Training in progress, step 900, checkpoint

checkpoint-900/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c3f0b1a0547175c19a1680ce564322547ef5e2733b885a76994c0c58f99c4522
+oid sha256:1b36c3689596a89a01d421d33d9935d3cadd42d25fef48b7d724429a9ce114ef
 size 14244

checkpoint-900/trainer_state.json

@@ -23,10 +23,14 @@
     {
       "epoch": 0.06,
       "eval_accuracy": 0.44479243019924036,
+      "eval_combined_score": 0.3403229798934559,
+      "eval_f1": 0.2738667532127967,
       "eval_loss": 1.065091609954834,
-      "eval_runtime": 50.348,
-      "eval_samples_per_second": 596.131,
-      "eval_steps_per_second": 1.172,
+      "eval_precision": 0.19784030596254612,
+      "eval_recall": 0.44479243019924036,
+      "eval_runtime": 49.6621,
+      "eval_samples_per_second": 604.364,
+      "eval_steps_per_second": 1.188,
       "step": 20
     },
     {
@@ -44,10 +48,14 @@
     {
       "epoch": 0.12,
       "eval_accuracy": 0.5033650962883988,
+      "eval_combined_score": 0.4633375988688383,
+      "eval_f1": 0.4137746858205889,
       "eval_loss": 1.0188277959823608,
-      "eval_runtime": 49.614,
-      "eval_samples_per_second": 604.95,
-      "eval_steps_per_second": 1.189,
+      "eval_precision": 0.43284551707796665,
+      "eval_recall": 0.5033650962883988,
+      "eval_runtime": 49.1985,
+      "eval_samples_per_second": 610.06,
+      "eval_steps_per_second": 1.199,
       "step": 40
     },
     {
@@ -65,10 +73,14 @@
     {
       "epoch": 0.18,
       "eval_accuracy": 0.5279203038581995,
+      "eval_combined_score": 0.48433586826626607,
+      "eval_f1": 0.4596383010419703,
       "eval_loss": 0.9871189594268799,
-      "eval_runtime": 49.5201,
-      "eval_samples_per_second": 606.098,
-      "eval_steps_per_second": 1.191,
+      "eval_precision": 0.421864564306695,
+      "eval_recall": 0.5279203038581995,
+      "eval_runtime": 49.0237,
+      "eval_samples_per_second": 612.235,
+      "eval_steps_per_second": 1.204,
       "step": 60
     },
     {
@@ -86,10 +98,14 @@
     {
       "epoch": 0.24,
       "eval_accuracy": 0.5308189511561271,
+      "eval_combined_score": 0.4866197693698143,
+      "eval_f1": 0.46528335920671143,
       "eval_loss": 0.9888613224029541,
-      "eval_runtime": 49.4469,
-      "eval_samples_per_second": 606.994,
-      "eval_steps_per_second": 1.193,
+      "eval_precision": 0.4195578159602916,
+      "eval_recall": 0.5308189511561271,
+      "eval_runtime": 49.518,
+      "eval_samples_per_second": 606.124,
+      "eval_steps_per_second": 1.191,
       "step": 80
     },
     {
@@ -107,10 +123,14 @@
     {
       "epoch": 0.3,
       "eval_accuracy": 0.5307856333710935,
+      "eval_combined_score": 0.48644275806230897,
+      "eval_f1": 0.4666072639999521,
       "eval_loss": 0.9762536883354187,
-      "eval_runtime": 49.4804,
-      "eval_samples_per_second": 606.583,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.4175925015070968,
+      "eval_recall": 0.5307856333710935,
+      "eval_runtime": 49.4629,
+      "eval_samples_per_second": 606.798,
+      "eval_steps_per_second": 1.193,
       "step": 100
     },
     {
@@ -128,10 +148,14 @@
     {
       "epoch": 0.36,
       "eval_accuracy": 0.5387819017791697,
+      "eval_combined_score": 0.4944622623129249,
