Training in progress, epoch 3, checkpoint

checkpoint-63/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:85f507b0d3859d5a42c970219587fa81011fa29adb74ba4ab5628dde56dafb73
+oid sha256:a9d07d05de52bc4d67d79c40ac25130d048bddc5285405105c5b7f0f63db644e
 size 3086640776

checkpoint-63/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4b640d9ef8b573c0450e41f463b65a9acae87a9a98f2afcfdd4f7a8d4d5dc601
+oid sha256:f35558c5b7388beac579ad45a309c9989258e292669efc82f02fe9faf9924f77
 size 6056331

checkpoint-63/trainer_state.json

@@ -11,443 +11,443 @@
   "log_history": [
     {
       "epoch": 0.04878048780487805,
-      "grad_norm": 126.0,
+      "grad_norm": 119.5,
       "learning_rate": 0.0,
-      "loss": 6.3712,
+      "loss": 6.2132,
       "step": 1
     },
     {
       "epoch": 0.0975609756097561,
-      "grad_norm": 139.0,
+      "grad_norm": 132.0,
       "learning_rate": 7.142857142857143e-06,
-      "loss": 6.9747,
+      "loss": 6.6559,
       "step": 2
     },
     {
       "epoch": 0.14634146341463414,
-      "grad_norm": 141.0,
+      "grad_norm": 128.0,
       "learning_rate": 1.4285714285714285e-05,
-      "loss": 7.2169,
+      "loss": 6.6652,
       "step": 3
     },
     {
       "epoch": 0.1951219512195122,
-      "grad_norm": 135.0,
+      "grad_norm": 129.0,
       "learning_rate": 2.1428571428571428e-05,
-      "loss": 6.3792,
+      "loss": 6.1414,
       "step": 4
     },
     {
       "epoch": 0.24390243902439024,
-      "grad_norm": 356.0,
+      "grad_norm": 348.0,
       "learning_rate": 2.857142857142857e-05,
-      "loss": 5.1872,
+      "loss": 5.1096,
       "step": 5
     },
     {
       "epoch": 0.2926829268292683,
-      "grad_norm": 158.0,
+      "grad_norm": 165.0,
       "learning_rate": 3.571428571428572e-05,
-      "loss": 5.7681,
+      "loss": 5.2497,
       "step": 6
     },
     {
       "epoch": 0.34146341463414637,
-      "grad_norm": 149.0,
+      "grad_norm": 136.0,
       "learning_rate": 4.2857142857142856e-05,
-      "loss": 4.4005,
+      "loss": 4.3006,
       "step": 7
     },
     {
       "epoch": 0.3902439024390244,
-      "grad_norm": 134.0,
+      "grad_norm": 150.0,
       "learning_rate": 5e-05,
-      "loss": 5.056,
+      "loss": 5.1854,
       "step": 8
     },
     {
       "epoch": 0.43902439024390244,
-      "grad_norm": 94.5,
+      "grad_norm": 213.0,
       "learning_rate": 4.996067037544542e-05,
-      "loss": 4.2935,
+      "loss": 8.1666,
       "step": 9
     },
     {
       "epoch": 0.4878048780487805,
-      "grad_norm": 212.0,
+      "grad_norm": 137.0,
       "learning_rate": 4.984280524733107e-05,
-      "loss": 8.9937,
+      "loss": 9.5804,
       "step": 10
     },
     {
       "epoch": 0.5365853658536586,
-      "grad_norm": 302.0,
+      "grad_norm": 65.0,
       "learning_rate": 4.96467754629559e-05,
-      "loss": 4.9127,
+      "loss": 5.2593,
       "step": 11
     },
     {
       "epoch": 0.5853658536585366,
-      "grad_norm": 44.5,
+      "grad_norm": 40.25,
       "learning_rate": 4.937319780454559e-05,
-      "loss": 4.4251,
+      "loss": 3.9807,
       "step": 12
     },
     {
       "epoch": 0.6341463414634146,
-      "grad_norm": 39.25,
+      "grad_norm": 39.75,
       "learning_rate": 4.9022933048627496e-05,
-      "loss": 3.6691,
+      "loss": 3.7199,
       "step": 13
     },
     {
       "epoch": 0.6829268292682927,
-      "grad_norm": 31.25,
+      "grad_norm": 50.0,
       "learning_rate": 4.8597083257709194e-05,
-      "loss": 3.232,
+      "loss": 3.3626,
       "step": 14
     },
     {
       "epoch": 0.7317073170731707,
