Training in progress, epoch 12, checkpoint

checkpoint-492/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cfe883009ca6eba3d193033ca6483a6c43d4f0310d57d368d86592627af508ee
+oid sha256:c53b5b5b30909241deca704a6ddd1b062bdb49e3a46df24a34685691cfa036c6
 size 592318676

checkpoint-492/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c3605a0c42f33fb0a181410f66b9fcb100880a05ff6a5e8e353bdd30deec106a
+oid sha256:bf752ed2e78d3694ee0d8ccab6933bd73d9a9cb27e7772cd2e19ad3ae0059748
 size 1014645498

checkpoint-492/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b6e53e706c63ba0c275c9b6332ceadb76c337b40858f3ddadb0ab10622a334eb
+oid sha256:1c189466643703424c77db24cb1c70bb1b76b65752ddeeaa3d7bb424b6ea3dcb
 size 14244

checkpoint-492/trainer_state.json

@@ -11,468 +11,468 @@
     {
       "epoch": 1.0,
       "eval_B": {
-        "f1-score": 0.6659475008989573,
-        "precision": 0.6905294556301268,
-        "recall": 0.6430555555555556,
-        "support": 1440.0
+        "f1-score": 0.5721362229102167,
+        "precision": 0.8076923076923077,
+        "recall": 0.4429530201342282,
+        "support": 1043.0
       },
       "eval_I": {
-        "f1-score": 0.9167863956473609,
-        "precision": 0.9325347388596071,
-        "recall": 0.9015611247510076,
-        "support": 21587.0
+        "f1-score": 0.927966217883682,
+        "precision": 0.8850358282336942,
+        "recall": 0.9752737752161383,
+        "support": 17350.0
       },
       "eval_O": {
-        "f1-score": 0.8443157797996509,
-        "precision": 0.8138010452653025,
-        "recall": 0.8772080588179128,
-        "support": 10473.0
+        "f1-score": 0.864171621779177,
+        "precision": 0.9349142280524723,
+        "recall": 0.8033817472360719,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.8828358208955224,
-      "eval_loss": 0.31183817982673645,
+      "eval_accuracy": 0.8977515478474963,
+      "eval_loss": 0.29473039507865906,
       "eval_macro avg": {
-        "f1-score": 0.8090165587819897,
-        "precision": 0.8122884132516788,
-        "recall": 0.8072749130414919,
-        "support": 33500.0
-      },
-      "eval_runtime": 1.3725,
-      "eval_samples_per_second": 59.018,
-      "eval_steps_per_second": 8.015,
+        "f1-score": 0.7880913541910252,
+        "precision": 0.8758807879928246,
+        "recall": 0.7405361808621462,
+        "support": 27619.0
+      },
+      "eval_runtime": 1.3384,
+      "eval_samples_per_second": 59.772,
+      "eval_steps_per_second": 7.472,
       "eval_weighted avg": {
-        "f1-score": 0.8833478055515172,
-        "precision": 0.8850127812218876,
-        "recall": 0.8828358208955224,
-        "support": 33500.0
+        "f1-score": 0.8932184128068332,
+        "precision": 0.8987766886849552,
+        "recall": 0.8977515478474963,
+        "support": 27619.0
       },
       "step": 41
     },
     {
       "epoch": 2.0,
       "eval_B": {
-        "f1-score": 0.8377561303325496,
-        "precision": 0.8113207547169812,
-        "recall": 0.8659722222222223,
-        "support": 1440.0
+        "f1-score": 0.8396183552930486,
+        "precision": 0.7979274611398963,
+        "recall": 0.8859060402684564,
+        "support": 1043.0
       },
       "eval_I": {
-        "f1-score": 0.9473755107538951,
-        "precision": 0.9191461555216729,
-        "recall": 0.9773938018251725,
-        "support": 21587.0
+        "f1-score": 0.9529475085691623,
+        "precision": 0.9369951534733441,
+        "recall": 0.9694524495677234,
+        "support": 17350.0
       },
       "eval_O": {
-        "f1-score": 0.8803449514911966,
-        "precision": 0.9519316163410302,
-        "recall": 0.8187720805881791,
-        "support": 10473.0
+        "f1-score": 0.9060667568786648,
