Training in progress, epoch 6, checkpoint

checkpoint-486/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e9cecea0118ec25df48c35406ac6c16abc16808e01a6405930bd7eea735e742d
+oid sha256:986df7aca58a24373770d92d17112daf69754ffd843342562bee2ae6b3e27706
 size 592324828

checkpoint-486/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f740e48933486ee38370159d550944a484e19f8c7698a390fa45e846e6a4552
+oid sha256:21314c6215b4ba1cc09de94c4e428b0eb3bdad409331d50ceaa5388ec1f93672
 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": 0.5724454649827784,
-        "precision": 0.6002889477486154,
-        "recall": 0.5470704410796576,
-        "support": 4557.0
+        "f1-score": 0.5385059461955894,
+        "precision": 0.6267132491265789,
+        "recall": 0.4720647773279352,
+        "support": 4940.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.7860447185813415,
-        "precision": 0.698526892771497,
-        "recall": 0.8986337593653592,
-        "support": 2269.0
+        "f1-score": 0.7814784727863526,
+        "precision": 0.9588039867109635,
+        "recall": 0.6595063985374772,
+        "support": 2188.0
       },
       "eval_O": {
-        "f1-score": 0.9987820830889982,
-        "precision": 0.9975671292124707,
-        "recall": 1.0,
-        "support": 11071.0
+        "f1-score": 0.9938608715302625,
+        "precision": 0.9996969237763298,
+        "recall": 0.9880925634688834,
+        "support": 13353.0
       },
       "eval_Premise": {
-        "f1-score": 0.8965445389824622,
-        "precision": 0.9051258677512096,
-        "recall": 0.8881243979633962,
-        "support": 14534.0
+        "f1-score": 0.8968837690149166,
+        "precision": 0.8455112497215416,
+        "recall": 0.954902824076986,
+        "support": 15899.0
       },
-      "eval_accuracy": 0.8791279948197712,
-      "eval_loss": 0.3157936632633209,
+      "eval_accuracy": 0.8837548103353491,
+      "eval_loss": 0.2860471308231354,
       "eval_macro avg": {
-        "f1-score": 0.8134542014088951,
-        "precision": 0.8003772093709481,
-        "recall": 0.8334571496021033,
-        "support": 32431.0
-      },
-      "eval_runtime": 4.8184,
-      "eval_samples_per_second": 16.603,
-      "eval_steps_per_second": 2.075,
+        "f1-score": 0.8026822648817803,
+        "precision": 0.8576813523338533,
+        "recall": 0.7686416408528204,
+        "support": 36380.0
+      },
+      "eval_runtime": 4.9572,
+      "eval_samples_per_second": 16.34,
+      "eval_steps_per_second": 2.219,
       "eval_weighted avg": {
-        "f1-score": 0.8781740995293698,
-        "precision": 0.8793943851252293,
-        "recall": 0.8791279948197712,
-        "support": 32431.0
+        "f1-score": 0.8768739289107891,
+        "precision": 0.8792073379911217,
+        "recall": 0.8837548103353491,
+        "support": 36380.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_Claim": {
-        "f1-score": 0.6090644692195831,
-        "precision": 0.680108254397835,
-        "recall": 0.5514592933947773,
-        "support": 4557.0
+        "f1-score": 0.6469459868753155,
+        "precision": 0.6453172205438067,
+        "recall": 0.648582995951417,
+        "support": 4940.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8585858585858586,
-        "precision": 0.8960229995208433,
-        "recall": 0.8241516086381666,
-        "support": 2269.0
+        "f1-score": 0.8435970110796187,
+        "precision": 0.9669226225634967,
+        "recall": 0.7481718464351006,
+        "support": 2188.0
       },
       "eval_O": {
-        "f1-score": 0.999458141425088,
-        "precision": 0.9992776523702032,
-        "recall": 0.9996386956914461,
-        "support": 11071.0
+        "f1-score": 0.998877245508982,
+        "precision": 0.9983541557567143,
