{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.2016460905349795,
  "eval_steps": 110,
  "global_step": 876,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0013717421124828531,
      "grad_norm": 13.75906753540039,
      "learning_rate": 1.8281535648994516e-06,
      "loss": 1.2468,
      "step": 1
    },
    {
      "epoch": 0.0027434842249657062,
      "grad_norm": 14.05058765411377,
      "learning_rate": 3.6563071297989032e-06,
      "loss": 1.4692,
      "step": 2
    },
    {
      "epoch": 0.00411522633744856,
      "grad_norm": 13.991771697998047,
      "learning_rate": 5.484460694698355e-06,
      "loss": 1.2457,
      "step": 3
    },
    {
      "epoch": 0.0054869684499314125,
      "grad_norm": 13.429465293884277,
      "learning_rate": 7.3126142595978065e-06,
      "loss": 1.1859,
      "step": 4
    },
    {
      "epoch": 0.006858710562414266,
      "grad_norm": 12.403002738952637,
      "learning_rate": 9.140767824497258e-06,
      "loss": 1.2404,
      "step": 5
    },
    {
      "epoch": 0.00823045267489712,
      "grad_norm": 1.5524662733078003,
      "learning_rate": 1.096892138939671e-05,
      "loss": 0.042,
      "step": 6
    },
    {
      "epoch": 0.009602194787379973,
      "grad_norm": 10.1494140625,
      "learning_rate": 1.2797074954296162e-05,
      "loss": 0.8856,
      "step": 7
    },
    {
      "epoch": 0.010973936899862825,
      "grad_norm": Infinity,
      "learning_rate": 1.2797074954296162e-05,
      "loss": 1.5417,
      "step": 8
    },
    {
      "epoch": 0.012345679012345678,
      "grad_norm": 0.4553964138031006,
      "learning_rate": 1.4625228519195613e-05,
      "loss": 0.023,
      "step": 9
    },
    {
      "epoch": 0.013717421124828532,
      "grad_norm": 10.332369804382324,
      "learning_rate": 1.6453382084095062e-05,
      "loss": 0.8655,
      "step": 10
    },
    {
      "epoch": 0.015089163237311385,
      "grad_norm": 7.544310092926025,
      "learning_rate": 1.8281535648994517e-05,
      "loss": 0.5894,
      "step": 11
    },
    {
      "epoch": 0.01646090534979424,
      "grad_norm": 8.16427230834961,
      "learning_rate": 2.0109689213893968e-05,
      "loss": 0.7053,
      "step": 12
    },
    {
      "epoch": 0.01783264746227709,
      "grad_norm": 7.403252601623535,
      "learning_rate": 2.193784277879342e-05,
      "loss": 0.5857,
      "step": 13
    },
    {
      "epoch": 0.019204389574759947,
      "grad_norm": 8.974674224853516,
      "learning_rate": 2.376599634369287e-05,
      "loss": 0.8375,
      "step": 14
    },
    {
      "epoch": 0.0205761316872428,
      "grad_norm": 13.417745590209961,
      "learning_rate": 2.5594149908592324e-05,
      "loss": 0.9043,
      "step": 15
    },
    {
      "epoch": 0.02194787379972565,
      "grad_norm": 12.881294250488281,
      "learning_rate": 2.742230347349177e-05,
      "loss": 0.8756,
      "step": 16
    },
    {
      "epoch": 0.023319615912208505,
      "grad_norm": 7.439205169677734,
      "learning_rate": 2.9250457038391226e-05,
      "loss": 0.5076,
      "step": 17
    },
    {
      "epoch": 0.024691358024691357,
      "grad_norm": 8.46964168548584,
      "learning_rate": 3.107861060329068e-05,
      "loss": 0.4757,
      "step": 18
    },
    {
      "epoch": 0.02606310013717421,
      "grad_norm": 17.02773666381836,
      "learning_rate": 3.2906764168190124e-05,
      "loss": 0.9993,
      "step": 19
    },
    {
      "epoch": 0.027434842249657063,
      "grad_norm": 6.2668776512146,
      "learning_rate": 3.473491773308958e-05,
      "loss": 0.2622,
      "step": 20
    },
    {
      "epoch": 0.02880658436213992,
      "grad_norm": 8.273824691772461,
      "learning_rate": 3.656307129798903e-05,
      "loss": 0.3497,
      "step": 21
    },
    {
      "epoch": 0.03017832647462277,
      "grad_norm": 5.5460944175720215,
      "learning_rate": 3.839122486288849e-05,
      "loss": 0.2514,
      "step": 22
    },
    {
      "epoch": 0.03155006858710562,
      "grad_norm": 4.283128261566162,
      "learning_rate": 4.0219378427787935e-05,
      "loss": 0.1673,
      "step": 23
    },
    {
      "epoch": 0.03292181069958848,
      "grad_norm": 4.708792209625244,
      "learning_rate": 4.204753199268738e-05,
      "loss": 0.203,
      "step": 24
    },
    {
      "epoch": 0.03429355281207133,
      "grad_norm": 14.491021156311035,
      "learning_rate": 4.387568555758684e-05,
      "loss": 0.698,
      "step": 25
    },
    {
      "epoch": 0.03566529492455418,
      "grad_norm": 7.903520584106445,
      "learning_rate": 4.570383912248629e-05,
      "loss": 0.3401,
      "step": 26
    },
    {
      "epoch": 0.037037037037037035,
      "grad_norm": 7.333080291748047,
      "learning_rate": 4.753199268738574e-05,
      "loss": 0.2185,
      "step": 27
    },
    {
      "epoch": 0.038408779149519894,
      "grad_norm": 8.625358581542969,
      "learning_rate": 4.936014625228519e-05,
      "loss": 0.4424,
      "step": 28
    },
    {
      "epoch": 0.039780521262002745,
      "grad_norm": 1.5588488578796387,
      "learning_rate": 5.118829981718465e-05,
      "loss": 0.0381,
      "step": 29
    },
    {
      "epoch": 0.0411522633744856,
      "grad_norm": 12.401138305664062,
      "learning_rate": 5.3016453382084095e-05,
      "loss": 0.8215,
      "step": 30
    },
    {
      "epoch": 0.04252400548696845,
      "grad_norm": 5.405845642089844,
      "learning_rate": 5.484460694698354e-05,
      "loss": 0.1542,
      "step": 31
    },
    {
      "epoch": 0.0438957475994513,
      "grad_norm": 8.558808326721191,
      "learning_rate": 5.6672760511883e-05,
      "loss": 0.6893,
      "step": 32
    },
    {
      "epoch": 0.04526748971193416,
      "grad_norm": 7.206741809844971,
      "learning_rate": 5.850091407678245e-05,
      "loss": 0.3773,
      "step": 33
    },
    {
      "epoch": 0.04663923182441701,
      "grad_norm": 8.300729751586914,
      "learning_rate": 6.0329067641681906e-05,
      "loss": 0.538,
      "step": 34
    },
    {
      "epoch": 0.04801097393689986,
      "grad_norm": 0.2500181496143341,
      "learning_rate": 6.215722120658135e-05,
      "loss": 0.0073,
      "step": 35
    },
    {
      "epoch": 0.04938271604938271,
      "grad_norm": 17.457223892211914,
      "learning_rate": 6.398537477148081e-05,
      "loss": 2.378,
      "step": 36
    },
    {
      "epoch": 0.05075445816186557,
      "grad_norm": 10.884990692138672,
      "learning_rate": 6.581352833638025e-05,
      "loss": 0.5949,
      "step": 37
    },
    {
      "epoch": 0.05212620027434842,
      "grad_norm": 10.013723373413086,
      "learning_rate": 6.764168190127972e-05,
      "loss": 0.7071,
      "step": 38
    },
    {
      "epoch": 0.053497942386831275,
      "grad_norm": 4.653324604034424,
      "learning_rate": 6.946983546617916e-05,
      "loss": 0.1607,
      "step": 39
    },
    {
      "epoch": 0.05486968449931413,
      "grad_norm": 9.527400970458984,
      "learning_rate": 7.129798903107861e-05,
      "loss": 0.7735,
      "step": 40
    },
    {
      "epoch": 0.056241426611796985,
      "grad_norm": 12.477531433105469,
      "learning_rate": 7.312614259597807e-05,
      "loss": 0.7594,
      "step": 41
    },
    {
      "epoch": 0.05761316872427984,
      "grad_norm": 5.369799613952637,
      "learning_rate": 7.495429616087751e-05,
      "loss": 0.3569,
      "step": 42
    },
    {
      "epoch": 0.05898491083676269,
      "grad_norm": 5.1385908126831055,
      "learning_rate": 7.678244972577697e-05,
      "loss": 0.2454,
      "step": 43
    },
    {
      "epoch": 0.06035665294924554,
      "grad_norm": 6.1807708740234375,
      "learning_rate": 7.861060329067642e-05,
      "loss": 0.2723,
      "step": 44
    },
    {
      "epoch": 0.06172839506172839,
      "grad_norm": 7.941879749298096,
      "learning_rate": 8.043875685557587e-05,
      "loss": 0.5338,
      "step": 45
    },
    {
      "epoch": 0.06310013717421124,
      "grad_norm": 5.015410423278809,
      "learning_rate": 8.226691042047532e-05,
      "loss": 0.1891,
      "step": 46
    },
    {
      "epoch": 0.0644718792866941,
      "grad_norm": 7.299699306488037,
      "learning_rate": 8.409506398537477e-05,
      "loss": 0.3647,
      "step": 47
    },
    {
      "epoch": 0.06584362139917696,
      "grad_norm": 8.421393394470215,
      "learning_rate": 8.592321755027423e-05,
      "loss": 0.383,
      "step": 48
    },
    {
      "epoch": 0.06721536351165981,
      "grad_norm": 5.5915937423706055,
      "learning_rate": 8.775137111517367e-05,
      "loss": 0.2353,
      "step": 49
    },
    {
      "epoch": 0.06858710562414266,
      "grad_norm": 8.187829971313477,
      "learning_rate": 8.957952468007313e-05,
      "loss": 0.5541,
      "step": 50
    },
    {
      "epoch": 0.06995884773662552,
      "grad_norm": 6.386786460876465,
      "learning_rate": 9.140767824497258e-05,
      "loss": 0.4908,
      "step": 51
    },
    {
      "epoch": 0.07133058984910837,
      "grad_norm": 8.64050006866455,
      "learning_rate": 9.323583180987204e-05,
      "loss": 0.586,
      "step": 52
    },
    {
      "epoch": 0.07270233196159122,
      "grad_norm": 5.879551410675049,
      "learning_rate": 9.506398537477148e-05,
      "loss": 0.2241,
      "step": 53
    },
    {
      "epoch": 0.07407407407407407,
      "grad_norm": 7.824138164520264,
      "learning_rate": 9.689213893967093e-05,
      "loss": 0.6046,
      "step": 54
    },
    {
      "epoch": 0.07544581618655692,
      "grad_norm": 6.351109504699707,
      "learning_rate": 9.872029250457039e-05,
      "loss": 0.231,
      "step": 55
    },
    {
      "epoch": 0.07681755829903979,
      "grad_norm": 9.437410354614258,
      "learning_rate": 0.00010054844606946984,
      "loss": 0.7105,
      "step": 56
    },
    {
      "epoch": 0.07818930041152264,
      "grad_norm": 8.40911865234375,
      "learning_rate": 0.0001023765996343693,
      "loss": 0.5591,
      "step": 57
    },
    {
      "epoch": 0.07956104252400549,
      "grad_norm": 7.631382942199707,
      "learning_rate": 0.00010420475319926874,
      "loss": 0.5194,
      "step": 58
    },
    {
      "epoch": 0.08093278463648834,
      "grad_norm": 5.773220062255859,
      "learning_rate": 0.00010603290676416819,
      "loss": 0.3297,
      "step": 59
    },
    {
      "epoch": 0.0823045267489712,
      "grad_norm": 1.3606321811676025,
      "learning_rate": 0.00010786106032906765,
      "loss": 0.0299,
      "step": 60
    },
    {
      "epoch": 0.08367626886145405,
      "grad_norm": 7.216275215148926,
      "learning_rate": 0.00010968921389396709,
      "loss": 0.3514,
      "step": 61
    },
    {
      "epoch": 0.0850480109739369,
      "grad_norm": 4.70477294921875,
      "learning_rate": 0.00011151736745886655,
      "loss": 0.1932,
      "step": 62
    },
    {
      "epoch": 0.08641975308641975,
      "grad_norm": 6.754104137420654,
      "learning_rate": 0.000113345521023766,
      "loss": 0.4035,
      "step": 63
    },
    {
      "epoch": 0.0877914951989026,
      "grad_norm": 0.19067375361919403,
      "learning_rate": 0.00011517367458866546,
      "loss": 0.0094,
      "step": 64
    },
    {
      "epoch": 0.08916323731138547,
      "grad_norm": 1.1715893745422363,
      "learning_rate": 0.0001170018281535649,
      "loss": 0.0148,
      "step": 65
    },
    {
      "epoch": 0.09053497942386832,
      "grad_norm": 1.6287739276885986,
      "learning_rate": 0.00011882998171846434,
      "loss": 0.0231,
      "step": 66
    },
    {
      "epoch": 0.09190672153635117,
      "grad_norm": 7.027708053588867,
      "learning_rate": 0.00012065813528336381,
      "loss": 0.3204,
      "step": 67
    },
    {
      "epoch": 0.09327846364883402,
      "grad_norm": 7.248253345489502,
      "learning_rate": 0.00012248628884826325,
      "loss": 0.3011,
      "step": 68
    },
    {
      "epoch": 0.09465020576131687,
      "grad_norm": 9.592718124389648,
      "learning_rate": 0.0001243144424131627,
      "loss": 0.3871,
      "step": 69
    },
    {
      "epoch": 0.09602194787379972,
      "grad_norm": 5.128874778747559,
      "learning_rate": 0.00012614259597806216,
      "loss": 0.1823,
      "step": 70
    },
    {
      "epoch": 0.09739368998628258,
      "grad_norm": 6.496853351593018,
      "learning_rate": 0.00012797074954296162,
      "loss": 0.3572,
      "step": 71
    },
    {
      "epoch": 0.09876543209876543,
      "grad_norm": 6.564659118652344,
      "learning_rate": 0.00012979890310786104,
      "loss": 0.5289,
      "step": 72
    },
    {
      "epoch": 0.10013717421124829,
      "grad_norm": 6.480371952056885,
      "learning_rate": 0.0001316270566727605,
      "loss": 0.3223,
      "step": 73
    },
    {
      "epoch": 0.10150891632373114,
      "grad_norm": 7.222306728363037,
      "learning_rate": 0.00013345521023765998,
      "loss": 0.3247,
      "step": 74
    },
    {
      "epoch": 0.102880658436214,
      "grad_norm": 5.406076431274414,
      "learning_rate": 0.00013528336380255943,
      "loss": 0.2133,
      "step": 75
    },
    {
      "epoch": 0.10425240054869685,
      "grad_norm": 11.029163360595703,
      "learning_rate": 0.00013711151736745886,
      "loss": 0.8249,
      "step": 76
    },
    {
      "epoch": 0.1056241426611797,
      "grad_norm": 7.284115314483643,
      "learning_rate": 0.00013893967093235832,
      "loss": 0.4341,
      "step": 77
    },
    {
      "epoch": 0.10699588477366255,
      "grad_norm": 6.240738868713379,
      "learning_rate": 0.00014076782449725777,
      "loss": 0.2932,
      "step": 78
    },
    {
      "epoch": 0.1083676268861454,
      "grad_norm": 0.1745665967464447,
      "learning_rate": 0.00014259597806215722,
      "loss": 0.0099,
      "step": 79
    },
    {
      "epoch": 0.10973936899862825,
      "grad_norm": 5.460353851318359,
      "learning_rate": 0.00014442413162705668,
      "loss": 0.3348,
      "step": 80
    },
    {
      "epoch": 0.1111111111111111,
      "grad_norm": 8.869246482849121,
      "learning_rate": 0.00014625228519195613,
      "loss": 0.6405,
      "step": 81
    },
    {
      "epoch": 0.11248285322359397,
      "grad_norm": 4.475996971130371,
      "learning_rate": 0.0001480804387568556,
      "loss": 0.1536,
      "step": 82
    },
    {
      "epoch": 0.11385459533607682,
      "grad_norm": 2.700299024581909,
      "learning_rate": 0.00014990859232175501,
      "loss": 0.1299,
      "step": 83
    },
    {
      "epoch": 0.11522633744855967,
      "grad_norm": 7.5515618324279785,
      "learning_rate": 0.00015173674588665447,
      "loss": 0.5863,
      "step": 84
    },
    {
      "epoch": 0.11659807956104253,
      "grad_norm": 9.869407653808594,
      "learning_rate": 0.00015356489945155395,
      "loss": 0.7205,
      "step": 85
    },
    {
      "epoch": 0.11796982167352538,
      "grad_norm": 8.208423614501953,
      "learning_rate": 0.00015539305301645338,
      "loss": 0.4052,
      "step": 86
    },
    {
      "epoch": 0.11934156378600823,
      "grad_norm": 6.408420562744141,
      "learning_rate": 0.00015722120658135283,
      "loss": 0.3953,
      "step": 87
    },
    {
      "epoch": 0.12071330589849108,
      "grad_norm": 7.050099849700928,
      "learning_rate": 0.00015904936014625229,
      "loss": 0.5598,
      "step": 88
    },
    {
      "epoch": 0.12208504801097393,
      "grad_norm": 5.326991558074951,
      "learning_rate": 0.00016087751371115174,
      "loss": 0.2856,
      "step": 89
    },
    {
      "epoch": 0.12345679012345678,
      "grad_norm": 4.510193347930908,
      "learning_rate": 0.0001627056672760512,
      "loss": 0.2277,
      "step": 90
    },
    {
      "epoch": 0.12482853223593965,
      "grad_norm": 5.769596576690674,
      "learning_rate": 0.00016453382084095065,
      "loss": 0.3296,
      "step": 91
    },
    {
      "epoch": 0.1262002743484225,
      "grad_norm": 6.066390037536621,
      "learning_rate": 0.0001663619744058501,
      "loss": 0.3079,
      "step": 92
    },
    {
      "epoch": 0.12757201646090535,
      "grad_norm": 6.80173921585083,
      "learning_rate": 0.00016819012797074953,
      "loss": 0.4867,
      "step": 93
    },
    {
      "epoch": 0.1289437585733882,
      "grad_norm": 6.219693183898926,
      "learning_rate": 0.00017001828153564899,
      "loss": 0.4319,
      "step": 94
    },
    {
      "epoch": 0.13031550068587106,
      "grad_norm": 5.316290855407715,
      "learning_rate": 0.00017184643510054847,
      "loss": 0.2952,
      "step": 95
    },
    {
      "epoch": 0.13168724279835392,
      "grad_norm": 6.86447811126709,
      "learning_rate": 0.00017367458866544792,
      "loss": 0.5531,
      "step": 96
    },
    {
      "epoch": 0.13305898491083676,
      "grad_norm": 1.2648167610168457,
      "learning_rate": 0.00017550274223034735,
      "loss": 0.0296,
      "step": 97
    },
    {
      "epoch": 0.13443072702331962,
      "grad_norm": 8.14661979675293,
      "learning_rate": 0.0001773308957952468,
      "loss": 0.8536,
      "step": 98
    },
    {
      "epoch": 0.13580246913580246,
      "grad_norm": 8.927884101867676,
      "learning_rate": 0.00017915904936014626,
      "loss": 0.4879,
      "step": 99
    },
    {
      "epoch": 0.13717421124828533,
      "grad_norm": 9.555243492126465,
      "learning_rate": 0.00018098720292504568,
      "loss": 0.67,
      "step": 100
    },
    {
      "epoch": 0.13854595336076816,
      "grad_norm": 7.783656120300293,
      "learning_rate": 0.00018281535648994517,
      "loss": 0.4813,
      "step": 101
    },
    {
      "epoch": 0.13991769547325103,
      "grad_norm": 0.5169872641563416,
      "learning_rate": 0.00018464351005484462,
      "loss": 0.0488,
      "step": 102
    },
    {
      "epoch": 0.1412894375857339,
      "grad_norm": 6.967692852020264,
      "learning_rate": 0.00018647166361974407,
      "loss": 0.5388,
      "step": 103
    },
    {
      "epoch": 0.14266117969821673,
      "grad_norm": 6.324373245239258,
      "learning_rate": 0.0001882998171846435,
      "loss": 0.376,
      "step": 104
    },
    {
      "epoch": 0.1440329218106996,
      "grad_norm": 0.7642683982849121,
      "learning_rate": 0.00019012797074954296,
      "loss": 0.017,
      "step": 105
    },
    {
      "epoch": 0.14540466392318244,
      "grad_norm": 8.600672721862793,
      "learning_rate": 0.00019195612431444244,
      "loss": 0.7542,
      "step": 106
    },
    {
      "epoch": 0.1467764060356653,
      "grad_norm": 7.111880302429199,
      "learning_rate": 0.00019378427787934186,
      "loss": 0.4063,
      "step": 107
    },
    {
      "epoch": 0.14814814814814814,
      "grad_norm": 6.075577735900879,
      "learning_rate": 0.00019561243144424132,
      "loss": 0.3658,
      "step": 108
    },
    {
      "epoch": 0.149519890260631,
      "grad_norm": 6.12313175201416,
      "learning_rate": 0.00019744058500914077,
      "loss": 0.4389,
      "step": 109
    },
    {
      "epoch": 0.15089163237311384,
      "grad_norm": 5.813235759735107,
      "learning_rate": 0.00019926873857404023,
      "loss": 0.3803,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_Qnli-dev_cosine_accuracy": 0.705078125,
      "eval_Qnli-dev_cosine_accuracy_threshold": 0.6866907477378845,
      "eval_Qnli-dev_cosine_ap": 0.7567018413685389,
      "eval_Qnli-dev_cosine_f1": 0.6931818181818182,
      "eval_Qnli-dev_cosine_f1_threshold": 0.6343963146209717,
      "eval_Qnli-dev_cosine_precision": 0.6267123287671232,
      "eval_Qnli-dev_cosine_recall": 0.7754237288135594,
      "eval_allNLI-dev_cosine_accuracy": 0.76953125,
      "eval_allNLI-dev_cosine_accuracy_threshold": 0.7752166986465454,
      "eval_allNLI-dev_cosine_ap": 0.6627175481841632,
      "eval_allNLI-dev_cosine_f1": 0.6624737945492662,
      "eval_allNLI-dev_cosine_f1_threshold": 0.6564935445785522,
      "eval_allNLI-dev_cosine_precision": 0.5197368421052632,
      "eval_allNLI-dev_cosine_recall": 0.9132947976878613,
      "eval_sequential_score": 0.7567018413685389,
      "eval_sts-test_pearson_cosine": 0.9026620207137961,
      "eval_sts-test_spearman_cosine": 0.913678627606199,
      "eval_vitaminc-pairs_loss": 2.009296178817749,
      "eval_vitaminc-pairs_runtime": 14.3224,
      "eval_vitaminc-pairs_samples_per_second": 8.937,
      "eval_vitaminc-pairs_steps_per_second": 0.07,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_negation-triplets_loss": 1.59572434425354,
      "eval_negation-triplets_runtime": 1.1528,
      "eval_negation-triplets_samples_per_second": 111.029,
      "eval_negation-triplets_steps_per_second": 0.867,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_scitail-pairs-pos_loss": 0.061776161193847656,
      "eval_scitail-pairs-pos_runtime": 1.5728,
      "eval_scitail-pairs-pos_samples_per_second": 81.383,
      "eval_scitail-pairs-pos_steps_per_second": 0.636,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_scitail-pairs-qa_loss": 0.009187542833387852,
      "eval_scitail-pairs-qa_runtime": 1.2102,
      "eval_scitail-pairs-qa_samples_per_second": 105.771,
      "eval_scitail-pairs-qa_steps_per_second": 0.826,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_xsum-pairs_loss": 0.37210211157798767,
      "eval_xsum-pairs_runtime": 6.2854,
      "eval_xsum-pairs_samples_per_second": 20.365,
      "eval_xsum-pairs_steps_per_second": 0.159,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_sciq_pairs_loss": 0.04122849553823471,
      "eval_sciq_pairs_runtime": 8.8116,
      "eval_sciq_pairs_samples_per_second": 14.526,
      "eval_sciq_pairs_steps_per_second": 0.113,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_qasc_pairs_loss": 0.4748501479625702,
      "eval_qasc_pairs_runtime": 1.3827,
      "eval_qasc_pairs_samples_per_second": 92.573,
      "eval_qasc_pairs_steps_per_second": 0.723,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_openbookqa_pairs_loss": 1.1540580987930298,
      "eval_openbookqa_pairs_runtime": 1.1788,
      "eval_openbookqa_pairs_samples_per_second": 108.581,
      "eval_openbookqa_pairs_steps_per_second": 0.848,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_nq_pairs_loss": 0.2363465428352356,
      "eval_nq_pairs_runtime": 7.8515,
      "eval_nq_pairs_samples_per_second": 16.303,
      "eval_nq_pairs_steps_per_second": 0.127,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_trivia_pairs_loss": 0.6520176529884338,
      "eval_trivia_pairs_runtime": 8.9067,
      "eval_trivia_pairs_samples_per_second": 14.371,
