552729547023773,
       "eval_macro avg": {
-        "f1-score": 0.8018111100871074,
-        "precision": 0.7735302223947104,
-        "recall": 0.852354837633045,
-        "support": 36380.0
-      },
-      "eval_runtime": 5.0862,
-      "eval_samples_per_second": 15.925,
-      "eval_steps_per_second": 2.163,
+        "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,
       "eval_weighted avg": {
-        "f1-score": 0.8920666591013163,
-        "precision": 0.9012548868310761,
-        "recall": 0.8894997251236944,
-        "support": 36380.0
+        "f1-score": 0.8867633844066056,
+        "precision": 0.886070631898416,
+        "recall": 0.8918700555462044,
+        "support": 29705.0
       },
       "step": 243
     },
     {
       "epoch": 4.0,
       "eval_B-Claim": {
-        "f1-score": 0.7034883720930232,
-        "precision": 0.6934097421203438,
-        "recall": 0.7138643067846607,
-        "support": 339.0
+        "f1-score": 0.6722408026755852,
+        "precision": 0.6146788990825688,
+        "recall": 0.7416974169741697,
+        "support": 271.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.8958990536277602,
-        "precision": 0.9044585987261147,
-        "recall": 0.8875,
-        "support": 160.0
+        "f1-score": 0.8664259927797834,
+        "precision": 0.8695652173913043,
+        "recall": 0.8633093525179856,
+        "support": 139.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.9051172707889126,
-        "precision": 0.9080213903743316,
-        "recall": 0.9022316684378321,
-        "support": 941.0
+        "f1-score": 0.8687035507844755,
+        "precision": 0.9100346020761245,
+        "recall": 0.8309636650868878,
+        "support": 633.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6932699265998139,
-        "precision": 0.6739698492462312,
-        "recall": 0.7137079608343976,
-        "support": 4698.0
+        "f1-score": 0.6483151400094921,
+        "precision": 0.6171222046532641,
+        "recall": 0.6828292926768308,
+        "support": 4001.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8904009720534629,
-        "precision": 0.877815045519885,
-        "recall": 0.903353057199211,
-        "support": 2028.0
+        "f1-score": 0.8696993060909791,
+        "precision": 0.9009584664536742,
+        "recall": 0.8405365126676602,
+        "support": 2013.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.9127480293558032,
-        "precision": 0.9218310342461052,
-        "recall": 0.9038422717179194,
-        "support": 14861.0
+        "f1-score": 0.8906159274643798,
+        "precision": 0.9020175517958925,
+        "recall": 0.879498941425547,
+        "support": 11336.0
       },
       "eval_O": {
-        "f1-score": 0.9981619715668254,
-        "precision": 0.9999248459341651,
-        "recall": 0.9964053021792856,
-        "support": 13353.0
+        "f1-score": 0.999557835160948,
+        "precision": 0.9999115357395613,
+        "recall": 0.9992043847241867,
+        "support": 11312.0
       },
-      "eval_accuracy": 0.9113523914238593,
-      "eval_loss": 0.2517726719379425,
+      "eval_accuracy": 0.8935869382258879,
+      "eval_loss": 0.29465603828430176,
       "eval_macro avg": {
-        "f1-score": 0.8570122280122289,
-        "precision": 0.8542043580238825,
-        "recall": 0.8601292238790438,
-        "support": 36380.0
-      },
-      "eval_runtime": 5.062,
-      "eval_samples_per_second": 16.002,
-      "eval_steps_per_second": 2.173,
+        "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,
       "eval_weighted avg": {
-        "f1-score": 0.9122886837716081,
-        "precision": 0.9134709768686261,
-        "recall": 0.9113523914238593,
-        "support": 36380.0
+        "f1-score": 0.8954766464091573,
+        "precision": 0.8982496227307208,
+        "recall": 0.8935869382258879,
+        "support": 29705.0
       },
       "step": 324
     },
     {
       "epoch": 5.0,
       "eval_B-Claim": {
-        "f1-score": 0.6590538336052201,
-        "precision": 0.7372262773722628,
-        "recall": 0.5958702064896755,
-        "support": 339.0
+        "f1-score": 0.6438095238095239,
+        "precision": 0.6653543307086615,
+        "recall": 0.6236162361623616,
+        "support": 271.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.9078947368421054,
-        "precision": 0.9583333333333334,
-        "recall": 0.8625,
-        "support": 160.0
+        "f1-score": 0.8785714285714286,
+        "precision": 0.8723404255319149,
+        "recall": 0.8848920863309353,
+        "support": 139.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.9066802651708312,
-        "precision": 0.8715686274509804,
-        "recall": 0.944739638682253,
-        "support": 941.0
+        "f1-score": 0.8783151326053042,
+        "precision": 0.8674884437596302,
+        "recall": 0.8894154818325435,
+        "support": 633.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6513589503280225,
-        "precision": 0.7243355914538823,
-        "recall": 0.5917411664538101,
-        "support": 4698.0
+        "f1-score": 0.6283729628640129,
+        "precision": 0.6748923959827834,
+        "recall": 0.5878530367408148,
+        "support": 4001.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8981975120588981,
-        "precision": 0.9256933542647828,