+        "precision": 0.9441833137485312,
+        "recall": 0.8709083026230219,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.9230149253731343,
-      "eval_loss": 0.22658132016658783,
+      "eval_accuracy": 0.9333791954813715,
+      "eval_loss": 0.19544227421283722,
       "eval_macro avg": {
-        "f1-score": 0.8884921975258804,
-        "precision": 0.8941328421932281,
-        "recall": 0.8873793682118579,
-        "support": 33500.0
-      },
-      "eval_runtime": 1.3801,
-      "eval_samples_per_second": 58.69,
-      "eval_steps_per_second": 7.97,
+        "f1-score": 0.8995442069136251,
+        "precision": 0.8930353094539237,
+        "recall": 0.9087555974864006,
+        "support": 27619.0
+      },
+      "eval_runtime": 1.3427,
+      "eval_samples_per_second": 59.582,
+      "eval_steps_per_second": 7.448,
       "eval_weighted avg": {
-        "f1-score": 0.9217079598594182,
-        "precision": 0.9247608884769675,
-        "recall": 0.9230149253731343,
-        "support": 33500.0
+        "f1-score": 0.933007462877301,
+        "precision": 0.9341445927577168,
+        "recall": 0.9333791954813715,
+        "support": 27619.0
       },
       "step": 82
     },
     {
       "epoch": 3.0,
       "eval_B": {
-        "f1-score": 0.8710106382978723,
-        "precision": 0.8354591836734694,
-        "recall": 0.9097222222222222,
-        "support": 1440.0
+        "f1-score": 0.8647365923113433,
+        "precision": 0.856203007518797,
+        "recall": 0.8734419942473634,
+        "support": 1043.0
       },
       "eval_I": {
-        "f1-score": 0.9543258273315707,
-        "precision": 0.9392974112791063,
-        "recall": 0.9698429610413675,
-        "support": 21587.0
+        "f1-score": 0.9557367501456028,
+        "precision": 0.9658622719246616,
+        "recall": 0.945821325648415,
+        "support": 17350.0
       },
       "eval_O": {
-        "f1-score": 0.8996818452972758,
-        "precision": 0.9384009125790729,
-        "recall": 0.8640313186288552,
-        "support": 10473.0
+        "f1-score": 0.9184183917833005,
+        "precision": 0.9021432305279665,
+        "recall": 0.9352915673097767,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.934179104477612,
-      "eval_loss": 0.20443089306354523,
+      "eval_accuracy": 0.9395705854665267,
+      "eval_loss": 0.17375266551971436,
       "eval_macro avg": {
-        "f1-score": 0.908339436975573,
-        "precision": 0.9043858358438829,
-        "recall": 0.9145321672974815,
-        "support": 33500.0
-      },
-      "eval_runtime": 1.3812,
-      "eval_samples_per_second": 58.646,
-      "eval_steps_per_second": 7.964,
+        "f1-score": 0.9129639114134155,
+        "precision": 0.9080695033238083,
+        "recall": 0.9181849624018517,
+        "support": 27619.0
+      },
+      "eval_runtime": 1.344,
+      "eval_samples_per_second": 59.525,
+      "eval_steps_per_second": 7.441,
       "eval_weighted avg": {
-        "f1-score": 0.9336613408822066,
-        "precision": 0.9345536477376863,
-        "recall": 0.934179104477612,
-        "support": 33500.0
+        "f1-score": 0.9398342070096555,
+        "precision": 0.940436062116152,
+        "recall": 0.9395705854665267,
+        "support": 27619.0
       },
       "step": 123
     },
     {
       "epoch": 4.0,
       "eval_B": {
-        "f1-score": 0.8771342484097756,
-        "precision": 0.8468002585649644,
-        "recall": 0.9097222222222222,
-        "support": 1440.0
+        "f1-score": 0.862493839329719,
+        "precision": 0.8874239350912779,
+        "recall": 0.8389261744966443,
+        "support": 1043.0
       },
       "eval_I": {
-        "f1-score": 0.959998160834981,
-        "precision": 0.952900369677331,