+      "eval_f1": 0.47105988260831677,
       "eval_loss": 0.9713281989097595,
-      "eval_runtime": 49.4983,
-      "eval_samples_per_second": 606.364,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.4292253630850433,
+      "eval_recall": 0.5387819017791697,
+      "eval_runtime": 49.0682,
+      "eval_samples_per_second": 611.68,
+      "eval_steps_per_second": 1.202,
       "step": 120
     },
     {
@@ -149,10 +173,14 @@
     {
       "epoch": 0.42,
       "eval_accuracy": 0.5312520823615646,
+      "eval_combined_score": 0.48712844254868687,
+      "eval_f1": 0.4673598197970026,
       "eval_loss": 0.9766249656677246,
-      "eval_runtime": 49.5213,
-      "eval_samples_per_second": 606.083,
-      "eval_steps_per_second": 1.191,
+      "eval_precision": 0.41864978567461564,
+      "eval_recall": 0.5312520823615646,
+      "eval_runtime": 49.1859,
+      "eval_samples_per_second": 610.215,
+      "eval_steps_per_second": 1.2,
       "step": 140
     },
     {
@@ -170,10 +198,14 @@
     {
       "epoch": 0.48,
       "eval_accuracy": 0.5398147531152129,
+      "eval_combined_score": 0.49477075635813,
+      "eval_f1": 0.4751263005883661,
       "eval_loss": 0.9589501619338989,
-      "eval_runtime": 49.4977,
-      "eval_samples_per_second": 606.372,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.4243272186137281,
+      "eval_recall": 0.5398147531152129,
+      "eval_runtime": 49.0847,
+      "eval_samples_per_second": 611.473,
+      "eval_steps_per_second": 1.202,
       "step": 160
     },
     {
@@ -191,10 +223,14 @@
     {
       "epoch": 0.54,
       "eval_accuracy": 0.5423469047777704,
+      "eval_combined_score": 0.49718028497336725,
+      "eval_f1": 0.4771728160733735,
       "eval_loss": 0.953514814376831,
-      "eval_runtime": 49.4511,
-      "eval_samples_per_second": 606.943,
-      "eval_steps_per_second": 1.193,
+      "eval_precision": 0.42685451426455484,
+      "eval_recall": 0.5423469047777704,
+      "eval_runtime": 49.0032,
+      "eval_samples_per_second": 612.491,
+      "eval_steps_per_second": 1.204,
       "step": 180
     },
     {
@@ -212,10 +248,14 @@
     {
       "epoch": 0.6,
       "eval_accuracy": 0.567201972412874,
+      "eval_combined_score": 0.5200447629299639,
+      "eval_f1": 0.49911443945569844,
       "eval_loss": 0.9273685812950134,
-      "eval_runtime": 49.4122,
-      "eval_samples_per_second": 607.421,
-      "eval_steps_per_second": 1.194,
+      "eval_precision": 0.44666066743840943,
+      "eval_recall": 0.567201972412874,
+      "eval_runtime": 49.1912,
+      "eval_samples_per_second": 610.15,
+      "eval_steps_per_second": 1.199,
       "step": 200
     },
     {
@@ -233,10 +273,14 @@
     {
       "epoch": 0.66,
       "eval_accuracy": 0.573598987139335,
+      "eval_combined_score": 0.5278568904198743,
+      "eval_f1": 0.5026102591641352,
       "eval_loss": 0.912590503692627,
-      "eval_runtime": 49.2827,
-      "eval_samples_per_second": 609.017,
-      "eval_steps_per_second": 1.197,
+      "eval_precision": 0.4616193282366919,
+      "eval_recall": 0.573598987139335,
+      "eval_runtime": 49.2154,
+      "eval_samples_per_second": 609.849,
+      "eval_steps_per_second": 1.199,
       "step": 220
     },
     {
@@ -254,10 +298,14 @@
     {
