Training in progress, epoch 4, checkpoint

checkpoint-324/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f5fe1adae011a782352680b73a860ddaad1ad1d087f47d3dadb9c901fc6d8cb2
+oid sha256:67bba726d5b93a5953d8c322309502f56f3d294303980872106da73402842fcd
 size 592324828

checkpoint-324/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2492b7297930901696c6034172664ac0dfc2ef8822581792774047f784fb6a67
+oid sha256:ad7419a2296c7b12859acd3f69a8a13a0da0939bc756ffeb1fd315ac1573d369
 size 1014657786

checkpoint-324/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bd95f0f86efa354df6a4222f1c9e390b588b8fcd802633d848de21a27eedcb23
+oid sha256:5474bfb178f1f5ac14d6be6fd7772f757be3e1728e1b25f0f54ac7eff70dac62
 size 14244

checkpoint-324/trainer_state.json

@@ -11,180 +11,180 @@
     {
       "epoch": 1.0,
       "eval_Claim": {
-        "f1-score": 0.4888068880688806,
-        "precision": 0.6228840125391849,
-        "recall": 0.40222672064777326,
-        "support": 4940.0
+        "f1-score": 0.41092129347162903,
+        "precision": 0.5640703517587939,
+        "recall": 0.32317658349328215,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.7325278370054489,
-        "precision": 0.7604525332021643,
-        "recall": 0.706581352833638,
-        "support": 2188.0
+        "f1-score": 0.6523102655237298,
+        "precision": 0.5929304446978335,
+        "recall": 0.724907063197026,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.8634848138844484,
-        "precision": 0.827613377692173,
-        "recall": 0.902606702950444,
-        "support": 10473.0
+        "f1-score": 0.8834432924913603,
+        "precision": 0.8545380875202593,
+        "recall": 0.9143724257533059,
+        "support": 9226.0
       },
       "eval_Premise": {
-        "f1-score": 0.8792819197655248,
-        "precision": 0.8543458914268763,
-        "recall": 0.9057173407132524,
-        "support": 15899.0
+        "f1-score": 0.8664166767469054,
+        "precision": 0.8441231929604023,
+        "recall": 0.8899196554294707,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.8174925373134329,
-      "eval_loss": 0.49806877970695496,
+      "eval_accuracy": 0.7997031029363844,
+      "eval_loss": 0.5104541182518005,
       "eval_macro avg": {
-        "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,
+        "f1-score": 0.7032728820584061,
+        "precision": 0.7139155192343223,
+        "recall": 0.7130939319682712,
+        "support": 27619.0
+      },
+      "eval_runtime": 4.7759,
+      "eval_samples_per_second": 16.751,
+      "eval_steps_per_second": 2.094,
       "eval_weighted avg": {
-        "f1-score": 0.8071778099281224,
-        "precision": 0.80572410140167,
-        "recall": 0.8174925373134329,
-        "support": 33500.0
+        "f1-score": 0.7866826459135919,
+        "precision": 0.7857670171690952,
+        "recall": 0.7997031029363844,
+        "support": 27619.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_Claim": {
-        "f1-score": 0.5993816254416962,
-        "precision": 0.6593780369290574,
-        "recall": 0.5493927125506073,
-        "support": 4940.0
+        "f1-score": 0.5200312174817897,
+        "precision": 0.5678977272727272,
+        "recall": 0.47960652591170827,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8017817371937639,
-        "precision": 0.7197382769901853,
-        "recall": 0.9049360146252285,
-        "support": 2188.0
+        "f1-score": 0.7526302911671151,
+        "precision": 0.7948320413436692,
+        "recall": 0.7146840148698885,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9058997050147494,
-        "precision": 0.9337184554575859,
-        "recall": 0.8796906330564308,
-        "support": 10473.0
+        "f1-score": 0.9116096466629278,
+        "precision": 0.944560669456067,
+        "recall": 0.8808801213960546,
+        "support": 9226.0
       },
       "eval_Premise": {
-        "f1-score": 0.8854737486606459,
-        "precision": 0.862579028987236,
-        "recall": 0.9096169570413234,
-        "support": 15899.0
+        "f1-score": 0.8738735224125151,
+        "precision": 0.8259587020648967,
+        "recall": 0.9276898865236478,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.8468358208955223,
-      "eval_loss": 0.4562840759754181,
+      "eval_accuracy": 0.8278359100619139,
+      "eval_loss": 0.44692954421043396,
       "eval_macro avg": {
-        "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,
+        "f1-score": 0.7645361694310869,
+        "precision": 0.7833122850343401,
+        "recall": 0.7507151371753248,
+        "support": 27619.0
+      },
+      "eval_runtime": 4.7439,
+      "eval_samples_per_second": 16.864,
+      "eval_steps_per_second": 2.108,
       "eval_weighted avg": {
-        "f1-score": 0.8442053257085675,
-        "precision": 0.8455251408465503,
-        "recall": 0.8468358208955223,
-        "support": 33500.0
+        "f1-score": 0.8236335905447046,