      "eval_trivia_pairs_steps_per_second": 0.112,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_gooaq_pairs_loss": 0.22620199620723724,
      "eval_gooaq_pairs_runtime": 2.067,
      "eval_gooaq_pairs_samples_per_second": 61.924,
      "eval_gooaq_pairs_steps_per_second": 0.484,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_paws-pos_loss": 0.02822125516831875,
      "eval_paws-pos_runtime": 1.5117,
      "eval_paws-pos_samples_per_second": 84.672,
      "eval_paws-pos_steps_per_second": 0.662,
      "step": 110
    },
    {
      "epoch": 0.15089163237311384,
      "eval_global_dataset_loss": 0.30668479204177856,
      "eval_global_dataset_runtime": 33.2591,
      "eval_global_dataset_samples_per_second": 11.546,
      "eval_global_dataset_steps_per_second": 0.06,
      "step": 110
    },
    {
      "epoch": 0.1522633744855967,
      "grad_norm": 4.30504846572876,
      "learning_rate": 0.00020109689213893968,
      "loss": 0.2478,
      "step": 111
    },
    {
      "epoch": 0.15363511659807957,
      "grad_norm": 6.559568881988525,
      "learning_rate": 0.00020292504570383914,
      "loss": 0.8402,
      "step": 112
    },
    {
      "epoch": 0.1550068587105624,
      "grad_norm": 5.812280654907227,
      "learning_rate": 0.0002047531992687386,
      "loss": 0.6608,
      "step": 113
    },
    {
      "epoch": 0.15637860082304528,
      "grad_norm": 2.0805885791778564,
      "learning_rate": 0.00020658135283363802,
      "loss": 0.0934,
      "step": 114
    },
    {
      "epoch": 0.15775034293552812,
      "grad_norm": 5.199294090270996,
      "learning_rate": 0.00020840950639853747,
      "loss": 0.3907,
      "step": 115
    },
    {
      "epoch": 0.15912208504801098,
      "grad_norm": 6.3685078620910645,
      "learning_rate": 0.00021023765996343693,
      "loss": 0.449,
      "step": 116
    },
    {
      "epoch": 0.16049382716049382,
      "grad_norm": 6.4199652671813965,
      "learning_rate": 0.00021206581352833638,
      "loss": 0.4041,
      "step": 117
    },
    {
      "epoch": 0.16186556927297668,
      "grad_norm": 6.015898704528809,
      "learning_rate": 0.00021389396709323584,
      "loss": 0.6749,
      "step": 118
    },
    {
      "epoch": 0.16323731138545952,
      "grad_norm": 7.721911430358887,
      "learning_rate": 0.0002157221206581353,
      "loss": 0.4847,
      "step": 119
    },
    {
      "epoch": 0.1646090534979424,
      "grad_norm": 1.8774610757827759,
      "learning_rate": 0.00021755027422303474,
      "loss": 0.0526,
      "step": 120
    },
    {
      "epoch": 0.16598079561042525,
      "grad_norm": 8.094359397888184,
      "learning_rate": 0.00021937842778793417,
      "loss": 0.6795,
      "step": 121
    },
    {
      "epoch": 0.1673525377229081,
      "grad_norm": 0.33090323209762573,
      "learning_rate": 0.00022120658135283365,
      "loss": 0.0064,
      "step": 122
    },
    {
      "epoch": 0.16872427983539096,
      "grad_norm": 7.3609418869018555,
      "learning_rate": 0.0002230347349177331,
      "loss": 0.5918,
      "step": 123
    },
    {
      "epoch": 0.1700960219478738,
      "grad_norm": 6.189216613769531,
      "learning_rate": 0.00022486288848263253,
      "loss": 0.3544,
      "step": 124
    },
    {
      "epoch": 0.17146776406035666,
      "grad_norm": 5.588890075683594,
      "learning_rate": 0.000226691042047532,
      "loss": 0.3849,
      "step": 125
    },
    {
      "epoch": 0.1728395061728395,
      "grad_norm": 3.4582345485687256,
      "learning_rate": 0.00022851919561243144,
      "loss": 0.2051,
      "step": 126
    },
    {
      "epoch": 0.17421124828532236,
      "grad_norm": 4.075862407684326,
      "learning_rate": 0.00023034734917733092,
      "loss": 0.2129,
      "step": 127
    },
    {
      "epoch": 0.1755829903978052,
      "grad_norm": 15.110091209411621,
      "learning_rate": 0.00023217550274223035,
      "loss": 2.7937,
      "step": 128
    },
    {
      "epoch": 0.17695473251028807,
      "grad_norm": 0.35791516304016113,
      "learning_rate": 0.0002340036563071298,
      "loss": 0.0166,
      "step": 129
    },
    {
      "epoch": 0.17832647462277093,
      "grad_norm": 7.5200090408325195,
      "learning_rate": 0.00023583180987202926,
      "loss": 0.7856,
      "step": 130
    },
    {
      "epoch": 0.17969821673525377,
      "grad_norm": 6.566864490509033,
      "learning_rate": 0.0002376599634369287,
      "loss": 0.8368,
      "step": 131
    },
    {
      "epoch": 0.18106995884773663,
      "grad_norm": 4.958701133728027,
      "learning_rate": 0.00023948811700182814,
      "loss": 0.3813,
      "step": 132
    },
    {
      "epoch": 0.18244170096021947,
      "grad_norm": 5.745133876800537,
      "learning_rate": 0.00024131627056672762,
      "loss": 0.5695,
      "step": 133
    },
    {
      "epoch": 0.18381344307270234,
      "grad_norm": 4.952736854553223,
      "learning_rate": 0.00024314442413162708,
      "loss": 0.351,
      "step": 134
    },
    {
      "epoch": 0.18518518518518517,
      "grad_norm": 5.733601093292236,
      "learning_rate": 0.0002449725776965265,
      "loss": 0.3821,
      "step": 135
    },
    {
      "epoch": 0.18655692729766804,
      "grad_norm": 5.019097328186035,
      "learning_rate": 0.00024680073126142596,
      "loss": 0.3249,
      "step": 136
    },
    {
      "epoch": 0.18792866941015088,
      "grad_norm": 5.300777912139893,
      "learning_rate": 0.0002486288848263254,
      "loss": 0.3404,
      "step": 137
    },
    {
      "epoch": 0.18930041152263374,
      "grad_norm": 4.518141269683838,
      "learning_rate": 0.00025045703839122487,
      "loss": 0.4535,
      "step": 138
    },
    {
      "epoch": 0.1906721536351166,
      "grad_norm": 1.0158088207244873,
      "learning_rate": 0.0002522851919561243,
      "loss": 0.0577,
      "step": 139
    },
    {
      "epoch": 0.19204389574759945,
      "grad_norm": 5.966796398162842,
      "learning_rate": 0.0002541133455210238,
      "loss": 0.7431,
      "step": 140
    },
    {
      "epoch": 0.1934156378600823,
      "grad_norm": 6.123642921447754,
      "learning_rate": 0.00025594149908592323,
      "loss": 0.6778,
      "step": 141
    },
    {
      "epoch": 0.19478737997256515,
      "grad_norm": 5.842874050140381,
      "learning_rate": 0.0002577696526508227,
      "loss": 0.5436,
      "step": 142
    },
    {
      "epoch": 0.19615912208504802,
      "grad_norm": 4.759068012237549,
      "learning_rate": 0.0002595978062157221,
      "loss": 0.3582,
      "step": 143
    },
    {
      "epoch": 0.19753086419753085,
      "grad_norm": 4.080338478088379,
      "learning_rate": 0.00026142595978062154,
      "loss": 0.316,
      "step": 144
    },
    {
      "epoch": 0.19890260631001372,
      "grad_norm": 4.1391448974609375,
      "learning_rate": 0.000263254113345521,
      "loss": 0.4446,
      "step": 145
    },
    {
      "epoch": 0.20027434842249658,
      "grad_norm": 5.856256008148193,
      "learning_rate": 0.0002650822669104205,
      "loss": 0.7792,
      "step": 146
    },
    {
      "epoch": 0.20164609053497942,
      "grad_norm": 7.747331142425537,
      "learning_rate": 0.00026691042047531996,
      "loss": 1.1147,
      "step": 147
    },
    {
      "epoch": 0.2030178326474623,
      "grad_norm": 6.825289249420166,
      "learning_rate": 0.0002687385740402194,
      "loss": 0.8267,
      "step": 148
    },
    {
      "epoch": 0.20438957475994513,
      "grad_norm": 7.336719512939453,
      "learning_rate": 0.00027056672760511887,
      "loss": 0.8149,
      "step": 149
    },
    {
      "epoch": 0.205761316872428,
      "grad_norm": 6.731626510620117,
      "learning_rate": 0.00027239488117001827,
      "loss": 0.942,
      "step": 150
    },
    {
      "epoch": 0.20713305898491083,
      "grad_norm": 10.727692604064941,
      "learning_rate": 0.0002742230347349177,
      "loss": 2.4865,
      "step": 151
    },
    {
      "epoch": 0.2085048010973937,
      "grad_norm": 8.583380699157715,
      "learning_rate": 0.0002760511882998172,
      "loss": 1.0715,
      "step": 152
    },
    {
      "epoch": 0.20987654320987653,
      "grad_norm": 6.236877918243408,
      "learning_rate": 0.00027787934186471663,
      "loss": 0.6219,
      "step": 153
    },
    {
      "epoch": 0.2112482853223594,
      "grad_norm": 6.254538536071777,
      "learning_rate": 0.0002797074954296161,
      "loss": 0.8705,
      "step": 154
    },
    {
      "epoch": 0.21262002743484226,
      "grad_norm": 3.0917959213256836,
      "learning_rate": 0.00028153564899451554,
      "loss": 0.2407,
      "step": 155
    },
    {
      "epoch": 0.2139917695473251,
      "grad_norm": 4.438024997711182,
      "learning_rate": 0.000283363802559415,
      "loss": 0.4925,
      "step": 156
    },
    {
      "epoch": 0.21536351165980797,
      "grad_norm": 0.43344631791114807,
      "learning_rate": 0.00028519195612431445,
      "loss": 0.0316,
      "step": 157
    },
    {
      "epoch": 0.2167352537722908,
      "grad_norm": 5.73934268951416,
      "learning_rate": 0.0002870201096892139,
      "loss": 0.3935,
      "step": 158
    },
    {
      "epoch": 0.21810699588477367,
      "grad_norm": 4.532804012298584,
      "learning_rate": 0.00028884826325411336,
      "loss": 0.2083,
      "step": 159
    },
    {
      "epoch": 0.2194787379972565,
      "grad_norm": 4.846848487854004,
      "learning_rate": 0.0002906764168190128,
      "loss": 0.2798,
      "step": 160
    },
    {
      "epoch": 0.22085048010973937,
      "grad_norm": 7.060863018035889,
      "learning_rate": 0.00029250457038391227,
      "loss": 0.8777,
      "step": 161
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 0.012754157185554504,
      "learning_rate": 0.0002943327239488117,
      "loss": 0.0002,
      "step": 162
    },
    {
      "epoch": 0.22359396433470508,
      "grad_norm": 4.094379901885986,
      "learning_rate": 0.0002961608775137112,
      "loss": 0.2736,
      "step": 163
    },
    {
      "epoch": 0.22496570644718794,
      "grad_norm": 10.741785049438477,
      "learning_rate": 0.0002979890310786106,
      "loss": 2.4185,
      "step": 164
    },
    {
      "epoch": 0.22633744855967078,
      "grad_norm": 4.820891380310059,
      "learning_rate": 0.00029981718464351003,
      "loss": 0.7767,
      "step": 165
    },
    {
      "epoch": 0.22770919067215364,
      "grad_norm": 6.423076152801514,
      "learning_rate": 0.0003016453382084095,
      "loss": 0.7971,
      "step": 166
    },
    {
      "epoch": 0.22908093278463648,
      "grad_norm": 4.492727756500244,
      "learning_rate": 0.00030347349177330894,
      "loss": 0.4535,
      "step": 167
    },
    {
      "epoch": 0.23045267489711935,
      "grad_norm": 5.301379680633545,
      "learning_rate": 0.00030530164533820845,
      "loss": 0.6654,
      "step": 168
    },
    {
      "epoch": 0.23182441700960219,
      "grad_norm": 5.155853748321533,
      "learning_rate": 0.0003071297989031079,
      "loss": 0.3985,
      "step": 169
    },
    {
      "epoch": 0.23319615912208505,
      "grad_norm": 0.4378865361213684,
      "learning_rate": 0.00030895795246800735,
      "loss": 0.0338,
      "step": 170
    },
    {
      "epoch": 0.2345679012345679,
      "grad_norm": 4.022473335266113,
      "learning_rate": 0.00031078610603290675,
      "loss": 0.1834,
      "step": 171
    },
    {
      "epoch": 0.23593964334705075,
      "grad_norm": 7.863429069519043,
      "learning_rate": 0.0003126142595978062,
      "loss": 0.603,
      "step": 172
    },
    {
      "epoch": 0.23731138545953362,
      "grad_norm": 8.951998710632324,
      "learning_rate": 0.00031444241316270566,
      "loss": 0.7871,
      "step": 173
    },
    {
      "epoch": 0.23868312757201646,
      "grad_norm": 6.265102386474609,
      "learning_rate": 0.0003162705667276051,
      "loss": 0.4304,
      "step": 174
    },
    {
      "epoch": 0.24005486968449932,
      "grad_norm": 6.6486005783081055,
      "learning_rate": 0.00031809872029250457,
      "loss": 0.649,
      "step": 175
    },
    {
      "epoch": 0.24142661179698216,
      "grad_norm": 0.47100114822387695,
      "learning_rate": 0.000319926873857404,
      "loss": 0.048,
      "step": 176
    },
    {
      "epoch": 0.24279835390946503,
      "grad_norm": 4.884115695953369,
      "learning_rate": 0.0003217550274223035,
      "loss": 0.4079,
      "step": 177
    },
    {
      "epoch": 0.24417009602194786,
      "grad_norm": 4.508667469024658,
      "learning_rate": 0.0003235831809872029,
      "loss": 0.4627,
      "step": 178
    },
    {
      "epoch": 0.24554183813443073,
      "grad_norm": 3.22367262840271,
      "learning_rate": 0.0003254113345521024,
      "loss": 0.3703,
      "step": 179
    },
    {
      "epoch": 0.24691358024691357,
      "grad_norm": 7.695303916931152,
      "learning_rate": 0.00032723948811700184,
      "loss": 0.8343,
      "step": 180
    },
    {
      "epoch": 0.24828532235939643,
      "grad_norm": 7.249318599700928,
      "learning_rate": 0.0003290676416819013,
      "loss": 0.692,
      "step": 181
    },
    {
      "epoch": 0.2496570644718793,
      "grad_norm": 11.686202049255371,
      "learning_rate": 0.00033089579524680075,
      "loss": 2.7071,
      "step": 182
    },
    {
      "epoch": 0.25102880658436216,
      "grad_norm": 6.061092376708984,
      "learning_rate": 0.0003327239488117002,
      "loss": 0.8451,
      "step": 183
    },
    {
      "epoch": 0.252400548696845,
      "grad_norm": 5.932607650756836,
      "learning_rate": 0.00033455210237659966,
      "loss": 0.635,
      "step": 184
    },
    {
      "epoch": 0.25377229080932784,
      "grad_norm": 3.491114616394043,
      "learning_rate": 0.00033638025594149906,
      "loss": 0.312,
      "step": 185
    },
    {
      "epoch": 0.2551440329218107,
      "grad_norm": 6.4914164543151855,
      "learning_rate": 0.0003382084095063985,
      "loss": 0.6996,
      "step": 186
    },
    {
      "epoch": 0.25651577503429357,
      "grad_norm": 6.15857458114624,
      "learning_rate": 0.00034003656307129797,
      "loss": 0.4432,
      "step": 187
    },
    {
      "epoch": 0.2578875171467764,
      "grad_norm": 4.767185211181641,
      "learning_rate": 0.0003418647166361974,
      "loss": 0.375,
      "step": 188
    },
    {
      "epoch": 0.25925925925925924,
      "grad_norm": 7.944342613220215,
      "learning_rate": 0.00034369287020109693,
      "loss": 0.9366,
      "step": 189
    },
    {
      "epoch": 0.2606310013717421,
      "grad_norm": 6.573953628540039,
      "learning_rate": 0.0003455210237659964,
      "loss": 0.755,
      "step": 190
    },
    {
      "epoch": 0.262002743484225,
      "grad_norm": 4.173367023468018,
      "learning_rate": 0.00034734917733089584,
      "loss": 0.6068,
      "step": 191
    },
    {
      "epoch": 0.26337448559670784,
      "grad_norm": 5.26171875,
      "learning_rate": 0.00034917733089579524,
      "loss": 0.5336,
      "step": 192
    },
    {
      "epoch": 0.26474622770919065,
      "grad_norm": 6.669304370880127,
      "learning_rate": 0.0003510054844606947,
      "loss": 0.8783,
      "step": 193
    },
    {
      "epoch": 0.2661179698216735,
      "grad_norm": 4.4192938804626465,
      "learning_rate": 0.00035283363802559415,
      "loss": 0.3576,
      "step": 194
    },
    {
      "epoch": 0.2674897119341564,
      "grad_norm": 10.117819786071777,
      "learning_rate": 0.0003546617915904936,
      "loss": 2.1854,
      "step": 195
    },
    {
      "epoch": 0.26886145404663925,
      "grad_norm": 5.256247520446777,
      "learning_rate": 0.00035648994515539306,
      "loss": 0.7835,
      "step": 196
    },
    {
      "epoch": 0.27023319615912206,
      "grad_norm": 5.784887313842773,
      "learning_rate": 0.0003583180987202925,
      "loss": 0.5668,
      "step": 197
    },
    {
      "epoch": 0.2716049382716049,
      "grad_norm": 4.977567672729492,
      "learning_rate": 0.00036014625228519197,
      "loss": 0.7033,
      "step": 198
    },
    {
      "epoch": 0.2729766803840878,
      "grad_norm": 0.011424711905419827,
      "learning_rate": 0.00036197440585009137,
      "loss": 0.0002,
      "step": 199
    },
    {
      "epoch": 0.27434842249657065,
      "grad_norm": 5.805008411407471,
      "learning_rate": 0.0003638025594149909,
      "loss": 0.5791,
      "step": 200
    },
    {
      "epoch": 0.2757201646090535,
      "grad_norm": 3.8826043605804443,
      "learning_rate": 0.00036563071297989033,
      "loss": 0.2697,
      "step": 201
    },
    {
      "epoch": 0.27709190672153633,
      "grad_norm": 6.563521385192871,
      "learning_rate": 0.0003674588665447898,
      "loss": 0.6261,
      "step": 202
    },
    {
      "epoch": 0.2784636488340192,
      "grad_norm": 4.584529399871826,
      "learning_rate": 0.00036928702010968924,
      "loss": 0.3253,
      "step": 203
    },
    {
      "epoch": 0.27983539094650206,
      "grad_norm": 6.636009216308594,
      "learning_rate": 0.0003711151736745887,
      "loss": 0.8323,
      "step": 204
    },
    {
      "epoch": 0.2812071330589849,
      "grad_norm": 5.0911359786987305,
      "learning_rate": 0.00037294332723948815,
      "loss": 0.4472,
      "step": 205
    },
    {
      "epoch": 0.2825788751714678,
      "grad_norm": 3.9219255447387695,
      "learning_rate": 0.00037477148080438755,
      "loss": 0.3342,
      "step": 206
    },
    {
      "epoch": 0.2839506172839506,
      "grad_norm": 5.114777565002441,
      "learning_rate": 0.000376599634369287,
      "loss": 0.6313,
      "step": 207
    },
    {
      "epoch": 0.28532235939643347,
      "grad_norm": 0.3298715353012085,
      "learning_rate": 0.00037842778793418646,
      "loss": 0.059,
      "step": 208
    },
    {
      "epoch": 0.28669410150891633,
      "grad_norm": 1.5965046882629395,
      "learning_rate": 0.0003802559414990859,
      "loss": 0.1195,
      "step": 209
    },
    {
      "epoch": 0.2880658436213992,
      "grad_norm": 0.39121323823928833,
      "learning_rate": 0.00038208409506398537,
      "loss": 0.0296,
      "step": 210
    },
    {
      "epoch": 0.289437585733882,
      "grad_norm": 4.317224025726318,
      "learning_rate": 0.0003839122486288849,
      "loss": 0.5316,
      "step": 211
    },
    {
      "epoch": 0.2908093278463649,
      "grad_norm": 4.000308036804199,
      "learning_rate": 0.00038574040219378433,
      "loss": 0.5201,
      "step": 212
    },
    {
      "epoch": 0.29218106995884774,
      "grad_norm": 6.2192301750183105,
      "learning_rate": 0.00038756855575868373,
      "loss": 0.6602,
      "step": 213
    },
    {
      "epoch": 0.2935528120713306,
      "grad_norm": 6.702320098876953,
      "learning_rate": 0.0003893967093235832,
      "loss": 0.9578,
      "step": 214
    },
    {
      "epoch": 0.29492455418381347,
      "grad_norm": 3.9136242866516113,
      "learning_rate": 0.00039122486288848264,
      "loss": 0.2089,
      "step": 215
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": 6.901303768157959,
      "learning_rate": 0.0003930530164533821,
      "loss": 1.2112,
      "step": 216
    },
    {
      "epoch": 0.29766803840877915,
      "grad_norm": 4.04884672164917,
      "learning_rate": 0.00039488117001828155,
      "loss": 0.3294,
      "step": 217
    },
    {
      "epoch": 0.299039780521262,
      "grad_norm": 5.46201753616333,
      "learning_rate": 0.000396709323583181,
      "loss": 0.867,
      "step": 218
    },
    {
      "epoch": 0.3004115226337449,
      "grad_norm": 5.559458255767822,
      "learning_rate": 0.00039853747714808046,
      "loss": 1.1745,
      "step": 219
    },
    {
      "epoch": 0.3017832647462277,
      "grad_norm": 4.930731296539307,
      "learning_rate": 0.00040036563071297986,
      "loss": 0.7287,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_Qnli-dev_cosine_accuracy": 0.736328125,
      "eval_Qnli-dev_cosine_accuracy_threshold": 0.8059661388397217,
      "eval_Qnli-dev_cosine_ap": 0.7693397822540732,
      "eval_Qnli-dev_cosine_f1": 0.720136518771331,
      "eval_Qnli-dev_cosine_f1_threshold": 0.7346209287643433,
      "eval_Qnli-dev_cosine_precision": 0.6028571428571429,
      "eval_Qnli-dev_cosine_recall": 0.8940677966101694,
      "eval_allNLI-dev_cosine_accuracy": 0.751953125,
      "eval_allNLI-dev_cosine_accuracy_threshold": 0.8584632873535156,
      "eval_allNLI-dev_cosine_ap": 0.6542482370347211,
      "eval_allNLI-dev_cosine_f1": 0.6681127982646421,
      "eval_allNLI-dev_cosine_f1_threshold": 0.7551147937774658,
      "eval_allNLI-dev_cosine_precision": 0.5347222222222222,
      "eval_allNLI-dev_cosine_recall": 0.8901734104046243,
      "eval_sequential_score": 0.7693397822540732,
      "eval_sts-test_pearson_cosine": 0.8635423204589071,
      "eval_sts-test_spearman_cosine": 0.8907274956890058,
      "eval_vitaminc-pairs_loss": 2.5131773948669434,
      "eval_vitaminc-pairs_runtime": 14.3452,
      "eval_vitaminc-pairs_samples_per_second": 8.923,
      "eval_vitaminc-pairs_steps_per_second": 0.07,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_negation-triplets_loss": 1.5992437601089478,
      "eval_negation-triplets_runtime": 1.1695,
      "eval_negation-triplets_samples_per_second": 109.451,
      "eval_negation-triplets_steps_per_second": 0.855,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_scitail-pairs-pos_loss": 0.12100159376859665,
      "eval_scitail-pairs-pos_runtime": 1.5843,
      "eval_scitail-pairs-pos_samples_per_second": 80.791,
      "eval_scitail-pairs-pos_steps_per_second": 0.631,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_scitail-pairs-qa_loss": 0.029047677293419838,
      "eval_scitail-pairs-qa_runtime": 1.2089,
      "eval_scitail-pairs-qa_samples_per_second": 105.881,
      "eval_scitail-pairs-qa_steps_per_second": 0.827,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_xsum-pairs_loss": 0.6063941717147827,
      "eval_xsum-pairs_runtime": 6.3291,
      "eval_xsum-pairs_samples_per_second": 20.224,
      "eval_xsum-pairs_steps_per_second": 0.158,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_sciq_pairs_loss": 0.05205194652080536,
      "eval_sciq_pairs_runtime": 8.8533,
      "eval_sciq_pairs_samples_per_second": 14.458,
      "eval_sciq_pairs_steps_per_second": 0.113,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_qasc_pairs_loss": 0.7244825959205627,
      "eval_qasc_pairs_runtime": 1.3845,
      "eval_qasc_pairs_samples_per_second": 92.449,
      "eval_qasc_pairs_steps_per_second": 0.722,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_openbookqa_pairs_loss": 1.4260488748550415,