       "epoch": 0.72,
       "eval_accuracy": 0.5759645498767242,
+      "eval_combined_score": 0.5280177627115059,
+      "eval_f1": 0.5069205876947673,
       "eval_loss": 0.9052607417106628,
-      "eval_runtime": 49.4457,
-      "eval_samples_per_second": 607.01,
-      "eval_steps_per_second": 1.193,
+      "eval_precision": 0.45322136339780783,
+      "eval_recall": 0.5759645498767242,
+      "eval_runtime": 49.1399,
+      "eval_samples_per_second": 610.787,
+      "eval_steps_per_second": 1.201,
       "step": 240
     },
     {
@@ -275,10 +323,14 @@
     {
       "epoch": 0.78,
       "eval_accuracy": 0.5766975411474645,
+      "eval_combined_score": 0.534067292009198,
+      "eval_f1": 0.5018221478654629,
       "eval_loss": 0.9178985953330994,
-      "eval_runtime": 49.3132,
-      "eval_samples_per_second": 608.641,
-      "eval_steps_per_second": 1.196,
+      "eval_precision": 0.4810519378764,
+      "eval_recall": 0.5766975411474645,
+      "eval_runtime": 49.2092,
+      "eval_samples_per_second": 609.927,
+      "eval_steps_per_second": 1.199,
       "step": 260
     },
     {
@@ -296,10 +348,14 @@
     {
       "epoch": 0.84,
       "eval_accuracy": 0.58915839275005,
+      "eval_combined_score": 0.5407092882328752,
+      "eval_f1": 0.5183161389995846,
       "eval_loss": 0.8937407732009888,
-      "eval_runtime": 49.4342,
-      "eval_samples_per_second": 607.15,
-      "eval_steps_per_second": 1.194,
+      "eval_precision": 0.46620422843181647,
+      "eval_recall": 0.58915839275005,
+      "eval_runtime": 49.2523,
+      "eval_samples_per_second": 609.393,
+      "eval_steps_per_second": 1.198,
       "step": 280
     },
     {
@@ -317,10 +373,14 @@
     {
       "epoch": 0.9,
       "eval_accuracy": 0.613013926834144,
+      "eval_combined_score": 0.5923759012239074,
+      "eval_f1": 0.5579282363395321,
       "eval_loss": 0.8468813300132751,
-      "eval_runtime": 49.5046,
-      "eval_samples_per_second": 606.287,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.5855475148878095,
+      "eval_recall": 0.613013926834144,
+      "eval_runtime": 49.1896,
+      "eval_samples_per_second": 610.17,
+      "eval_steps_per_second": 1.199,
       "step": 300
     },
     {
@@ -338,10 +398,14 @@
     {
       "epoch": 0.96,
       "eval_accuracy": 0.6046511627906976,
+      "eval_combined_score": 0.590510828847508,
+      "eval_f1": 0.5352474275558012,
       "eval_loss": 0.8615403771400452,
-      "eval_runtime": 49.5041,
-      "eval_samples_per_second": 606.294,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.6174935622528357,
+      "eval_recall": 0.6046511627906976,
+      "eval_runtime": 49.37,
+      "eval_samples_per_second": 607.94,
+      "eval_steps_per_second": 1.195,
       "step": 320
     },
     {
@@ -359,10 +423,14 @@
     {
       "epoch": 1.02,
       "eval_accuracy": 0.6439328313453722,
+      "eval_combined_score": 0.6332615383344228,
+      "eval_f1": 0.6212200513859825,
       "eval_loss": 0.7896137237548828,
-      "eval_runtime": 49.4448,
-      "eval_samples_per_second": 607.02,
-      "eval_steps_per_second": 1.193,
+      "eval_precision": 0.6239604392609642,
+      "eval_recall": 0.6439328313453722,
+      "eval_runtime": 49.1146,
+      "eval_samples_per_second": 611.102,