+        "recall": 0.9994008836965476,
+        "support": 13353.0
       },
       "eval_Premise": {
-        "f1-score": 0.9105885478942475,
-        "precision": 0.8801849235905997,
-        "recall": 0.9431677445988716,
-        "support": 14534.0
+        "f1-score": 0.9099026477336144,
+        "precision": 0.8972179761540813,
+        "recall": 0.922951129001824,
+        "support": 15899.0
       },
-      "eval_accuracy": 0.8990780426135487,
-      "eval_loss": 0.24451124668121338,
+      "eval_accuracy": 0.903243540406817,
+      "eval_loss": 0.24885737895965576,
       "eval_macro avg": {
-        "f1-score": 0.8444242542811943,
-        "precision": 0.8638984574698704,
-        "recall": 0.8296043355808155,
-        "support": 32431.0
-      },
-      "eval_runtime": 4.8529,
-      "eval_samples_per_second": 16.485,
-      "eval_steps_per_second": 2.061,
+        "f1-score": 0.8498307227993828,
+        "precision": 0.8769529937545246,
+        "recall": 0.8297767137712222,
+        "support": 36380.0
+      },
+      "eval_runtime": 5.0097,
+      "eval_samples_per_second": 16.169,
+      "eval_steps_per_second": 2.196,
       "eval_weighted avg": {
-        "f1-score": 0.8949194640367888,
-        "precision": 0.8938342965082798,
-        "recall": 0.8990780426135487,
-        "support": 32431.0
+        "f1-score": 0.9028656814459712,
+        "precision": 0.9043261520711511,
+        "recall": 0.903243540406817,
+        "support": 36380.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_Claim": {
-        "f1-score": 0.6542320859402667,
-        "precision": 0.7079713847726111,
-        "recall": 0.60807548825982,
-        "support": 4557.0
+        "f1-score": 0.7040630102775911,
+        "precision": 0.6698958142935478,
+        "recall": 0.7419028340080972,
+        "support": 4940.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8915187376725839,
-        "precision": 0.8866608544027899,
-        "recall": 0.8964301454385192,
-        "support": 2269.0
+        "f1-score": 0.8888411094388303,
+        "precision": 0.8392204628501827,
+        "recall": 0.9446983546617916,
+        "support": 2188.0
       },
       "eval_O": {
-        "f1-score": 0.9995936611133684,
-        "precision": 0.9992778479870013,
-        "recall": 0.9999096739228616,
-        "support": 11071.0
+        "f1-score": 0.998989104796136,
+        "precision": 0.9988769092542678,
+        "recall": 0.9991013255448213,
+        "support": 13353.0
       },
       "eval_Premise": {
-        "f1-score": 0.9157990498332153,
-        "precision": 0.8973258501155497,
-        "recall": 0.935048850970139,
-        "support": 14534.0
+        "f1-score": 0.914582593823615,
+        "precision": 0.9390987408880053,
+        "recall": 0.8913139191144097,
+        "support": 15899.0
       },
-      "eval_accuracy": 0.9085442940396534,
-      "eval_loss": 0.22665414214134216,
+      "eval_accuracy": 0.9137987905442551,
+      "eval_loss": 0.24298855662345886,
       "eval_macro avg": {
-        "f1-score": 0.8652858836398586,
-        "precision": 0.872808984319488,
-        "recall": 0.8598660396478349,
-        "support": 32431.0
-      },
-      "eval_runtime": 4.8494,
-      "eval_samples_per_second": 16.497,
-      "eval_steps_per_second": 2.062,
+        "f1-score": 0.8766189545840432,
+        "precision": 0.8617729818215009,
+        "recall": 0.8942541083322799,
+        "support": 36380.0
+      },
+      "eval_runtime": 4.9876,
+      "eval_samples_per_second": 16.24,
+      "eval_steps_per_second": 2.205,
       "eval_weighted avg": {
-        "f1-score": 0.9059516032151627,
-        "precision": 0.9047762338408377,
-        "recall": 0.9085442940396534,
-        "support": 32431.0
+        "f1-score": 0.9154289662937856,
+        "precision": 0.9184781183611038,
