Training in progress, epoch 9, checkpoint

checkpoint-729/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3f717aacc8edf06064bfe1f7adf262c9a4a3a4cd5e0723dca3d139e639cf39f
+oid sha256:e4cb71a392cefead6f7be4c24d9b5751ecd3091b01406ccab27107182124ce31
 size 592318676

checkpoint-729/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b9943acac61e958c384f83a55d7be97c4c5fc2ade72d0fe881961c737eeb403d
+oid sha256:484cf65f309c3af6c48483c760039d3010b2f0dfcee4e73bffd25a15a77a4588
 size 1014645498

checkpoint-729/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0968d0d979e679458f54b3ce479d877ab7f754e313e39c8158ce8426e6dbe221
+oid sha256:33e201bb4bf305b6dcac90ce17e6d2f74219c833c6d3bc2a1eb84fd26a97a92a
 size 14244

checkpoint-729/trainer_state.json

@@ -11,358 +11,358 @@
     {
       "epoch": 1.0,
       "eval_B": {
-        "f1-score": 0.8266764382557715,
-        "precision": 0.7866108786610879,
-        "recall": 0.871042471042471,
-        "support": 1295.0
+        "f1-score": 0.7905866302864938,
+        "precision": 0.7768096514745308,
+        "recall": 0.8048611111111111,
+        "support": 1440.0
       },
       "eval_I": {
-        "f1-score": 0.9410658617447671,
-        "precision": 0.9344077040239768,
-        "recall": 0.9478195863443808,
-        "support": 20065.0
+        "f1-score": 0.9005331227169513,
+        "precision": 0.9637593111099372,
+        "recall": 0.8450919534905267,
+        "support": 21587.0
       },
       "eval_O": {
-        "f1-score": 0.8562443302086484,
-        "precision": 0.8789421405512788,
-        "recall": 0.8346893055064262,
-        "support": 8481.0
+        "f1-score": 0.8288892663043479,
+        "precision": 0.7463108800367001,
+        "recall": 0.9320156593144275,
+        "support": 10473.0
       },
-      "eval_accuracy": 0.912335377500754,
-      "eval_loss": 0.2450982630252838,
+      "eval_accuracy": 0.8705373134328358,
+      "eval_loss": 0.32930415868759155,
       "eval_macro avg": {
-        "f1-score": 0.8746622100697291,
-        "precision": 0.8666535744121145,
-        "recall": 0.8845171209644261,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.6269,
-      "eval_samples_per_second": 17.29,
-      "eval_steps_per_second": 2.161,
+        "f1-score": 0.8400030064359308,
+        "precision": 0.8289599475403894,
+        "recall": 0.8606562413053552,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.7349,
+      "eval_samples_per_second": 17.107,
+      "eval_steps_per_second": 2.323,
       "eval_weighted avg": {
-        "f1-score": 0.9119949287205363,
-        "precision": 0.9122301519092053,
-        "recall": 0.912335377500754,
-        "support": 29841.0
+        "f1-score": 0.8734092702599646,
+        "precision": 0.8877430445874,
+        "recall": 0.8705373134328358,
+        "support": 33500.0
       },
       "step": 81
     },
     {
       "epoch": 2.0,
       "eval_B": {
-        "f1-score": 0.8833395452851286,
-        "precision": 0.8537463976945245,
-        "recall": 0.915057915057915,
-        "support": 1295.0
+        "f1-score": 0.8656218739579861,
+        "precision": 0.8325849903784477,
+        "recall": 0.9013888888888889,
+        "support": 1440.0
       },
       "eval_I": {
-        "f1-score": 0.9549518227873249,
-        "precision": 0.9516903430183636,
-        "recall": 0.958235733864939,
-        "support": 20065.0
+        "f1-score": 0.9547168948727912,
+        "precision": 0.9423736462093862,
+        "recall": 0.9673877796822161,
+        "support": 21587.0
       },
       "eval_O": {
-        "f1-score": 0.88924750463212,
-        "precision": 0.9016969696969697,
-        "recall": 0.877137130055418,
-        "support": 8481.0
+        "f1-score": 0.9013528191962081,
+        "precision": 0.9332379102341274,
+        "recall": 0.8715745249689678,
+        "support": 10473.0
       },
-      "eval_accuracy": 0.9333132267685399,
-      "eval_loss": 0.20076116919517517,
+      "eval_accuracy": 0.9345970149253732,
+      "eval_loss": 0.2162298709154129,