      "eval_openbookqa_pairs_runtime": 1.1784,
      "eval_openbookqa_pairs_samples_per_second": 108.623,
      "eval_openbookqa_pairs_steps_per_second": 0.849,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_nq_pairs_loss": 0.7104523777961731,
      "eval_nq_pairs_runtime": 7.8911,
      "eval_nq_pairs_samples_per_second": 16.221,
      "eval_nq_pairs_steps_per_second": 0.127,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_trivia_pairs_loss": 0.8537120223045349,
      "eval_trivia_pairs_runtime": 8.9305,
      "eval_trivia_pairs_samples_per_second": 14.333,
      "eval_trivia_pairs_steps_per_second": 0.112,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_gooaq_pairs_loss": 0.5029886960983276,
      "eval_gooaq_pairs_runtime": 2.0675,
      "eval_gooaq_pairs_samples_per_second": 61.91,
      "eval_gooaq_pairs_steps_per_second": 0.484,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_paws-pos_loss": 0.03150199353694916,
      "eval_paws-pos_runtime": 1.5078,
      "eval_paws-pos_samples_per_second": 84.892,
      "eval_paws-pos_steps_per_second": 0.663,
      "step": 220
    },
    {
      "epoch": 0.3017832647462277,
      "eval_global_dataset_loss": 0.18923546373844147,
      "eval_global_dataset_runtime": 33.3421,
      "eval_global_dataset_samples_per_second": 11.517,
      "eval_global_dataset_steps_per_second": 0.06,
      "step": 220
    },
    {
      "epoch": 0.30315500685871055,
      "grad_norm": 3.582907199859619,
      "learning_rate": 0.00040219378427787936,
      "loss": 0.5484,
      "step": 221
    },
    {
      "epoch": 0.3045267489711934,
      "grad_norm": 4.960206031799316,
      "learning_rate": 0.0004040219378427788,
      "loss": 0.9396,
      "step": 222
    },
    {
      "epoch": 0.3058984910836763,
      "grad_norm": 4.219746112823486,
      "learning_rate": 0.0004058500914076783,
      "loss": 0.4335,
      "step": 223
    },
    {
      "epoch": 0.30727023319615915,
      "grad_norm": 6.449894428253174,
      "learning_rate": 0.00040767824497257773,
      "loss": 0.9026,
      "step": 224
    },
    {
      "epoch": 0.30864197530864196,
      "grad_norm": 6.239223003387451,
      "learning_rate": 0.0004095063985374772,
      "loss": 0.7214,
      "step": 225
    },
    {
      "epoch": 0.3100137174211248,
      "grad_norm": 4.418921947479248,
      "learning_rate": 0.00041133455210237664,
      "loss": 0.4794,
      "step": 226
    },
    {
      "epoch": 0.3113854595336077,
      "grad_norm": 0.012527555227279663,
      "learning_rate": 0.00041316270566727604,
      "loss": 0.0003,
      "step": 227
    },
    {
      "epoch": 0.31275720164609055,
      "grad_norm": 2.672603130340576,
      "learning_rate": 0.0004149908592321755,
      "loss": 0.3003,
      "step": 228
    },
    {
      "epoch": 0.31412894375857336,
      "grad_norm": 4.433743476867676,
      "learning_rate": 0.00041681901279707495,
      "loss": 0.4667,
      "step": 229
    },
    {
      "epoch": 0.31550068587105623,
      "grad_norm": 4.458980083465576,
      "learning_rate": 0.0004186471663619744,
      "loss": 0.5006,
      "step": 230
    },
    {
      "epoch": 0.3168724279835391,
      "grad_norm": 5.0898237228393555,
      "learning_rate": 0.00042047531992687385,
      "loss": 0.5555,
      "step": 231
    },
    {
      "epoch": 0.31824417009602196,
      "grad_norm": 4.338139533996582,
      "learning_rate": 0.00042230347349177336,
      "loss": 0.4437,
      "step": 232
    },
    {
      "epoch": 0.3196159122085048,
      "grad_norm": 5.023694038391113,
      "learning_rate": 0.00042413162705667276,
      "loss": 0.8813,
      "step": 233
    },
    {
      "epoch": 0.32098765432098764,
      "grad_norm": 6.410233497619629,
      "learning_rate": 0.0004259597806215722,
      "loss": 1.1836,
      "step": 234
    },
    {
      "epoch": 0.3223593964334705,
      "grad_norm": 3.8459813594818115,
      "learning_rate": 0.00042778793418647167,
      "loss": 0.3176,
      "step": 235
    },
    {
      "epoch": 0.32373113854595337,
      "grad_norm": 5.539570331573486,
      "learning_rate": 0.0004296160877513711,
      "loss": 0.6248,
      "step": 236
    },
    {
      "epoch": 0.32510288065843623,
      "grad_norm": 3.5191774368286133,
      "learning_rate": 0.0004314442413162706,
      "loss": 0.3623,
      "step": 237
    },
    {
      "epoch": 0.32647462277091904,
      "grad_norm": 3.2997043132781982,
      "learning_rate": 0.00043327239488117003,
      "loss": 0.3205,
      "step": 238
    },
    {
      "epoch": 0.3278463648834019,
      "grad_norm": 3.9236536026000977,
      "learning_rate": 0.0004351005484460695,
      "loss": 0.439,
      "step": 239
    },
    {
      "epoch": 0.3292181069958848,
      "grad_norm": 3.9750499725341797,
      "learning_rate": 0.0004369287020109689,
      "loss": 0.653,
      "step": 240
    },
    {
      "epoch": 0.33058984910836764,
      "grad_norm": 4.344120502471924,
      "learning_rate": 0.00043875685557586834,
      "loss": 0.5743,
      "step": 241
    },
    {
      "epoch": 0.3319615912208505,
      "grad_norm": 3.2905893325805664,
      "learning_rate": 0.0004405850091407678,
      "loss": 0.5844,
      "step": 242
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 5.996461391448975,
      "learning_rate": 0.0004424131627056673,
      "loss": 0.9451,
      "step": 243
    },
    {
      "epoch": 0.3347050754458162,
      "grad_norm": 4.085718631744385,
      "learning_rate": 0.00044424131627056676,
      "loss": 0.7071,
      "step": 244
    },
    {
      "epoch": 0.33607681755829905,
      "grad_norm": 0.26297450065612793,
      "learning_rate": 0.0004460694698354662,
      "loss": 0.0226,
      "step": 245
    },
    {
      "epoch": 0.3374485596707819,
      "grad_norm": 5.191401481628418,
      "learning_rate": 0.00044789762340036567,
      "loss": 1.0585,
      "step": 246
    },
    {
      "epoch": 0.3388203017832647,
      "grad_norm": 5.426116466522217,
      "learning_rate": 0.00044972577696526507,
      "loss": 1.0764,
      "step": 247
    },
    {
      "epoch": 0.3401920438957476,
      "grad_norm": 0.17406082153320312,
      "learning_rate": 0.0004515539305301645,
      "loss": 0.0289,
      "step": 248
    },
    {
      "epoch": 0.34156378600823045,
      "grad_norm": 4.900349140167236,
      "learning_rate": 0.000453382084095064,
      "loss": 0.5588,
      "step": 249
    },
    {
      "epoch": 0.3429355281207133,
      "grad_norm": 5.373581409454346,
      "learning_rate": 0.00045521023765996343,
      "loss": 0.7509,
      "step": 250
    },
    {
      "epoch": 0.3443072702331962,
      "grad_norm": 5.4629106521606445,
      "learning_rate": 0.0004570383912248629,
      "loss": 0.8388,
      "step": 251
    },
    {
      "epoch": 0.345679012345679,
      "grad_norm": 4.140360355377197,
      "learning_rate": 0.00045886654478976234,
      "loss": 0.5444,
      "step": 252
    },
    {
      "epoch": 0.34705075445816186,
      "grad_norm": 5.176646709442139,
      "learning_rate": 0.00046069469835466185,
      "loss": 1.2432,
      "step": 253
    },
    {
      "epoch": 0.3484224965706447,
      "grad_norm": 5.172772407531738,
      "learning_rate": 0.00046252285191956125,
      "loss": 0.8329,
      "step": 254
    },
    {
      "epoch": 0.3497942386831276,
      "grad_norm": 4.968120574951172,
      "learning_rate": 0.0004643510054844607,
      "loss": 0.7158,
      "step": 255
    },
    {
      "epoch": 0.3511659807956104,
      "grad_norm": 5.91867208480835,
      "learning_rate": 0.00046617915904936016,
      "loss": 0.935,
      "step": 256
    },
    {
      "epoch": 0.35253772290809327,
      "grad_norm": 6.223313808441162,
      "learning_rate": 0.0004680073126142596,
      "loss": 0.8777,
      "step": 257
    },
    {
      "epoch": 0.35390946502057613,
      "grad_norm": 11.474114418029785,
      "learning_rate": 0.00046983546617915907,
      "loss": 3.1178,
      "step": 258
    },
    {
      "epoch": 0.355281207133059,
      "grad_norm": 6.462806701660156,
      "learning_rate": 0.0004716636197440585,
      "loss": 0.9418,
      "step": 259
    },
    {
      "epoch": 0.35665294924554186,
      "grad_norm": 5.5286173820495605,
      "learning_rate": 0.000473491773308958,
      "loss": 0.6252,
      "step": 260
    },
    {
      "epoch": 0.35802469135802467,
      "grad_norm": 8.872392654418945,
      "learning_rate": 0.0004753199268738574,
      "loss": 2.9074,
      "step": 261
    },
    {
      "epoch": 0.35939643347050754,
      "grad_norm": 2.4954137802124023,
      "learning_rate": 0.00047714808043875683,
      "loss": 0.303,
      "step": 262
    },
    {
      "epoch": 0.3607681755829904,
      "grad_norm": 3.22896671295166,
      "learning_rate": 0.0004789762340036563,
      "loss": 0.7146,
      "step": 263
    },
    {
      "epoch": 0.36213991769547327,
      "grad_norm": 4.280979156494141,
      "learning_rate": 0.0004808043875685558,
      "loss": 0.6848,
      "step": 264
    },
    {
      "epoch": 0.3635116598079561,
      "grad_norm": 3.9235236644744873,
      "learning_rate": 0.00048263254113345525,
      "loss": 0.47,
      "step": 265
    },
    {
      "epoch": 0.36488340192043894,
      "grad_norm": 3.460500717163086,
      "learning_rate": 0.0004844606946983547,
      "loss": 0.5762,
      "step": 266
    },
    {
      "epoch": 0.3662551440329218,
      "grad_norm": 11.026795387268066,
      "learning_rate": 0.00048628884826325416,
      "loss": 3.3872,
      "step": 267
    },
    {
      "epoch": 0.3676268861454047,
      "grad_norm": 3.2708449363708496,
      "learning_rate": 0.00048811700182815356,
      "loss": 0.4195,
      "step": 268
    },
    {
      "epoch": 0.36899862825788754,
      "grad_norm": 5.177340030670166,
      "learning_rate": 0.000489945155393053,
      "loss": 1.2292,
      "step": 269
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 5.413723945617676,
      "learning_rate": 0.0004917733089579525,
      "loss": 1.1249,
      "step": 270
    },
    {
      "epoch": 0.3717421124828532,
      "grad_norm": 4.922053337097168,
      "learning_rate": 0.0004936014625228519,
      "loss": 1.0863,
      "step": 271
    },
    {
      "epoch": 0.3731138545953361,
      "grad_norm": 4.433996677398682,
      "learning_rate": 0.0004954296160877514,
      "loss": 0.9361,
      "step": 272
    },
    {
      "epoch": 0.37448559670781895,
      "grad_norm": 5.205246448516846,
      "learning_rate": 0.0004972577696526508,
      "loss": 0.7965,
      "step": 273
    },
    {
      "epoch": 0.37585733882030176,
      "grad_norm": 4.139344215393066,
      "learning_rate": 0.0004990859232175503,
      "loss": 0.7914,
      "step": 274
    },
    {
      "epoch": 0.3772290809327846,
      "grad_norm": 0.1397838592529297,
      "learning_rate": 0.0005009140767824497,
      "loss": 0.0027,
      "step": 275
    },
    {
      "epoch": 0.3786008230452675,
      "grad_norm": 3.3880808353424072,
      "learning_rate": 0.0005027422303473492,
      "loss": 0.6585,
      "step": 276
    },
    {
      "epoch": 0.37997256515775035,
      "grad_norm": 4.524999141693115,
      "learning_rate": 0.0005045703839122486,
      "loss": 0.5388,
      "step": 277
    },
    {
      "epoch": 0.3813443072702332,
      "grad_norm": 6.445588111877441,
      "learning_rate": 0.000506398537477148,
      "loss": 1.238,
      "step": 278
    },
    {
      "epoch": 0.38271604938271603,
      "grad_norm": 2.8341169357299805,
      "learning_rate": 0.0005082266910420476,
      "loss": 0.3782,
      "step": 279
    },
    {
      "epoch": 0.3840877914951989,
      "grad_norm": 5.0054240226745605,
      "learning_rate": 0.0005100548446069469,
      "loss": 1.1769,
      "step": 280
    },
    {
      "epoch": 0.38545953360768176,
      "grad_norm": 3.6371870040893555,
      "learning_rate": 0.0005118829981718465,
      "loss": 0.4808,
      "step": 281
    },
    {
      "epoch": 0.3868312757201646,
      "grad_norm": 8.928922653198242,
      "learning_rate": 0.0005137111517367459,
      "loss": 3.2104,
      "step": 282
    },
    {
      "epoch": 0.38820301783264743,
      "grad_norm": 2.7686235904693604,
      "learning_rate": 0.0005155393053016454,
      "loss": 0.3027,
      "step": 283
    },
    {
      "epoch": 0.3895747599451303,
      "grad_norm": 1.498235821723938,
      "learning_rate": 0.0005173674588665448,
      "loss": 0.1422,
      "step": 284
    },
    {
      "epoch": 0.39094650205761317,
      "grad_norm": 3.578543186187744,
      "learning_rate": 0.0005191956124314442,
      "loss": 0.6059,
      "step": 285
    },
    {
      "epoch": 0.39231824417009603,
      "grad_norm": 2.900531053543091,
      "learning_rate": 0.0005210237659963437,
      "loss": 0.3491,
      "step": 286
    },
    {
      "epoch": 0.3936899862825789,
      "grad_norm": 5.693866729736328,
      "learning_rate": 0.0005228519195612431,
      "loss": 1.1603,
      "step": 287
    },
    {
      "epoch": 0.3950617283950617,
      "grad_norm": 3.7944750785827637,
      "learning_rate": 0.0005246800731261426,
      "loss": 0.5784,
      "step": 288
    },
    {
      "epoch": 0.39643347050754457,
      "grad_norm": 4.433256149291992,
      "learning_rate": 0.000526508226691042,
      "loss": 0.6532,
      "step": 289
    },
    {
      "epoch": 0.39780521262002744,
      "grad_norm": 3.986520290374756,
      "learning_rate": 0.0005283363802559416,
      "loss": 0.5613,
      "step": 290
    },
    {
      "epoch": 0.3991769547325103,
      "grad_norm": 4.399818420410156,
      "learning_rate": 0.000530164533820841,
      "loss": 0.8469,
      "step": 291
    },
    {
      "epoch": 0.40054869684499317,
      "grad_norm": 3.0586366653442383,
      "learning_rate": 0.0005319926873857404,
      "loss": 0.4484,
      "step": 292
    },
    {
      "epoch": 0.401920438957476,
      "grad_norm": 0.10376634448766708,
      "learning_rate": 0.0005338208409506399,
      "loss": 0.0034,
      "step": 293
    },
    {
      "epoch": 0.40329218106995884,
      "grad_norm": 8.303990364074707,
      "learning_rate": 0.0005356489945155393,
      "loss": 3.0798,
      "step": 294
    },
    {
      "epoch": 0.4046639231824417,
      "grad_norm": 4.507086277008057,
      "learning_rate": 0.0005374771480804388,
      "loss": 0.8632,
      "step": 295
    },
    {
      "epoch": 0.4060356652949246,
      "grad_norm": 5.233419895172119,
      "learning_rate": 0.0005393053016453382,
      "loss": 1.2788,
      "step": 296
    },
    {
      "epoch": 0.4074074074074074,
      "grad_norm": 4.281100273132324,
      "learning_rate": 0.0005411334552102377,
      "loss": 1.2439,
      "step": 297
    },
    {
      "epoch": 0.40877914951989025,
      "grad_norm": 1.219199299812317,
      "learning_rate": 0.0005429616087751371,
      "loss": 0.1067,
      "step": 298
    },
    {
      "epoch": 0.4101508916323731,
      "grad_norm": 3.0503711700439453,
      "learning_rate": 0.0005447897623400365,
      "loss": 0.4197,
      "step": 299
    },
    {
      "epoch": 0.411522633744856,
      "grad_norm": 3.5557351112365723,
      "learning_rate": 0.000546617915904936,
      "loss": 0.56,
      "step": 300
    },
    {
      "epoch": 0.41289437585733885,
      "grad_norm": 4.1112470626831055,
      "learning_rate": 0.0005484460694698354,
      "loss": 0.7235,
      "step": 301
    },
    {
      "epoch": 0.41426611796982166,
      "grad_norm": 2.915947675704956,
      "learning_rate": 0.000550274223034735,
      "loss": 0.3506,
      "step": 302
    },
    {
      "epoch": 0.4156378600823045,
      "grad_norm": 4.125770568847656,
      "learning_rate": 0.0005521023765996344,
      "loss": 0.6808,
      "step": 303
    },
    {
      "epoch": 0.4170096021947874,
      "grad_norm": 5.084654808044434,
      "learning_rate": 0.0005539305301645339,
      "loss": 0.9406,
      "step": 304
    },
    {
      "epoch": 0.41838134430727025,
      "grad_norm": 4.542891025543213,
      "learning_rate": 0.0005557586837294333,
      "loss": 0.6707,
      "step": 305
    },
    {
      "epoch": 0.41975308641975306,
      "grad_norm": 4.285159587860107,
      "learning_rate": 0.0005575868372943327,
      "loss": 1.0718,
      "step": 306
    },
    {
      "epoch": 0.42112482853223593,
      "grad_norm": 5.053350925445557,
      "learning_rate": 0.0005594149908592322,
      "loss": 0.9847,
      "step": 307
    },
    {
      "epoch": 0.4224965706447188,
      "grad_norm": 1.4923471212387085,
      "learning_rate": 0.0005612431444241316,
      "loss": 0.122,
      "step": 308
    },
    {
      "epoch": 0.42386831275720166,
      "grad_norm": 4.332481384277344,
      "learning_rate": 0.0005630712979890311,
      "loss": 0.8221,
      "step": 309
    },
    {
      "epoch": 0.4252400548696845,
      "grad_norm": 4.23899507522583,
      "learning_rate": 0.0005648994515539305,
      "loss": 0.9891,
      "step": 310
    },
    {
      "epoch": 0.42661179698216734,
      "grad_norm": 4.370994567871094,
      "learning_rate": 0.00056672760511883,
      "loss": 0.926,
      "step": 311
    },
    {
      "epoch": 0.4279835390946502,
      "grad_norm": 0.23886460065841675,
      "learning_rate": 0.0005685557586837294,
      "loss": 0.0151,
      "step": 312
    },
    {
      "epoch": 0.42935528120713307,
      "grad_norm": 7.701839447021484,
      "learning_rate": 0.0005703839122486289,
      "loss": 2.8429,
      "step": 313
    },
    {
      "epoch": 0.43072702331961593,
      "grad_norm": 4.738073825836182,
      "learning_rate": 0.0005722120658135283,
      "loss": 0.9917,
      "step": 314
    },
    {
      "epoch": 0.43209876543209874,
      "grad_norm": 3.7907347679138184,
      "learning_rate": 0.0005740402193784278,
      "loss": 0.9199,
      "step": 315
    },
    {
      "epoch": 0.4334705075445816,
      "grad_norm": 2.7892837524414062,
      "learning_rate": 0.0005758683729433273,
      "loss": 0.3931,
      "step": 316
    },
    {
      "epoch": 0.4348422496570645,
      "grad_norm": 3.414641857147217,
      "learning_rate": 0.0005776965265082267,
      "loss": 0.7119,
      "step": 317
    },
    {
      "epoch": 0.43621399176954734,
      "grad_norm": 0.023860761895775795,
      "learning_rate": 0.0005795246800731262,
      "loss": 0.0008,
      "step": 318
    },
    {
      "epoch": 0.4375857338820302,
      "grad_norm": 3.0215470790863037,
      "learning_rate": 0.0005813528336380256,
      "loss": 0.4985,
      "step": 319
    },
    {
      "epoch": 0.438957475994513,
      "grad_norm": 3.1884336471557617,
      "learning_rate": 0.000583180987202925,
      "loss": 0.7732,
      "step": 320
    },
    {
      "epoch": 0.4403292181069959,
      "grad_norm": 3.4848649501800537,
      "learning_rate": 0.0005850091407678245,
      "loss": 0.7515,
      "step": 321
    },
    {
      "epoch": 0.44170096021947874,
      "grad_norm": 0.9857578873634338,
      "learning_rate": 0.0005868372943327239,
      "loss": 0.0885,
      "step": 322
    },
    {
      "epoch": 0.4430727023319616,
      "grad_norm": 4.590999126434326,
      "learning_rate": 0.0005886654478976234,
      "loss": 1.1677,
      "step": 323
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 4.161310195922852,
      "learning_rate": 0.0005904936014625229,
      "loss": 0.7439,
      "step": 324
    },
    {
      "epoch": 0.4458161865569273,
      "grad_norm": 4.842323303222656,
      "learning_rate": 0.0005923217550274223,
      "loss": 1.3473,
      "step": 325
    },
    {
      "epoch": 0.44718792866941015,
      "grad_norm": 3.332562208175659,
      "learning_rate": 0.0005941499085923218,
      "loss": 0.7273,
      "step": 326
    },
    {
      "epoch": 0.448559670781893,
      "grad_norm": 4.295160293579102,
      "learning_rate": 0.0005959780621572211,
      "loss": 0.843,
      "step": 327
    },
    {
      "epoch": 0.4499314128943759,
      "grad_norm": 3.641636848449707,
      "learning_rate": 0.0005978062157221207,
      "loss": 0.7881,
      "step": 328
    },
    {
      "epoch": 0.4513031550068587,
      "grad_norm": 2.767233371734619,
      "learning_rate": 0.0005996343692870201,
      "loss": 0.5319,
      "step": 329
    },
    {
      "epoch": 0.45267489711934156,
      "grad_norm": 2.925886869430542,
      "learning_rate": 0.0006014625228519196,
      "loss": 0.4826,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_Qnli-dev_cosine_accuracy": 0.708984375,
      "eval_Qnli-dev_cosine_accuracy_threshold": 0.8169091939926147,
      "eval_Qnli-dev_cosine_ap": 0.7472481805376167,
      "eval_Qnli-dev_cosine_f1": 0.7189781021897811,
      "eval_Qnli-dev_cosine_f1_threshold": 0.7571755051612854,
      "eval_Qnli-dev_cosine_precision": 0.6314102564102564,
      "eval_Qnli-dev_cosine_recall": 0.8347457627118644,
      "eval_allNLI-dev_cosine_accuracy": 0.748046875,
      "eval_allNLI-dev_cosine_accuracy_threshold": 0.8621972799301147,
      "eval_allNLI-dev_cosine_ap": 0.6527104471447597,
      "eval_allNLI-dev_cosine_f1": 0.6606334841628959,
      "eval_allNLI-dev_cosine_f1_threshold": 0.7864561080932617,
      "eval_allNLI-dev_cosine_precision": 0.5427509293680297,
      "eval_allNLI-dev_cosine_recall": 0.8439306358381503,
      "eval_sequential_score": 0.7472481805376167,
      "eval_sts-test_pearson_cosine": 0.8465015878560311,
      "eval_sts-test_spearman_cosine": 0.8833058569973334,
      "eval_vitaminc-pairs_loss": 2.5387091636657715,
      "eval_vitaminc-pairs_runtime": 14.3065,
      "eval_vitaminc-pairs_samples_per_second": 8.947,
      "eval_vitaminc-pairs_steps_per_second": 0.07,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_negation-triplets_loss": 1.8608626127243042,
      "eval_negation-triplets_runtime": 1.1519,
      "eval_negation-triplets_samples_per_second": 111.119,
      "eval_negation-triplets_steps_per_second": 0.868,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_scitail-pairs-pos_loss": 0.07322188466787338,
      "eval_scitail-pairs-pos_runtime": 1.5539,
      "eval_scitail-pairs-pos_samples_per_second": 82.372,
      "eval_scitail-pairs-pos_steps_per_second": 0.644,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_scitail-pairs-qa_loss": 0.026681702584028244,
      "eval_scitail-pairs-qa_runtime": 1.2098,
      "eval_scitail-pairs-qa_samples_per_second": 105.799,
      "eval_scitail-pairs-qa_steps_per_second": 0.827,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_xsum-pairs_loss": 0.6542444825172424,
      "eval_xsum-pairs_runtime": 6.2906,
      "eval_xsum-pairs_samples_per_second": 20.348,