-      "grad_norm": 44.25,
+      "grad_norm": 44.5,
       "learning_rate": 4.8096988312782174e-05,
-      "loss": 3.1154,
+      "loss": 3.156,
       "step": 15
     },
     {
       "epoch": 0.7804878048780488,
-      "grad_norm": 43.75,
+      "grad_norm": 43.5,
       "learning_rate": 4.752422169756048e-05,
-      "loss": 2.6809,
+      "loss": 2.62,
       "step": 16
     },
     {
       "epoch": 0.8292682926829268,
-      "grad_norm": 41.0,
+      "grad_norm": 47.5,
       "learning_rate": 4.6880585547718845e-05,
-      "loss": 2.1384,
+      "loss": 2.4872,
       "step": 17
     },
     {
       "epoch": 0.8780487804878049,
-      "grad_norm": 50.0,
+      "grad_norm": 44.5,
       "learning_rate": 4.6168104980707107e-05,
-      "loss": 2.0084,
+      "loss": 2.0436,
       "step": 18
     },
     {
       "epoch": 0.926829268292683,
-      "grad_norm": 48.25,
+      "grad_norm": 50.5,
       "learning_rate": 4.538902172398151e-05,
-      "loss": 1.8335,
+      "loss": 1.9184,
       "step": 19
     },
     {
       "epoch": 0.975609756097561,
-      "grad_norm": 50.0,
+      "grad_norm": 55.5,
       "learning_rate": 4.454578706170075e-05,
-      "loss": 1.6659,
+      "loss": 1.8052,
       "step": 20
     },
     {
       "epoch": 1.0,
-      "grad_norm": 50.75,
+      "grad_norm": 41.75,
       "learning_rate": 4.364105412207914e-05,
-      "loss": 1.5889,
+      "loss": 1.4898,
       "step": 21
     },
     {
       "epoch": 1.048780487804878,
-      "grad_norm": 44.0,
+      "grad_norm": 47.25,
       "learning_rate": 4.267766952966369e-05,
-      "loss": 1.4364,
+      "loss": 1.3246,
       "step": 22
     },
     {
       "epoch": 1.0975609756097562,
-      "grad_norm": 44.25,
+      "grad_norm": 40.25,
       "learning_rate": 4.16586644488001e-05,
-      "loss": 1.3537,
+      "loss": 1.1731,
       "step": 23
     },
     {
       "epoch": 1.146341463414634,
-      "grad_norm": 37.0,
+      "grad_norm": 44.75,
       "learning_rate": 4.058724504646834e-05,
-      "loss": 1.1017,
+      "loss": 1.14,
       "step": 24
     },
     {
       "epoch": 1.1951219512195121,
-      "grad_norm": 40.25,
+      "grad_norm": 33.75,
       "learning_rate": 3.946678240449515e-05,
-      "loss": 1.0232,
+      "loss": 1.0197,
       "step": 25
     },
     {
       "epoch": 1.2439024390243902,
-      "grad_norm": 34.25,
+      "grad_norm": 31.25,
       "learning_rate": 3.830080191288342e-05,
-      "loss": 0.9218,
+      "loss": 0.9634,
       "step": 26
     },
     {
       "epoch": 1.2926829268292683,
-      "grad_norm": 27.125,
+      "grad_norm": 26.75,
       "learning_rate": 3.7092972177631e-05,
-      "loss": 0.8892,
+      "loss": 1.0227,
       "step": 27
     },
     {
       "epoch": 1.3414634146341464,
-      "grad_norm": 25.0,
+      "grad_norm": 22.875,
       "learning_rate": 3.5847093477938956e-05,
-      "loss": 0.9091,
+      "loss": 0.8732,
       "step": 28
     },
     {
       "epoch": 1.3902439024390243,
-      "grad_norm": 22.5,
+      "grad_norm": 19.0,
       "learning_rate": 3.456708580912725e-05,
-      "loss": 0.9093,
+      "loss": 0.8081,
       "step": 29
     },
     {
       "epoch": 1.4390243902439024,
-      "grad_norm": 23.625,
+      "grad_norm": 18.25,
       "learning_rate": 3.3256976548879184e-05,
-      "loss": 0.846,
+      "loss": 0.7939,
       "step": 30
     },
     {
       "epoch": 1.4878048780487805,
-      "grad_norm": 19.375,
+      "grad_norm": 13.0,
       "learning_rate": 3.1920887785621235e-05,
-      "loss": 0.7321,
+      "loss": 0.7528,
       "step": 31