+        "recall": 0.9137987905442551,
+        "support": 36380.0
       },
       "step": 243
     },
     {
       "epoch": 4.0,
       "eval_Claim": {
-        "f1-score": 0.6861642294713161,
-        "precision": 0.6424741478360781,
-        "recall": 0.7362299758613122,
-        "support": 4557.0
+        "f1-score": 0.6981611893583725,
+        "precision": 0.6754352763058289,
+        "recall": 0.7224696356275304,
+        "support": 4940.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8995515695067264,
-        "precision": 0.9155636695572797,
-        "recall": 0.884089907448215,
-        "support": 2269.0
+        "f1-score": 0.901231802911534,
+        "precision": 0.8836187966622749,
+        "recall": 0.9195612431444241,
+        "support": 2188.0
       },
       "eval_O": {
-        "f1-score": 0.9976481230212574,
-        "precision": 0.9990941208442794,
-        "recall": 0.9962063047601842,
-        "support": 11071.0
+        "f1-score": 0.9989130842172332,
+        "precision": 0.9998499399759904,
+        "recall": 0.9979779824758481,
+        "support": 13353.0
       },
       "eval_Premise": {
-        "f1-score": 0.9062532879739067,
-        "precision": 0.9242435081193219,
-        "recall": 0.8889500481629283,
-        "support": 14534.0
+        "f1-score": 0.9146224912392481,
+        "precision": 0.9266670970240785,
+        "recall": 0.9028869740235235,
+        "support": 15899.0
       },
-      "eval_accuracy": 0.9037649162838025,
-      "eval_loss": 0.2452327460050583,
+      "eval_accuracy": 0.9142935678944475,
+      "eval_loss": 0.23980534076690674,
       "eval_macro avg": {
-        "f1-score": 0.8724043024933017,
-        "precision": 0.8703438615892398,
-        "recall": 0.8763690590581599,
-        "support": 32431.0
-      },
-      "eval_runtime": 4.8393,
-      "eval_samples_per_second": 16.531,
-      "eval_steps_per_second": 2.066,
+        "f1-score": 0.8782321419315969,
+        "precision": 0.8713927774920431,
+        "recall": 0.8857239588178316,
+        "support": 36380.0
+      },
+      "eval_runtime": 4.9726,
+      "eval_samples_per_second": 16.289,
+      "eval_steps_per_second": 2.212,
       "eval_weighted avg": {
-        "f1-score": 0.9060584182508294,
-        "precision": 0.9095955973231696,
-        "recall": 0.9037649162838025,
-        "support": 32431.0
+        "f1-score": 0.9153623106642749,
+        "precision": 0.9168247558035478,
+        "recall": 0.9142935678944475,
+        "support": 36380.0
       },
       "step": 324
     },
     {
       "epoch": 5.0,
       "eval_Claim": {
-        "f1-score": 0.6640249759846301,
-        "precision": 0.7332272606735614,
-        "recall": 0.6067588325652842,
-        "support": 4557.0
+        "f1-score": 0.6650318783717509,
+        "precision": 0.6450999048525214,
+        "recall": 0.6862348178137652,
+        "support": 4940.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8848007008322383,
-        "precision": 0.8794079233783195,
-        "recall": 0.8902600264433671,
-        "support": 2269.0
+        "f1-score": 0.9172031076581575,
+        "precision": 0.8916702632714717,
+        "recall": 0.9442413162705667,
+        "support": 2188.0
       },
       "eval_O": {
-        "f1-score": 0.999367774566474,
-        "precision": 0.9992775219001174,
-        "recall": 0.9994580435371692,
-        "support": 11071.0
+        "f1-score": 0.9982383147794145,
+        "precision": 0.9992495872730002,
+        "recall": 0.9972290870965326,
+        "support": 13353.0
       },
       "eval_Premise": {
-        "f1-score": 0.9201314377682404,
-        "precision": 0.8973839110529758,
-        "recall": 0.9440621989816981,
-        "support": 14534.0
+        "f1-score": 0.8998438545616774,
+        "precision": 0.9119622787753521,