      "eval_xsum-pairs_steps_per_second": 0.159,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_sciq_pairs_loss": 0.06421030312776566,
      "eval_sciq_pairs_runtime": 8.8514,
      "eval_sciq_pairs_samples_per_second": 14.461,
      "eval_sciq_pairs_steps_per_second": 0.113,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_qasc_pairs_loss": 0.8813464641571045,
      "eval_qasc_pairs_runtime": 1.3875,
      "eval_qasc_pairs_samples_per_second": 92.249,
      "eval_qasc_pairs_steps_per_second": 0.721,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_openbookqa_pairs_loss": 1.4074363708496094,
      "eval_openbookqa_pairs_runtime": 1.1826,
      "eval_openbookqa_pairs_samples_per_second": 108.237,
      "eval_openbookqa_pairs_steps_per_second": 0.846,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_nq_pairs_loss": 0.62897789478302,
      "eval_nq_pairs_runtime": 7.8684,
      "eval_nq_pairs_samples_per_second": 16.268,
      "eval_nq_pairs_steps_per_second": 0.127,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_trivia_pairs_loss": 1.084182620048523,
      "eval_trivia_pairs_runtime": 8.9262,
      "eval_trivia_pairs_samples_per_second": 14.34,
      "eval_trivia_pairs_steps_per_second": 0.112,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_gooaq_pairs_loss": 0.6594768762588501,
      "eval_gooaq_pairs_runtime": 2.0651,
      "eval_gooaq_pairs_samples_per_second": 61.982,
      "eval_gooaq_pairs_steps_per_second": 0.484,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_paws-pos_loss": 0.03268549218773842,
      "eval_paws-pos_runtime": 1.5103,
      "eval_paws-pos_samples_per_second": 84.754,
      "eval_paws-pos_steps_per_second": 0.662,
      "step": 330
    },
    {
      "epoch": 0.45267489711934156,
      "eval_global_dataset_loss": 0.21171291172504425,
      "eval_global_dataset_runtime": 33.2988,
      "eval_global_dataset_samples_per_second": 11.532,
      "eval_global_dataset_steps_per_second": 0.06,
      "step": 330
    },
    {
      "epoch": 0.4540466392318244,
      "grad_norm": 2.6243412494659424,
      "learning_rate": 0.000603290676416819,
      "loss": 0.6096,
      "step": 331
    },
    {
      "epoch": 0.4554183813443073,
      "grad_norm": 2.776013135910034,
      "learning_rate": 0.0006051188299817185,
      "loss": 0.3687,
      "step": 332
    },
    {
      "epoch": 0.4567901234567901,
      "grad_norm": 4.418542385101318,
      "learning_rate": 0.0006069469835466179,
      "loss": 0.9713,
      "step": 333
    },
    {
      "epoch": 0.45816186556927296,
      "grad_norm": 5.112300872802734,
      "learning_rate": 0.0006087751371115173,
      "loss": 1.3203,
      "step": 334
    },
    {
      "epoch": 0.45953360768175583,
      "grad_norm": 2.1659023761749268,
      "learning_rate": 0.0006106032906764169,
      "loss": 0.3443,
      "step": 335
    },
    {
      "epoch": 0.4609053497942387,
      "grad_norm": 4.783431529998779,
      "learning_rate": 0.0006124314442413162,
      "loss": 1.4592,
      "step": 336
    },
    {
      "epoch": 0.46227709190672156,
      "grad_norm": 3.507357597351074,
      "learning_rate": 0.0006142595978062158,
      "loss": 0.8,
      "step": 337
    },
    {
      "epoch": 0.46364883401920437,
      "grad_norm": 3.118370771408081,
      "learning_rate": 0.0006160877513711151,
      "loss": 0.5481,
      "step": 338
    },
    {
      "epoch": 0.46502057613168724,
      "grad_norm": 4.21981143951416,
      "learning_rate": 0.0006179159049360147,
      "loss": 1.4286,
      "step": 339
    },
    {
      "epoch": 0.4663923182441701,
      "grad_norm": 2.675670862197876,
      "learning_rate": 0.0006197440585009141,
      "loss": 0.4012,
      "step": 340
    },
    {
      "epoch": 0.46776406035665297,
      "grad_norm": 1.6964771747589111,
      "learning_rate": 0.0006215722120658135,
      "loss": 0.1474,
      "step": 341
    },
    {
      "epoch": 0.4691358024691358,
      "grad_norm": 4.83234167098999,
      "learning_rate": 0.000623400365630713,
      "loss": 1.2755,
      "step": 342
    },
    {
      "epoch": 0.47050754458161864,
      "grad_norm": 2.7494046688079834,
      "learning_rate": 0.0006252285191956124,
      "loss": 0.4935,
      "step": 343
    },
    {
      "epoch": 0.4718792866941015,
      "grad_norm": 4.708520412445068,
      "learning_rate": 0.0006270566727605119,
      "loss": 1.0101,
      "step": 344
    },
    {
      "epoch": 0.4732510288065844,
      "grad_norm": 2.6878857612609863,
      "learning_rate": 0.0006288848263254113,
      "loss": 0.4529,
      "step": 345
    },
    {
      "epoch": 0.47462277091906724,
      "grad_norm": 2.571988105773926,
      "learning_rate": 0.0006307129798903109,
      "loss": 0.3516,
      "step": 346
    },
    {
      "epoch": 0.47599451303155005,
      "grad_norm": 3.210439443588257,
      "learning_rate": 0.0006325411334552102,
      "loss": 0.4045,
      "step": 347
    },
    {
      "epoch": 0.4773662551440329,
      "grad_norm": 4.047224521636963,
      "learning_rate": 0.0006343692870201097,
      "loss": 1.2326,
      "step": 348
    },
    {
      "epoch": 0.4787379972565158,
      "grad_norm": 4.121635437011719,
      "learning_rate": 0.0006361974405850091,
      "loss": 0.8951,
      "step": 349
    },
    {
      "epoch": 0.48010973936899864,
      "grad_norm": 3.480602741241455,
      "learning_rate": 0.0006380255941499086,
      "loss": 0.6783,
      "step": 350
    },
    {
      "epoch": 0.48148148148148145,
      "grad_norm": 4.286612033843994,
      "learning_rate": 0.000639853747714808,
      "loss": 1.1821,
      "step": 351
    },
    {
      "epoch": 0.4828532235939643,
      "grad_norm": 3.077362537384033,
      "learning_rate": 0.0006416819012797075,
      "loss": 0.5101,
      "step": 352
    },
    {
      "epoch": 0.4842249657064472,
      "grad_norm": 3.127272844314575,
      "learning_rate": 0.000643510054844607,
      "loss": 0.844,
      "step": 353
    },
    {
      "epoch": 0.48559670781893005,
      "grad_norm": 2.6955454349517822,
      "learning_rate": 0.0006453382084095064,
      "loss": 0.5413,
      "step": 354
    },
    {
      "epoch": 0.4869684499314129,
      "grad_norm": 3.285903215408325,
      "learning_rate": 0.0006471663619744058,
      "loss": 0.9982,
      "step": 355
    },
    {
      "epoch": 0.4883401920438957,
      "grad_norm": 3.6762568950653076,
      "learning_rate": 0.0006489945155393053,
      "loss": 0.937,
      "step": 356
    },
    {
      "epoch": 0.4897119341563786,
      "grad_norm": 3.365633964538574,
      "learning_rate": 0.0006508226691042048,
      "loss": 0.7977,
      "step": 357
    },
    {
      "epoch": 0.49108367626886146,
      "grad_norm": 3.7572262287139893,
      "learning_rate": 0.0006526508226691042,
      "loss": 0.8697,
      "step": 358
    },
    {
      "epoch": 0.4924554183813443,
      "grad_norm": 3.309539794921875,
      "learning_rate": 0.0006544789762340037,
      "loss": 1.1136,
      "step": 359
    },
    {
      "epoch": 0.49382716049382713,
      "grad_norm": 4.250339508056641,
      "learning_rate": 0.0006563071297989031,
      "loss": 1.3018,
      "step": 360
    },
    {
      "epoch": 0.49519890260631,
      "grad_norm": 0.07554444670677185,
      "learning_rate": 0.0006581352833638026,
      "loss": 0.0023,
      "step": 361
    },
    {
      "epoch": 0.49657064471879286,
      "grad_norm": 3.6443750858306885,
      "learning_rate": 0.0006599634369287019,
      "loss": 0.7019,
      "step": 362
    },
    {
      "epoch": 0.49794238683127573,
      "grad_norm": 0.19442614912986755,
      "learning_rate": 0.0006617915904936015,
      "loss": 0.0481,
      "step": 363
    },
    {
      "epoch": 0.4993141289437586,
      "grad_norm": 3.7148313522338867,
      "learning_rate": 0.0006636197440585009,
      "loss": 0.891,
      "step": 364
    },
    {
      "epoch": 0.5006858710562414,
      "grad_norm": 2.7239511013031006,
      "learning_rate": 0.0006654478976234004,
      "loss": 0.6353,
      "step": 365
    },
    {
      "epoch": 0.5020576131687243,
      "grad_norm": 2.5572762489318848,
      "learning_rate": 0.0006672760511882999,
      "loss": 0.5181,
      "step": 366
    },
    {
      "epoch": 0.5034293552812071,
      "grad_norm": 3.162834405899048,
      "learning_rate": 0.0006691042047531993,
      "loss": 0.8311,
      "step": 367
    },
    {
      "epoch": 0.50480109739369,
      "grad_norm": 0.06661587208509445,
      "learning_rate": 0.0006709323583180988,
      "loss": 0.0026,
      "step": 368
    },
    {
      "epoch": 0.5061728395061729,
      "grad_norm": 2.4806196689605713,
      "learning_rate": 0.0006727605118829981,
      "loss": 0.5004,
      "step": 369
    },
    {
      "epoch": 0.5075445816186557,
      "grad_norm": 2.5774953365325928,
      "learning_rate": 0.0006745886654478977,
      "loss": 0.4511,
      "step": 370
    },
    {
      "epoch": 0.5089163237311386,
      "grad_norm": 2.2276206016540527,
      "learning_rate": 0.000676416819012797,
      "loss": 0.4558,
      "step": 371
    },
    {
      "epoch": 0.5102880658436214,
      "grad_norm": 2.652674674987793,
      "learning_rate": 0.0006782449725776966,
      "loss": 0.5073,
      "step": 372
    },
    {
      "epoch": 0.5116598079561042,
      "grad_norm": 2.2147669792175293,
      "learning_rate": 0.0006800731261425959,
      "loss": 0.3979,
      "step": 373
    },
    {
      "epoch": 0.5130315500685871,
      "grad_norm": 1.5278068780899048,
      "learning_rate": 0.0006819012797074955,
      "loss": 0.1665,
      "step": 374
    },
    {
      "epoch": 0.51440329218107,
      "grad_norm": 2.9153432846069336,
      "learning_rate": 0.0006837294332723948,
      "loss": 0.8231,
      "step": 375
    },
    {
      "epoch": 0.5157750342935528,
      "grad_norm": 4.252976894378662,
      "learning_rate": 0.0006855575868372943,
      "loss": 1.0406,
      "step": 376
    },
    {
      "epoch": 0.5171467764060357,
      "grad_norm": 3.768838405609131,
      "learning_rate": 0.0006873857404021939,
      "loss": 0.725,
      "step": 377
    },
    {
      "epoch": 0.5185185185185185,
      "grad_norm": 1.8634569644927979,
      "learning_rate": 0.0006892138939670932,
      "loss": 0.2603,
      "step": 378
    },
    {
      "epoch": 0.5198902606310014,
      "grad_norm": 5.095807075500488,
      "learning_rate": 0.0006910420475319928,
      "loss": 1.7357,
      "step": 379
    },
    {
      "epoch": 0.5212620027434842,
      "grad_norm": 2.8866710662841797,
      "learning_rate": 0.0006928702010968921,
      "loss": 0.9147,
      "step": 380
    },
    {
      "epoch": 0.522633744855967,
      "grad_norm": 2.369819402694702,
      "learning_rate": 0.0006946983546617917,
      "loss": 0.4277,
      "step": 381
    },
    {
      "epoch": 0.52400548696845,
      "grad_norm": 1.7865092754364014,
      "learning_rate": 0.000696526508226691,
      "loss": 0.4788,
      "step": 382
    },
    {
      "epoch": 0.5253772290809328,
      "grad_norm": 3.0729360580444336,
      "learning_rate": 0.0006983546617915905,
      "loss": 0.7666,
      "step": 383
    },
    {
      "epoch": 0.5267489711934157,
      "grad_norm": 3.023810386657715,
      "learning_rate": 0.0007001828153564899,
      "loss": 0.7728,
      "step": 384
    },
    {
      "epoch": 0.5281207133058985,
      "grad_norm": 2.5811986923217773,
      "learning_rate": 0.0007020109689213894,
      "loss": 0.4744,
      "step": 385
    },
    {
      "epoch": 0.5294924554183813,
      "grad_norm": 2.748720407485962,
      "learning_rate": 0.0007038391224862888,
      "loss": 1.1014,
      "step": 386
    },
    {
      "epoch": 0.5308641975308642,
      "grad_norm": 3.950869560241699,
      "learning_rate": 0.0007056672760511883,
      "loss": 1.5588,
      "step": 387
    },
    {
      "epoch": 0.532235939643347,
      "grad_norm": 1.595503330230713,
      "learning_rate": 0.0007074954296160879,
      "loss": 0.3185,
      "step": 388
    },
    {
      "epoch": 0.53360768175583,
      "grad_norm": 3.505636692047119,
      "learning_rate": 0.0007093235831809872,
      "loss": 1.3348,
      "step": 389
    },
    {
      "epoch": 0.5349794238683128,
      "grad_norm": 1.0456945896148682,
      "learning_rate": 0.0007111517367458867,
      "loss": 0.1656,
      "step": 390
    },
    {
      "epoch": 0.5363511659807956,
      "grad_norm": 2.747938871383667,
      "learning_rate": 0.0007129798903107861,
      "loss": 0.9375,
      "step": 391
    },
    {
      "epoch": 0.5377229080932785,
      "grad_norm": 4.741049289703369,
      "learning_rate": 0.0007148080438756856,
      "loss": 1.4665,
      "step": 392
    },
    {
      "epoch": 0.5390946502057613,
      "grad_norm": 4.830301284790039,
      "learning_rate": 0.000716636197440585,
      "loss": 1.4635,
      "step": 393
    },
    {
      "epoch": 0.5404663923182441,
      "grad_norm": 2.938199758529663,
      "learning_rate": 0.0007184643510054845,
      "loss": 0.8677,
      "step": 394
    },
    {
      "epoch": 0.541838134430727,
      "grad_norm": 6.291453838348389,
      "learning_rate": 0.0007202925045703839,
      "loss": 3.033,
      "step": 395
    },
    {
      "epoch": 0.5432098765432098,
      "grad_norm": 3.0533947944641113,
      "learning_rate": 0.0007221206581352834,
      "loss": 1.4375,
      "step": 396
    },
    {
      "epoch": 0.5445816186556928,
      "grad_norm": 3.13800048828125,
      "learning_rate": 0.0007239488117001827,
      "loss": 0.9762,
      "step": 397
    },
    {
      "epoch": 0.5459533607681756,
      "grad_norm": 1.1184616088867188,
      "learning_rate": 0.0007257769652650823,
      "loss": 0.1333,
      "step": 398
    },
    {
      "epoch": 0.5473251028806584,
      "grad_norm": 3.610217571258545,
      "learning_rate": 0.0007276051188299818,
      "loss": 1.1823,
      "step": 399
    },
    {
      "epoch": 0.5486968449931413,
      "grad_norm": 3.6696395874023438,
      "learning_rate": 0.0007294332723948812,
      "loss": 1.2443,
      "step": 400
    },
    {
      "epoch": 0.5500685871056241,
      "grad_norm": 2.967648506164551,
      "learning_rate": 0.0007312614259597807,
      "loss": 0.662,
      "step": 401
    },
    {
      "epoch": 0.551440329218107,
      "grad_norm": 0.8750410079956055,
      "learning_rate": 0.0007330895795246801,
      "loss": 0.0709,
      "step": 402
    },
    {
      "epoch": 0.5528120713305898,
      "grad_norm": 3.0801315307617188,
      "learning_rate": 0.0007349177330895796,
      "loss": 0.5822,
      "step": 403
    },
    {
      "epoch": 0.5541838134430727,
      "grad_norm": 3.701993227005005,
      "learning_rate": 0.0007367458866544789,
      "loss": 1.0826,
      "step": 404
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 3.439502716064453,
      "learning_rate": 0.0007385740402193785,
      "loss": 0.7953,
      "step": 405
    },
    {
      "epoch": 0.5569272976680384,
      "grad_norm": 4.8917341232299805,
      "learning_rate": 0.0007404021937842778,
      "loss": 1.6109,
      "step": 406
    },
    {
      "epoch": 0.5582990397805213,
      "grad_norm": 3.793834924697876,
      "learning_rate": 0.0007422303473491774,
      "loss": 1.2505,
      "step": 407
    },
    {
      "epoch": 0.5596707818930041,
      "grad_norm": 3.0604796409606934,
      "learning_rate": 0.0007440585009140767,
      "loss": 1.0019,
      "step": 408
    },
    {
      "epoch": 0.5610425240054869,
      "grad_norm": 2.872400999069214,
      "learning_rate": 0.0007458866544789763,
      "loss": 0.8224,
      "step": 409
    },
    {
      "epoch": 0.5624142661179699,
      "grad_norm": 2.631157398223877,
      "learning_rate": 0.0007477148080438758,
      "loss": 0.6592,
      "step": 410
    },
    {
      "epoch": 0.5637860082304527,
      "grad_norm": 2.843379020690918,
      "learning_rate": 0.0007495429616087751,
      "loss": 0.7099,
      "step": 411
    },
    {
      "epoch": 0.5651577503429356,
      "grad_norm": 2.3366591930389404,
      "learning_rate": 0.0007513711151736747,
      "loss": 0.5484,
      "step": 412
    },
    {
      "epoch": 0.5665294924554184,
      "grad_norm": 3.4202780723571777,
      "learning_rate": 0.000753199268738574,
      "loss": 1.3019,
      "step": 413
    },
    {
      "epoch": 0.5679012345679012,
      "grad_norm": 0.9125491976737976,
      "learning_rate": 0.0007550274223034736,
      "loss": 0.1266,
      "step": 414
    },
    {
      "epoch": 0.5692729766803841,
      "grad_norm": 2.8945682048797607,
      "learning_rate": 0.0007568555758683729,
      "loss": 0.8932,
      "step": 415
    },
    {
      "epoch": 0.5706447187928669,
      "grad_norm": 2.898399591445923,
      "learning_rate": 0.0007586837294332725,
      "loss": 0.7547,
      "step": 416
    },
    {
      "epoch": 0.5720164609053497,
      "grad_norm": 2.9406332969665527,
      "learning_rate": 0.0007605118829981718,
      "loss": 0.7614,
      "step": 417
    },
    {
      "epoch": 0.5733882030178327,
      "grad_norm": 2.1753182411193848,
      "learning_rate": 0.0007623400365630713,
      "loss": 0.7364,
      "step": 418
    },
    {
      "epoch": 0.5747599451303155,
      "grad_norm": 2.5140652656555176,
      "learning_rate": 0.0007641681901279707,
      "loss": 0.6539,
      "step": 419
    },
    {
      "epoch": 0.5761316872427984,
      "grad_norm": 2.0174620151519775,
      "learning_rate": 0.0007659963436928702,
      "loss": 0.3848,
      "step": 420
    },
    {
      "epoch": 0.5775034293552812,
      "grad_norm": 0.11105114966630936,
      "learning_rate": 0.0007678244972577697,
      "loss": 0.0394,
      "step": 421
    },
    {
      "epoch": 0.578875171467764,
      "grad_norm": 1.8194284439086914,
      "learning_rate": 0.0007696526508226691,
      "loss": 0.4623,
      "step": 422
    },
    {
      "epoch": 0.5802469135802469,
      "grad_norm": 0.7781994342803955,
      "learning_rate": 0.0007714808043875687,
      "loss": 0.0783,
      "step": 423
    },
    {
      "epoch": 0.5816186556927297,
      "grad_norm": 3.7422642707824707,
      "learning_rate": 0.000773308957952468,
      "loss": 1.4366,
      "step": 424
    },
    {
      "epoch": 0.5829903978052127,
      "grad_norm": 3.9761717319488525,
      "learning_rate": 0.0007751371115173675,
      "loss": 1.3568,
      "step": 425
    },
    {
      "epoch": 0.5843621399176955,
      "grad_norm": 2.948404550552368,
      "learning_rate": 0.0007769652650822669,
      "loss": 0.9065,
      "step": 426
    },
    {
      "epoch": 0.5857338820301783,
      "grad_norm": 2.7700140476226807,
      "learning_rate": 0.0007787934186471664,
      "loss": 0.6723,
      "step": 427
    },
    {
      "epoch": 0.5871056241426612,
      "grad_norm": 2.8618569374084473,
      "learning_rate": 0.0007806215722120658,
      "loss": 0.7596,
      "step": 428
    },
    {
      "epoch": 0.588477366255144,
      "grad_norm": 0.956656813621521,
      "learning_rate": 0.0007824497257769653,
      "loss": 0.1426,
      "step": 429
    },
    {
      "epoch": 0.5898491083676269,
      "grad_norm": 2.8430604934692383,
      "learning_rate": 0.0007842778793418648,
      "loss": 1.3264,
      "step": 430
    },
    {
      "epoch": 0.5912208504801097,
      "grad_norm": 0.11261007934808731,
      "learning_rate": 0.0007861060329067642,
      "loss": 0.0442,
      "step": 431
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": 2.7201974391937256,
      "learning_rate": 0.0007879341864716636,
      "loss": 0.6046,
      "step": 432
    },
    {
      "epoch": 0.5939643347050755,
      "grad_norm": 2.5923287868499756,
      "learning_rate": 0.0007897623400365631,
      "loss": 0.6173,
      "step": 433
    },
    {
      "epoch": 0.5953360768175583,
      "grad_norm": 4.77182674407959,
      "learning_rate": 0.0007915904936014625,
      "loss": 2.8892,
      "step": 434
    },
    {
      "epoch": 0.5967078189300411,
      "grad_norm": 3.1731245517730713,
      "learning_rate": 0.000793418647166362,
      "loss": 1.3149,
      "step": 435
    },
    {
      "epoch": 0.598079561042524,
      "grad_norm": 2.849473237991333,
      "learning_rate": 0.0007952468007312615,
      "loss": 1.1886,
      "step": 436
    },
    {
      "epoch": 0.5994513031550068,
      "grad_norm": 3.5986573696136475,
      "learning_rate": 0.0007970749542961609,
      "loss": 1.2799,
      "step": 437
    },
    {
      "epoch": 0.6008230452674898,
      "grad_norm": 2.668875217437744,
      "learning_rate": 0.0007989031078610604,
      "loss": 0.7527,
      "step": 438
    },
    {
      "epoch": 0.6021947873799726,
      "grad_norm": 2.7584633827209473,
      "learning_rate": 0.0008007312614259597,
      "loss": 0.7853,
      "step": 439
    },
    {
      "epoch": 0.6035665294924554,
      "grad_norm": 2.521331787109375,
      "learning_rate": 0.0008025594149908593,
      "loss": 0.7291,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_Qnli-dev_cosine_accuracy": 0.71875,
      "eval_Qnli-dev_cosine_accuracy_threshold": 0.8342527151107788,
      "eval_Qnli-dev_cosine_ap": 0.7590651825705997,
      "eval_Qnli-dev_cosine_f1": 0.6824034334763949,
      "eval_Qnli-dev_cosine_f1_threshold": 0.830267071723938,
      "eval_Qnli-dev_cosine_precision": 0.691304347826087,
      "eval_Qnli-dev_cosine_recall": 0.673728813559322,
      "eval_allNLI-dev_cosine_accuracy": 0.75,
      "eval_allNLI-dev_cosine_accuracy_threshold": 0.8865965604782104,
      "eval_allNLI-dev_cosine_ap": 0.6185049182990069,
      "eval_allNLI-dev_cosine_f1": 0.6239669421487604,
      "eval_allNLI-dev_cosine_f1_threshold": 0.7850924730300903,
      "eval_allNLI-dev_cosine_precision": 0.4855305466237942,
      "eval_allNLI-dev_cosine_recall": 0.8728323699421965,
      "eval_sequential_score": 0.7590651825705997,
      "eval_sts-test_pearson_cosine": 0.8447585777536966,
      "eval_sts-test_spearman_cosine": 0.8940651790173606,
      "eval_vitaminc-pairs_loss": 3.1624042987823486,
      "eval_vitaminc-pairs_runtime": 14.3967,
      "eval_vitaminc-pairs_samples_per_second": 8.891,
      "eval_vitaminc-pairs_steps_per_second": 0.069,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_negation-triplets_loss": 2.1101934909820557,
      "eval_negation-triplets_runtime": 1.1572,
      "eval_negation-triplets_samples_per_second": 110.608,
      "eval_negation-triplets_steps_per_second": 0.864,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_scitail-pairs-pos_loss": 0.14219464361667633,
      "eval_scitail-pairs-pos_runtime": 1.5903,
