Training in progress, epoch 6, checkpoint

checkpoint-486/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:27755321fe7f6ef741130ea5db96e69e03c3cfee0ab1bc88b3bc8406b24921f9
+oid sha256:3d0d86217593f3352db806724a7edc920070629ce3fcfcaceef8dd5ef4f2d963
 size 592324828

checkpoint-486/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dde09801335c1408d4c51ef18ea070f8217d05ff2b7a7610b88c0878a75a099c
+oid sha256:6fb4c2a7e0a13c7bb22dee8b505ea2703d87aba86e5d4b7d44eafb83fdfd48f9
 size 1014657786

checkpoint-486/rng_state.pth

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

checkpoint-486/trainer_state.json

@@ -11,270 +11,270 @@
     {
       "epoch": 1.0,
       "eval_Claim": {
-        "f1-score": 0.38183790485045715,
-        "precision": 0.6497890295358649,
-        "recall": 0.27035330261136714,
-        "support": 4557.0
+        "f1-score": 0.4888068880688806,
+        "precision": 0.6228840125391849,
+        "recall": 0.40222672064777326,
+        "support": 4940.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.6266258607498087,
-        "precision": 0.5535653937140926,
-        "recall": 0.7219039224327898,
-        "support": 2269.0
+        "f1-score": 0.7325278370054489,
+        "precision": 0.7604525332021643,
+        "recall": 0.706581352833638,
+        "support": 2188.0
       },
       "eval_O": {
-        "f1-score": 0.8545803971812939,
-        "precision": 0.8443037974683544,
-        "recall": 0.8651102464332037,
-        "support": 8481.0
+        "f1-score": 0.8634848138844484,
+        "precision": 0.827613377692173,
+        "recall": 0.902606702950444,
+        "support": 10473.0
       },
       "eval_Premise": {
-        "f1-score": 0.878949075575738,
-        "precision": 0.8314310260186549,
-        "recall": 0.9322278794550709,
-        "support": 14534.0
+        "f1-score": 0.8792819197655248,
+        "precision": 0.8543458914268763,
+        "recall": 0.9057173407132524,
+        "support": 15899.0
       },
-      "eval_accuracy": 0.7960859220535504,
-      "eval_loss": 0.5359914898872375,
+      "eval_accuracy": 0.8174925373134329,
+      "eval_loss": 0.49806877970695496,
       "eval_macro avg": {
-        "f1-score": 0.6854983095893245,
-        "precision": 0.7197723116842417,
-        "recall": 0.697398837733108,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.8191,
-      "eval_samples_per_second": 16.601,
-      "eval_steps_per_second": 2.075,
+        "f1-score": 0.7410253646810757,
+        "precision": 0.7663239537150996,
+        "recall": 0.7292830292862769,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.9194,
+      "eval_samples_per_second": 16.465,
+      "eval_steps_per_second": 2.236,
       "eval_weighted avg": {
-        "f1-score": 0.7769240850962494,
-        "precision": 0.7862232339538372,
-        "recall": 0.7960859220535504,
-        "support": 29841.0
+        "f1-score": 0.8071778099281224,
+        "precision": 0.80572410140167,
+        "recall": 0.8174925373134329,
+        "support": 33500.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_Claim": {
-        "f1-score": 0.6161300742974051,
-        "precision": 0.6048625792811839,
-        "recall": 0.6278253236778583,
-        "support": 4557.0
+        "f1-score": 0.5993816254416962,
+        "precision": 0.6593780369290574,
+        "recall": 0.5493927125506073,
+        "support": 4940.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.7920215150156881,
-        "precision": 0.8057455540355677,
-        "recall": 0.7787571617452622,
-        "support": 2269.0
+        "f1-score": 0.8017817371937639,
+        "precision": 0.7197382769901853,
+        "recall": 0.9049360146252285,
+        "support": 2188.0
       },
       "eval_O": {
-        "f1-score": 0.8922796352583586,
-        "precision": 0.9209436566695947,
-        "recall": 0.865346067680698,