+      "eval_steps_per_second": 1.201,
       "step": 340
     },
     {
@@ -380,10 +448,14 @@
     {
       "epoch": 1.08,
       "eval_accuracy": 0.5315852602119011,
+      "eval_combined_score": 0.5500800253024862,
+      "eval_f1": 0.4943799031015135,
       "eval_loss": 1.0123510360717773,
-      "eval_runtime": 49.4838,
-      "eval_samples_per_second": 606.542,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.642769677684629,
+      "eval_recall": 0.5315852602119011,
+      "eval_runtime": 49.1941,
+      "eval_samples_per_second": 610.114,
+      "eval_steps_per_second": 1.199,
       "step": 360
     },
     {
@@ -401,10 +473,14 @@
     {
       "epoch": 1.14,
       "eval_accuracy": 0.6488638635303525,
+      "eval_combined_score": 0.648329789222037,
+      "eval_f1": 0.6371420074616733,
       "eval_loss": 0.791083574295044,
-      "eval_runtime": 49.4701,
-      "eval_samples_per_second": 606.71,
-      "eval_steps_per_second": 1.193,
+      "eval_precision": 0.6584494223657698,
+      "eval_recall": 0.6488638635303525,
+      "eval_runtime": 49.3484,
+      "eval_samples_per_second": 608.207,
+      "eval_steps_per_second": 1.196,
       "step": 380
     },
     {
@@ -422,10 +498,14 @@
     {
       "epoch": 1.2,
       "eval_accuracy": 0.6699873392416872,
+      "eval_combined_score": 0.6621103162794577,
+      "eval_f1": 0.6459453754294268,
       "eval_loss": 0.7472424507141113,
-      "eval_runtime": 49.4938,
-      "eval_samples_per_second": 606.419,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.6625212112050299,
+      "eval_recall": 0.6699873392416872,
+      "eval_runtime": 49.3153,
+      "eval_samples_per_second": 608.614,
+      "eval_steps_per_second": 1.196,
       "step": 400
     },
     {
@@ -443,10 +523,14 @@
     {
       "epoch": 1.26,
       "eval_accuracy": 0.6580595721996402,
+      "eval_combined_score": 0.6513132521613654,
+      "eval_f1": 0.6095365334747936,
       "eval_loss": 0.7663838863372803,
-      "eval_runtime": 49.5064,
-      "eval_samples_per_second": 606.265,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.6795973307713873,
+      "eval_recall": 0.6580595721996402,
+      "eval_runtime": 49.2143,
+      "eval_samples_per_second": 609.863,
+      "eval_steps_per_second": 1.199,
       "step": 420
     },
     {
@@ -464,10 +548,14 @@
     {
       "epoch": 1.32,
       "eval_accuracy": 0.6991737189311654,
+      "eval_combined_score": 0.6905198116505921,
+      "eval_f1": 0.653333567856404,
       "eval_loss": 0.6993714570999146,
-      "eval_runtime": 49.4888,
-      "eval_samples_per_second": 606.481,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.7103982408836337,
+      "eval_recall": 0.6991737189311654,
+      "eval_runtime": 49.1465,
+      "eval_samples_per_second": 610.704,
+      "eval_steps_per_second": 1.2,
       "step": 440
     },
     {
@@ -485,10 +573,14 @@
     {
       "epoch": 1.38,
       "eval_accuracy": 0.7283600986206437,
+      "eval_combined_score": 0.7200537958840363,
+      "eval_f1": 0.6890059409106147,
       "eval_loss": 0.6510820984840393,
-      "eval_runtime": 49.5999,
-      "eval_samples_per_second": 605.122,
-      "eval_steps_per_second": 1.19,
+      "eval_precision": 0.7344890453842432,
+      "eval_recall": 0.7283600986206437,