      "eval_scitail-pairs-pos_samples_per_second": 80.488,
      "eval_scitail-pairs-pos_steps_per_second": 0.629,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_scitail-pairs-qa_loss": 0.03170065954327583,
      "eval_scitail-pairs-qa_runtime": 1.2167,
      "eval_scitail-pairs-qa_samples_per_second": 105.199,
      "eval_scitail-pairs-qa_steps_per_second": 0.822,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_xsum-pairs_loss": 0.8414545655250549,
      "eval_xsum-pairs_runtime": 6.3264,
      "eval_xsum-pairs_samples_per_second": 20.233,
      "eval_xsum-pairs_steps_per_second": 0.158,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_sciq_pairs_loss": 0.06330692023038864,
      "eval_sciq_pairs_runtime": 8.8936,
      "eval_sciq_pairs_samples_per_second": 14.392,
      "eval_sciq_pairs_steps_per_second": 0.112,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_qasc_pairs_loss": 0.9395630359649658,
      "eval_qasc_pairs_runtime": 1.3863,
      "eval_qasc_pairs_samples_per_second": 92.332,
      "eval_qasc_pairs_steps_per_second": 0.721,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_openbookqa_pairs_loss": 1.5656604766845703,
      "eval_openbookqa_pairs_runtime": 1.1858,
      "eval_openbookqa_pairs_samples_per_second": 107.947,
      "eval_openbookqa_pairs_steps_per_second": 0.843,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_nq_pairs_loss": 0.8197879195213318,
      "eval_nq_pairs_runtime": 7.9153,
      "eval_nq_pairs_samples_per_second": 16.171,
      "eval_nq_pairs_steps_per_second": 0.126,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_trivia_pairs_loss": 1.1605956554412842,
      "eval_trivia_pairs_runtime": 8.9789,
      "eval_trivia_pairs_samples_per_second": 14.256,
      "eval_trivia_pairs_steps_per_second": 0.111,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_gooaq_pairs_loss": 0.7199620604515076,
      "eval_gooaq_pairs_runtime": 2.0759,
      "eval_gooaq_pairs_samples_per_second": 61.66,
      "eval_gooaq_pairs_steps_per_second": 0.482,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_paws-pos_loss": 0.034769680351018906,
      "eval_paws-pos_runtime": 1.5142,
      "eval_paws-pos_samples_per_second": 84.536,
      "eval_paws-pos_steps_per_second": 0.66,
      "step": 440
    },
    {
      "epoch": 0.6035665294924554,
      "eval_global_dataset_loss": 0.2699156403541565,
      "eval_global_dataset_runtime": 33.4517,
      "eval_global_dataset_samples_per_second": 11.479,
      "eval_global_dataset_steps_per_second": 0.06,
      "step": 440
    },
    {
      "epoch": 0.6049382716049383,
      "grad_norm": 2.2343204021453857,
      "learning_rate": 0.0008043875685557587,
      "loss": 0.6532,
      "step": 441
    },
    {
      "epoch": 0.6063100137174211,
      "grad_norm": 2.0353541374206543,
      "learning_rate": 0.0008062157221206582,
      "loss": 0.5846,
      "step": 442
    },
    {
      "epoch": 0.607681755829904,
      "grad_norm": 2.2103233337402344,
      "learning_rate": 0.0008080438756855576,
      "loss": 0.6766,
      "step": 443
    },
    {
      "epoch": 0.6090534979423868,
      "grad_norm": 0.8574596047401428,
      "learning_rate": 0.0008098720292504571,
      "loss": 0.1179,
      "step": 444
    },
    {
      "epoch": 0.6104252400548696,
      "grad_norm": 2.2213659286499023,
      "learning_rate": 0.0008117001828153565,
      "loss": 0.5953,
      "step": 445
    },
    {
      "epoch": 0.6117969821673526,
      "grad_norm": 2.4394009113311768,
      "learning_rate": 0.0008135283363802559,
      "loss": 0.7219,
      "step": 446
    },
    {
      "epoch": 0.6131687242798354,
      "grad_norm": 2.216285228729248,
      "learning_rate": 0.0008153564899451555,
      "loss": 0.9445,
      "step": 447
    },
    {
      "epoch": 0.6145404663923183,
      "grad_norm": 2.296591281890869,
      "learning_rate": 0.0008171846435100548,
      "loss": 0.5726,
      "step": 448
    },
    {
      "epoch": 0.6159122085048011,
      "grad_norm": 2.837359666824341,
      "learning_rate": 0.0008190127970749544,
      "loss": 0.8687,
      "step": 449
    },
    {
      "epoch": 0.6172839506172839,
      "grad_norm": 2.9849822521209717,
      "learning_rate": 0.0008208409506398537,
      "loss": 0.7864,
      "step": 450
    },
    {
      "epoch": 0.6186556927297668,
      "grad_norm": 2.474256753921509,
      "learning_rate": 0.0008226691042047533,
      "loss": 0.6835,
      "step": 451
    },
    {
      "epoch": 0.6200274348422496,
      "grad_norm": 2.815581798553467,
      "learning_rate": 0.0008244972577696527,
      "loss": 1.2349,
      "step": 452
    },
    {
      "epoch": 0.6213991769547325,
      "grad_norm": 2.5145702362060547,
      "learning_rate": 0.0008263254113345521,
      "loss": 0.5894,
      "step": 453
    },
    {
      "epoch": 0.6227709190672154,
      "grad_norm": 3.4624481201171875,
      "learning_rate": 0.0008281535648994516,
      "loss": 1.2882,
      "step": 454
    },
    {
      "epoch": 0.6241426611796982,
      "grad_norm": 3.0444629192352295,
      "learning_rate": 0.000829981718464351,
      "loss": 1.1063,
      "step": 455
    },
    {
      "epoch": 0.6255144032921811,
      "grad_norm": 3.4558730125427246,
      "learning_rate": 0.0008318098720292505,
      "loss": 1.6217,
      "step": 456
    },
    {
      "epoch": 0.6268861454046639,
      "grad_norm": 1.9938262701034546,
      "learning_rate": 0.0008336380255941499,
      "loss": 0.6361,
      "step": 457
    },
    {
      "epoch": 0.6282578875171467,
      "grad_norm": 2.8914825916290283,
      "learning_rate": 0.0008354661791590493,
      "loss": 1.2991,
      "step": 458
    },
    {
      "epoch": 0.6296296296296297,
      "grad_norm": 1.987092137336731,
      "learning_rate": 0.0008372943327239488,
      "loss": 0.6638,
      "step": 459
    },
    {
      "epoch": 0.6310013717421125,
      "grad_norm": 1.9341247081756592,
      "learning_rate": 0.0008391224862888483,
      "loss": 0.6688,
      "step": 460
    },
    {
      "epoch": 0.6323731138545954,
      "grad_norm": 2.6572768688201904,
      "learning_rate": 0.0008409506398537477,
      "loss": 0.9422,
      "step": 461
    },
    {
      "epoch": 0.6337448559670782,
      "grad_norm": 2.504377841949463,
      "learning_rate": 0.0008427787934186472,
      "loss": 0.8578,
      "step": 462
    },
    {
      "epoch": 0.635116598079561,
      "grad_norm": 1.6810640096664429,
      "learning_rate": 0.0008446069469835467,
      "loss": 0.4707,
      "step": 463
    },
    {
      "epoch": 0.6364883401920439,
      "grad_norm": 2.0036683082580566,
      "learning_rate": 0.0008464351005484461,
      "loss": 0.4281,
      "step": 464
    },
    {
      "epoch": 0.6378600823045267,
      "grad_norm": 2.055009365081787,
      "learning_rate": 0.0008482632541133455,
      "loss": 0.5323,
      "step": 465
    },
    {
      "epoch": 0.6392318244170097,
      "grad_norm": 1.8697012662887573,
      "learning_rate": 0.000850091407678245,
      "loss": 0.4676,
      "step": 466
    },
    {
      "epoch": 0.6406035665294925,
      "grad_norm": 0.3099338114261627,
      "learning_rate": 0.0008519195612431444,
      "loss": 0.0323,
      "step": 467
    },
    {
      "epoch": 0.6419753086419753,
      "grad_norm": 2.843210458755493,
      "learning_rate": 0.0008537477148080439,
      "loss": 1.0748,
      "step": 468
    },
    {
      "epoch": 0.6433470507544582,
      "grad_norm": 5.503166198730469,
      "learning_rate": 0.0008555758683729433,
      "loss": 3.3347,
      "step": 469
    },
    {
      "epoch": 0.644718792866941,
      "grad_norm": 3.0612828731536865,
      "learning_rate": 0.0008574040219378428,
      "loss": 1.4944,
      "step": 470
    },
    {
      "epoch": 0.6460905349794238,
      "grad_norm": 1.8079982995986938,
      "learning_rate": 0.0008592321755027423,
      "loss": 0.666,
      "step": 471
    },
    {
      "epoch": 0.6474622770919067,
      "grad_norm": 2.734410285949707,
      "learning_rate": 0.0008610603290676416,
      "loss": 1.1204,
      "step": 472
    },
    {
      "epoch": 0.6488340192043895,
      "grad_norm": 2.612880229949951,
      "learning_rate": 0.0008628884826325412,
      "loss": 1.452,
      "step": 473
    },
    {
      "epoch": 0.6502057613168725,
      "grad_norm": 1.7585951089859009,
      "learning_rate": 0.0008647166361974406,
      "loss": 0.5587,
      "step": 474
    },
    {
      "epoch": 0.6515775034293553,
      "grad_norm": 2.7308504581451416,
      "learning_rate": 0.0008665447897623401,
      "loss": 1.189,
      "step": 475
    },
    {
      "epoch": 0.6529492455418381,
      "grad_norm": 1.9969093799591064,
      "learning_rate": 0.0008683729433272395,
      "loss": 0.5736,
      "step": 476
    },
    {
      "epoch": 0.654320987654321,
      "grad_norm": 2.545562505722046,
      "learning_rate": 0.000870201096892139,
      "loss": 0.7508,
      "step": 477
    },
    {
      "epoch": 0.6556927297668038,
      "grad_norm": 2.3430142402648926,
      "learning_rate": 0.0008720292504570384,
      "loss": 0.7138,
      "step": 478
    },
    {
      "epoch": 0.6570644718792867,
      "grad_norm": 2.081550359725952,
      "learning_rate": 0.0008738574040219378,
      "loss": 0.5152,
      "step": 479
    },
    {
      "epoch": 0.6584362139917695,
      "grad_norm": 3.031339645385742,
      "learning_rate": 0.0008756855575868373,
      "loss": 1.0291,
      "step": 480
    },
    {
      "epoch": 0.6598079561042524,
      "grad_norm": 3.382559299468994,
      "learning_rate": 0.0008775137111517367,
      "loss": 1.2339,
      "step": 481
    },
    {
      "epoch": 0.6611796982167353,
      "grad_norm": 2.266045570373535,
      "learning_rate": 0.0008793418647166362,
      "loss": 0.6402,
      "step": 482
    },
    {
      "epoch": 0.6625514403292181,
      "grad_norm": 4.375522136688232,
      "learning_rate": 0.0008811700182815356,
      "loss": 3.3593,
      "step": 483
    },
    {
      "epoch": 0.663923182441701,
      "grad_norm": 1.7218207120895386,
      "learning_rate": 0.0008829981718464352,
      "loss": 0.4424,
      "step": 484
    },
    {
      "epoch": 0.6652949245541838,
      "grad_norm": 1.930618405342102,
      "learning_rate": 0.0008848263254113346,
      "loss": 0.79,
      "step": 485
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 2.2200028896331787,
      "learning_rate": 0.000886654478976234,
      "loss": 0.9445,
      "step": 486
    },
    {
      "epoch": 0.6680384087791496,
      "grad_norm": 1.67755925655365,
      "learning_rate": 0.0008884826325411335,
      "loss": 0.5046,
      "step": 487
    },
    {
      "epoch": 0.6694101508916324,
      "grad_norm": 1.119525671005249,
      "learning_rate": 0.0008903107861060329,
      "loss": 0.2215,
      "step": 488
    },
    {
      "epoch": 0.6707818930041153,
      "grad_norm": 1.9932385683059692,
      "learning_rate": 0.0008921389396709324,
      "loss": 0.9744,
      "step": 489
    },
    {
      "epoch": 0.6721536351165981,
      "grad_norm": 1.830539345741272,
      "learning_rate": 0.0008939670932358318,
      "loss": 0.5102,
      "step": 490
    },
    {
      "epoch": 0.6735253772290809,
      "grad_norm": 2.152487277984619,
      "learning_rate": 0.0008957952468007313,
      "loss": 0.5452,
      "step": 491
    },
    {
      "epoch": 0.6748971193415638,
      "grad_norm": 2.4515275955200195,
      "learning_rate": 0.0008976234003656307,
      "loss": 1.1853,
      "step": 492
    },
    {
      "epoch": 0.6762688614540466,
      "grad_norm": 2.1000306606292725,
      "learning_rate": 0.0008994515539305301,
      "loss": 0.5773,
      "step": 493
    },
    {
      "epoch": 0.6776406035665294,
      "grad_norm": 2.6680173873901367,
      "learning_rate": 0.0009012797074954296,
      "loss": 0.8209,
      "step": 494
    },
    {
      "epoch": 0.6790123456790124,
      "grad_norm": 2.963197708129883,
      "learning_rate": 0.000903107861060329,
      "loss": 0.961,
      "step": 495
    },
    {
      "epoch": 0.6803840877914952,
      "grad_norm": 1.574116826057434,
      "learning_rate": 0.0009049360146252286,
      "loss": 0.3303,
      "step": 496
    },
    {
      "epoch": 0.6817558299039781,
      "grad_norm": 2.57684063911438,
      "learning_rate": 0.000906764168190128,
      "loss": 0.8192,
      "step": 497
    },
    {
      "epoch": 0.6831275720164609,
      "grad_norm": 3.053014039993286,
      "learning_rate": 0.0009085923217550275,
      "loss": 1.3271,
      "step": 498
    },
    {
      "epoch": 0.6844993141289437,
      "grad_norm": 2.6430559158325195,
      "learning_rate": 0.0009104204753199269,
      "loss": 0.9859,
      "step": 499
    },
    {
      "epoch": 0.6858710562414266,
      "grad_norm": 0.7626333236694336,
      "learning_rate": 0.0009122486288848263,
      "loss": 0.1419,
      "step": 500
    },
    {
      "epoch": 0.6872427983539094,
      "grad_norm": 0.8805200457572937,
      "learning_rate": 0.0009140767824497258,
      "loss": 0.1522,
      "step": 501
    },
    {
      "epoch": 0.6886145404663924,
      "grad_norm": 1.4140968322753906,
      "learning_rate": 0.0009159049360146252,
      "loss": 0.3983,
      "step": 502
    },
    {
      "epoch": 0.6899862825788752,
      "grad_norm": 0.16096271574497223,
      "learning_rate": 0.0009177330895795247,
      "loss": 0.0167,
      "step": 503
    },
    {
      "epoch": 0.691358024691358,
      "grad_norm": 2.1687278747558594,
      "learning_rate": 0.0009195612431444241,
      "loss": 0.7314,
      "step": 504
    },
    {
      "epoch": 0.6927297668038409,
      "grad_norm": 2.1703848838806152,
      "learning_rate": 0.0009213893967093237,
      "loss": 0.5175,
      "step": 505
    },
    {
      "epoch": 0.6941015089163237,
      "grad_norm": 3.221038341522217,
      "learning_rate": 0.000923217550274223,
      "loss": 1.6745,
      "step": 506
    },
    {
      "epoch": 0.6954732510288066,
      "grad_norm": 2.9030301570892334,
      "learning_rate": 0.0009250457038391225,
      "loss": 1.2944,
      "step": 507
    },
    {
      "epoch": 0.6968449931412894,
      "grad_norm": 2.1682050228118896,
      "learning_rate": 0.000926873857404022,
      "loss": 0.5509,
      "step": 508
    },
    {
      "epoch": 0.6982167352537723,
      "grad_norm": 2.4524831771850586,
      "learning_rate": 0.0009287020109689214,
      "loss": 0.8585,
      "step": 509
    },
    {
      "epoch": 0.6995884773662552,
      "grad_norm": 2.954036235809326,
      "learning_rate": 0.0009305301645338209,
      "loss": 1.1137,
      "step": 510
    },
    {
      "epoch": 0.700960219478738,
      "grad_norm": 0.5421282052993774,
      "learning_rate": 0.0009323583180987203,
      "loss": 0.08,
      "step": 511
    },
    {
      "epoch": 0.7023319615912208,
      "grad_norm": 1.943343997001648,
      "learning_rate": 0.0009341864716636198,
      "loss": 0.6881,
      "step": 512
    },
    {
      "epoch": 0.7037037037037037,
      "grad_norm": 2.1507110595703125,
      "learning_rate": 0.0009360146252285192,
      "loss": 0.8449,
      "step": 513
    },
    {
      "epoch": 0.7050754458161865,
      "grad_norm": 2.5957894325256348,
      "learning_rate": 0.0009378427787934186,
      "loss": 1.3807,
      "step": 514
    },
    {
      "epoch": 0.7064471879286695,
      "grad_norm": 2.7923741340637207,
      "learning_rate": 0.0009396709323583181,
      "loss": 1.2345,
      "step": 515
    },
    {
      "epoch": 0.7078189300411523,
      "grad_norm": 2.737544298171997,
      "learning_rate": 0.0009414990859232176,
      "loss": 1.3458,
      "step": 516
    },
    {
      "epoch": 0.7091906721536351,
      "grad_norm": 3.436458110809326,
      "learning_rate": 0.000943327239488117,
      "loss": 1.7297,
      "step": 517
    },
    {
      "epoch": 0.710562414266118,
      "grad_norm": 0.19496861100196838,
      "learning_rate": 0.0009451553930530165,
      "loss": 0.0337,
      "step": 518
    },
    {
      "epoch": 0.7119341563786008,
      "grad_norm": 1.9844475984573364,
      "learning_rate": 0.000946983546617916,
      "loss": 0.7494,
      "step": 519
    },
    {
      "epoch": 0.7133058984910837,
      "grad_norm": 2.368131637573242,
      "learning_rate": 0.0009488117001828154,
      "loss": 1.2839,
      "step": 520
    },
    {
      "epoch": 0.7146776406035665,
      "grad_norm": 1.8333654403686523,
      "learning_rate": 0.0009506398537477148,
      "loss": 0.7679,
      "step": 521
    },
    {
      "epoch": 0.7160493827160493,
      "grad_norm": 2.5262162685394287,
      "learning_rate": 0.0009524680073126143,
      "loss": 1.1101,
      "step": 522
    },
    {
      "epoch": 0.7174211248285323,
      "grad_norm": 1.7962846755981445,
      "learning_rate": 0.0009542961608775137,
      "loss": 0.7259,
      "step": 523
    },
    {
      "epoch": 0.7187928669410151,
      "grad_norm": 1.6876013278961182,
      "learning_rate": 0.0009561243144424132,
      "loss": 0.8531,
      "step": 524
    },
    {
      "epoch": 0.720164609053498,
      "grad_norm": 0.15522260963916779,
      "learning_rate": 0.0009579524680073126,
      "loss": 0.0376,
      "step": 525
    },
    {
      "epoch": 0.7215363511659808,
      "grad_norm": 0.10816041380167007,
      "learning_rate": 0.0009597806215722121,
      "loss": 0.0368,
      "step": 526
    },
    {
      "epoch": 0.7229080932784636,
      "grad_norm": 0.5582392811775208,
      "learning_rate": 0.0009616087751371116,
      "loss": 0.0679,
      "step": 527
    },
    {
      "epoch": 0.7242798353909465,
      "grad_norm": 0.790245532989502,
      "learning_rate": 0.0009634369287020109,
      "loss": 0.1483,
      "step": 528
    },
    {
      "epoch": 0.7256515775034293,
      "grad_norm": 0.5912861227989197,
      "learning_rate": 0.0009652650822669105,
      "loss": 0.086,
      "step": 529
    },
    {
      "epoch": 0.7270233196159122,
      "grad_norm": 1.9863473176956177,
      "learning_rate": 0.0009670932358318098,
      "loss": 0.4599,
      "step": 530
    },
    {
      "epoch": 0.7283950617283951,
      "grad_norm": 2.432748794555664,
      "learning_rate": 0.0009689213893967094,
      "loss": 0.6046,
      "step": 531
    },
    {
      "epoch": 0.7297668038408779,
      "grad_norm": 2.2815091609954834,
      "learning_rate": 0.0009707495429616088,
      "loss": 0.5801,
      "step": 532
    },
    {
      "epoch": 0.7311385459533608,
      "grad_norm": 6.126629829406738,
      "learning_rate": 0.0009725776965265083,
      "loss": 4.0638,
      "step": 533
    },
    {
      "epoch": 0.7325102880658436,
      "grad_norm": 2.440173864364624,
      "learning_rate": 0.0009744058500914077,
      "loss": 0.9248,
      "step": 534
    },
    {
      "epoch": 0.7338820301783264,
      "grad_norm": 2.2785732746124268,
      "learning_rate": 0.0009762340036563071,
      "loss": 0.7769,
      "step": 535
    },
    {
      "epoch": 0.7352537722908093,
      "grad_norm": 1.7889460325241089,
      "learning_rate": 0.0009780621572212066,
      "loss": 0.5365,
      "step": 536
    },
    {
      "epoch": 0.7366255144032922,
      "grad_norm": 1.9316608905792236,
      "learning_rate": 0.000979890310786106,
      "loss": 0.7444,
      "step": 537
    },
    {
      "epoch": 0.7379972565157751,
      "grad_norm": 1.945172905921936,
      "learning_rate": 0.0009817184643510055,
      "loss": 0.7193,
      "step": 538
    },
    {
      "epoch": 0.7393689986282579,
      "grad_norm": 2.1771185398101807,
      "learning_rate": 0.000983546617915905,
      "loss": 1.5464,
      "step": 539
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 2.8104841709136963,
      "learning_rate": 0.0009853747714808044,
      "loss": 1.7129,
      "step": 540
    },
    {
      "epoch": 0.7421124828532236,
      "grad_norm": 2.2292134761810303,
      "learning_rate": 0.0009872029250457038,
      "loss": 1.0137,
      "step": 541
    },
    {
      "epoch": 0.7434842249657064,
      "grad_norm": 1.7931193113327026,
      "learning_rate": 0.0009890310786106033,
      "loss": 0.6174,
      "step": 542
    },
    {
      "epoch": 0.7448559670781894,
      "grad_norm": 1.5730267763137817,
      "learning_rate": 0.0009908592321755027,
      "loss": 1.1872,
      "step": 543
    },
    {
      "epoch": 0.7462277091906722,
      "grad_norm": 1.9772863388061523,
      "learning_rate": 0.0009926873857404022,
      "loss": 1.2298,
      "step": 544
    },
    {
      "epoch": 0.747599451303155,
      "grad_norm": 3.7602641582489014,
      "learning_rate": 0.0009945155393053017,
      "loss": 3.6099,
      "step": 545
    },
    {
      "epoch": 0.7489711934156379,
      "grad_norm": 1.6600819826126099,
      "learning_rate": 0.0009963436928702011,
      "loss": 0.9129,
      "step": 546
    },
    {
      "epoch": 0.7503429355281207,
      "grad_norm": 1.4769471883773804,
      "learning_rate": 0.0009981718464351006,
      "loss": 0.4851,
      "step": 547
    },
    {
      "epoch": 0.7517146776406035,
      "grad_norm": 1.5102423429489136,
      "learning_rate": 0.001,
      "loss": 0.5724,
      "step": 548
    },
    {
      "epoch": 0.7530864197530864,
      "grad_norm": 1.6490839719772339,
      "learning_rate": 0.000999999174352948,
      "loss": 0.6958,
      "step": 549
    },
    {
      "epoch": 0.7544581618655692,
      "grad_norm": 1.4960081577301025,
      "learning_rate": 0.0009999966974148216,
      "loss": 0.9115,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_Qnli-dev_cosine_accuracy": 0.685546875,
      "eval_Qnli-dev_cosine_accuracy_threshold": 0.8639271259307861,
      "eval_Qnli-dev_cosine_ap": 0.7427350367644162,
      "eval_Qnli-dev_cosine_f1": 0.6996336996336996,
      "eval_Qnli-dev_cosine_f1_threshold": 0.7733485698699951,
      "eval_Qnli-dev_cosine_precision": 0.6161290322580645,
      "eval_Qnli-dev_cosine_recall": 0.809322033898305,
      "eval_allNLI-dev_cosine_accuracy": 0.755859375,
      "eval_allNLI-dev_cosine_accuracy_threshold": 0.8583443760871887,
      "eval_allNLI-dev_cosine_ap": 0.6513030662407349,
      "eval_allNLI-dev_cosine_f1": 0.6559633027522936,
      "eval_allNLI-dev_cosine_f1_threshold": 0.8035047054290771,
      "eval_allNLI-dev_cosine_precision": 0.5437262357414449,
      "eval_allNLI-dev_cosine_recall": 0.8265895953757225,
      "eval_sequential_score": 0.7427350367644162,
      "eval_sts-test_pearson_cosine": 0.8397477513546239,
      "eval_sts-test_spearman_cosine": 0.8842177086994444,
      "eval_vitaminc-pairs_loss": 2.9589016437530518,