-        "support": 8481.0
+        "f1-score": 0.9058997050147494,
+        "precision": 0.9337184554575859,
+        "recall": 0.8796906330564308,
+        "support": 10473.0
       },
       "eval_Premise": {
-        "f1-score": 0.8914289590611538,
-        "precision": 0.8790554552143957,
-        "recall": 0.9041557726709784,
-        "support": 14534.0
+        "f1-score": 0.8854737486606459,
+        "precision": 0.862579028987236,
+        "recall": 0.9096169570413234,
+        "support": 15899.0
       },
-      "eval_accuracy": 0.8413927147213566,
-      "eval_loss": 0.443877637386322,
+      "eval_accuracy": 0.8468358208955223,
+      "eval_loss": 0.4562840759754181,
       "eval_macro avg": {
-        "f1-score": 0.7979650459081514,
-        "precision": 0.8026518113001855,
-        "recall": 0.7940210814436992,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.8087,
-      "eval_samples_per_second": 16.637,
-      "eval_steps_per_second": 2.08,
+        "f1-score": 0.7981342040777137,
+        "precision": 0.7938534495910161,
+        "recall": 0.8109090793183975,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.904,
+      "eval_samples_per_second": 16.517,
+      "eval_steps_per_second": 2.243,
       "eval_weighted avg": {
-        "f1-score": 0.842071433389123,
-        "precision": 0.8435143116581857,
-        "recall": 0.8413927147213566,
-        "support": 29841.0
+        "f1-score": 0.8442053257085675,
+        "precision": 0.8455251408465503,
+        "recall": 0.8468358208955223,
+        "support": 33500.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_Claim": {
-        "f1-score": 0.6042799305957202,
-        "precision": 0.6389432485322897,
-        "recall": 0.5731841123546193,
-        "support": 4557.0
+        "f1-score": 0.6235462150581514,
+        "precision": 0.6285479226655697,
+        "recall": 0.6186234817813765,
+        "support": 4940.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8349557522123894,
-        "precision": 0.8382940915148823,
-        "recall": 0.8316438959894227,
-        "support": 2269.0
+        "f1-score": 0.8338848444738584,
+        "precision": 0.8059701492537313,
+        "recall": 0.8638025594149908,
+        "support": 2188.0
       },
       "eval_O": {
-        "f1-score": 0.8941106484235574,
-        "precision": 0.9022691799735862,
-        "recall": 0.8860983374602052,
-        "support": 8481.0
+        "f1-score": 0.9066250826953974,
+        "precision": 0.8974646833193002,
+        "recall": 0.9159744103886184,
+        "support": 10473.0
       },
       "eval_Premise": {
-        "f1-score": 0.8917763490086511,
-        "precision": 0.8729980887102089,
-        "recall": 0.9113802119168846,
-        "support": 14534.0
+        "f1-score": 0.8827730692315017,
+        "precision": 0.8911176621379133,
+        "recall": 0.8745833071262343,
+        "support": 15899.0
       },
-      "eval_accuracy": 0.8464863778023525,
-      "eval_loss": 0.4555923044681549,
+      "eval_accuracy": 0.8490746268656716,
+      "eval_loss": 0.43820467591285706,
       "eval_macro avg": {
-        "f1-score": 0.8062806700600795,
-        "precision": 0.8131261521827418,
-        "recall": 0.8005766394302829,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.7811,
-      "eval_samples_per_second": 16.733,
-      "eval_steps_per_second": 2.092,
+        "f1-score": 0.8117073028647273,
+        "precision": 0.8057751043441286,
+        "recall": 0.818245939677805,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.9179,
+      "eval_samples_per_second": 16.47,
+      "eval_steps_per_second": 2.237,
       "eval_weighted avg": {
-        "f1-score": 0.8442159482345275,
-        "precision": 0.8429359878649869,
-        "recall": 0.8464863778023525,
-        "support": 29841.0
+        "f1-score": 0.8488105033097496,