+      "eval_runtime": 49.0446,
+      "eval_samples_per_second": 611.974,
+      "eval_steps_per_second": 1.203,
       "step": 460
     },
     {
@@ -506,10 +598,14 @@
     {
       "epoch": 1.44,
       "eval_accuracy": 0.7576797494502565,
+      "eval_combined_score": 0.7514829004826655,
+      "eval_f1": 0.7336051419487455,
       "eval_loss": 0.5987845063209534,
-      "eval_runtime": 49.5123,
-      "eval_samples_per_second": 606.192,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.7569669610814035,
+      "eval_recall": 0.7576797494502565,
+      "eval_runtime": 49.3939,
+      "eval_samples_per_second": 607.646,
+      "eval_steps_per_second": 1.194,
       "step": 480
     },
     {
@@ -527,10 +623,14 @@
     {
       "epoch": 1.5,
       "eval_accuracy": 0.7564136736189778,
+      "eval_combined_score": 0.7674838523650971,
+      "eval_f1": 0.766902713706533,
       "eval_loss": 0.5992804765701294,
-      "eval_runtime": 49.5641,
-      "eval_samples_per_second": 605.56,
-      "eval_steps_per_second": 1.19,
+      "eval_precision": 0.7902053485158999,
+      "eval_recall": 0.7564136736189778,
+      "eval_runtime": 49.1622,
+      "eval_samples_per_second": 610.509,
+      "eval_steps_per_second": 1.2,
       "step": 500
     },
     {
@@ -548,10 +648,14 @@
     {
       "epoch": 1.57,
       "eval_accuracy": 0.8125874591857133,
+      "eval_combined_score": 0.8104607682202668,
+      "eval_f1": 0.7941544044601274,
       "eval_loss": 0.5067983865737915,
-      "eval_runtime": 49.3712,
-      "eval_samples_per_second": 607.925,
-      "eval_steps_per_second": 1.195,
+      "eval_precision": 0.8225137500495135,
+      "eval_recall": 0.8125874591857133,
+      "eval_runtime": 49.0571,
+      "eval_samples_per_second": 611.817,
+      "eval_steps_per_second": 1.203,
       "step": 520
     },
     {
@@ -569,10 +673,14 @@
     {
       "epoch": 1.63,
       "eval_accuracy": 0.8520023988805224,
+      "eval_combined_score": 0.8500052190439381,
+      "eval_f1": 0.8448837390352126,
       "eval_loss": 0.42726200819015503,
-      "eval_runtime": 49.3448,
-      "eval_samples_per_second": 608.251,
-      "eval_steps_per_second": 1.196,
+      "eval_precision": 0.8511323393794948,
+      "eval_recall": 0.8520023988805224,
+      "eval_runtime": 49.0407,
+      "eval_samples_per_second": 612.023,
+      "eval_steps_per_second": 1.203,
       "step": 540
     },
     {
@@ -590,10 +698,14 @@
     {
       "epoch": 1.69,
       "eval_accuracy": 0.8721929766109149,
+      "eval_combined_score": 0.8711433017214012,
+      "eval_f1": 0.8698654313719106,
       "eval_loss": 0.37960025668144226,
-      "eval_runtime": 49.2893,
-      "eval_samples_per_second": 608.936,
-      "eval_steps_per_second": 1.197,
+      "eval_precision": 0.8703218222918644,
+      "eval_recall": 0.8721929766109149,
+      "eval_runtime": 49.6519,
+      "eval_samples_per_second": 604.489,
+      "eval_steps_per_second": 1.188,
       "step": 560
     },
     {
@@ -611,10 +723,14 @@
     {
       "epoch": 1.75,
       "eval_accuracy": 0.8756580262544146,
+      "eval_combined_score": 0.876232023467392,
+      "eval_f1": 0.8758457824110393,
       "eval_loss": 0.3854842483997345,
-      "eval_runtime": 49.5207,
-      "eval_samples_per_second": 606.09,