-        "recall": 0.9672024829758651,
-        "support": 21587.0
+        "f1-score": 0.9584169773444221,
+        "precision": 0.9537671232876712,
+        "recall": 0.9631123919308358,
+        "support": 17350.0
       },
       "eval_O": {
-        "f1-score": 0.9145503290275409,
-        "precision": 0.9341764588727345,
-        "recall": 0.8957318819822401,
-        "support": 10473.0
+        "f1-score": 0.9205518294345384,
+        "precision": 0.926259190167892,
+        "recall": 0.9149143724257534,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.9423880597014925,
-      "eval_loss": 0.18545031547546387,
+      "eval_accuracy": 0.9423223143488179,
+      "eval_loss": 0.17503595352172852,
       "eval_macro avg": {
-        "f1-score": 0.9172275794240993,
-        "precision": 0.9112923623716767,
-        "recall": 0.9242188623934425,
-        "support": 33500.0
-      },
-      "eval_runtime": 1.377,
-      "eval_samples_per_second": 58.824,
-      "eval_steps_per_second": 7.988,
+        "f1-score": 0.9138208820362266,
+        "precision": 0.9224834161822804,
+        "recall": 0.9056509796177444,
+        "support": 27619.0
+      },
+      "eval_runtime": 1.3496,
+      "eval_samples_per_second": 59.277,
+      "eval_steps_per_second": 7.41,
       "eval_weighted avg": {
-        "f1-score": 0.9422280361659775,
-        "precision": 0.9424860509352908,
-        "recall": 0.9423880597014925,
-        "support": 33500.0
+        "f1-score": 0.9421458709478862,
+        "precision": 0.9420728499160095,
+        "recall": 0.9423223143488179,
+        "support": 27619.0
       },
       "step": 164
     },
     {
       "epoch": 5.0,
       "eval_B": {
-        "f1-score": 0.8835341365461846,
-        "precision": 0.8527131782945736,
-        "recall": 0.9166666666666666,
-        "support": 1440.0
+        "f1-score": 0.8739573679332716,
+        "precision": 0.8457399103139014,
+        "recall": 0.9041227229146692,
+        "support": 1043.0
       },
       "eval_I": {
-        "f1-score": 0.9602170394181884,
-        "precision": 0.9453672113485365,
-        "recall": 0.9755408347616621,
-        "support": 21587.0
+        "f1-score": 0.954658525554048,
+        "precision": 0.9367580161988239,
+        "recall": 0.9732564841498559,
+        "support": 17350.0
       },
       "eval_O": {
-        "f1-score": 0.9125018611345476,
-        "precision": 0.9500826787928897,
-        "recall": 0.877780960565263,
-        "support": 10473.0
+        "f1-score": 0.9086082241301401,
+        "precision": 0.948690728945506,
+        "recall": 0.8717754172989378,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.9424477611940298,
-      "eval_loss": 0.19699566066265106,
+      "eval_accuracy": 0.9367464426662805,
+      "eval_loss": 0.20350226759910583,
       "eval_macro avg": {
-        "f1-score": 0.9187510123663069,
-        "precision": 0.9160543561453333,
-        "recall": 0.9233294873311971,
-        "support": 33500.0
-      },
-      "eval_runtime": 1.3779,
-      "eval_samples_per_second": 58.784,
-      "eval_steps_per_second": 7.983,
+        "f1-score": 0.9124080392058199,
+        "precision": 0.9103962184860771,
+        "recall": 0.916384874787821,
+        "support": 27619.0
+      },
+      "eval_runtime": 1.3385,
+      "eval_samples_per_second": 59.767,
+      "eval_steps_per_second": 7.471,
       "eval_weighted avg": {
-        "f1-score": 0.9420037724838524,
-        "precision": 0.9428586526305366,
-        "recall": 0.9424477611940298,
-        "support": 33500.0
+        "f1-score": 0.9362280469583188,
+        "precision": 0.9373068891979518,
+        "recall": 0.9367464426662805,
+        "support": 27619.0
       },
       "step": 205
     },
     {
       "epoch": 6.0,
       "eval_B": {