+        "precision": 0.8488213959184715,
+        "recall": 0.8490746268656716,
+        "support": 33500.0
       },
       "step": 243
     },
     {
       "epoch": 4.0,
       "eval_Claim": {
-        "f1-score": 0.6278026905829596,
-        "precision": 0.5981717011128775,
-        "recall": 0.6605222734254992,
-        "support": 4557.0
+        "f1-score": 0.6415287144900879,
+        "precision": 0.6477507222451506,
+        "recall": 0.6354251012145749,
+        "support": 4940.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8182207014542344,
-        "precision": 0.7947652679684254,
-        "recall": 0.8431026884089907,
-        "support": 2269.0
+        "f1-score": 0.8005126014523707,
+        "precision": 0.7514033680834001,
+        "recall": 0.856489945155393,
+        "support": 2188.0
       },
       "eval_O": {
-        "f1-score": 0.9069328626209752,
-        "precision": 0.9373427277302466,
-        "recall": 0.8784341469166372,
-        "support": 8481.0
+        "f1-score": 0.9065169636319258,
+        "precision": 0.9273871354374751,
+        "recall": 0.8865654540246348,
+        "support": 10473.0
       },
       "eval_Premise": {
-        "f1-score": 0.888919552918449,
-        "precision": 0.8913795489137954,
-        "recall": 0.8864730975643319,
-        "support": 14534.0
+        "f1-score": 0.8932505382719131,
+        "precision": 0.8863636363636364,
+        "recall": 0.9002452984464432,
+        "support": 15899.0
       },
-      "eval_accuracy": 0.8463858449783854,
-      "eval_loss": 0.4377317428588867,
+      "eval_accuracy": 0.8540597014925373,
+      "eval_loss": 0.4749625325202942,
       "eval_macro avg": {
-        "f1-score": 0.8104689518941546,
-        "precision": 0.8054148114313363,
-        "recall": 0.8171330515788647,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.8109,
-      "eval_samples_per_second": 16.629,
-      "eval_steps_per_second": 2.079,
+        "f1-score": 0.8104522044615743,
+        "precision": 0.8032262155324155,
+        "recall": 0.8196814497102615,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.8972,
+      "eval_samples_per_second": 16.54,
+      "eval_steps_per_second": 2.246,
       "eval_weighted avg": {
-        "f1-score": 0.8487883791626096,
-        "precision": 0.8523207959781866,
-        "recall": 0.8463858449783854,
-        "support": 29841.0
+        "f1-score": 0.8542213698406307,
+        "precision": 0.8551874645146165,
+        "recall": 0.8540597014925373,
+        "support": 33500.0
       },
       "step": 324
     },
     {
       "epoch": 5.0,
       "eval_Claim": {
-        "f1-score": 0.6247990569070839,
-        "precision": 0.6105990783410138,
-        "recall": 0.6396752249286811,
-        "support": 4557.0
+        "f1-score": 0.6099687080911935,
+        "precision": 0.6808882235528942,
+        "recall": 0.5524291497975709,
+        "support": 4940.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.7998289136013687,
-        "precision": 0.7769007062733694,
-        "recall": 0.8241516086381666,
-        "support": 2269.0
+        "f1-score": 0.8315018315018317,
+        "precision": 0.8330275229357799,
+        "recall": 0.829981718464351,
+        "support": 2188.0
       },
       "eval_O": {
-        "f1-score": 0.9028039585296891,
-        "precision": 0.9020600353148911,
-        "recall": 0.9035491097747907,
-        "support": 8481.0
+        "f1-score": 0.9103454839630722,
+        "precision": 0.9074091642159188,
+        "recall": 0.9133008689009835,
+        "support": 10473.0
       },
       "eval_Premise": {
-        "f1-score": 0.8885327014878567,
-        "precision": 0.9001058948111542,
-        "recall": 0.8772533370028898,
-        "support": 14534.0
+        "f1-score": 0.8913988368533824,
+        "precision": 0.8682249120505635,
+        "recall": 0.915843763758727,