-      "eval_steps_per_second": 1.191,
+      "eval_precision": 0.8777662589496996,
+      "eval_recall": 0.8756580262544146,
+      "eval_runtime": 49.2229,
+      "eval_samples_per_second": 609.757,
+      "eval_steps_per_second": 1.199,
       "step": 580
     },
     {
@@ -632,10 +748,14 @@
     {
       "epoch": 1.81,
       "eval_accuracy": 0.8893849536882789,
+      "eval_combined_score": 0.8889004932680924,
+      "eval_f1": 0.8883136621852403,
       "eval_loss": 0.35855188965797424,
-      "eval_runtime": 49.3852,
-      "eval_samples_per_second": 607.753,
-      "eval_steps_per_second": 1.195,
+      "eval_precision": 0.8885184035105714,
+      "eval_recall": 0.8893849536882789,
+      "eval_runtime": 49.2611,
+      "eval_samples_per_second": 609.285,
+      "eval_steps_per_second": 1.198,
       "step": 600
     },
     {
@@ -653,10 +773,14 @@
     {
       "epoch": 1.87,
       "eval_accuracy": 0.8971813153861531,
+      "eval_combined_score": 0.8972621407363919,
+      "eval_f1": 0.8972738562660447,
       "eval_loss": 0.321043998003006,
-      "eval_runtime": 49.4982,
-      "eval_samples_per_second": 606.366,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.8974120759072168,
+      "eval_recall": 0.8971813153861531,
+      "eval_runtime": 49.4334,
+      "eval_samples_per_second": 607.16,
+      "eval_steps_per_second": 1.194,
       "step": 620
     },
     {
@@ -674,10 +798,14 @@
     {
       "epoch": 1.93,
       "eval_accuracy": 0.9035450123275804,
+      "eval_combined_score": 0.9033726098964578,
+      "eval_f1": 0.9030761902479666,
       "eval_loss": 0.3006099760532379,
-      "eval_runtime": 49.5179,
-      "eval_samples_per_second": 606.124,
-      "eval_steps_per_second": 1.191,
+      "eval_precision": 0.903324224682704,
+      "eval_recall": 0.9035450123275804,
+      "eval_runtime": 49.1842,
+      "eval_samples_per_second": 610.237,
+      "eval_steps_per_second": 1.2,
       "step": 640
     },
     {
@@ -695,10 +823,14 @@
     {
       "epoch": 1.99,
       "eval_accuracy": 0.9014126740854268,
+      "eval_combined_score": 0.9026415340332592,
+      "eval_f1": 0.9025066760774267,
       "eval_loss": 0.30540063977241516,
-      "eval_runtime": 49.5044,
-      "eval_samples_per_second": 606.289,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.9052341118847564,
+      "eval_recall": 0.9014126740854268,
+      "eval_runtime": 49.4063,
+      "eval_samples_per_second": 607.494,
+      "eval_steps_per_second": 1.194,
       "step": 660
     },
     {
@@ -716,10 +848,14 @@
     {
       "epoch": 2.05,
       "eval_accuracy": 0.8912507496501633,
+      "eval_combined_score": 0.8911549520027604,
+      "eval_f1": 0.8866402874920362,
       "eval_loss": 0.3174073100090027,
-      "eval_runtime": 49.4656,
-      "eval_samples_per_second": 606.765,
-      "eval_steps_per_second": 1.193,
+      "eval_precision": 0.895478021218679,
+      "eval_recall": 0.8912507496501633,
+      "eval_runtime": 49.3463,
+      "eval_samples_per_second": 608.232,
+      "eval_steps_per_second": 1.196,
       "step": 680
     },
     {
@@ -737,10 +873,14 @@
     {
       "epoch": 2.11,
       "eval_accuracy": 0.9122076364363297,
+      "eval_combined_score": 0.9124808443428073,
+      "eval_f1": 0.9125293911738805,