       "eval_macro avg": {
-        "f1-score": 0.9091796242348579,
-        "precision": 0.9023779034699526,
-        "recall": 0.916810259659424,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.6524,
-      "eval_samples_per_second": 17.195,
-      "eval_steps_per_second": 2.149,
+        "f1-score": 0.9072305293423284,
+        "precision": 0.902732182273987,
+        "recall": 0.9134503978466909,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.753,
+      "eval_samples_per_second": 17.042,
+      "eval_steps_per_second": 2.314,
       "eval_weighted avg": {
-        "f1-score": 0.9331705077630417,
-        "precision": 0.9332314707173981,
-        "recall": 0.9333132267685399,
-        "support": 29841.0
+        "f1-score": 0.9342040950316517,
+        "precision": 0.9347982961417612,
+        "recall": 0.9345970149253732,
+        "support": 33500.0
       },
       "step": 162
     },
     {
       "epoch": 3.0,
       "eval_B": {
-        "f1-score": 0.8927335640138409,
-        "precision": 0.8889739663093414,
-        "recall": 0.8965250965250965,
-        "support": 1295.0
+        "f1-score": 0.8603587759409076,
+        "precision": 0.8717034925160371,
+        "recall": 0.8493055555555555,
+        "support": 1440.0
       },
       "eval_I": {
-        "f1-score": 0.9590235396687009,
-        "precision": 0.9586653386454184,
-        "recall": 0.9593820084724645,
-        "support": 20065.0
+        "f1-score": 0.9434195218082959,
+        "precision": 0.9677057963955188,
+        "recall": 0.9203224162690509,
+        "support": 21587.0
       },
       "eval_O": {
-        "f1-score": 0.9006849315068494,
-        "precision": 0.9020697811945594,
-        "recall": 0.8993043273198915,
-        "support": 8481.0
+        "f1-score": 0.8879310344827586,
+        "precision": 0.8459410391631365,
+        "recall": 0.9343072663038289,
+        "support": 10473.0
       },
-      "eval_accuracy": 0.9395797727958178,
-      "eval_loss": 0.19863653182983398,
+      "eval_accuracy": 0.9216417910447762,
+      "eval_loss": 0.24082203209400177,
       "eval_macro avg": {
-        "f1-score": 0.9174806783964637,
-        "precision": 0.9165696953831063,
-        "recall": 0.9184038107724842,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.655,
-      "eval_samples_per_second": 17.186,
-      "eval_steps_per_second": 2.148,
+        "f1-score": 0.8972364440773206,
+        "precision": 0.8951167760248975,
+        "recall": 0.9013117460428117,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.7554,
+      "eval_samples_per_second": 17.033,
+      "eval_steps_per_second": 2.313,
       "eval_weighted avg": {
-        "f1-score": 0.9395665759512081,
-        "precision": 0.9395561516236713,
-        "recall": 0.9395797727958178,
-        "support": 29841.0
+        "f1-score": 0.9225019575751797,
+        "precision": 0.9255121957960801,
+        "recall": 0.9216417910447762,
+        "support": 33500.0
       },
       "step": 243
     },
     {
       "epoch": 4.0,
       "eval_B": {
-        "f1-score": 0.8999616711383672,
-        "precision": 0.893455098934551,
-        "recall": 0.9065637065637066,
-        "support": 1295.0
+        "f1-score": 0.8804020100502512,
+        "precision": 0.8504854368932039,
+        "recall": 0.9125,
+        "support": 1440.0
       },
       "eval_I": {
-        "f1-score": 0.9612657113512985,
-        "precision": 0.9436362763454799,
-        "recall": 0.9795664091701969,
-        "support": 20065.0
+        "f1-score": 0.9589184992350025,
+        "precision": 0.945595388218339,
+        "recall": 0.9726224116366332,
+        "support": 21587.0
       },
       "eval_O": {
-        "f1-score": 0.8996847765622102,
-        "precision": 0.9454403741231489,
-        "recall": 0.8581535196321188,
-        "support": 8481.0
+        "f1-score": 0.9091178797468354,
+        "precision": 0.9427750999897446,
+        "recall": 0.877780960565263,
+        "support": 10473.0
       },
-      "eval_accuracy": 0.9418920277470594,