+        "precision": 0.8242076985653165,
+        "recall": 0.8278359100619139,
+        "support": 27619.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_Claim": {
-        "f1-score": 0.6235462150581514,
-        "precision": 0.6285479226655697,
-        "recall": 0.6186234817813765,
-        "support": 4940.0
+        "f1-score": 0.6218357913273168,
+        "precision": 0.5744204961366409,
+        "recall": 0.6777831094049904,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8338848444738584,
-        "precision": 0.8059701492537313,
-        "recall": 0.8638025594149908,
-        "support": 2188.0
+        "f1-score": 0.7799868909766223,
+        "precision": 0.7360824742268042,
+        "recall": 0.8294609665427509,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9066250826953974,
-        "precision": 0.8974646833193002,
-        "recall": 0.9159744103886184,
-        "support": 10473.0
+        "f1-score": 0.9093637121422617,
+        "precision": 0.9330596419204014,
+        "recall": 0.8868415347929763,
+        "support": 9226.0
       },
       "eval_Premise": {
-        "f1-score": 0.8827730692315017,
-        "precision": 0.8911176621379133,
-        "recall": 0.8745833071262343,
-        "support": 15899.0
+        "f1-score": 0.876420695504665,
+        "precision": 0.89797514556357,
+        "recall": 0.855876749772219,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.8490746268656716,
-      "eval_loss": 0.43820467591285706,
+      "eval_accuracy": 0.8372859263550454,
+      "eval_loss": 0.40934404730796814,
       "eval_macro avg": {
-        "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,
+        "f1-score": 0.7969017724877165,
+        "precision": 0.7853844394618541,
+        "recall": 0.8124905901282342,
+        "support": 27619.0
+      },
+      "eval_runtime": 4.7941,
+      "eval_samples_per_second": 16.687,
+      "eval_steps_per_second": 2.086,
       "eval_weighted avg": {
-        "f1-score": 0.8488105033097496,
-        "precision": 0.8488213959184715,
-        "recall": 0.8490746268656716,
-        "support": 33500.0
+        "f1-score": 0.8414917278933443,
+        "precision": 0.8482528803063183,
+        "recall": 0.8372859263550454,
+        "support": 27619.0
       },
       "step": 243
     },
     {
       "epoch": 4.0,
       "eval_Claim": {
-        "f1-score": 0.6415287144900879,
-        "precision": 0.6477507222451506,
-        "recall": 0.6354251012145749,
-        "support": 4940.0
+        "f1-score": 0.608644284729743,
+        "precision": 0.516112873601603,
+        "recall": 0.7416026871401151,
+        "support": 4168.0
       },
       "eval_MajorClaim": {
-        "f1-score": 0.8005126014523707,
-        "precision": 0.7514033680834001,
-        "recall": 0.856489945155393,
-        "support": 2188.0
+        "f1-score": 0.788337924701561,
+        "precision": 0.7790381125226861,
+        "recall": 0.7978624535315985,
+        "support": 2152.0
       },
       "eval_O": {
-        "f1-score": 0.9065169636319258,
-        "precision": 0.9273871354374751,
-        "recall": 0.8865654540246348,
-        "support": 10473.0
+        "f1-score": 0.9068679925629612,
+        "precision": 0.9442684500762643,
+        "recall": 0.8723173639713853,
+        "support": 9226.0
       },
       "eval_Premise": {
-        "f1-score": 0.8932505382719131,
-        "precision": 0.8863636363636364,
-        "recall": 0.9002452984464432,
-        "support": 15899.0
+        "f1-score": 0.8528029247910863,
+        "precision": 0.8985600293497202,
+        "recall": 0.8114801623457302,
+        "support": 12073.0
       },
-      "eval_accuracy": 0.8540597014925373,
-      "eval_loss": 0.4749625325202942,
+      "eval_accuracy": 0.8201962417176581,
+      "eval_loss": 0.5160673260688782,
       "eval_macro avg": {
-        "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,
+        "f1-score": 0.789163281696338,
+        "precision": 0.7844948663875684,
+        "recall": 0.8058156667472073,
+        "support": 27619.0
+      },
+      "eval_runtime": 4.7669,
+      "eval_samples_per_second": 16.782,
+      "eval_steps_per_second": 2.098,
       "eval_weighted avg": {
-        "f1-score": 0.8542213698406307,
-        "precision": 0.8551874645146165,
-        "recall": 0.8540597014925373,
-        "support": 33500.0
+        "f1-score": 0.8289940404467937,
+        "precision": 0.8468005514342695,
+        "recall": 0.8201962417176581,
+        "support": 27619.0
       },
       "step": 324
     }
@@ -194,7 +194,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 573368735469600.0,
+  "total_flos": 575154930907200.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

checkpoint-324/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da575726aa6fb9f10d3d3191c16916e8d9dc133cfc573c1d27486b346adcb4eb
+oid sha256:1e651c478af6098730eda01c59bbe4834fbad7ed476e7ec6e4265141ee6647e9
 size 4856