     },
     {
       "epoch": 1.5365853658536586,
-      "grad_norm": 15.6875,
+      "grad_norm": 31.0,
       "learning_rate": 3.056302334890786e-05,
-      "loss": 0.7505,
+      "loss": 0.8375,
       "step": 32
     },
     {
       "epoch": 1.5853658536585367,
-      "grad_norm": 16.5,
+      "grad_norm": 47.5,
       "learning_rate": 2.918765558261841e-05,
-      "loss": 0.7204,
+      "loss": 1.1202,
       "step": 33
     },
     {
       "epoch": 1.6341463414634148,
-      "grad_norm": 13.25,
+      "grad_norm": 40.75,
       "learning_rate": 2.7799111902582696e-05,
-      "loss": 0.6878,
+      "loss": 0.9477,
       "step": 34
     },
     {
       "epoch": 1.6829268292682928,
-      "grad_norm": 15.0,
+      "grad_norm": 22.0,
       "learning_rate": 2.6401761180929797e-05,
-      "loss": 0.6356,
+      "loss": 0.7566,
       "step": 35
     },
     {
       "epoch": 1.7317073170731707,
-      "grad_norm": 14.3125,
+      "grad_norm": 9.6875,
       "learning_rate": 2.5e-05,
-      "loss": 0.6332,
+      "loss": 0.7021,
       "step": 36
     },
     {
       "epoch": 1.7804878048780488,
-      "grad_norm": 9.9375,
+      "grad_norm": 9.5,
       "learning_rate": 2.3598238819070202e-05,
-      "loss": 0.6806,
+      "loss": 0.6245,
       "step": 37
     },
     {
       "epoch": 1.8292682926829267,
-      "grad_norm": 11.5,
+      "grad_norm": 7.28125,
       "learning_rate": 2.2200888097417307e-05,
-      "loss": 0.6415,
+      "loss": 0.575,
       "step": 38
     },
     {
       "epoch": 1.8780487804878048,
-      "grad_norm": 9.75,
+      "grad_norm": 5.78125,
       "learning_rate": 2.0812344417381595e-05,
-      "loss": 0.5846,
+      "loss": 0.6548,
       "step": 39
     },
     {
       "epoch": 1.9268292682926829,
-      "grad_norm": 10.25,
+      "grad_norm": 5.1875,
       "learning_rate": 1.9436976651092144e-05,
-      "loss": 0.5771,
+      "loss": 0.603,
       "step": 40
     },
     {
       "epoch": 1.975609756097561,
-      "grad_norm": 7.75,
+      "grad_norm": 4.71875,
       "learning_rate": 1.8079112214378768e-05,
-      "loss": 0.6536,
+      "loss": 0.5833,
       "step": 41
     },
     {
       "epoch": 2.0,
-      "grad_norm": 8.125,
+      "grad_norm": 5.125,
       "learning_rate": 1.6743023451120832e-05,
-      "loss": 0.6518,
+      "loss": 0.541,
       "step": 42
     },
     {
       "epoch": 2.048780487804878,
-      "grad_norm": 7.96875,
+      "grad_norm": 4.03125,
       "learning_rate": 1.5432914190872757e-05,
-      "loss": 0.5361,
+      "loss": 0.5195,
       "step": 43
     },
     {
       "epoch": 2.097560975609756,
-      "grad_norm": 8.0,
+      "grad_norm": 4.1875,
       "learning_rate": 1.4152906522061048e-05,
-      "loss": 0.6016,
+      "loss": 0.4844,
       "step": 44
     },
     {
       "epoch": 2.1463414634146343,
-      "grad_norm": 7.0625,
+      "grad_norm": 3.265625,
       "learning_rate": 1.2907027822369005e-05,
-      "loss": 0.5721,
+      "loss": 0.5301,
       "step": 45
     },
     {
       "epoch": 2.1951219512195124,
-      "grad_norm": 8.0,
+      "grad_norm": 3.734375,
       "learning_rate": 1.1699198087116589e-05,
-      "loss": 0.5486,
+      "loss": 0.4798,
       "step": 46
     },
     {
       "epoch": 2.2439024390243905,
-      "grad_norm": 7.65625,
+      "grad_norm": 3.390625,
       "learning_rate": 1.0533217595504858e-05,
-      "loss": 0.5682,
+      "loss": 0.5796,
       "step": 47
     },
     {
       "epoch": 2.292682926829268,
-      "grad_norm": 7.28125,
+      "grad_norm": 3.109375,
       "learning_rate": 9.412754953531663e-06,