-      "eval_loss": 0.19743724167346954,
+      "eval_accuracy": 0.9403880597014925,
+      "eval_loss": 0.22083505988121033,
       "eval_macro avg": {
-        "f1-score": 0.920304053017292,
-        "precision": 0.9275105831343932,
-        "recall": 0.9147612117886741,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.6615,
-      "eval_samples_per_second": 17.162,
-      "eval_steps_per_second": 2.145,
+        "f1-score": 0.9161461296773631,
+        "precision": 0.9129519750337626,
+        "recall": 0.9209677907339654,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.7449,
+      "eval_samples_per_second": 17.071,
+      "eval_steps_per_second": 2.318,
       "eval_weighted avg": {
-        "f1-score": 0.9411036309913238,
-        "precision": 0.9419713163409645,
-        "recall": 0.9418920277470594,
-        "support": 29841.0
+        "f1-score": 0.9399744505088647,
+        "precision": 0.9406253819936745,
+        "recall": 0.9403880597014925,
+        "support": 33500.0
       },
       "step": 324
     },
     {
       "epoch": 5.0,
       "eval_B": {
-        "f1-score": 0.8953846153846153,
-        "precision": 0.8919540229885058,
-        "recall": 0.8988416988416988,
-        "support": 1295.0
+        "f1-score": 0.8879781420765027,
+        "precision": 0.8736559139784946,
+        "recall": 0.9027777777777778,
+        "support": 1440.0
       },
       "eval_I": {
-        "f1-score": 0.9602426713742261,
-        "precision": 0.9581224570804803,
-        "recall": 0.9623722900573137,
-        "support": 20065.0
+        "f1-score": 0.9613201259451634,
+        "precision": 0.9539317642765919,
+        "recall": 0.9688238291564367,
+        "support": 21587.0
       },
       "eval_O": {
-        "f1-score": 0.9022119433078337,
-        "precision": 0.9075399665950847,
-        "recall": 0.8969461148449476,
-        "support": 8481.0
+        "f1-score": 0.916006030835076,
+        "precision": 0.9334853291038858,
+        "recall": 0.899169292466342,
+        "support": 10473.0
       },
-      "eval_accuracy": 0.9410207432726785,
-      "eval_loss": 0.2340461015701294,
+      "eval_accuracy": 0.9442089552238806,
+      "eval_loss": 0.2257838249206543,
       "eval_macro avg": {
-        "f1-score": 0.9192797433555584,
-        "precision": 0.9192054822213569,
-        "recall": 0.9193867012479867,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.6719,
-      "eval_samples_per_second": 17.124,
-      "eval_steps_per_second": 2.14,
+        "f1-score": 0.921768099618914,
+        "precision": 0.9203576691196576,
+        "recall": 0.9235902998001855,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.7845,
+      "eval_samples_per_second": 16.93,
+      "eval_steps_per_second": 2.299,
       "eval_weighted avg": {
-        "f1-score": 0.9409353496612266,
-        "precision": 0.9408751053176122,
-        "recall": 0.9410207432726785,
-        "support": 29841.0
+        "f1-score": 0.9440011117702735,
+        "precision": 0.9440889959185916,
+        "recall": 0.9442089552238806,
+        "support": 33500.0
       },
       "step": 405
     },
     {
       "epoch": 6.0,
       "eval_B": {
-        "f1-score": 0.8929831438651511,
-        "precision": 0.9068471337579618,
-        "recall": 0.8795366795366796,
-        "support": 1295.0
+        "f1-score": 0.8873579056148813,
+        "precision": 0.8803827751196173,
+        "recall": 0.8944444444444445,
+        "support": 1440.0
       },
       "eval_I": {
-        "f1-score": 0.9522293798155866,
-        "precision": 0.9655225409836066,
-        "recall": 0.9392972838275604,
-        "support": 20065.0
+        "f1-score": 0.9622060984570168,
+        "precision": 0.9539267015706806,
+        "recall": 0.9706304720433594,
+        "support": 21587.0
       },
       "eval_O": {
-        "f1-score": 0.8885216003647556,
-        "precision": 0.8599007170435742,
-        "recall": 0.919113312109421,
-        "support": 8481.0
+        "f1-score": 0.9182769530299342,
+        "precision": 0.9365567911040509,