-        "f1-score": 0.882471457353929,
-        "precision": 0.8543563068920677,
-        "recall": 0.9125,
-        "support": 1440.0
+        "f1-score": 0.8744710860366715,
+        "precision": 0.8579335793357934,
+        "recall": 0.8916586768935763,
+        "support": 1043.0
       },
       "eval_I": {
-        "f1-score": 0.9563333022648896,
-        "precision": 0.9621173050775939,
-        "recall": 0.9506184277574466,
-        "support": 21587.0
+        "f1-score": 0.9548456588905582,
+        "precision": 0.9394277427631212,
+        "recall": 0.970778097982709,
+        "support": 17350.0
       },
       "eval_O": {
-        "f1-score": 0.9094096465460059,
-        "precision": 0.9025674786043449,
-        "recall": 0.9163563448868519,
-        "support": 10473.0
+        "f1-score": 0.9091520861372813,
+        "precision": 0.941900999302812,
+        "recall": 0.8786039453717754,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.9382686567164179,
-      "eval_loss": 0.22578871250152588,
+      "eval_accuracy": 0.9369998913791231,
+      "eval_loss": 0.18961849808692932,
       "eval_macro avg": {
-        "f1-score": 0.9160714687216082,
-        "precision": 0.9063470301913354,
-        "recall": 0.9264915908814327,
-        "support": 33500.0
-      },
-      "eval_runtime": 1.3827,
-      "eval_samples_per_second": 58.582,
-      "eval_steps_per_second": 7.956,
+        "f1-score": 0.9128229436881702,
+        "precision": 0.9130874404672422,
+        "recall": 0.9136802400826869,
+        "support": 27619.0
+      },
+      "eval_runtime": 1.3446,
+      "eval_samples_per_second": 59.497,
+      "eval_steps_per_second": 7.437,
       "eval_weighted avg": {
-        "f1-score": 0.9384887499360641,
-        "precision": 0.9388683149271014,
-        "recall": 0.9382686567164179,
-        "support": 33500.0
+        "f1-score": 0.9365466769683909,
+        "precision": 0.9371763887090455,
+        "recall": 0.9369998913791231,
+        "support": 27619.0
       },
       "step": 246
     },
     {
       "epoch": 7.0,
       "eval_B": {
-        "f1-score": 0.8892651540805959,
-        "precision": 0.8678122934567085,
-        "recall": 0.9118055555555555,
-        "support": 1440.0
+        "f1-score": 0.8734770384254921,
+        "precision": 0.854262144821265,
+        "recall": 0.8935762224352828,
+        "support": 1043.0
       },
       "eval_I": {
-        "f1-score": 0.9597268994787103,
-        "precision": 0.9557587173243901,
-        "recall": 0.963728169731783,
-        "support": 21587.0
+        "f1-score": 0.9571367703451216,
+        "precision": 0.9436012321478577,
+        "recall": 0.9710662824207493,
+        "support": 17350.0
       },
       "eval_O": {
-        "f1-score": 0.914125549702798,
-        "precision": 0.925440313111546,
-        "recall": 0.903084121073236,
-        "support": 10473.0
+        "f1-score": 0.9144644952231968,
+        "precision": 0.9436181252161882,
+        "recall": 0.8870583134619553,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.9425373134328359,
-      "eval_loss": 0.22205175459384918,
+      "eval_accuracy": 0.9400774828922119,
+      "eval_loss": 0.19739200174808502,
       "eval_macro avg": {
-        "f1-score": 0.9210392010873681,
-        "precision": 0.9163371079642149,
-        "recall": 0.9262059487868582,
-        "support": 33500.0
-      },
-      "eval_runtime": 1.3808,
-      "eval_samples_per_second": 58.663,
-      "eval_steps_per_second": 7.967,
+        "f1-score": 0.9150261013312702,
+        "precision": 0.9138271673951035,
+        "recall": 0.9172336061059957,
+        "support": 27619.0
+      },
+      "eval_runtime": 1.3403,
+      "eval_samples_per_second": 59.69,
+      "eval_steps_per_second": 7.461,