       "eval_loss": 0.2769884169101715,
-      "eval_runtime": 49.5225,
-      "eval_samples_per_second": 606.068,
-      "eval_steps_per_second": 1.191,
+      "eval_precision": 0.9129787133246888,
+      "eval_recall": 0.9122076364363297,
+      "eval_runtime": 49.3634,
+      "eval_samples_per_second": 608.021,
+      "eval_steps_per_second": 1.195,
       "step": 700
     },
     {
@@ -758,10 +898,14 @@
     {
       "epoch": 2.17,
       "eval_accuracy": 0.9062437529153062,
+      "eval_combined_score": 0.9068222505646787,
+      "eval_f1": 0.9054896753410417,
       "eval_loss": 0.2979203760623932,
-      "eval_runtime": 49.2758,
-      "eval_samples_per_second": 609.102,
-      "eval_steps_per_second": 1.197,
+      "eval_precision": 0.9093118210870608,
+      "eval_recall": 0.9062437529153062,
+      "eval_runtime": 49.0419,
+      "eval_samples_per_second": 612.007,
+      "eval_steps_per_second": 1.203,
       "step": 720
     },
     {
@@ -779,10 +923,14 @@
     {
       "epoch": 2.23,
       "eval_accuracy": 0.8997801026187779,
+      "eval_combined_score": 0.9003031124352014,
+      "eval_f1": 0.8971406374938125,
       "eval_loss": 0.29734131693840027,
-      "eval_runtime": 49.4882,
-      "eval_samples_per_second": 606.488,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.9045116070094374,
+      "eval_recall": 0.8997801026187779,
+      "eval_runtime": 49.1818,
+      "eval_samples_per_second": 610.267,
+      "eval_steps_per_second": 1.2,
       "step": 740
     },
     {
@@ -800,10 +948,14 @@
     {
       "epoch": 2.29,
       "eval_accuracy": 0.9221696541613914,
+      "eval_combined_score": 0.9221674217032323,
+      "eval_f1": 0.922071694968714,
       "eval_loss": 0.24671417474746704,
-      "eval_runtime": 49.5036,
-      "eval_samples_per_second": 606.299,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.9222586835214323,
+      "eval_recall": 0.9221696541613914,
+      "eval_runtime": 49.2126,
+      "eval_samples_per_second": 609.884,
+      "eval_steps_per_second": 1.199,
       "step": 760
     },
     {
@@ -821,10 +973,14 @@
     {
       "epoch": 2.35,
       "eval_accuracy": 0.9113080562404211,
+      "eval_combined_score": 0.9131963179862455,
+      "eval_f1": 0.9128361841568111,
       "eval_loss": 0.2760636806488037,
-      "eval_runtime": 49.4644,
-      "eval_samples_per_second": 606.78,
-      "eval_steps_per_second": 1.193,
+      "eval_precision": 0.917332975307328,
+      "eval_recall": 0.9113080562404211,
+      "eval_runtime": 49.2008,
+      "eval_samples_per_second": 610.031,
+      "eval_steps_per_second": 1.199,
       "step": 780
     },
     {
@@ -842,10 +998,14 @@
     {
       "epoch": 2.41,
       "eval_accuracy": 0.9260011994402612,
+      "eval_combined_score": 0.9259453001449679,
+      "eval_f1": 0.9257439757889774,
       "eval_loss": 0.24102580547332764,
-      "eval_runtime": 49.464,
-      "eval_samples_per_second": 606.785,
-      "eval_steps_per_second": 1.193,
+      "eval_precision": 0.9260348259103722,
+      "eval_recall": 0.9260011994402612,
+      "eval_runtime": 49.2957,
+      "eval_samples_per_second": 608.856,
+      "eval_steps_per_second": 1.197,
       "step": 800
     },
     {
@@ -863,10 +1023,14 @@
     {
       "epoch": 2.47,