+        "support": 15899.0
       },
-      "eval_accuracy": 0.8444086994403672,
-      "eval_loss": 0.5236712694168091,
+      "eval_accuracy": 0.8558507462686568,
+      "eval_loss": 0.5162495374679565,
       "eval_macro avg": {
-        "f1-score": 0.8039911576314996,
-        "precision": 0.7974164286851071,
-        "recall": 0.8111573200861321,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.8153,
-      "eval_samples_per_second": 16.614,
-      "eval_steps_per_second": 2.077,
+        "f1-score": 0.8108037151023699,
+        "precision": 0.8223874556887891,
+        "recall": 0.8028888752304081,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.9413,
+      "eval_samples_per_second": 16.392,
+      "eval_steps_per_second": 2.226,
       "eval_weighted avg": {
-        "f1-score": 0.8455693764619782,
-        "precision": 0.8470828034323642,
-        "recall": 0.8444086994403672,
-        "support": 29841.0
+        "f1-score": 0.8519095458499607,
+        "precision": 0.8505509283002989,
+        "recall": 0.8558507462686568,
+        "support": 33500.0
       },
       "step": 405
     },
     {
       "epoch": 6.0,
       "eval_Claim": {
-        "f1-score": 0.6166529266281945,
-        "precision": 0.5813095006800078,
-        "recall": 0.6565723063418916,
-        "support": 4557.0
+        "f1-score": 0.646279703934554,
+        "precision": 0.6227477477477478,
+        "recall": 0.6716599190283401,
+        "support": 4940.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8074107959743824,
-        "precision": 0.8392772230147408,
-        "recall": 0.7778757161745262,
-        "support": 2269.0
+        "f1-score": 0.8414067684140676,
+        "precision": 0.8152593227603944,
+        "recall": 0.8692870201096892,
+        "support": 2188.0
       },
       "eval_O": {
-        "f1-score": 0.8913107165875667,
-        "precision": 0.8858607034707664,
-        "recall": 0.8968282042212004,
-        "support": 8481.0
+        "f1-score": 0.9112506693929215,
+        "precision": 0.9295788637266588,
+        "recall": 0.8936312422419556,
+        "support": 10473.0
       },
       "eval_Premise": {
-        "f1-score": 0.8815305371596762,
-        "precision": 0.8981792217065334,
-        "recall": 0.865487821659557,
-        "support": 14534.0
+        "f1-score": 0.8861383012314494,
+        "precision": 0.8897343224906474,
+        "recall": 0.8825712308950249,
+        "support": 15899.0
       },
-      "eval_accuracy": 0.8358298984618477,
-      "eval_loss": 0.5537309050559998,
+      "eval_accuracy": 0.8540597014925373,
+      "eval_loss": 0.552895724773407,
       "eval_macro avg": {
-        "f1-score": 0.7992262440874549,
-        "precision": 0.8011566622180121,
-        "recall": 0.7991910120992938,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.7996,
-      "eval_samples_per_second": 16.668,
-      "eval_steps_per_second": 2.084,
+        "f1-score": 0.8212688607432481,
+        "precision": 0.8143300641813621,
+        "recall": 0.8292873530687525,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.9336,
+      "eval_samples_per_second": 16.418,
+      "eval_steps_per_second": 2.23,
       "eval_weighted avg": {
-        "f1-score": 0.838225042631562,
-        "precision": 0.8418105575563007,
-        "recall": 0.8358298984618477,
-        "support": 29841.0
+        "f1-score": 0.8556973390614198,
+        "precision": 0.8579560210496036,
+        "recall": 0.8540597014925373,
+        "support": 33500.0
       },
       "step": 486
     }
@@ -284,7 +284,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 862732396360800.0,
+  "total_flos": 860053103204400.0,
   "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 sha256:19f05e33d3889e7c1ce1bf18bf9e16659f52e74109a304d618b5d09955736add
+oid sha256:da575726aa6fb9f10d3d3191c16916e8d9dc133cfc573c1d27486b346adcb4eb
 size 4856