      "eval_vitaminc-pairs_runtime": 14.2988,
      "eval_vitaminc-pairs_samples_per_second": 8.952,
      "eval_vitaminc-pairs_steps_per_second": 0.07,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_negation-triplets_loss": 2.106090784072876,
      "eval_negation-triplets_runtime": 1.1528,
      "eval_negation-triplets_samples_per_second": 111.031,
      "eval_negation-triplets_steps_per_second": 0.867,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_scitail-pairs-pos_loss": 0.25539278984069824,
      "eval_scitail-pairs-pos_runtime": 1.5749,
      "eval_scitail-pairs-pos_samples_per_second": 81.276,
      "eval_scitail-pairs-pos_steps_per_second": 0.635,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_scitail-pairs-qa_loss": 0.02696998417377472,
      "eval_scitail-pairs-qa_runtime": 1.2219,
      "eval_scitail-pairs-qa_samples_per_second": 104.753,
      "eval_scitail-pairs-qa_steps_per_second": 0.818,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_xsum-pairs_loss": 1.035770297050476,
      "eval_xsum-pairs_runtime": 6.3034,
      "eval_xsum-pairs_samples_per_second": 20.307,
      "eval_xsum-pairs_steps_per_second": 0.159,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_sciq_pairs_loss": 0.07481610774993896,
      "eval_sciq_pairs_runtime": 8.8401,
      "eval_sciq_pairs_samples_per_second": 14.48,
      "eval_sciq_pairs_steps_per_second": 0.113,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_qasc_pairs_loss": 0.9381294846534729,
      "eval_qasc_pairs_runtime": 1.3844,
      "eval_qasc_pairs_samples_per_second": 92.461,
      "eval_qasc_pairs_steps_per_second": 0.722,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_openbookqa_pairs_loss": 1.4043291807174683,
      "eval_openbookqa_pairs_runtime": 1.1842,
      "eval_openbookqa_pairs_samples_per_second": 108.088,
      "eval_openbookqa_pairs_steps_per_second": 0.844,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_nq_pairs_loss": 0.8262844085693359,
      "eval_nq_pairs_runtime": 7.8627,
      "eval_nq_pairs_samples_per_second": 16.279,
      "eval_nq_pairs_steps_per_second": 0.127,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_trivia_pairs_loss": 1.2341746091842651,
      "eval_trivia_pairs_runtime": 8.9261,
      "eval_trivia_pairs_samples_per_second": 14.34,
      "eval_trivia_pairs_steps_per_second": 0.112,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_gooaq_pairs_loss": 0.8736229538917542,
      "eval_gooaq_pairs_runtime": 2.07,
      "eval_gooaq_pairs_samples_per_second": 61.836,
      "eval_gooaq_pairs_steps_per_second": 0.483,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_paws-pos_loss": 0.03411751613020897,
      "eval_paws-pos_runtime": 1.5071,
      "eval_paws-pos_samples_per_second": 84.931,
      "eval_paws-pos_steps_per_second": 0.664,
      "step": 550
    },
    {
      "epoch": 0.7544581618655692,
      "eval_global_dataset_loss": 0.26876401901245117,
      "eval_global_dataset_runtime": 33.2849,
      "eval_global_dataset_samples_per_second": 11.537,
      "eval_global_dataset_steps_per_second": 0.06,
      "step": 550
    },
    {
      "epoch": 0.7558299039780522,
      "grad_norm": 1.4109474420547485,
      "learning_rate": 0.0009999925691947101,
      "loss": 0.6115,
      "step": 551
    },
    {
      "epoch": 0.757201646090535,
      "grad_norm": 0.21552617847919464,
      "learning_rate": 0.000999986789707762,
      "loss": 0.0239,
      "step": 552
    },
    {
      "epoch": 0.7585733882030178,
      "grad_norm": 2.0374319553375244,
      "learning_rate": 0.000999979358975186,
      "loss": 0.8398,
      "step": 553
    },
    {
      "epoch": 0.7599451303155007,
      "grad_norm": 2.699413776397705,
      "learning_rate": 0.0009999702770242486,
      "loss": 1.0964,
      "step": 554
    },
    {
      "epoch": 0.7613168724279835,
      "grad_norm": 1.9054522514343262,
      "learning_rate": 0.000999959543888277,
      "loss": 0.5098,
      "step": 555
    },
    {
      "epoch": 0.7626886145404664,
      "grad_norm": 1.9262897968292236,
      "learning_rate": 0.0009999471596066566,
      "loss": 0.5578,
      "step": 556
    },
    {
      "epoch": 0.7640603566529492,
      "grad_norm": 2.377450942993164,
      "learning_rate": 0.0009999331242248324,
      "loss": 1.3982,
      "step": 557
    },
    {
      "epoch": 0.7654320987654321,
      "grad_norm": 2.6483099460601807,
      "learning_rate": 0.0009999174377943074,
      "loss": 1.1118,
      "step": 558
    },
    {
      "epoch": 0.766803840877915,
      "grad_norm": 1.7998634576797485,
      "learning_rate": 0.000999900100372644,
      "loss": 0.508,
      "step": 559
    },
    {
      "epoch": 0.7681755829903978,
      "grad_norm": 2.284386157989502,
      "learning_rate": 0.0009998811120234623,
      "loss": 1.4645,
      "step": 560
    },
    {
      "epoch": 0.7695473251028807,
      "grad_norm": 2.0943026542663574,
      "learning_rate": 0.0009998604728164411,
      "loss": 0.8549,
      "step": 561
    },
    {
      "epoch": 0.7709190672153635,
      "grad_norm": 1.3964064121246338,
      "learning_rate": 0.0009998381828273167,
      "loss": 0.4917,
      "step": 562
    },
    {
      "epoch": 0.7722908093278463,
      "grad_norm": 1.6305673122406006,
      "learning_rate": 0.0009998142421378832,
      "loss": 0.6253,
      "step": 563
    },
    {
      "epoch": 0.7736625514403292,
      "grad_norm": 1.6184250116348267,
      "learning_rate": 0.0009997886508359918,
      "loss": 0.8867,
      "step": 564
    },
    {
      "epoch": 0.7750342935528121,
      "grad_norm": 1.9399840831756592,
      "learning_rate": 0.0009997614090155513,
      "loss": 0.7135,
      "step": 565
    },
    {
      "epoch": 0.7764060356652949,
      "grad_norm": 2.2604219913482666,
      "learning_rate": 0.0009997325167765264,
      "loss": 1.222,
      "step": 566
    },
    {
      "epoch": 0.7777777777777778,
      "grad_norm": 1.4148451089859009,
      "learning_rate": 0.0009997019742249383,
      "loss": 0.4528,
      "step": 567
    },
    {
      "epoch": 0.7791495198902606,
      "grad_norm": 0.7912073731422424,
      "learning_rate": 0.0009996697814728644,
      "loss": 0.1328,
      "step": 568
    },
    {
      "epoch": 0.7805212620027435,
      "grad_norm": 2.348604917526245,
      "learning_rate": 0.0009996359386384374,
      "loss": 0.8546,
      "step": 569
    },
    {
      "epoch": 0.7818930041152263,
      "grad_norm": 1.507812261581421,
      "learning_rate": 0.000999600445845845,
      "loss": 0.4279,
      "step": 570
    },
    {
      "epoch": 0.7832647462277091,
      "grad_norm": 2.647885799407959,
      "learning_rate": 0.0009995633032253294,
      "loss": 1.425,
      "step": 571
    },
    {
      "epoch": 0.7846364883401921,
      "grad_norm": 1.7565921545028687,
      "learning_rate": 0.0009995245109131869,
      "loss": 0.4775,
      "step": 572
    },
    {
      "epoch": 0.7860082304526749,
      "grad_norm": 2.1025984287261963,
      "learning_rate": 0.000999484069051768,
      "loss": 0.6325,
      "step": 573
    },
    {
      "epoch": 0.7873799725651578,
      "grad_norm": 2.177631378173828,
      "learning_rate": 0.0009994419777894753,
      "loss": 0.7343,
      "step": 574
    },
    {
      "epoch": 0.7887517146776406,
      "grad_norm": 2.1677980422973633,
      "learning_rate": 0.0009993982372807648,
      "loss": 0.8758,
      "step": 575
    },
    {
      "epoch": 0.7901234567901234,
      "grad_norm": 2.5000686645507812,
      "learning_rate": 0.000999352847686144,
      "loss": 1.4886,
      "step": 576
    },
    {
      "epoch": 0.7914951989026063,
      "grad_norm": 2.379132032394409,
      "learning_rate": 0.000999305809172172,
      "loss": 1.1258,
      "step": 577
    },
    {
      "epoch": 0.7928669410150891,
      "grad_norm": 2.348276376724243,
      "learning_rate": 0.0009992571219114585,
      "loss": 1.0294,
      "step": 578
    },
    {
      "epoch": 0.7942386831275721,
      "grad_norm": 2.1541614532470703,
      "learning_rate": 0.000999206786082664,
      "loss": 1.0091,
      "step": 579
    },
    {
      "epoch": 0.7956104252400549,
      "grad_norm": 2.221259593963623,
      "learning_rate": 0.000999154801870497,
      "loss": 1.4008,
      "step": 580
    },
    {
      "epoch": 0.7969821673525377,
      "grad_norm": 1.3551172018051147,
      "learning_rate": 0.0009991011694657167,
      "loss": 0.4384,
      "step": 581
    },
    {
      "epoch": 0.7983539094650206,
      "grad_norm": 2.4766149520874023,
      "learning_rate": 0.000999045889065129,
      "loss": 1.2668,
      "step": 582
    },
    {
      "epoch": 0.7997256515775034,
      "grad_norm": 1.6486444473266602,
      "learning_rate": 0.000998988960871588,
      "loss": 0.7217,
      "step": 583
    },
    {
      "epoch": 0.8010973936899863,
      "grad_norm": 2.368161201477051,
      "learning_rate": 0.0009989303850939937,
      "loss": 1.242,
      "step": 584
    },
    {
      "epoch": 0.8024691358024691,
      "grad_norm": 1.684938907623291,
      "learning_rate": 0.0009988701619472926,
      "loss": 0.8439,
      "step": 585
    },
    {
      "epoch": 0.803840877914952,
      "grad_norm": 2.2783966064453125,
      "learning_rate": 0.0009988082916524762,
      "loss": 0.876,
      "step": 586
    },
    {
      "epoch": 0.8052126200274349,
      "grad_norm": 2.324449300765991,
      "learning_rate": 0.00099874477443658,
      "loss": 1.194,
      "step": 587
    },
    {
      "epoch": 0.8065843621399177,
      "grad_norm": 0.8597037196159363,
      "learning_rate": 0.0009986796105326831,
      "loss": 0.2045,
      "step": 588
    },
    {
      "epoch": 0.8079561042524005,
      "grad_norm": 2.2256205081939697,
      "learning_rate": 0.0009986128001799076,
      "loss": 1.2798,
      "step": 589
    },
    {
      "epoch": 0.8093278463648834,
      "grad_norm": 2.1167430877685547,
      "learning_rate": 0.0009985443436234167,
      "loss": 0.8574,
      "step": 590
    },
    {
      "epoch": 0.8106995884773662,
      "grad_norm": 0.10849010199308395,
      "learning_rate": 0.0009984742411144143,
      "loss": 0.0082,
      "step": 591
    },
    {
      "epoch": 0.8120713305898491,
      "grad_norm": 1.1498011350631714,
      "learning_rate": 0.0009984024929101448,
      "loss": 0.4044,
      "step": 592
    },
    {
      "epoch": 0.813443072702332,
      "grad_norm": 1.5021452903747559,
      "learning_rate": 0.0009983290992738915,
      "loss": 0.4919,
      "step": 593
    },
    {
      "epoch": 0.8148148148148148,
      "grad_norm": 2.5541698932647705,
      "learning_rate": 0.0009982540604749751,
      "loss": 1.2524,
      "step": 594
    },
    {
      "epoch": 0.8161865569272977,
      "grad_norm": 2.190497636795044,
      "learning_rate": 0.000998177376788754,
      "loss": 0.795,
      "step": 595
    },
    {
      "epoch": 0.8175582990397805,
      "grad_norm": 2.115995407104492,
      "learning_rate": 0.000998099048496622,
      "loss": 0.8756,
      "step": 596
    },
    {
      "epoch": 0.8189300411522634,
      "grad_norm": 0.6140339374542236,
      "learning_rate": 0.0009980190758860081,
      "loss": 0.1396,
      "step": 597
    },
    {
      "epoch": 0.8203017832647462,
      "grad_norm": 1.3098580837249756,
      "learning_rate": 0.0009979374592503753,
      "loss": 0.37,
      "step": 598
    },
    {
      "epoch": 0.821673525377229,
      "grad_norm": 1.9168777465820312,
      "learning_rate": 0.0009978541988892192,
      "loss": 0.8108,
      "step": 599
    },
    {
      "epoch": 0.823045267489712,
      "grad_norm": 1.329971432685852,
      "learning_rate": 0.0009977692951080672,
      "loss": 0.4655,
      "step": 600
    },
    {
      "epoch": 0.8244170096021948,
      "grad_norm": 1.5312412977218628,
      "learning_rate": 0.0009976827482184776,
      "loss": 0.6358,
      "step": 601
    },
    {
      "epoch": 0.8257887517146777,
      "grad_norm": 2.3435771465301514,
      "learning_rate": 0.0009975945585380377,
      "loss": 1.4994,
      "step": 602
    },
    {
      "epoch": 0.8271604938271605,
      "grad_norm": 1.913093090057373,
      "learning_rate": 0.0009975047263903632,
      "loss": 0.7349,
      "step": 603
    },
    {
      "epoch": 0.8285322359396433,
      "grad_norm": 2.202099323272705,
      "learning_rate": 0.000997413252105097,
      "loss": 1.3005,
      "step": 604
    },
    {
      "epoch": 0.8299039780521262,
      "grad_norm": 1.8408820629119873,
      "learning_rate": 0.000997320136017908,
      "loss": 1.019,
      "step": 605
    },
    {
      "epoch": 0.831275720164609,
      "grad_norm": 2.17401385307312,
      "learning_rate": 0.0009972253784704896,
      "loss": 1.0097,
      "step": 606
    },
    {
      "epoch": 0.8326474622770919,
      "grad_norm": 3.5524911880493164,
      "learning_rate": 0.0009971289798105585,
      "loss": 3.3517,
      "step": 607
    },
    {
      "epoch": 0.8340192043895748,
      "grad_norm": 2.2348220348358154,
      "learning_rate": 0.0009970309403918537,
      "loss": 1.3093,
      "step": 608
    },
    {
      "epoch": 0.8353909465020576,
      "grad_norm": 2.220083713531494,
      "learning_rate": 0.0009969312605741353,
      "loss": 1.4763,
      "step": 609
    },
    {
      "epoch": 0.8367626886145405,
      "grad_norm": 1.5978267192840576,
      "learning_rate": 0.0009968299407231824,
      "loss": 0.8629,
      "step": 610
    },
    {
      "epoch": 0.8381344307270233,
      "grad_norm": 1.6459195613861084,
      "learning_rate": 0.0009967269812107924,
      "loss": 0.8401,
      "step": 611
    },
    {
      "epoch": 0.8395061728395061,
      "grad_norm": 1.993955373764038,
      "learning_rate": 0.0009966223824147797,
      "loss": 1.1366,
      "step": 612
    },
    {
      "epoch": 0.840877914951989,
      "grad_norm": 2.3398549556732178,
      "learning_rate": 0.0009965161447189742,
      "loss": 1.4177,
      "step": 613
    },
    {
      "epoch": 0.8422496570644719,
      "grad_norm": 2.820521116256714,
      "learning_rate": 0.0009964082685132193,
      "loss": 2.9207,
      "step": 614
    },
    {
      "epoch": 0.8436213991769548,
      "grad_norm": 1.454741358757019,
      "learning_rate": 0.0009962987541933717,
      "loss": 0.6808,
      "step": 615
    },
    {
      "epoch": 0.8449931412894376,
      "grad_norm": 1.4422433376312256,
      "learning_rate": 0.0009961876021612984,
      "loss": 0.5965,
      "step": 616
    },
    {
      "epoch": 0.8463648834019204,
      "grad_norm": 0.2472972720861435,
      "learning_rate": 0.0009960748128248769,
      "loss": 0.0313,
      "step": 617
    },
    {
      "epoch": 0.8477366255144033,
      "grad_norm": 1.2811288833618164,
      "learning_rate": 0.000995960386597992,
      "loss": 0.8359,
      "step": 618
    },
    {
      "epoch": 0.8491083676268861,
      "grad_norm": 1.7258317470550537,
      "learning_rate": 0.0009958443239005362,
      "loss": 0.9549,
      "step": 619
    },
    {
      "epoch": 0.850480109739369,
      "grad_norm": 2.052578926086426,
      "learning_rate": 0.000995726625158406,
      "loss": 1.2424,
      "step": 620
    },
    {
      "epoch": 0.8518518518518519,
      "grad_norm": 2.598790407180786,
      "learning_rate": 0.0009956072908035023,
      "loss": 1.5772,
      "step": 621
    },
    {
      "epoch": 0.8532235939643347,
      "grad_norm": 1.8372243642807007,
      "learning_rate": 0.000995486321273727,
      "loss": 0.8351,
      "step": 622
    },
    {
      "epoch": 0.8545953360768176,
      "grad_norm": 2.4052348136901855,
      "learning_rate": 0.0009953637170129837,
      "loss": 0.726,
      "step": 623
    },
    {
      "epoch": 0.8559670781893004,
      "grad_norm": 1.429192066192627,
      "learning_rate": 0.0009952394784711735,
      "loss": 0.4982,
      "step": 624
    },
    {
      "epoch": 0.8573388203017832,
      "grad_norm": 1.8847312927246094,
      "learning_rate": 0.000995113606104195,
      "loss": 1.0102,
      "step": 625
    },
    {
      "epoch": 0.8587105624142661,
      "grad_norm": 1.5976717472076416,
      "learning_rate": 0.0009949861003739422,
      "loss": 0.7013,
      "step": 626
    },
    {
      "epoch": 0.8600823045267489,
      "grad_norm": 0.0,
      "learning_rate": 0.000994856961748303,
      "loss": 0.0,
      "step": 627
    },
    {
      "epoch": 0.8614540466392319,
      "grad_norm": 2.0918030738830566,
      "learning_rate": 0.000994726190701157,
      "loss": 0.9771,
      "step": 628
    },
    {
      "epoch": 0.8628257887517147,
      "grad_norm": 1.5623385906219482,
      "learning_rate": 0.0009945937877123738,
      "loss": 0.6401,
      "step": 629
    },
    {
      "epoch": 0.8641975308641975,
      "grad_norm": 1.988359808921814,
      "learning_rate": 0.0009944597532678121,
      "loss": 0.8453,
      "step": 630
    },
    {
      "epoch": 0.8655692729766804,
      "grad_norm": 1.7483891248703003,
      "learning_rate": 0.0009943240878593168,
      "loss": 0.6376,
      "step": 631
    },
    {
      "epoch": 0.8669410150891632,
      "grad_norm": 1.747367024421692,
      "learning_rate": 0.000994186791984718,
      "loss": 0.8945,
      "step": 632
    },
    {
      "epoch": 0.8683127572016461,
      "grad_norm": 1.7018707990646362,
      "learning_rate": 0.0009940478661478289,
      "loss": 0.7919,
      "step": 633
    },
    {
      "epoch": 0.869684499314129,
      "grad_norm": 2.589437484741211,
      "learning_rate": 0.0009939073108584432,
      "loss": 2.7374,
      "step": 634
    },
    {
      "epoch": 0.8710562414266118,
      "grad_norm": 1.5699372291564941,
      "learning_rate": 0.0009937651266323352,
      "loss": 0.6152,
      "step": 635
    },
    {
      "epoch": 0.8724279835390947,
      "grad_norm": 1.5574616193771362,
      "learning_rate": 0.0009936213139912556,
      "loss": 0.7544,
      "step": 636
    },
    {
      "epoch": 0.8737997256515775,
      "grad_norm": 1.4045825004577637,
      "learning_rate": 0.0009934758734629308,
      "loss": 0.764,
      "step": 637
    },
    {
      "epoch": 0.8751714677640604,
      "grad_norm": 0.0,
      "learning_rate": 0.0009933288055810616,
      "loss": 0.0,
      "step": 638
    },
    {
      "epoch": 0.8765432098765432,
      "grad_norm": 0.1518259048461914,
      "learning_rate": 0.0009931801108853191,
      "loss": 0.0341,
      "step": 639
    },
    {
      "epoch": 0.877914951989026,
      "grad_norm": 1.7389847040176392,
      "learning_rate": 0.0009930297899213452,
      "loss": 0.8661,
      "step": 640
    },
    {
      "epoch": 0.879286694101509,
      "grad_norm": 1.4264923334121704,
      "learning_rate": 0.0009928778432407492,
      "loss": 0.5543,
      "step": 641
    },
    {
      "epoch": 0.8806584362139918,
      "grad_norm": 1.8644518852233887,
      "learning_rate": 0.0009927242714011055,
      "loss": 0.7215,
      "step": 642
    },
    {
      "epoch": 0.8820301783264746,
      "grad_norm": 1.5277243852615356,
      "learning_rate": 0.0009925690749659527,
      "loss": 0.5738,
      "step": 643
    },
    {
      "epoch": 0.8834019204389575,
      "grad_norm": 1.8840594291687012,
      "learning_rate": 0.0009924122545047907,
      "loss": 1.2774,
      "step": 644
    },
    {
      "epoch": 0.8847736625514403,
      "grad_norm": 1.5248987674713135,
      "learning_rate": 0.000992253810593079,
      "loss": 0.5194,
      "step": 645
    },
    {
      "epoch": 0.8861454046639232,
      "grad_norm": 1.2852954864501953,
      "learning_rate": 0.000992093743812234,
      "loss": 0.4957,
      "step": 646
    },
    {
      "epoch": 0.887517146776406,
      "grad_norm": 1.4410088062286377,
      "learning_rate": 0.0009919320547496276,
      "loss": 0.6336,
      "step": 647
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 2.2397258281707764,
      "learning_rate": 0.0009917687439985848,
      "loss": 1.3606,
      "step": 648
    },
    {
      "epoch": 0.8902606310013718,
      "grad_norm": 1.6791517734527588,
      "learning_rate": 0.000991603812158381,
      "loss": 0.5841,
      "step": 649
    },
    {
      "epoch": 0.8916323731138546,
      "grad_norm": 1.8580983877182007,
      "learning_rate": 0.000991437259834241,
      "loss": 0.7388,
      "step": 650
    },
    {
      "epoch": 0.8930041152263375,
      "grad_norm": 0.7454937696456909,
      "learning_rate": 0.0009912690876373357,
      "loss": 0.1784,
      "step": 651
    },
    {
      "epoch": 0.8943758573388203,
      "grad_norm": 1.311299443244934,
      "learning_rate": 0.0009910992961847797,
      "loss": 0.8583,
      "step": 652
    },
    {
      "epoch": 0.8957475994513031,
      "grad_norm": 0.6724503040313721,
      "learning_rate": 0.00099092788609963,
      "loss": 0.16,
      "step": 653
    },
    {
      "epoch": 0.897119341563786,
      "grad_norm": 0.3248959481716156,
      "learning_rate": 0.0009907548580108834,
      "loss": 0.0596,
      "step": 654
    },
    {
      "epoch": 0.8984910836762688,
      "grad_norm": 1.593714714050293,
      "learning_rate": 0.0009905802125534738,
      "loss": 0.5839,
      "step": 655
    },
    {
      "epoch": 0.8998628257887518,
      "grad_norm": 1.5883684158325195,
      "learning_rate": 0.0009904039503682701,
      "loss": 0.5885,
      "step": 656
    },
    {
      "epoch": 0.9012345679012346,
      "grad_norm": 2.72161602973938,
      "learning_rate": 0.0009902260721020737,
      "loss": 1.712,
      "step": 657
    },
    {
      "epoch": 0.9026063100137174,
      "grad_norm": 2.3555116653442383,
      "learning_rate": 0.0009900465784076168,
      "loss": 1.1589,
      "step": 658
    },
    {
      "epoch": 0.9039780521262003,
      "grad_norm": 1.8374806642532349,
      "learning_rate": 0.0009898654699435586,
      "loss": 1.062,
      "step": 659
    },
    {
      "epoch": 0.9053497942386831,
      "grad_norm": 1.2965621948242188,
      "learning_rate": 0.0009896827473744848,
      "loss": 0.5203,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_Qnli-dev_cosine_accuracy": 0.705078125,
      "eval_Qnli-dev_cosine_accuracy_threshold": 0.8212183713912964,
      "eval_Qnli-dev_cosine_ap": 0.7186148814161074,
      "eval_Qnli-dev_cosine_f1": 0.685823754789272,
      "eval_Qnli-dev_cosine_f1_threshold": 0.7696558237075806,
      "eval_Qnli-dev_cosine_precision": 0.6258741258741258,
      "eval_Qnli-dev_cosine_recall": 0.7584745762711864,
      "eval_allNLI-dev_cosine_accuracy": 0.765625,
      "eval_allNLI-dev_cosine_accuracy_threshold": 0.8777807950973511,
