{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.992,
  "eval_steps": 400,
  "global_step": 936,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0064,
      "grad_norm": 0.23147642612457275,
      "learning_rate": 0.001997860962566845,
      "loss": 29.0583,
      "step": 2
    },
    {
      "epoch": 0.0128,
      "grad_norm": 1.181070327758789,
      "learning_rate": 0.001993582887700535,
      "loss": 28.9384,
      "step": 4
    },
    {
      "epoch": 0.0192,
      "grad_norm": 1.9193989038467407,
      "learning_rate": 0.0019893048128342247,
      "loss": 28.118,
      "step": 6
    },
    {
      "epoch": 0.0256,
      "grad_norm": 4.55759334564209,
      "learning_rate": 0.0019850267379679146,
      "loss": 28.2661,
      "step": 8
    },
    {
      "epoch": 0.032,
      "grad_norm": 6.3390374183654785,
      "learning_rate": 0.0019807486631016045,
      "loss": 27.4742,
      "step": 10
    },
    {
      "epoch": 0.0384,
      "grad_norm": 4.0472869873046875,
      "learning_rate": 0.0019764705882352944,
      "loss": 27.5481,
      "step": 12
    },
    {
      "epoch": 0.0448,
      "grad_norm": Infinity,
      "learning_rate": 0.0019764705882352944,
      "loss": 691.2838,
      "step": 14
    },
    {
      "epoch": 0.0512,
      "grad_norm": 596.6735229492188,
      "learning_rate": 0.001972192513368984,
      "loss": 614.0375,
      "step": 16
    },
    {
      "epoch": 0.0576,
      "grad_norm": 374.73675537109375,
      "learning_rate": 0.001967914438502674,
      "loss": 234.3992,
      "step": 18
    },
    {
      "epoch": 0.064,
      "grad_norm": 28.9688777923584,
      "learning_rate": 0.0019636363636363636,
      "loss": 41.9122,
      "step": 20
    },
    {
      "epoch": 0.0704,
      "grad_norm": 169.22000122070312,
      "learning_rate": 0.0019593582887700535,
      "loss": 41.3339,
      "step": 22
    },
    {
      "epoch": 0.0768,
      "grad_norm": 12.487041473388672,
      "learning_rate": 0.0019550802139037433,
      "loss": 30.189,
      "step": 24
    },
    {
      "epoch": 0.0832,
      "grad_norm": 8.636120796203613,
      "learning_rate": 0.001950802139037433,
      "loss": 27.9389,
      "step": 26
    },
    {
      "epoch": 0.0896,
      "grad_norm": 4.744749069213867,
      "learning_rate": 0.001946524064171123,
      "loss": 27.1649,
      "step": 28
    },
    {
      "epoch": 0.096,
      "grad_norm": 7.722327709197998,
      "learning_rate": 0.0019422459893048128,
      "loss": 27.3547,
      "step": 30
    },
    {
      "epoch": 0.1024,
      "grad_norm": 9.445578575134277,
      "learning_rate": 0.0019379679144385026,
      "loss": 26.9296,
      "step": 32
    },
    {
      "epoch": 0.1088,
      "grad_norm": 6.174092769622803,
      "learning_rate": 0.0019336898395721925,
      "loss": 27.1266,
      "step": 34
    },
    {
      "epoch": 0.1152,
      "grad_norm": 356.22625732421875,
      "learning_rate": 0.0019294117647058824,
      "loss": 45.6243,
      "step": 36
    },
    {
      "epoch": 0.1216,
      "grad_norm": 23.33014488220215,
      "learning_rate": 0.001925133689839572,
      "loss": 32.972,
      "step": 38
    },
    {
      "epoch": 0.128,
      "grad_norm": 10.953542709350586,
      "learning_rate": 0.0019208556149732622,
      "loss": 27.1236,
      "step": 40
    },
    {
      "epoch": 0.1344,
      "grad_norm": 6.473776340484619,
      "learning_rate": 0.0019165775401069518,
      "loss": 26.9502,
      "step": 42
    },
    {
      "epoch": 0.1408,
      "grad_norm": Infinity,
      "learning_rate": 0.001914438502673797,
      "loss": 44.8032,
      "step": 44
    },
    {
      "epoch": 0.1472,
      "grad_norm": 182.4479217529297,
      "learning_rate": 0.0019101604278074866,