       "eval_accuracy": 0.922103018591324,
+      "eval_combined_score": 0.9220769784769673,
+      "eval_f1": 0.9212994658547122,
       "eval_loss": 0.2446586638689041,
-      "eval_runtime": 49.2591,
-      "eval_samples_per_second": 609.308,
-      "eval_steps_per_second": 1.198,
+      "eval_precision": 0.9228024108705088,
+      "eval_recall": 0.922103018591324,
+      "eval_runtime": 48.9946,
+      "eval_samples_per_second": 612.598,
+      "eval_steps_per_second": 1.204,
       "step": 820
     },
     {
@@ -884,10 +1048,14 @@
     {
       "epoch": 2.53,
       "eval_accuracy": 0.9237355900579729,
+      "eval_combined_score": 0.9240135925879284,
+      "eval_f1": 0.9231767824872119,
       "eval_loss": 0.2475174069404602,
-      "eval_runtime": 49.5127,
-      "eval_samples_per_second": 606.188,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.9254064077485559,
+      "eval_recall": 0.9237355900579729,
+      "eval_runtime": 49.3973,
+      "eval_samples_per_second": 607.604,
+      "eval_steps_per_second": 1.194,
       "step": 840
     },
     {
@@ -905,10 +1073,14 @@
     {
       "epoch": 2.59,
       "eval_accuracy": 0.9265342840007996,
+      "eval_combined_score": 0.9265368337935618,
+      "eval_f1": 0.9258759190762016,
       "eval_loss": 0.2590079605579376,
-      "eval_runtime": 49.5242,
-      "eval_samples_per_second": 606.047,
-      "eval_steps_per_second": 1.191,
+      "eval_precision": 0.927202848096446,
+      "eval_recall": 0.9265342840007996,
+      "eval_runtime": 48.961,
+      "eval_samples_per_second": 613.019,
+      "eval_steps_per_second": 1.205,
       "step": 860
     },
     {
@@ -926,10 +1098,14 @@
     {
       "epoch": 2.65,
       "eval_accuracy": 0.930032651429333,
+      "eval_combined_score": 0.9305237768064085,
+      "eval_f1": 0.9305263249699511,
       "eval_loss": 0.22479340434074402,
-      "eval_runtime": 49.4946,
-      "eval_samples_per_second": 606.409,
-      "eval_steps_per_second": 1.192,
+      "eval_precision": 0.9315034793970168,
+      "eval_recall": 0.930032651429333,
+      "eval_runtime": 49.2223,
+      "eval_samples_per_second": 609.764,
+      "eval_steps_per_second": 1.199,
       "step": 880
     },
     {
@@ -947,18 +1123,22 @@
     {
       "epoch": 2.71,
       "eval_accuracy": 0.9273005930565736,
+      "eval_combined_score": 0.9281516215920332,
+      "eval_f1": 0.9280796047956419,
       "eval_loss": 0.22854498028755188,
-      "eval_runtime": 49.2942,
-      "eval_samples_per_second": 608.875,
-      "eval_steps_per_second": 1.197,
+      "eval_precision": 0.9299256954593439,
+      "eval_recall": 0.9273005930565736,
+      "eval_runtime": 49.2291,
+      "eval_samples_per_second": 609.68,
+      "eval_steps_per_second": 1.198,
       "step": 900
     }
   ],
   "logging_steps": 10,
-  "max_steps": 996,
-  "num_train_epochs": 3,
+  "max_steps": 1660,
+  "num_train_epochs": 5,
   "save_steps": 100,
-  "total_flos": 1.212384490375086e+17,
+  "total_flos": 1.2123845307437875e+17,
   "trial_name": null,
   "trial_params": null
 }

checkpoint-900/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1f1be80ebe52f6e43af0b8aa087e72fad77310d5998b6e0b8f66a6a1d53be7b7
+oid sha256:ae98e7de1b05a570517ae68653e3c31b639a52c739d05197601f467f38b01c66
 size 4536