+        "recall": 0.9006970304592762,
+        "support": 10473.0
       },
-      "eval_accuracy": 0.930967460875976,
-      "eval_loss": 0.21774999797344208,
+      "eval_accuracy": 0.9454925373134329,
+      "eval_loss": 0.2242102175951004,
       "eval_macro avg": {
-        "f1-score": 0.9112447080151643,
-        "precision": 0.9107567972617142,
-        "recall": 0.9126490918245537,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.647,
-      "eval_samples_per_second": 17.215,
-      "eval_steps_per_second": 2.152,
+        "f1-score": 0.9226136523672773,
+        "precision": 0.923622089264783,
+        "recall": 0.9219239823156933,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.7715,
+      "eval_samples_per_second": 16.976,
+      "eval_steps_per_second": 2.305,
       "eval_weighted avg": {
-        "f1-score": 0.9315521386682287,
-        "precision": 0.9329578366776978,
-        "recall": 0.930967460875976,
-        "support": 29841.0
+        "f1-score": 0.945255312255509,
+        "precision": 0.9453351097376493,
+        "recall": 0.9454925373134329,
+        "support": 33500.0
       },
       "step": 486
     },
     {
       "epoch": 6.17,
-      "grad_norm": 9.310327529907227,
+      "grad_norm": 1.8705081939697266,
       "learning_rate": 1.7530864197530865e-05,
-      "loss": 0.1627,
+      "loss": 0.1601,
       "step": 500
     },
     {
       "epoch": 7.0,
       "eval_B": {
-        "f1-score": 0.8936490041337843,
-        "precision": 0.8704245973645681,
-        "recall": 0.9181467181467181,
-        "support": 1295.0
+        "f1-score": 0.8876749743939911,
+        "precision": 0.8730691739422431,
+        "recall": 0.9027777777777778,
+        "support": 1440.0
       },
       "eval_I": {
-        "f1-score": 0.9599271850630982,
-        "precision": 0.9477800446905664,
-        "recall": 0.9723897333665587,
-        "support": 20065.0
+        "f1-score": 0.9611415113943341,
+        "precision": 0.9581108451482232,
+        "recall": 0.9641914114976606,
+        "support": 21587.0
       },
       "eval_O": {
-        "f1-score": 0.8979841172877214,
-        "precision": 0.9316770186335404,
-        "recall": 0.8666430845419172,
-        "support": 8481.0
+        "f1-score": 0.9164739884393064,
+        "precision": 0.9247594050743657,
+        "recall": 0.9083357204239473,
+        "support": 10473.0
       },
-      "eval_accuracy": 0.9399819040916859,
-      "eval_loss": 0.27376389503479004,
+      "eval_accuracy": 0.944089552238806,
+      "eval_loss": 0.22517083585262299,
       "eval_macro avg": {
-        "f1-score": 0.9171867688282013,
-        "precision": 0.9166272202295582,
-        "recall": 0.9190598453517312,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.6671,
-      "eval_samples_per_second": 17.141,
-      "eval_steps_per_second": 2.143,
+        "f1-score": 0.9217634914092105,
+        "precision": 0.9186464747216107,
+        "recall": 0.9251016365664619,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.7926,
+      "eval_samples_per_second": 16.901,
+      "eval_steps_per_second": 2.295,
       "eval_weighted avg": {
-        "f1-score": 0.9394463230910989,
-        "precision": 0.9398464945991887,
-        "recall": 0.9399819040916859,
-        "support": 29841.0
+        "f1-score": 0.9440192791200506,
+        "precision": 0.9440287663891153,
+        "recall": 0.944089552238806,
+        "support": 33500.0
       },
       "step": 567
     },
     {
       "epoch": 8.0,
       "eval_B": {
-        "f1-score": 0.8948545861297539,
-        "precision": 0.8651766402307137,
-        "recall": 0.9266409266409267,
-        "support": 1295.0
+        "f1-score": 0.894540522210919,
+        "precision": 0.8740888005301524,
+        "recall": 0.9159722222222222,
+        "support": 1440.0
       },
       "eval_I": {
-        "f1-score": 0.9590139880661254,
-        "precision": 0.9414702069429106,
-        "recall": 0.9772240219287316,
-        "support": 20065.0
+        "f1-score": 0.9638018385639943,