      "eval_allNLI-dev_cosine_ap": 0.6686332553712844,
      "eval_allNLI-dev_cosine_f1": 0.671201814058957,
      "eval_allNLI-dev_cosine_f1_threshold": 0.811836838722229,
      "eval_allNLI-dev_cosine_precision": 0.5522388059701493,
      "eval_allNLI-dev_cosine_recall": 0.8554913294797688,
      "eval_sequential_score": 0.7186148814161074,
      "eval_sts-test_pearson_cosine": 0.8424999179514703,
      "eval_sts-test_spearman_cosine": 0.8868519917012493,
      "eval_vitaminc-pairs_loss": 2.4104244709014893,
      "eval_vitaminc-pairs_runtime": 14.3546,
      "eval_vitaminc-pairs_samples_per_second": 8.917,
      "eval_vitaminc-pairs_steps_per_second": 0.07,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_negation-triplets_loss": 2.0854151248931885,
      "eval_negation-triplets_runtime": 1.1542,
      "eval_negation-triplets_samples_per_second": 110.903,
      "eval_negation-triplets_steps_per_second": 0.866,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_scitail-pairs-pos_loss": 0.174000546336174,
      "eval_scitail-pairs-pos_runtime": 1.5642,
      "eval_scitail-pairs-pos_samples_per_second": 81.828,
      "eval_scitail-pairs-pos_steps_per_second": 0.639,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_scitail-pairs-qa_loss": 0.024205714464187622,
      "eval_scitail-pairs-qa_runtime": 1.2073,
      "eval_scitail-pairs-qa_samples_per_second": 106.02,
      "eval_scitail-pairs-qa_steps_per_second": 0.828,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_xsum-pairs_loss": 0.6651791930198669,
      "eval_xsum-pairs_runtime": 6.3082,
      "eval_xsum-pairs_samples_per_second": 20.291,
      "eval_xsum-pairs_steps_per_second": 0.159,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_sciq_pairs_loss": 0.08112659305334091,
      "eval_sciq_pairs_runtime": 8.8181,
      "eval_sciq_pairs_samples_per_second": 14.516,
      "eval_sciq_pairs_steps_per_second": 0.113,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_qasc_pairs_loss": 0.8805878162384033,
      "eval_qasc_pairs_runtime": 1.3897,
      "eval_qasc_pairs_samples_per_second": 92.104,
      "eval_qasc_pairs_steps_per_second": 0.72,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_openbookqa_pairs_loss": 1.3323636054992676,
      "eval_openbookqa_pairs_runtime": 1.1839,
      "eval_openbookqa_pairs_samples_per_second": 108.119,
      "eval_openbookqa_pairs_steps_per_second": 0.845,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_nq_pairs_loss": 0.882255494594574,
      "eval_nq_pairs_runtime": 7.889,
      "eval_nq_pairs_samples_per_second": 16.225,
      "eval_nq_pairs_steps_per_second": 0.127,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_trivia_pairs_loss": 1.2122423648834229,
      "eval_trivia_pairs_runtime": 8.9233,
      "eval_trivia_pairs_samples_per_second": 14.344,
      "eval_trivia_pairs_steps_per_second": 0.112,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_gooaq_pairs_loss": 0.7754307389259338,
      "eval_gooaq_pairs_runtime": 2.0694,
      "eval_gooaq_pairs_samples_per_second": 61.854,
      "eval_gooaq_pairs_steps_per_second": 0.483,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_paws-pos_loss": 0.03314030170440674,
      "eval_paws-pos_runtime": 1.5045,
      "eval_paws-pos_samples_per_second": 85.076,
      "eval_paws-pos_steps_per_second": 0.665,
      "step": 660
    },
    {
      "epoch": 0.9053497942386831,
      "eval_global_dataset_loss": 0.24366270005702972,
      "eval_global_dataset_runtime": 33.3186,
      "eval_global_dataset_samples_per_second": 11.525,
      "eval_global_dataset_steps_per_second": 0.06,
      "step": 660
    },
    {
      "epoch": 0.906721536351166,
      "grad_norm": 1.8050487041473389,
      "learning_rate": 0.0009894984113709032,
      "loss": 0.8125,
      "step": 661
    },
    {
      "epoch": 0.9080932784636488,
      "grad_norm": 1.6021864414215088,
      "learning_rate": 0.0009893124626092428,
      "loss": 0.6125,
      "step": 662
    },
    {
      "epoch": 0.9094650205761317,
      "grad_norm": 1.8087327480316162,
      "learning_rate": 0.0009891249017718508,
      "loss": 0.7289,
      "step": 663
    },
    {
      "epoch": 0.9108367626886146,
      "grad_norm": 1.9543635845184326,
      "learning_rate": 0.0009889357295469893,
      "loss": 1.237,
      "step": 664
    },
    {
      "epoch": 0.9122085048010974,
      "grad_norm": 2.4027888774871826,
      "learning_rate": 0.0009887449466288336,
      "loss": 2.274,
      "step": 665
    },
    {
      "epoch": 0.9135802469135802,
      "grad_norm": 1.611427664756775,
      "learning_rate": 0.00098855255371747,
      "loss": 0.7283,
      "step": 666
    },
    {
      "epoch": 0.9149519890260631,
      "grad_norm": 1.0573492050170898,
      "learning_rate": 0.0009883585515188926,
      "loss": 0.5324,
      "step": 667
    },
    {
      "epoch": 0.9163237311385459,
      "grad_norm": 0.8396095633506775,
      "learning_rate": 0.0009881629407450007,
      "loss": 0.3173,
      "step": 668
    },
    {
      "epoch": 0.9176954732510288,
      "grad_norm": 1.3495731353759766,
      "learning_rate": 0.0009879657221135959,
      "loss": 0.7001,
      "step": 669
    },
    {
      "epoch": 0.9190672153635117,
      "grad_norm": 1.5316470861434937,
      "learning_rate": 0.0009877668963483807,
      "loss": 1.0469,
      "step": 670
    },
    {
      "epoch": 0.9204389574759945,
      "grad_norm": 0.7262068390846252,
      "learning_rate": 0.0009875664641789544,
      "loss": 0.1802,
      "step": 671
    },
    {
      "epoch": 0.9218106995884774,
      "grad_norm": 1.9269604682922363,
      "learning_rate": 0.0009873644263408117,
      "loss": 1.0294,
      "step": 672
    },
    {
      "epoch": 0.9231824417009602,
      "grad_norm": 1.9547969102859497,
      "learning_rate": 0.0009871607835753391,
      "loss": 1.1394,
      "step": 673
    },
    {
      "epoch": 0.9245541838134431,
      "grad_norm": 1.395339846611023,
      "learning_rate": 0.0009869555366298118,
      "loss": 0.6619,
      "step": 674
    },
    {
      "epoch": 0.9259259259259259,
      "grad_norm": 1.4688735008239746,
      "learning_rate": 0.0009867486862573929,
      "loss": 0.606,
      "step": 675
    },
    {
      "epoch": 0.9272976680384087,
      "grad_norm": 3.08390474319458,
      "learning_rate": 0.0009865402332171278,
      "loss": 2.8903,
      "step": 676
    },
    {
      "epoch": 0.9286694101508917,
      "grad_norm": 1.559369683265686,
      "learning_rate": 0.0009863301782739445,
      "loss": 0.9967,
      "step": 677
    },
    {
      "epoch": 0.9300411522633745,
      "grad_norm": 1.3444671630859375,
      "learning_rate": 0.0009861185221986483,
      "loss": 0.668,
      "step": 678
    },
    {
      "epoch": 0.9314128943758574,
      "grad_norm": 1.212170124053955,
      "learning_rate": 0.00098590526576792,
      "loss": 0.6055,
      "step": 679
    },
    {
      "epoch": 0.9327846364883402,
      "grad_norm": 0.26228439807891846,
      "learning_rate": 0.0009856904097643134,
      "loss": 0.0423,
      "step": 680
    },
    {
      "epoch": 0.934156378600823,
      "grad_norm": 1.117007851600647,
      "learning_rate": 0.0009854739549762517,
      "loss": 0.5199,
      "step": 681
    },
    {
      "epoch": 0.9355281207133059,
      "grad_norm": 0.14136677980422974,
      "learning_rate": 0.0009852559021980244,
      "loss": 0.027,
      "step": 682
    },
    {
      "epoch": 0.9368998628257887,
      "grad_norm": 1.8655521869659424,
      "learning_rate": 0.000985036252229786,
      "loss": 1.2006,
      "step": 683
    },
    {
      "epoch": 0.9382716049382716,
      "grad_norm": 2.7191097736358643,
      "learning_rate": 0.0009848150058775513,
      "loss": 2.7359,
      "step": 684
    },
    {
      "epoch": 0.9396433470507545,
      "grad_norm": 2.323805093765259,
      "learning_rate": 0.000984592163953193,
      "loss": 1.2884,
      "step": 685
    },
    {
      "epoch": 0.9410150891632373,
      "grad_norm": 1.5158648490905762,
      "learning_rate": 0.0009843677272744393,
      "loss": 0.8306,
      "step": 686
    },
    {
      "epoch": 0.9423868312757202,
      "grad_norm": 0.14592251181602478,
      "learning_rate": 0.0009841416966648697,
      "loss": 0.0149,
      "step": 687
    },
    {
      "epoch": 0.943758573388203,
      "grad_norm": 1.3281621932983398,
      "learning_rate": 0.0009839140729539135,
      "loss": 0.5484,
      "step": 688
    },
    {
      "epoch": 0.9451303155006858,
      "grad_norm": 1.3377467393875122,
      "learning_rate": 0.0009836848569768453,
      "loss": 0.3777,
      "step": 689
    },
    {
      "epoch": 0.9465020576131687,
      "grad_norm": 2.10798716545105,
      "learning_rate": 0.000983454049574783,
      "loss": 1.0078,
      "step": 690
    },
    {
      "epoch": 0.9478737997256516,
      "grad_norm": 0.0,
      "learning_rate": 0.0009832216515946843,
      "loss": 0.0,
      "step": 691
    },
    {
      "epoch": 0.9492455418381345,
      "grad_norm": 1.8169300556182861,
      "learning_rate": 0.000982987663889343,
      "loss": 0.4982,
      "step": 692
    },
    {
      "epoch": 0.9506172839506173,
      "grad_norm": 4.212324142456055,
      "learning_rate": 0.0009827520873173872,
      "loss": 0.9299,
      "step": 693
    },
    {
      "epoch": 0.9519890260631001,
      "grad_norm": 2.0838160514831543,
      "learning_rate": 0.0009825149227432753,
      "loss": 0.9821,
      "step": 694
    },
    {
      "epoch": 0.953360768175583,
      "grad_norm": 0.9287631511688232,
      "learning_rate": 0.0009822761710372926,
      "loss": 0.126,
      "step": 695
    },
    {
      "epoch": 0.9547325102880658,
      "grad_norm": 1.845534086227417,
      "learning_rate": 0.0009820358330755485,
      "loss": 0.9357,
      "step": 696
    },
    {
      "epoch": 0.9561042524005487,
      "grad_norm": 1.9858702421188354,
      "learning_rate": 0.0009817939097399738,
      "loss": 0.9327,
      "step": 697
    },
    {
      "epoch": 0.9574759945130316,
      "grad_norm": 2.7823987007141113,
      "learning_rate": 0.000981550401918316,
      "loss": 0.5134,
      "step": 698
    },
    {
      "epoch": 0.9588477366255144,
      "grad_norm": 1.6842830181121826,
      "learning_rate": 0.0009813053105041374,
      "loss": 0.9199,
      "step": 699
    },
    {
      "epoch": 0.9602194787379973,
      "grad_norm": 2.740748882293701,
      "learning_rate": 0.0009810586363968115,
      "loss": 0.4819,
      "step": 700
    },
    {
      "epoch": 0.9615912208504801,
      "grad_norm": 1.9738132953643799,
      "learning_rate": 0.0009808103805015193,
      "loss": 1.3836,
      "step": 701
    },
    {
      "epoch": 0.9629629629629629,
      "grad_norm": 1.3568272590637207,
      "learning_rate": 0.0009805605437292466,
      "loss": 0.5922,
      "step": 702
    },
    {
      "epoch": 0.9643347050754458,
      "grad_norm": 1.3056647777557373,
      "learning_rate": 0.0009803091269967798,
      "loss": 0.6897,
      "step": 703
    },
    {
      "epoch": 0.9657064471879286,
      "grad_norm": 1.3313848972320557,
      "learning_rate": 0.0009800561312267033,
      "loss": 0.6079,
      "step": 704
    },
    {
      "epoch": 0.9670781893004116,
      "grad_norm": 0.29244112968444824,
      "learning_rate": 0.0009798015573473959,
      "loss": 0.0456,
      "step": 705
    },
    {
      "epoch": 0.9684499314128944,
      "grad_norm": 1.086900234222412,
      "learning_rate": 0.0009795454062930273,
      "loss": 0.5054,
      "step": 706
    },
    {
      "epoch": 0.9698216735253772,
      "grad_norm": 2.065800666809082,
      "learning_rate": 0.0009792876790035545,
      "loss": 1.1145,
      "step": 707
    },
    {
      "epoch": 0.9711934156378601,
      "grad_norm": 1.428763508796692,
      "learning_rate": 0.0009790283764247187,
      "loss": 0.6606,
      "step": 708
    },
    {
      "epoch": 0.9725651577503429,
      "grad_norm": 1.5778110027313232,
      "learning_rate": 0.000978767499508042,
      "loss": 0.824,
      "step": 709
    },
    {
      "epoch": 0.9739368998628258,
      "grad_norm": 1.6329480409622192,
      "learning_rate": 0.000978505049210823,
      "loss": 0.7261,
      "step": 710
    },
    {
      "epoch": 0.9753086419753086,
      "grad_norm": 0.0,
      "learning_rate": 0.0009782410264961346,
      "loss": 0.0,
      "step": 711
    },
    {
      "epoch": 0.9766803840877915,
      "grad_norm": 1.2380640506744385,
      "learning_rate": 0.0009779754323328191,
      "loss": 0.4709,
      "step": 712
    },
    {
      "epoch": 0.9780521262002744,
      "grad_norm": 1.931005835533142,
      "learning_rate": 0.0009777082676954854,
      "loss": 0.8106,
      "step": 713
    },
    {
      "epoch": 0.9794238683127572,
      "grad_norm": 3.596893072128296,
      "learning_rate": 0.0009774395335645058,
      "loss": 0.9323,
      "step": 714
    },
    {
      "epoch": 0.9807956104252401,
      "grad_norm": 2.934410333633423,
      "learning_rate": 0.0009771692309260116,
      "loss": 0.4457,
      "step": 715
    },
    {
      "epoch": 0.9821673525377229,
      "grad_norm": 1.522354245185852,
      "learning_rate": 0.0009768973607718896,
      "loss": 0.892,
      "step": 716
    },
    {
      "epoch": 0.9835390946502057,
      "grad_norm": 0.008752044290304184,
      "learning_rate": 0.000976623924099779,
      "loss": 0.0004,
      "step": 717
    },
    {
      "epoch": 0.9849108367626886,
      "grad_norm": 1.9101074934005737,
      "learning_rate": 0.0009763489219130669,
      "loss": 0.4401,
      "step": 718
    },
    {
      "epoch": 0.9862825788751715,
      "grad_norm": 1.5685993432998657,
      "learning_rate": 0.000976072355220886,
      "loss": 0.6879,
      "step": 719
    },
    {
      "epoch": 0.9876543209876543,
      "grad_norm": 1.2950056791305542,
      "learning_rate": 0.0009757942250381094,
      "loss": 0.5459,
      "step": 720
    },
    {
      "epoch": 0.9890260631001372,
      "grad_norm": 1.4739328622817993,
      "learning_rate": 0.0009755145323853476,
      "loss": 0.5961,
      "step": 721
    },
    {
      "epoch": 0.99039780521262,
      "grad_norm": 1.3605318069458008,
      "learning_rate": 0.0009752332782889445,
      "loss": 0.525,
      "step": 722
    },
    {
      "epoch": 0.9917695473251029,
      "grad_norm": 1.3376563787460327,
      "learning_rate": 0.0009749504637809742,
      "loss": 0.4761,
      "step": 723
    },
    {
      "epoch": 0.9931412894375857,
      "grad_norm": 1.6065219640731812,
      "learning_rate": 0.0009746660898992362,
      "loss": 0.6615,
      "step": 724
    },
    {
      "epoch": 0.9945130315500685,
      "grad_norm": 2.057121753692627,
      "learning_rate": 0.0009743801576872529,
      "loss": 0.2234,
      "step": 725
    },
    {
      "epoch": 0.9958847736625515,
      "grad_norm": 1.6144764423370361,
      "learning_rate": 0.0009740926681942644,
      "loss": 0.7386,
      "step": 726
    },
    {
      "epoch": 0.9972565157750343,
      "grad_norm": 1.6106828451156616,
      "learning_rate": 0.000973803622475226,
      "loss": 0.6143,
      "step": 727
    },
    {
      "epoch": 0.9986282578875172,
      "grad_norm": 2.7121546268463135,
      "learning_rate": 0.0009735130215908027,
      "loss": 0.2098,
      "step": 728
    },
    {
      "epoch": 1.0,
      "grad_norm": 3.613818407058716,
      "learning_rate": 0.000973220866607367,
      "loss": 0.9473,
      "step": 729
    },
    {
      "epoch": 1.0013717421124828,
      "grad_norm": 1.3742542266845703,
      "learning_rate": 0.0009729271585969944,
      "loss": 1.0674,
      "step": 730
    },
    {
      "epoch": 1.0027434842249656,
      "grad_norm": 1.2879219055175781,
      "learning_rate": 0.0009726318986374583,
      "loss": 0.6253,
      "step": 731
    },
    {
      "epoch": 1.0041152263374487,
      "grad_norm": 1.6723568439483643,
      "learning_rate": 0.0009723350878122281,
      "loss": 1.1695,
      "step": 732
    },
    {
      "epoch": 1.0054869684499315,
      "grad_norm": 1.352607011795044,
      "learning_rate": 0.0009720367272104635,
      "loss": 0.7821,
      "step": 733
    },
    {
      "epoch": 1.0068587105624143,
      "grad_norm": 1.9459525346755981,
      "learning_rate": 0.0009717368179270114,
      "loss": 1.3219,
      "step": 734
    },
    {
      "epoch": 1.008230452674897,
      "grad_norm": 1.2547602653503418,
      "learning_rate": 0.0009714353610624018,
      "loss": 0.5477,
      "step": 735
    },
    {
      "epoch": 1.00960219478738,
      "grad_norm": 1.3829255104064941,
      "learning_rate": 0.0009711323577228433,
      "loss": 0.586,
      "step": 736
    },
    {
      "epoch": 1.010973936899863,
      "grad_norm": 2.543593168258667,
      "learning_rate": 0.0009708278090202196,
      "loss": 2.7577,
      "step": 737
    },
    {
      "epoch": 1.0123456790123457,
      "grad_norm": 1.2957632541656494,
      "learning_rate": 0.000970521716072085,
      "loss": 0.5702,
      "step": 738
    },
    {
      "epoch": 1.0137174211248285,
      "grad_norm": 1.436959147453308,
      "learning_rate": 0.0009702140800016606,
      "loss": 0.7072,
      "step": 739
    },
    {
      "epoch": 1.0150891632373114,
      "grad_norm": 1.2317531108856201,
      "learning_rate": 0.0009699049019378303,
      "loss": 0.6435,
      "step": 740
    },
    {
      "epoch": 1.0164609053497942,
      "grad_norm": 1.547977089881897,
      "learning_rate": 0.0009695941830151357,
      "loss": 1.0398,
      "step": 741
    },
    {
      "epoch": 1.0178326474622772,
      "grad_norm": 0.8504058122634888,
      "learning_rate": 0.0009692819243737734,
      "loss": 0.3255,
      "step": 742
    },
    {
      "epoch": 1.01920438957476,
      "grad_norm": 0.9661345481872559,
      "learning_rate": 0.0009689681271595893,
      "loss": 0.4854,
      "step": 743
    },
    {
      "epoch": 1.0205761316872428,
      "grad_norm": 1.3813225030899048,
      "learning_rate": 0.0009686527925240763,
      "loss": 0.7538,
      "step": 744
    },
    {
      "epoch": 1.0219478737997256,
      "grad_norm": 1.557174563407898,
      "learning_rate": 0.0009683359216243677,
      "loss": 0.8678,
      "step": 745
    },
    {
      "epoch": 1.0233196159122084,
      "grad_norm": 1.7113001346588135,
      "learning_rate": 0.0009680175156232348,
      "loss": 1.1826,
      "step": 746
    },
    {
      "epoch": 1.0246913580246915,
      "grad_norm": 0.5500476360321045,
      "learning_rate": 0.000967697575689082,
      "loss": 0.1444,
      "step": 747
    },
    {
      "epoch": 1.0260631001371743,
      "grad_norm": 1.592626929283142,
      "learning_rate": 0.0009673761029959427,
      "loss": 0.6853,
      "step": 748
    },
    {
      "epoch": 1.027434842249657,
      "grad_norm": 0.5350927114486694,
      "learning_rate": 0.0009670530987234742,
      "loss": 0.0938,
      "step": 749
    },
    {
      "epoch": 1.02880658436214,
      "grad_norm": 1.7295268774032593,
      "learning_rate": 0.0009667285640569544,
      "loss": 1.1651,
      "step": 750
    },
    {
      "epoch": 1.0301783264746227,
      "grad_norm": 2.3311941623687744,
      "learning_rate": 0.0009664025001872773,
      "loss": 1.3879,
      "step": 751
    },
    {
      "epoch": 1.0315500685871055,
      "grad_norm": 1.8849881887435913,
      "learning_rate": 0.0009660749083109482,
      "loss": 0.8814,
      "step": 752
    },
    {
      "epoch": 1.0329218106995885,
      "grad_norm": 2.384006977081299,
      "learning_rate": 0.000965745789630079,
      "loss": 1.4394,
      "step": 753
    },
    {
      "epoch": 1.0342935528120714,
      "grad_norm": 2.0547990798950195,
      "learning_rate": 0.0009654151453523849,
      "loss": 1.0803,
      "step": 754
    },
    {
      "epoch": 1.0356652949245542,
      "grad_norm": 1.5164790153503418,
      "learning_rate": 0.0009650829766911791,
      "loss": 0.999,
      "step": 755
    },
    {
      "epoch": 1.037037037037037,
      "grad_norm": 0.14709718525409698,
      "learning_rate": 0.0009647492848653688,
      "loss": 0.0263,
      "step": 756
    },
    {
      "epoch": 1.0384087791495198,
      "grad_norm": 1.7678200006484985,
      "learning_rate": 0.0009644140710994499,
      "loss": 1.3087,
      "step": 757
    },
    {
      "epoch": 1.0397805212620028,
      "grad_norm": 1.449784517288208,
      "learning_rate": 0.0009640773366235037,
      "loss": 0.8582,
      "step": 758
    },
    {
      "epoch": 1.0411522633744856,
      "grad_norm": 1.5782291889190674,
      "learning_rate": 0.000963739082673192,
      "loss": 1.2331,
      "step": 759
    },
    {
      "epoch": 1.0425240054869684,
      "grad_norm": 1.7318320274353027,
      "learning_rate": 0.0009633993104897516,
      "loss": 1.153,
      "step": 760
    },
    {
      "epoch": 1.0438957475994513,
      "grad_norm": 1.2774981260299683,
      "learning_rate": 0.000963058021319991,
      "loss": 0.7105,
      "step": 761
    },
    {
      "epoch": 1.045267489711934,
      "grad_norm": 1.151254653930664,
      "learning_rate": 0.0009627152164162854,
      "loss": 0.4483,
      "step": 762
    },
    {
      "epoch": 1.046639231824417,
      "grad_norm": 1.1768347024917603,
      "learning_rate": 0.000962370897036572,
      "loss": 0.509,
      "step": 763
    },
    {
      "epoch": 1.0480109739369,
      "grad_norm": 1.2135556936264038,
      "learning_rate": 0.0009620250644443453,
      "loss": 0.5256,
      "step": 764
    },
    {
      "epoch": 1.0493827160493827,
      "grad_norm": 1.44373619556427,
      "learning_rate": 0.0009616777199086528,
      "loss": 1.0534,
      "step": 765
    },
    {
      "epoch": 1.0507544581618655,
      "grad_norm": 0.7816022634506226,
      "learning_rate": 0.0009613288647040898,
      "loss": 0.3023,
      "step": 766
    },
    {
      "epoch": 1.0521262002743483,
      "grad_norm": 1.0203887224197388,
      "learning_rate": 0.0009609785001107956,
      "loss": 0.485,
      "step": 767
    },
    {
      "epoch": 1.0534979423868314,
      "grad_norm": 2.1880743503570557,
      "learning_rate": 0.0009606266274144476,
      "loss": 2.504,
      "step": 768
    },
    {
      "epoch": 1.0548696844993142,
      "grad_norm": 1.410413384437561,
      "learning_rate": 0.0009602732479062577,
      "loss": 0.6644,
      "step": 769
    },
    {
      "epoch": 1.056241426611797,
      "grad_norm": 0.8773303627967834,
      "learning_rate": 0.0009599183628829671,
      "loss": 0.315,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_Qnli-dev_cosine_accuracy": 0.671875,