       "eval_weighted avg": {
-        "f1-score": 0.9424418890435934,
-        "precision": 0.9424999860500445,
-        "recall": 0.9425373134328359,
-        "support": 33500.0
+        "f1-score": 0.9397229787282255,
+        "precision": 0.9402330865729558,
+        "recall": 0.9400774828922119,
+        "support": 27619.0
       },
       "step": 287
     },
     {
       "epoch": 8.0,
       "eval_B": {
-        "f1-score": 0.8742837883383889,
-        "precision": 0.8493778650949574,
-        "recall": 0.9006944444444445,
-        "support": 1440.0
+        "f1-score": 0.875,
+        "precision": 0.851952770208901,
+        "recall": 0.8993288590604027,
+        "support": 1043.0
       },
       "eval_I": {
-        "f1-score": 0.954344074989507,
-        "precision": 0.9607962815155641,
-        "recall": 0.9479779496919443,
-        "support": 21587.0
+        "f1-score": 0.9546683185043361,
+        "precision": 0.9332269074094462,
+        "recall": 0.9771181556195966,
+        "support": 17350.0
       },
       "eval_O": {
-        "f1-score": 0.9060386816096845,
-        "precision": 0.8975079632752483,
-        "recall": 0.9147331232693593,
-        "support": 10473.0
+        "f1-score": 0.90670155876664,
+        "precision": 0.9541427203065134,
+        "recall": 0.8637546065467158,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.9355522388059702,
-      "eval_loss": 0.2697383165359497,
+      "eval_accuracy": 0.9363119591585503,
+      "eval_loss": 0.2391989678144455,
       "eval_macro avg": {
-        "f1-score": 0.9115555149791934,
-        "precision": 0.9025607032952566,
-        "recall": 0.9211351724685827,
-        "support": 33500.0
-      },
-      "eval_runtime": 1.3839,
-      "eval_samples_per_second": 58.529,
-      "eval_steps_per_second": 7.948,
+        "f1-score": 0.9121232924236587,
+        "precision": 0.9131074659749535,
+        "recall": 0.913400540408905,
+        "support": 27619.0
+      },
+      "eval_runtime": 1.3469,
+      "eval_samples_per_second": 59.396,
+      "eval_steps_per_second": 7.424,
       "eval_weighted avg": {
-        "f1-score": 0.9358011138657909,
-        "precision": 0.9362213240058178,
-        "recall": 0.9355522388059702,
-        "support": 33500.0
+        "f1-score": 0.9356366598077863,
+        "precision": 0.937144513575063,
+        "recall": 0.9363119591585503,
+        "support": 27619.0
       },
       "step": 328
     },
     {
       "epoch": 9.0,
       "eval_B": {
-        "f1-score": 0.8867732063923836,
-        "precision": 0.8687541638907396,
-        "recall": 0.9055555555555556,
-        "support": 1440.0
+        "f1-score": 0.8698850574712644,
+        "precision": 0.8356890459363958,
+        "recall": 0.9069990412272292,
+        "support": 1043.0
       },
       "eval_I": {
-        "f1-score": 0.9593785402168635,
-        "precision": 0.9576294655220161,
-        "recall": 0.9611340158428684,
-        "support": 21587.0
+        "f1-score": 0.9540004482294935,
+        "precision": 0.9281042189033032,
+        "recall": 0.9813832853025937,
+        "support": 17350.0
       },
       "eval_O": {
-        "f1-score": 0.9133903681630298,
-        "precision": 0.9195780509048679,
-        "recall": 0.907285400553805,
-        "support": 10473.0
+        "f1-score": 0.9027465883572292,
+        "precision": 0.9629038201695124,
+        "recall": 0.8496639930630826,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.941910447761194,
-      "eval_loss": 0.236963152885437,
+      "eval_accuracy": 0.9345740251276295,
+      "eval_loss": 0.25876709818840027,
       "eval_macro avg": {
-        "f1-score": 0.9198473715907589,
-        "precision": 0.9153205601058745,
-        "recall": 0.9246583239840763,