+        "precision": 0.9540255968049378,
+        "recall": 0.9737805160513272,
+        "support": 21587.0
       },
       "eval_O": {
-        "f1-score": 0.8927241122423641,
-        "precision": 0.9427035531663825,
-        "recall": 0.8477773847423653,
-        "support": 8481.0
+        "f1-score": 0.9203132648066569,
+        "precision": 0.9441598875163202,
+        "recall": 0.8976415544734078,
+        "support": 10473.0
       },
-      "eval_accuracy": 0.9382393351429241,
-      "eval_loss": 0.3103167414665222,
+      "eval_accuracy": 0.9474925373134329,
+      "eval_loss": 0.24869142472743988,
       "eval_macro avg": {
-        "f1-score": 0.9155308954794145,
-        "precision": 0.916450133446669,
-        "recall": 0.9172141111040079,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.6339,
-      "eval_samples_per_second": 17.264,
-      "eval_steps_per_second": 2.158,
+        "f1-score": 0.9262185418605234,
+        "precision": 0.9240914282838034,
+        "recall": 0.9291314309156524,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.7743,
+      "eval_samples_per_second": 16.966,
+      "eval_steps_per_second": 2.304,
       "eval_weighted avg": {
-        "f1-score": 0.9373896838414373,
-        "precision": 0.9385098450391195,
-        "recall": 0.9382393351429241,
-        "support": 29841.0
+        "f1-score": 0.947228939205516,
+        "precision": 0.9475052218791054,
+        "recall": 0.9474925373134329,
+        "support": 33500.0
       },
       "step": 648
     },
     {
       "epoch": 9.0,
       "eval_B": {
-        "f1-score": 0.9011235955056179,
-        "precision": 0.8749090909090909,
-        "recall": 0.9289575289575289,
-        "support": 1295.0
+        "f1-score": 0.8851236868858015,
+        "precision": 0.8643282594308405,
+        "recall": 0.9069444444444444,
+        "support": 1440.0
       },
       "eval_I": {
-        "f1-score": 0.9615119937082186,
-        "precision": 0.9485040973670175,
-        "recall": 0.9748816346872664,
-        "support": 20065.0
+        "f1-score": 0.9617001723147618,
+        "precision": 0.9540067462849849,
+        "recall": 0.9695186918052532,
+        "support": 21587.0
       },
       "eval_O": {
-        "f1-score": 0.9013722126929674,
-        "precision": 0.9380339155935229,
-        "recall": 0.8674684589081476,
-        "support": 8481.0
+        "f1-score": 0.9165854609237966,
+        "precision": 0.9358272808675754,
+        "recall": 0.8981189725961998,
+        "support": 10473.0
       },
-      "eval_accuracy": 0.9423611809255722,
-      "eval_loss": 0.32883673906326294,
+      "eval_accuracy": 0.9445074626865672,
+      "eval_loss": 0.3045330345630646,
       "eval_macro avg": {
-        "f1-score": 0.9213359339689347,
-        "precision": 0.9204823679565438,
-        "recall": 0.9237692075176476,
-        "support": 29841.0
-      },
-      "eval_runtime": 4.6714,
-      "eval_samples_per_second": 17.126,
-      "eval_steps_per_second": 2.141,
+        "f1-score": 0.9211364400414532,
+        "precision": 0.9180540955278002,
+        "recall": 0.9248607029486324,
+        "support": 33500.0
+      },
+      "eval_runtime": 4.7599,
+      "eval_samples_per_second": 17.017,
+      "eval_steps_per_second": 2.311,
       "eval_weighted avg": {
-        "f1-score": 0.9417992341337168,
-        "precision": 0.9423346276781992,
-        "recall": 0.9423611809255722,
-        "support": 29841.0
+        "f1-score": 0.9443044555560967,
+        "precision": 0.9444685205421044,
+        "recall": 0.9445074626865672,
+        "support": 33500.0
       },
       "step": 729
     }
@@ -372,7 +372,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 1294079874620400.0,
+  "total_flos": 1290060993022200.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null

checkpoint-729/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d229d7bc9127d222c880edfd3fa9418a9ccede656fe27f7f0aa04990c9605be4
+oid sha256:b8ca363922308e3892528dd031742ffedc372734b464a042d09444af9700901e
 size 4856