+        "recall": 0.8880432731618341,
+        "support": 15899.0
       },
-      "eval_accuracy": 0.9118127717307515,
-      "eval_loss": 0.28411003947257996,
+      "eval_accuracy": 0.9040956569543706,
+      "eval_loss": 0.2932502031326294,
       "eval_macro avg": {
-        "f1-score": 0.8670812222878956,
-        "precision": 0.8773241542512435,
-        "recall": 0.8601347753818797,
-        "support": 32431.0
-      },
-      "eval_runtime": 4.8228,
-      "eval_samples_per_second": 16.588,
-      "eval_steps_per_second": 2.073,
+        "f1-score": 0.87007928884275,
+        "precision": 0.8619955085430864,
+        "recall": 0.8789371235856747,
+        "support": 36380.0
+      },
+      "eval_runtime": 4.9595,
+      "eval_samples_per_second": 16.332,
+      "eval_steps_per_second": 2.218,
       "eval_weighted avg": {
-        "f1-score": 0.9087220731552942,
-        "precision": 0.9078434958291443,
-        "recall": 0.9118127717307515,
-        "support": 32431.0
+        "f1-score": 0.9051179642561333,
+        "precision": 0.9065430476942037,
+        "recall": 0.9040956569543706,
+        "support": 36380.0
       },
       "step": 405
     },
     {
       "epoch": 6.0,
       "eval_Claim": {
-        "f1-score": 0.6496087442553721,
-        "precision": 0.7484258729250143,
-        "recall": 0.5738424402018872,
-        "support": 4557.0
+        "f1-score": 0.6622698072805139,
+        "precision": 0.5740163325909429,
+        "recall": 0.7825910931174089,
+        "support": 4940.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8865800865800866,
-        "precision": 0.8711186729051468,
-        "recall": 0.9026002644336713,
-        "support": 2269.0
+        "f1-score": 0.9223704351629127,
+        "precision": 0.8842767295597485,
+        "recall": 0.9638939670932358,
+        "support": 2188.0
       },
       "eval_O": {
-        "f1-score": 0.9986892655367232,
-        "precision": 0.9994572100597069,
-        "recall": 0.9979225002258152,
-        "support": 11071.0
+        "f1-score": 0.9982004948639124,
+        "precision": 0.9993994444861497,
+        "recall": 0.9970044184827379,
+        "support": 13353.0
       },
       "eval_Premise": {
-        "f1-score": 0.9191778088205947,
-        "precision": 0.889647180015452,
-        "recall": 0.9507362047612495,
-        "support": 14534.0
+        "f1-score": 0.8780749379985253,
+        "precision": 0.9398091685199799,
+        "recall": 0.8239511918988616,
+        "support": 15899.0
       },
-      "eval_accuracy": 0.9105177145323918,
-      "eval_loss": 0.3082066774368286,
+      "eval_accuracy": 0.8902693787795491,
+      "eval_loss": 0.3604726195335388,
       "eval_macro avg": {
-        "f1-score": 0.8635139762981943,
-        "precision": 0.87716223397633,
-        "recall": 0.8562753524056559,
-        "support": 32431.0
-      },
-      "eval_runtime": 4.8389,
-      "eval_samples_per_second": 16.533,
-      "eval_steps_per_second": 2.067,
+        "f1-score": 0.865228918826466,
+        "precision": 0.8493754187892053,
+        "recall": 0.8918601676480611,
+        "support": 36380.0
+      },
+      "eval_runtime": 4.991,
+      "eval_samples_per_second": 16.229,
+      "eval_steps_per_second": 2.204,
       "eval_weighted avg": {
-        "f1-score": 0.9061618943658084,
-        "precision": 0.9059932736781678,
-        "recall": 0.9105177145323918,
-        "support": 32431.0
+        "f1-score": 0.8955262233990811,
+        "precision": 0.9086708334194257,
+        "recall": 0.8902693787795491,
+        "support": 36380.0
       },
       "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:b2dd7b387b7349843b62a9d5b4facde1d14a6a35003d55f9e21a477105ecd1d3
+oid sha256:95e8ce1619d475b315f2cb3494747dc2424fb9856a35d6759fcf75e79995f1f1
 size 4856