-        "support": 33500.0
-      },
-      "eval_runtime": 1.3821,
-      "eval_samples_per_second": 58.608,
-      "eval_steps_per_second": 7.959,
+        "f1-score": 0.9088773646859957,
+        "precision": 0.9088990283364038,
+        "recall": 0.9126821065309684,
+        "support": 27619.0
+      },
+      "eval_runtime": 1.3397,
+      "eval_samples_per_second": 59.713,
+      "eval_steps_per_second": 7.464,
       "eval_weighted avg": {
-        "f1-score": 0.9418804564369516,
-        "precision": 0.9419132595627793,
-        "recall": 0.941910447761194,
-        "support": 33500.0
+        "f1-score": 0.9337028102359982,
+        "precision": 0.9362389122621345,
+        "recall": 0.9345740251276295,
+        "support": 27619.0
       },
       "step": 369
     },
     {
       "epoch": 10.0,
       "eval_B": {
-        "f1-score": 0.8820989704417137,
-        "precision": 0.8453214513049013,
-        "recall": 0.9222222222222223,
-        "support": 1440.0
+        "f1-score": 0.8675496688741721,
+        "precision": 0.8562091503267973,
+        "recall": 0.8791946308724832,
+        "support": 1043.0
       },
       "eval_I": {
-        "f1-score": 0.9598799630655587,
-        "precision": 0.956655776929094,
-        "recall": 0.9631259554361421,
-        "support": 21587.0
+        "f1-score": 0.9541242937853106,
+        "precision": 0.9356232686980609,
+        "recall": 0.973371757925072,
+        "support": 17350.0
       },
       "eval_O": {
-        "f1-score": 0.914896705210702,
-        "precision": 0.9273244409572381,
-        "recall": 0.9027976701995608,
-        "support": 10473.0
+        "f1-score": 0.9069058903182126,
+        "precision": 0.9457519416333255,
+        "recall": 0.8711250812920008,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.9425074626865672,
-      "eval_loss": 0.27435505390167236,
+      "eval_accuracy": 0.935660233896955,
+      "eval_loss": 0.2737439274787903,
       "eval_macro avg": {
-        "f1-score": 0.9189585462393248,
-        "precision": 0.9097672230637445,
-        "recall": 0.9293819492859751,
-        "support": 33500.0
-      },
-      "eval_runtime": 1.375,
-      "eval_samples_per_second": 58.909,
-      "eval_steps_per_second": 8.0,
+        "f1-score": 0.9095266176592318,
+        "precision": 0.9125281202193946,
+        "recall": 0.9078971566965187,
+        "support": 27619.0
+      },
+      "eval_runtime": 1.3436,
+      "eval_samples_per_second": 59.54,
+      "eval_steps_per_second": 7.443,
       "eval_weighted avg": {
-        "f1-score": 0.9424735663822079,
-        "precision": 0.9427002990027631,
-        "recall": 0.9425074626865672,
-        "support": 33500.0
+        "f1-score": 0.9350818112852286,
+        "precision": 0.9360077218295835,
+        "recall": 0.935660233896955,
+        "support": 27619.0
       },
       "step": 410
     },
     {
       "epoch": 11.0,
       "eval_B": {
-        "f1-score": 0.8886590830748018,
-        "precision": 0.8822724161533196,
-        "recall": 0.8951388888888889,
-        "support": 1440.0
+        "f1-score": 0.8654028436018957,
+        "precision": 0.8556701030927835,
+        "recall": 0.87535953978907,
+        "support": 1043.0
       },
       "eval_I": {
-        "f1-score": 0.9554279336882978,
-        "precision": 0.9591074596209505,
-        "recall": 0.9517765321721406,
-        "support": 21587.0
+        "f1-score": 0.9553462854557281,
+        "precision": 0.9378157792460163,
+        "recall": 0.9735446685878962,
+        "support": 17350.0
       },
       "eval_O": {
-        "f1-score": 0.9066856330014225,
-        "precision": 0.9005368748233964,
-        "recall": 0.91291893440275,
-        "support": 10473.0
+        "f1-score": 0.9094388473011763,