      "loss": 56.3821,
      "step": 46
    },
    {
      "epoch": 0.1536,
      "grad_norm": 24.16608238220215,
      "learning_rate": 0.0019058823529411763,
      "loss": 31.7551,
      "step": 48
    },
    {
      "epoch": 0.16,
      "grad_norm": 13.420609474182129,
      "learning_rate": 0.0019016042780748664,
      "loss": 26.8265,
      "step": 50
    },
    {
      "epoch": 0.1664,
      "grad_norm": 6.592583179473877,
      "learning_rate": 0.001897326203208556,
      "loss": 26.9479,
      "step": 52
    },
    {
      "epoch": 0.1728,
      "grad_norm": 18.330659866333008,
      "learning_rate": 0.001893048128342246,
      "loss": 26.9247,
      "step": 54
    },
    {
      "epoch": 0.1792,
      "grad_norm": 10.113327980041504,
      "learning_rate": 0.0018887700534759358,
      "loss": 26.654,
      "step": 56
    },
    {
      "epoch": 0.1856,
      "grad_norm": Infinity,
      "learning_rate": 0.0018866310160427808,
      "loss": 158.5872,
      "step": 58
    },
    {
      "epoch": 0.192,
      "grad_norm": 714.9314575195312,
      "learning_rate": 0.0018823529411764706,
      "loss": 245.1731,
      "step": 60
    },
    {
      "epoch": 0.1984,
      "grad_norm": 20.074459075927734,
      "learning_rate": 0.0018780748663101605,
      "loss": 54.0979,
      "step": 62
    },
    {
      "epoch": 0.2048,
      "grad_norm": 6.04124641418457,
      "learning_rate": 0.0018737967914438502,
      "loss": 26.655,
      "step": 64
    },
    {
      "epoch": 0.2112,
      "grad_norm": 4.1345906257629395,
      "learning_rate": 0.0018695187165775403,
      "loss": 26.5224,
      "step": 66
    },
    {
      "epoch": 0.2176,
      "grad_norm": 5.253076553344727,
      "learning_rate": 0.00186524064171123,
      "loss": 26.2523,
      "step": 68
    },
    {
      "epoch": 0.224,
      "grad_norm": 4.040248394012451,
      "learning_rate": 0.00186096256684492,
      "loss": 26.5548,
      "step": 70
    },
    {
      "epoch": 0.2304,
      "grad_norm": 4.1401262283325195,
      "learning_rate": 0.0018566844919786097,
      "loss": 26.3328,
      "step": 72
    },
    {
      "epoch": 0.2368,
      "grad_norm": 4.546834945678711,
      "learning_rate": 0.0018524064171122994,
      "loss": 26.1653,
      "step": 74
    },
    {
      "epoch": 0.2432,
      "grad_norm": 3.73606276512146,
      "learning_rate": 0.0018481283422459895,
      "loss": 26.4023,
      "step": 76
    },
    {
      "epoch": 0.2496,
      "grad_norm": 3.841369390487671,
      "learning_rate": 0.0018438502673796791,
      "loss": 26.2402,
      "step": 78
    },
    {
      "epoch": 0.256,
      "grad_norm": 3.3648502826690674,
      "learning_rate": 0.001839572192513369,
      "loss": 26.143,
      "step": 80
    },
    {
      "epoch": 0.2624,
      "grad_norm": 3.7182576656341553,
      "learning_rate": 0.0018352941176470589,
      "loss": 26.5052,
      "step": 82
    },
    {
      "epoch": 0.2688,
      "grad_norm": 3.4634881019592285,
      "learning_rate": 0.0018310160427807488,
      "loss": 26.2055,
      "step": 84
    },
    {
      "epoch": 0.2752,
      "grad_norm": 3.5876669883728027,
      "learning_rate": 0.0018267379679144384,
      "loss": 26.3232,
      "step": 86
    },
    {
      "epoch": 0.2816,
      "grad_norm": 3.0892839431762695,
      "learning_rate": 0.0018224598930481285,
      "loss": 26.0274,
      "step": 88
    },
    {
      "epoch": 0.288,
      "grad_norm": 3.4667418003082275,
      "learning_rate": 0.0018181818181818182,
      "loss": 26.0062,
      "step": 90
    },
    {
      "epoch": 0.2944,
      "grad_norm": 3.052298069000244,
      "learning_rate": 0.001813903743315508,
      "loss": 26.2114,
      "step": 92
    },