-      "loss": 0.54,
+      "loss": 0.5007,
       "step": 48
     },
     {
       "epoch": 2.341463414634146,
-      "grad_norm": 7.25,
+      "grad_norm": 2.890625,
       "learning_rate": 8.341335551199902e-06,
-      "loss": 0.5295,
+      "loss": 0.5081,
       "step": 49
     },
     {
       "epoch": 2.3902439024390243,
-      "grad_norm": 7.53125,
+      "grad_norm": 2.8125,
       "learning_rate": 7.3223304703363135e-06,
-      "loss": 0.5384,
+      "loss": 0.5446,
       "step": 50
     },
     {
       "epoch": 2.4390243902439024,
-      "grad_norm": 6.40625,
+      "grad_norm": 3.421875,
       "learning_rate": 6.358945877920861e-06,
-      "loss": 0.5395,
+      "loss": 0.4891,
       "step": 51
     },
     {
       "epoch": 2.4878048780487805,
-      "grad_norm": 6.96875,
+      "grad_norm": 2.953125,
       "learning_rate": 5.454212938299255e-06,
-      "loss": 0.5124,
+      "loss": 0.5342,
       "step": 52
     },
     {
       "epoch": 2.5365853658536586,
-      "grad_norm": 6.875,
+      "grad_norm": 2.859375,
       "learning_rate": 4.610978276018496e-06,
-      "loss": 0.4981,
+      "loss": 0.5187,
       "step": 53
     },
     {
       "epoch": 2.5853658536585367,
-      "grad_norm": 6.78125,
+      "grad_norm": 2.84375,
       "learning_rate": 3.831895019292897e-06,
-      "loss": 0.5473,
+      "loss": 0.5401,
       "step": 54
     },
     {
       "epoch": 2.6341463414634148,
-      "grad_norm": 6.15625,
+      "grad_norm": 2.890625,
       "learning_rate": 3.119414452281158e-06,
-      "loss": 0.527,
+      "loss": 0.5199,
       "step": 55
     },
     {
       "epoch": 2.682926829268293,
-      "grad_norm": 6.75,
+      "grad_norm": 3.09375,
       "learning_rate": 2.475778302439524e-06,
-      "loss": 0.5601,
+      "loss": 0.4834,
       "step": 56
     },
     {
       "epoch": 2.7317073170731705,
-      "grad_norm": 6.15625,
+      "grad_norm": 2.984375,
       "learning_rate": 1.9030116872178316e-06,
-      "loss": 0.5356,
+      "loss": 0.4964,
       "step": 57
     },
     {
       "epoch": 2.7804878048780486,
-      "grad_norm": 6.28125,
+      "grad_norm": 2.921875,
       "learning_rate": 1.4029167422908107e-06,
-      "loss": 0.5391,
+      "loss": 0.5161,
       "step": 58
     },
     {
       "epoch": 2.8292682926829267,
-      "grad_norm": 6.625,
+      "grad_norm": 3.5,
       "learning_rate": 9.770669513725128e-07,
-      "loss": 0.5177,
+      "loss": 0.4928,
       "step": 59
     },
     {
       "epoch": 2.8780487804878048,
-      "grad_norm": 6.625,
+      "grad_norm": 2.890625,
       "learning_rate": 6.268021954544096e-07,
-      "loss": 0.5542,
+      "loss": 0.518,
       "step": 60
     },
     {
       "epoch": 2.926829268292683,
-      "grad_norm": 6.84375,
+      "grad_norm": 2.984375,
       "learning_rate": 3.5322453704410286e-07,
-      "loss": 0.5601,
+      "loss": 0.5304,
       "step": 61
     },
     {
       "epoch": 2.975609756097561,
-      "grad_norm": 6.1875,
+      "grad_norm": 2.59375,
       "learning_rate": 1.571947526689349e-07,
-      "loss": 0.5629,
+      "loss": 0.5375,
       "step": 62
     },
     {
       "epoch": 3.0,
-      "grad_norm": 8.375,
+      "grad_norm": 3.0625,
       "learning_rate": 3.9329624554584884e-08,
-      "loss": 0.5202,
+      "loss": 0.5582,
       "step": 63
     }
   ],

checkpoint-63/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:807175855d5dd743e81254531a1a0148b2ade3b9521c50fa1999cea9a77e0a70
+oid sha256:841e48b365d43c1bb533dc5dd460b128556952c5da32bc814c2b5072cc6bb501
 size 5969