      "eval_Qnli-dev_cosine_accuracy_threshold": 0.814854085445404,
      "eval_Qnli-dev_cosine_ap": 0.7068053601409796,
      "eval_Qnli-dev_cosine_f1": 0.674074074074074,
      "eval_Qnli-dev_cosine_f1_threshold": 0.7766687273979187,
      "eval_Qnli-dev_cosine_precision": 0.5986842105263158,
      "eval_Qnli-dev_cosine_recall": 0.7711864406779662,
      "eval_allNLI-dev_cosine_accuracy": 0.73828125,
      "eval_allNLI-dev_cosine_accuracy_threshold": 0.8861159086227417,
      "eval_allNLI-dev_cosine_ap": 0.6420317420284644,
      "eval_allNLI-dev_cosine_f1": 0.6635071090047393,
      "eval_allNLI-dev_cosine_f1_threshold": 0.8311915397644043,
      "eval_allNLI-dev_cosine_precision": 0.5622489959839357,
      "eval_allNLI-dev_cosine_recall": 0.8092485549132948,
      "eval_sequential_score": 0.7068053601409796,
      "eval_sts-test_pearson_cosine": 0.8217207063377605,
      "eval_sts-test_spearman_cosine": 0.8811758759601482,
      "eval_vitaminc-pairs_loss": 2.6132569313049316,
      "eval_vitaminc-pairs_runtime": 14.304,
      "eval_vitaminc-pairs_samples_per_second": 8.949,
      "eval_vitaminc-pairs_steps_per_second": 0.07,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_negation-triplets_loss": 2.132938861846924,
      "eval_negation-triplets_runtime": 1.1478,
      "eval_negation-triplets_samples_per_second": 111.519,
      "eval_negation-triplets_steps_per_second": 0.871,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_scitail-pairs-pos_loss": 0.07428418844938278,
      "eval_scitail-pairs-pos_runtime": 1.5461,
      "eval_scitail-pairs-pos_samples_per_second": 82.787,
      "eval_scitail-pairs-pos_steps_per_second": 0.647,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_scitail-pairs-qa_loss": 0.08091621100902557,
      "eval_scitail-pairs-qa_runtime": 1.2038,
      "eval_scitail-pairs-qa_samples_per_second": 106.326,
      "eval_scitail-pairs-qa_steps_per_second": 0.831,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_xsum-pairs_loss": 0.5966969132423401,
      "eval_xsum-pairs_runtime": 6.2886,
      "eval_xsum-pairs_samples_per_second": 20.354,
      "eval_xsum-pairs_steps_per_second": 0.159,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_sciq_pairs_loss": 0.07557538151741028,
      "eval_sciq_pairs_runtime": 8.8359,
      "eval_sciq_pairs_samples_per_second": 14.486,
      "eval_sciq_pairs_steps_per_second": 0.113,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_qasc_pairs_loss": 0.7843579649925232,
      "eval_qasc_pairs_runtime": 1.3845,
      "eval_qasc_pairs_samples_per_second": 92.455,
      "eval_qasc_pairs_steps_per_second": 0.722,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_openbookqa_pairs_loss": 1.2132459878921509,
      "eval_openbookqa_pairs_runtime": 1.1796,
      "eval_openbookqa_pairs_samples_per_second": 108.51,
      "eval_openbookqa_pairs_steps_per_second": 0.848,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_nq_pairs_loss": 0.6117019653320312,
      "eval_nq_pairs_runtime": 7.8794,
      "eval_nq_pairs_samples_per_second": 16.245,
      "eval_nq_pairs_steps_per_second": 0.127,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_trivia_pairs_loss": 1.201032042503357,
      "eval_trivia_pairs_runtime": 8.9312,
      "eval_trivia_pairs_samples_per_second": 14.332,
      "eval_trivia_pairs_steps_per_second": 0.112,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_gooaq_pairs_loss": 0.6824722290039062,
      "eval_gooaq_pairs_runtime": 2.0669,
      "eval_gooaq_pairs_samples_per_second": 61.929,
      "eval_gooaq_pairs_steps_per_second": 0.484,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_paws-pos_loss": 0.0346776507794857,
      "eval_paws-pos_runtime": 1.5045,
      "eval_paws-pos_samples_per_second": 85.081,
      "eval_paws-pos_steps_per_second": 0.665,
      "step": 770
    },
    {
      "epoch": 1.056241426611797,
      "eval_global_dataset_loss": 0.2154615968465805,
      "eval_global_dataset_runtime": 33.2725,
      "eval_global_dataset_samples_per_second": 11.541,
      "eval_global_dataset_steps_per_second": 0.06,
      "step": 770
    },
    {
      "epoch": 1.0576131687242798,
      "grad_norm": 2.182539939880371,
      "learning_rate": 0.0009595619736468414,
      "loss": 1.2877,
      "step": 771
    },
    {
      "epoch": 1.0589849108367626,
      "grad_norm": 0.7996990084648132,
      "learning_rate": 0.0009592040815056662,
      "loss": 0.2366,
      "step": 772
    },
    {
      "epoch": 1.0603566529492456,
      "grad_norm": 1.3453471660614014,
      "learning_rate": 0.0009588446877727416,
      "loss": 0.7266,
      "step": 773
    },
    {
      "epoch": 1.0617283950617284,
      "grad_norm": 1.4371206760406494,
      "learning_rate": 0.0009584837937668784,
      "loss": 0.7014,
      "step": 774
    },
    {
      "epoch": 1.0631001371742113,
      "grad_norm": 1.5218498706817627,
      "learning_rate": 0.0009581214008123924,
      "loss": 1.1267,
      "step": 775
    },
    {
      "epoch": 1.064471879286694,
      "grad_norm": 1.5613763332366943,
      "learning_rate": 0.0009577575102390998,
      "loss": 1.118,
      "step": 776
    },
    {
      "epoch": 1.0658436213991769,
      "grad_norm": 1.1349512338638306,
      "learning_rate": 0.0009573921233823129,
      "loss": 0.9615,
      "step": 777
    },
    {
      "epoch": 1.06721536351166,
      "grad_norm": 1.3341729640960693,
      "learning_rate": 0.0009570252415828337,
      "loss": 0.4946,
      "step": 778
    },
    {
      "epoch": 1.0685871056241427,
      "grad_norm": 1.9713042974472046,
      "learning_rate": 0.0009566568661869511,
      "loss": 1.3275,
      "step": 779
    },
    {
      "epoch": 1.0699588477366255,
      "grad_norm": 0.9707056283950806,
      "learning_rate": 0.000956286998546434,
      "loss": 0.3784,
      "step": 780
    },
    {
      "epoch": 1.0713305898491083,
      "grad_norm": 2.2192482948303223,
      "learning_rate": 0.0009559156400185271,
      "loss": 2.5881,
      "step": 781
    },
    {
      "epoch": 1.0727023319615911,
      "grad_norm": 1.0046429634094238,
      "learning_rate": 0.0009555427919659469,
      "loss": 0.4091,
      "step": 782
    },
    {
      "epoch": 1.074074074074074,
      "grad_norm": 1.3024638891220093,
      "learning_rate": 0.0009551684557568748,
      "loss": 0.5982,
      "step": 783
    },
    {
      "epoch": 1.075445816186557,
      "grad_norm": 0.0879034623503685,
      "learning_rate": 0.0009547926327649534,
      "loss": 0.0227,
      "step": 784
    },
    {
      "epoch": 1.0768175582990398,
      "grad_norm": 1.1643495559692383,
      "learning_rate": 0.0009544153243692813,
      "loss": 0.6449,
      "step": 785
    },
    {
      "epoch": 1.0781893004115226,
      "grad_norm": 1.0957930088043213,
      "learning_rate": 0.0009540365319544075,
      "loss": 0.5379,
      "step": 786
    },
    {
      "epoch": 1.0795610425240054,
      "grad_norm": 1.4604713916778564,
      "learning_rate": 0.000953656256910327,
      "loss": 1.2099,
      "step": 787
    },
    {
      "epoch": 1.0809327846364885,
      "grad_norm": 1.1882197856903076,
      "learning_rate": 0.000953274500632475,
      "loss": 0.5034,
      "step": 788
    },
    {
      "epoch": 1.0823045267489713,
      "grad_norm": 1.0302724838256836,
      "learning_rate": 0.0009528912645217226,
      "loss": 0.4975,
      "step": 789
    },
    {
      "epoch": 1.083676268861454,
      "grad_norm": 0.8519008755683899,
      "learning_rate": 0.000952506549984371,
      "loss": 0.2687,
      "step": 790
    },
    {
      "epoch": 1.0850480109739369,
      "grad_norm": 1.4866607189178467,
      "learning_rate": 0.0009521203584321461,
      "loss": 0.9273,
      "step": 791
    },
    {
      "epoch": 1.0864197530864197,
      "grad_norm": 1.2431771755218506,
      "learning_rate": 0.0009517326912821946,
      "loss": 0.5536,
      "step": 792
    },
    {
      "epoch": 1.0877914951989025,
      "grad_norm": 0.11001141369342804,
      "learning_rate": 0.0009513435499570777,
      "loss": 0.0154,
      "step": 793
    },
    {
      "epoch": 1.0891632373113855,
      "grad_norm": 1.7603157758712769,
      "learning_rate": 0.0009509529358847655,
      "loss": 0.9278,
      "step": 794
    },
    {
      "epoch": 1.0905349794238683,
      "grad_norm": 1.3704885244369507,
      "learning_rate": 0.0009505608504986334,
      "loss": 0.5601,
      "step": 795
    },
    {
      "epoch": 1.0919067215363512,
      "grad_norm": 1.5036773681640625,
      "learning_rate": 0.0009501672952374551,
      "loss": 0.5198,
      "step": 796
    },
    {
      "epoch": 1.093278463648834,
      "grad_norm": 1.9286489486694336,
      "learning_rate": 0.0009497722715453985,
      "loss": 1.1777,
      "step": 797
    },
    {
      "epoch": 1.0946502057613168,
      "grad_norm": 1.4159934520721436,
      "learning_rate": 0.0009493757808720196,
      "loss": 0.8035,
      "step": 798
    },
    {
      "epoch": 1.0960219478737998,
      "grad_norm": 1.782509207725525,
      "learning_rate": 0.0009489778246722578,
      "loss": 1.0265,
      "step": 799
    },
    {
      "epoch": 1.0973936899862826,
      "grad_norm": 0.5141792297363281,
      "learning_rate": 0.0009485784044064304,
      "loss": 0.092,
      "step": 800
    },
    {
      "epoch": 1.0987654320987654,
      "grad_norm": 1.107999324798584,
      "learning_rate": 0.000948177521540227,
      "loss": 0.3972,
      "step": 801
    },
    {
      "epoch": 1.1001371742112482,
      "grad_norm": 1.4391822814941406,
      "learning_rate": 0.0009477751775447043,
      "loss": 0.7188,
      "step": 802
    },
    {
      "epoch": 1.101508916323731,
      "grad_norm": 0.36844563484191895,
      "learning_rate": 0.0009473713738962807,
      "loss": 0.0339,
      "step": 803
    },
    {
      "epoch": 1.102880658436214,
      "grad_norm": 0.1568753868341446,
      "learning_rate": 0.0009469661120767307,
      "loss": 0.0561,
      "step": 804
    },
    {
      "epoch": 1.1042524005486969,
      "grad_norm": 1.6684329509735107,
      "learning_rate": 0.0009465593935731799,
      "loss": 0.9105,
      "step": 805
    },
    {
      "epoch": 1.1056241426611797,
      "grad_norm": 1.4347095489501953,
      "learning_rate": 0.0009461512198780989,
      "loss": 1.0564,
      "step": 806
    },
    {
      "epoch": 1.1069958847736625,
      "grad_norm": 1.244839072227478,
      "learning_rate": 0.0009457415924892987,
      "loss": 0.4831,
      "step": 807
    },
    {
      "epoch": 1.1083676268861453,
      "grad_norm": 0.11359336227178574,
      "learning_rate": 0.0009453305129099241,
      "loss": 0.0249,
      "step": 808
    },
    {
      "epoch": 1.1097393689986284,
      "grad_norm": 1.7081202268600464,
      "learning_rate": 0.0009449179826484494,
      "loss": 1.028,
      "step": 809
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 1.129709243774414,
      "learning_rate": 0.0009445040032186714,
      "loss": 0.8074,
      "step": 810
    },
    {
      "epoch": 1.112482853223594,
      "grad_norm": 1.416955828666687,
      "learning_rate": 0.0009440885761397055,
      "loss": 0.7905,
      "step": 811
    },
    {
      "epoch": 1.1138545953360768,
      "grad_norm": 1.422277569770813,
      "learning_rate": 0.0009436717029359794,
      "loss": 0.6485,
      "step": 812
    },
    {
      "epoch": 1.1152263374485596,
      "grad_norm": 1.5454785823822021,
      "learning_rate": 0.0009432533851372263,
      "loss": 0.9483,
      "step": 813
    },
    {
      "epoch": 1.1165980795610426,
      "grad_norm": 1.5955668687820435,
      "learning_rate": 0.0009428336242784815,
      "loss": 0.7743,
      "step": 814
    },
    {
      "epoch": 1.1179698216735254,
      "grad_norm": 1.4997410774230957,
      "learning_rate": 0.0009424124219000753,
      "loss": 0.9345,
      "step": 815
    },
    {
      "epoch": 1.1193415637860082,
      "grad_norm": 1.551510214805603,
      "learning_rate": 0.0009419897795476276,
      "loss": 0.994,
      "step": 816
    },
    {
      "epoch": 1.120713305898491,
      "grad_norm": 1.4581773281097412,
      "learning_rate": 0.0009415656987720426,
      "loss": 0.763,
      "step": 817
    },
    {
      "epoch": 1.1220850480109739,
      "grad_norm": 0.4800634980201721,
      "learning_rate": 0.0009411401811295024,
      "loss": 0.0999,
      "step": 818
    },
    {
      "epoch": 1.123456790123457,
      "grad_norm": 1.3740246295928955,
      "learning_rate": 0.0009407132281814625,
      "loss": 0.866,
      "step": 819
    },
    {
      "epoch": 1.1248285322359397,
      "grad_norm": 0.04215388000011444,
      "learning_rate": 0.0009402848414946445,
      "loss": 0.0044,
      "step": 820
    },
    {
      "epoch": 1.1262002743484225,
      "grad_norm": 1.014538049697876,
      "learning_rate": 0.0009398550226410315,
      "loss": 0.4619,
      "step": 821
    },
    {
      "epoch": 1.1275720164609053,
      "grad_norm": 1.8286973237991333,
      "learning_rate": 0.0009394237731978624,
      "loss": 0.9776,
      "step": 822
    },
    {
      "epoch": 1.1289437585733881,
      "grad_norm": 1.2532325983047485,
      "learning_rate": 0.0009389910947476251,
      "loss": 0.5966,
      "step": 823
    },
    {
      "epoch": 1.130315500685871,
      "grad_norm": 0.9711207747459412,
      "learning_rate": 0.0009385569888780516,
      "loss": 0.3389,
      "step": 824
    },
    {
      "epoch": 1.131687242798354,
      "grad_norm": 1.085077166557312,
      "learning_rate": 0.0009381214571821118,
      "loss": 0.4415,
      "step": 825
    },
    {
      "epoch": 1.1330589849108368,
      "grad_norm": 1.7634426355361938,
      "learning_rate": 0.0009376845012580078,
      "loss": 1.0464,
      "step": 826
    },
    {
      "epoch": 1.1344307270233196,
      "grad_norm": 1.824653148651123,
      "learning_rate": 0.0009372461227091678,
      "loss": 1.2186,
      "step": 827
    },
    {
      "epoch": 1.1358024691358024,
      "grad_norm": 1.3176405429840088,
      "learning_rate": 0.0009368063231442405,
      "loss": 0.8022,
      "step": 828
    },
    {
      "epoch": 1.1371742112482854,
      "grad_norm": 1.768389105796814,
      "learning_rate": 0.0009363651041770892,
      "loss": 0.9649,
      "step": 829
    },
    {
      "epoch": 1.1385459533607682,
      "grad_norm": 1.4625869989395142,
      "learning_rate": 0.0009359224674267856,
      "loss": 0.7304,
      "step": 830
    },
    {
      "epoch": 1.139917695473251,
      "grad_norm": 1.2580175399780273,
      "learning_rate": 0.000935478414517604,
      "loss": 0.6431,
      "step": 831
    },
    {
      "epoch": 1.1412894375857339,
      "grad_norm": 1.6227208375930786,
      "learning_rate": 0.0009350329470790153,
      "loss": 1.0951,
      "step": 832
    },
    {
      "epoch": 1.1426611796982167,
      "grad_norm": 0.8904606699943542,
      "learning_rate": 0.0009345860667456814,
      "loss": 0.3192,
      "step": 833
    },
    {
      "epoch": 1.1440329218106995,
      "grad_norm": 1.1826539039611816,
      "learning_rate": 0.0009341377751574485,
      "loss": 0.5565,
      "step": 834
    },
    {
      "epoch": 1.1454046639231825,
      "grad_norm": 1.4180018901824951,
      "learning_rate": 0.0009336880739593415,
      "loss": 0.9978,
      "step": 835
    },
    {
      "epoch": 1.1467764060356653,
      "grad_norm": 0.8679835200309753,
      "learning_rate": 0.0009332369648015582,
      "loss": 0.6324,
      "step": 836
    },
    {
      "epoch": 1.1481481481481481,
      "grad_norm": 1.2708864212036133,
      "learning_rate": 0.0009327844493394632,
      "loss": 0.7331,
      "step": 837
    },
    {
      "epoch": 1.149519890260631,
      "grad_norm": 1.229516625404358,
      "learning_rate": 0.0009323305292335806,
      "loss": 0.5068,
      "step": 838
    },
    {
      "epoch": 1.1508916323731138,
      "grad_norm": 2.050509452819824,
      "learning_rate": 0.0009318752061495895,
      "loss": 1.303,
      "step": 839
    },
    {
      "epoch": 1.1522633744855968,
      "grad_norm": 1.1115632057189941,
      "learning_rate": 0.0009314184817583175,
      "loss": 0.4929,
      "step": 840
    },
    {
      "epoch": 1.1536351165980796,
      "grad_norm": 1.490692377090454,
      "learning_rate": 0.0009309603577357343,
      "loss": 0.7712,
      "step": 841
    },
    {
      "epoch": 1.1550068587105624,
      "grad_norm": 1.0925594568252563,
      "learning_rate": 0.0009305008357629452,
      "loss": 0.9389,
      "step": 842
    },
    {
      "epoch": 1.1563786008230452,
      "grad_norm": 0.39707836508750916,
      "learning_rate": 0.0009300399175261858,
      "loss": 0.0754,
      "step": 843
    },
    {
      "epoch": 1.157750342935528,
      "grad_norm": 1.3847075700759888,
      "learning_rate": 0.0009295776047168147,
      "loss": 0.7039,
      "step": 844
    },
    {
      "epoch": 1.159122085048011,
      "grad_norm": 0.7433251142501831,
      "learning_rate": 0.0009291138990313089,
      "loss": 0.3222,
      "step": 845
    },
    {
      "epoch": 1.1604938271604939,
      "grad_norm": 1.616162657737732,
      "learning_rate": 0.0009286488021712558,
      "loss": 1.268,
      "step": 846
    },
    {
      "epoch": 1.1618655692729767,
      "grad_norm": 1.4475303888320923,
      "learning_rate": 0.0009281823158433483,
      "loss": 0.7463,
      "step": 847
    },
    {
      "epoch": 1.1632373113854595,
      "grad_norm": 0.5591118931770325,
      "learning_rate": 0.0009277144417593776,
      "loss": 0.1299,
      "step": 848
    },
    {
      "epoch": 1.1646090534979423,
      "grad_norm": 1.0664170980453491,
      "learning_rate": 0.0009272451816362275,
      "loss": 0.485,
      "step": 849
    },
    {
      "epoch": 1.1659807956104253,
      "grad_norm": 1.5471608638763428,
      "learning_rate": 0.0009267745371958679,
      "loss": 0.8352,
      "step": 850
    },
    {
      "epoch": 1.1673525377229081,
      "grad_norm": 0.8688656687736511,
      "learning_rate": 0.0009263025101653488,
      "loss": 0.3851,
      "step": 851
    },
    {
      "epoch": 1.168724279835391,
      "grad_norm": 0.22162756323814392,
      "learning_rate": 0.0009258291022767931,
      "loss": 0.0474,
      "step": 852
    },
    {
      "epoch": 1.1700960219478738,
      "grad_norm": 0.2703648507595062,
      "learning_rate": 0.0009253543152673915,
      "loss": 0.0516,
      "step": 853
    },
    {
      "epoch": 1.1714677640603566,
      "grad_norm": 2.8573837280273438,
      "learning_rate": 0.0009248781508793949,
      "loss": 3.1361,
      "step": 854
    },
    {
      "epoch": 1.1728395061728394,
      "grad_norm": 2.351766586303711,
      "learning_rate": 0.0009244006108601086,
      "loss": 1.6477,
      "step": 855
    },
    {
      "epoch": 1.1742112482853224,
      "grad_norm": 1.4621500968933105,
      "learning_rate": 0.0009239216969618862,
      "loss": 1.0224,
      "step": 856
    },
    {
      "epoch": 1.1755829903978052,
      "grad_norm": 0.7923703789710999,
      "learning_rate": 0.0009234414109421227,
      "loss": 0.2316,
      "step": 857
    },
    {
      "epoch": 1.176954732510288,
      "grad_norm": 1.5916928052902222,
      "learning_rate": 0.0009229597545632475,
      "loss": 1.3128,
      "step": 858
    },
    {
      "epoch": 1.1783264746227708,
      "grad_norm": 1.5192334651947021,
      "learning_rate": 0.0009224767295927197,
      "loss": 0.8875,
      "step": 859
    },
    {
      "epoch": 1.1796982167352539,
      "grad_norm": 1.1825644969940186,
      "learning_rate": 0.0009219923378030198,
      "loss": 0.5464,
      "step": 860
    },
    {
      "epoch": 1.1810699588477367,
      "grad_norm": 1.6619049310684204,
      "learning_rate": 0.0009215065809716435,
      "loss": 1.0535,
      "step": 861
    },
    {
      "epoch": 1.1824417009602195,
      "grad_norm": 1.2359375953674316,
      "learning_rate": 0.0009210194608810971,
      "loss": 0.6007,
      "step": 862
    },
    {
      "epoch": 1.1838134430727023,
      "grad_norm": 1.2766097784042358,
      "learning_rate": 0.0009205309793188875,
      "loss": 0.6007,
      "step": 863
    },
    {
      "epoch": 1.1851851851851851,
      "grad_norm": 0.8956004977226257,
      "learning_rate": 0.0009200411380775192,
      "loss": 0.5336,
      "step": 864
    },
    {
      "epoch": 1.186556927297668,
      "grad_norm": 1.2947998046875,
      "learning_rate": 0.000919549938954485,
      "loss": 0.5458,
      "step": 865
    },
    {
      "epoch": 1.187928669410151,
      "grad_norm": 2.1609323024749756,
      "learning_rate": 0.0009190573837522612,
      "loss": 2.6501,
      "step": 866
    },
    {
      "epoch": 1.1893004115226338,
      "grad_norm": 1.3712961673736572,
      "learning_rate": 0.0009185634742782999,
      "loss": 0.8157,
      "step": 867
    },
    {
      "epoch": 1.1906721536351166,
      "grad_norm": 0.14107005298137665,
      "learning_rate": 0.000918068212345023,
      "loss": 0.0466,
      "step": 868
    },
    {
      "epoch": 1.1920438957475994,
      "grad_norm": 1.436710238456726,
      "learning_rate": 0.0009175715997698149,
      "loss": 1.2872,
      "step": 869
    },
    {
      "epoch": 1.1934156378600824,
      "grad_norm": 0.8854731321334839,
      "learning_rate": 0.000917073638375017,
      "loss": 0.4272,
      "step": 870
    },
    {
      "epoch": 1.1947873799725652,
      "grad_norm": 0.01813948154449463,
      "learning_rate": 0.0009165743299879194,
      "loss": 0.0017,
      "step": 871
    },
    {
      "epoch": 1.196159122085048,
      "grad_norm": 1.189801573753357,
      "learning_rate": 0.0009160736764407554,
      "loss": 0.452,
      "step": 872
    },
    {
      "epoch": 1.1975308641975309,
      "grad_norm": 1.2418094873428345,
      "learning_rate": 0.0009155716795706948,
      "loss": 0.5653,
      "step": 873
    },
    {
      "epoch": 1.1989026063100137,
      "grad_norm": 0.8849117755889893,
      "learning_rate": 0.0009150683412198361,
      "loss": 0.3144,
      "step": 874
    },
    {
      "epoch": 1.2002743484224965,
      "grad_norm": 1.9851653575897217,
      "learning_rate": 0.0009145636632352008,
      "loss": 2.2342,
      "step": 875
    },
    {
      "epoch": 1.2016460905349795,
      "grad_norm": 1.5564216375350952,
      "learning_rate": 0.0009140576474687263,
      "loss": 1.0388,
      "step": 876
    }
  ],
  "logging_steps": 1,
  "max_steps": 2187,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 219,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}