+        "precision": 0.9459079733052336,
+        "recall": 0.8756774333405593,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.9371940298507463,
-      "eval_loss": 0.29653802514076233,
+      "eval_accuracy": 0.9371447192150332,
+      "eval_loss": 0.27220794558525085,
       "eval_macro avg": {
-        "f1-score": 0.916924216588174,
-        "precision": 0.9139722501992221,
-        "recall": 0.9199447851545933,
-        "support": 33500.0
-      },
-      "eval_runtime": 1.3782,
-      "eval_samples_per_second": 58.771,
-      "eval_steps_per_second": 7.981,
+        "f1-score": 0.9100626587862667,
+        "precision": 0.9131312852146779,
+        "recall": 0.9081938805725085,
+        "support": 27619.0
+      },
+      "eval_runtime": 1.3435,
+      "eval_samples_per_second": 59.547,
+      "eval_steps_per_second": 7.443,
       "eval_weighted avg": {
-        "f1-score": 0.9373197169725641,
-        "precision": 0.9374939611977214,
-        "recall": 0.9371940298507463,
-        "support": 33500.0
+        "f1-score": 0.9366145053671137,
+        "precision": 0.937416801808836,
+        "recall": 0.9371447192150332,
+        "support": 27619.0
       },
       "step": 451
     },
     {
       "epoch": 12.0,
       "eval_B": {
-        "f1-score": 0.8851788756388416,
-        "precision": 0.8688963210702341,
-        "recall": 0.9020833333333333,
-        "support": 1440.0
+        "f1-score": 0.8710900473933649,
+        "precision": 0.8612933458294283,
+        "recall": 0.8811121764141898,
+        "support": 1043.0
       },
       "eval_I": {
-        "f1-score": 0.956586582154592,
-        "precision": 0.9624402138235019,
-        "recall": 0.9508037244637977,
-        "support": 21587.0
+        "f1-score": 0.9563688940549427,
+        "precision": 0.9410812921943871,
+        "recall": 0.9721613832853025,
+        "support": 17350.0
       },
       "eval_O": {
-        "f1-score": 0.9096066565809379,
-        "precision": 0.9008334113681056,
-        "recall": 0.9185524682516948,
-        "support": 10473.0
+        "f1-score": 0.9124614953794455,
+        "precision": 0.9440259589755475,
+        "recall": 0.8829395187513549,
+        "support": 9226.0
       },
-      "eval_accuracy": 0.9386268656716418,
-      "eval_loss": 0.3317875564098358,
+      "eval_accuracy": 0.9389188602049314,
+      "eval_loss": 0.27493152022361755,
       "eval_macro avg": {
-        "f1-score": 0.9171240381247904,
-        "precision": 0.9107233154206137,
-        "recall": 0.9238131753496086,
-        "support": 33500.0
-      },
-      "eval_runtime": 1.3866,
-      "eval_samples_per_second": 58.418,
-      "eval_steps_per_second": 7.933,
+        "f1-score": 0.9133068122759177,
+        "precision": 0.9154668656664544,
+        "recall": 0.9120710261502823,
+        "support": 27619.0
+      },
+      "eval_runtime": 1.3368,
+      "eval_samples_per_second": 59.844,
+      "eval_steps_per_second": 7.48,
       "eval_weighted avg": {
-        "f1-score": 0.9388299296795006,
-        "precision": 0.9391592810569325,
-        "recall": 0.9386268656716418,
-        "support": 33500.0
+        "f1-score": 0.938481371072642,
+        "precision": 0.9390518439038746,
+        "recall": 0.9389188602049314,
+        "support": 27619.0
       },
       "step": 492
     }
@@ -482,7 +482,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 16,
   "save_steps": 500,
-  "total_flos": 1720081324029600.0,
+  "total_flos": 1725439832827200.0,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null

checkpoint-492/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9afc9010526d77714317ce08b144ea519027c2bf7e4a39f3188a1467d7fdf528
+oid sha256:aac2e52662466408a70f6e06b318a723873ee5b7ae3588f54b82f465a458153d
 size 4664