-        "recall": 0.8722879684418146,
-        "support": 2028.0
+        "f1-score": 0.885450461692039,
+        "precision": 0.8896690070210632,
+        "recall": 0.8812717337307501,
+        "support": 2013.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.9158848058378942,
-        "precision": 0.8877170824123777,
-        "recall": 0.9458986609245676,
-        "support": 14861.0
+        "f1-score": 0.8989315871101154,
+        "precision": 0.878494442573257,
+        "recall": 0.9203422724064926,
+        "support": 11336.0
       },
       "eval_O": {
-        "f1-score": 0.9994384336041331,
-        "precision": 0.9992513849378649,
-        "recall": 0.9996255523103422,
-        "support": 13353.0
+        "f1-score": 0.9997347245556637,
+        "precision": 1.0,
+        "recall": 0.9994695898161244,
+        "support": 11312.0
       },
-      "eval_accuracy": 0.9121220450797142,
-      "eval_loss": 0.2739432156085968,
+      "eval_accuracy": 0.8995118666891095,
+      "eval_loss": 0.3176642060279846,
       "eval_macro avg": {
-        "f1-score": 0.8483583624924435,
-        "precision": 0.8720179501750691,
-        "recall": 0.8303804561860663,
-        "support": 36380.0
-      },
-      "eval_runtime": 5.0932,
-      "eval_samples_per_second": 15.904,
-      "eval_steps_per_second": 2.16,
+        "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,
       "eval_weighted avg": {
-        "f1-score": 0.9087400479269403,
-        "precision": 0.9081638580455985,
-        "recall": 0.9121220450797142,
-        "support": 36380.0
+        "f1-score": 0.8971010593476484,
+        "precision": 0.8958911872123141,
+        "recall": 0.8995118666891095,
+        "support": 29705.0
       },
       "step": 405
     },
     {
       "epoch": 6.0,
       "eval_B-Claim": {
-        "f1-score": 0.6675461741424802,
-        "precision": 0.60381861575179,
-        "recall": 0.7463126843657817,
-        "support": 339.0
+        "f1-score": 0.6150712830957231,
+        "precision": 0.6863636363636364,
+        "recall": 0.5571955719557196,
+        "support": 271.0
       },
       "eval_B-MajorClaim": {
-        "f1-score": 0.887608069164265,
-        "precision": 0.8235294117647058,
-        "recall": 0.9625,
-        "support": 160.0
+        "f1-score": 0.8686131386861314,
+        "precision": 0.8814814814814815,
+        "recall": 0.8561151079136691,
+        "support": 139.0
       },
       "eval_B-Premise": {
-        "f1-score": 0.8737316798196167,
-        "precision": 0.9303721488595438,
-        "recall": 0.8235919234856536,
-        "support": 941.0
+        "f1-score": 0.8804841149773072,
+        "precision": 0.8447024673439768,
+        "recall": 0.919431279620853,
+        "support": 633.0
       },
       "eval_I-Claim": {
-        "f1-score": 0.6529236868186323,
-        "precision": 0.6109050445103857,
-        "recall": 0.7011494252873564,
-        "support": 4698.0
+        "f1-score": 0.5758052970651396,
+        "precision": 0.6739276139410187,
+        "recall": 0.5026243439140214,
+        "support": 4001.0
       },
       "eval_I-MajorClaim": {
-        "f1-score": 0.8689320388349515,
-        "precision": 0.7863418530351438,
-        "recall": 0.9709072978303748,
-        "support": 2028.0
+        "f1-score": 0.8627552339105711,
+        "precision": 0.8992456896551724,
+        "recall": 0.829110779930452,
+        "support": 2013.0
       },
       "eval_I-Premise": {
-        "f1-score": 0.8937734066010992,
-        "precision": 0.9312910284463894,
-        "recall": 0.8591615638247763,
-        "support": 14861.0
+        "f1-score": 0.894107779408681,
+        "precision": 0.8521864257734432,
+        "recall": 0.9403669724770642,
+        "support": 11336.0
       },
       "eval_O": {
-        "f1-score": 0.9988009592326139,
-        "precision": 0.9994750656167979,
-        "recall": 0.9981277615517112,
-        "support": 13353.0
+        "f1-score": 1.0,
+        "precision": 1.0,
+        "recall": 1.0,
+        "support": 11312.0
       },
-      "eval_accuracy": 0.8944749862561847,
-      "eval_loss": 0.37784042954444885,
+      "eval_accuracy": 0.8922403635751557,
+      "eval_loss": 0.41071853041648865,
       "eval_macro avg": {
-        "f1-score": 0.8347594306590942,
-        "precision": 0.8122475954263937,
-        "recall": 0.8659643794779507,
-        "support": 36380.0
-      },
-      "eval_runtime": 5.0734,
-      "eval_samples_per_second": 15.966,
-      "eval_steps_per_second": 2.168,
+        "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,
       "eval_weighted avg": {
-        "f1-score": 0.8971820344797716,
-        "precision": 0.9033144340485707,
-        "recall": 0.8944749862561847,
-        "support": 36380.0
+        "f1-score": 0.8864802913843919,
+        "precision": 0.8861194550557427,
+        "recall": 0.8922403635751557,
+        "support": 29705.0
       },
       "step": 486
     }
@@ -392,7 +392,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 860065544394000.0,
+  "total_flos": 862744876308000.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

checkpoint-486/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:db3de2ea286aed4cce53d86abf84d503faba79a292e98a5c89e444e2125b5ea5
+oid sha256:e3aa7d1dcb1ecb96052506e9849f1d3fa19e688784f8ea37f784dfaca9546785
 size 4920