    {
      "epoch": 0.3008,
      "grad_norm": 2.7124950885772705,
      "learning_rate": 0.001809625668449198,
      "loss": 25.9583,
      "step": 94
    },
    {
      "epoch": 0.3072,
      "grad_norm": 2.713899612426758,
      "learning_rate": 0.0018053475935828878,
      "loss": 26.0525,
      "step": 96
    },
    {
      "epoch": 0.3136,
      "grad_norm": 2.6182234287261963,
      "learning_rate": 0.0018010695187165775,
      "loss": 25.4642,
      "step": 98
    },
    {
      "epoch": 0.32,
      "grad_norm": 2.3585636615753174,
      "learning_rate": 0.0017967914438502676,
      "loss": 26.0951,
      "step": 100
    },
    {
      "epoch": 0.3264,
      "grad_norm": 2.5994300842285156,
      "learning_rate": 0.0017925133689839572,
      "loss": 26.1542,
      "step": 102
    },
    {
      "epoch": 0.3328,
      "grad_norm": 2.327500581741333,
      "learning_rate": 0.0017882352941176471,
      "loss": 26.0515,
      "step": 104
    },
    {
      "epoch": 0.3392,
      "grad_norm": 2.2460238933563232,
      "learning_rate": 0.001783957219251337,
      "loss": 25.722,
      "step": 106
    },
    {
      "epoch": 0.3456,
      "grad_norm": 2.0208640098571777,
      "learning_rate": 0.0017796791443850269,
      "loss": 25.8647,
      "step": 108
    },
    {
      "epoch": 0.352,
      "grad_norm": 2.1272950172424316,
      "learning_rate": 0.0017754010695187165,
      "loss": 25.8338,
      "step": 110
    },
    {
      "epoch": 0.3584,
      "grad_norm": 1.743446946144104,
      "learning_rate": 0.0017711229946524066,
      "loss": 25.6917,
      "step": 112
    },
    {
      "epoch": 0.3648,
      "grad_norm": 1.884177565574646,
      "learning_rate": 0.0017668449197860963,
      "loss": 25.492,
      "step": 114
    },
    {
      "epoch": 0.3712,
      "grad_norm": 1.5283397436141968,
      "learning_rate": 0.001762566844919786,
      "loss": 25.7483,
      "step": 116
    },
    {
      "epoch": 0.3776,
      "grad_norm": 1.838241457939148,
      "learning_rate": 0.001758288770053476,
      "loss": 26.1345,
      "step": 118
    },
    {
      "epoch": 0.384,
      "grad_norm": 1.7873716354370117,
      "learning_rate": 0.0017540106951871657,
      "loss": 25.5023,
      "step": 120
    },
    {
      "epoch": 0.3904,
      "grad_norm": 1.707553505897522,
      "learning_rate": 0.0017497326203208556,
      "loss": 25.6723,
      "step": 122
    },
    {
      "epoch": 0.3968,
      "grad_norm": 1.5225554704666138,
      "learning_rate": 0.0017454545454545455,
      "loss": 25.4451,
      "step": 124
    },
    {
      "epoch": 0.4032,
      "grad_norm": 1.5387077331542969,
      "learning_rate": 0.0017411764705882354,
      "loss": 24.9427,
      "step": 126
    },
    {
      "epoch": 0.4096,
      "grad_norm": 1.4218844175338745,
      "learning_rate": 0.001736898395721925,
      "loss": 25.2159,
      "step": 128
    },
    {
      "epoch": 0.416,
      "grad_norm": 1.4706506729125977,
      "learning_rate": 0.0017326203208556151,
      "loss": 25.5617,
      "step": 130
    },
    {
      "epoch": 0.4224,
      "grad_norm": 1.6522170305252075,
      "learning_rate": 0.0017283422459893048,
      "loss": 25.4537,
      "step": 132
    },
    {
      "epoch": 0.4288,
      "grad_norm": 1.5547223091125488,
      "learning_rate": 0.0017240641711229947,
      "loss": 25.126,
      "step": 134
    },
    {
      "epoch": 0.4352,
      "grad_norm": 1.2542740106582642,
      "learning_rate": 0.0017197860962566845,
      "loss": 25.3198,
      "step": 136
    },
    {
      "epoch": 0.4416,
      "grad_norm": 1.3993158340454102,
      "learning_rate": 0.0017155080213903744,
      "loss": 25.4372,
      "step": 138
    },
    {
      "epoch": 0.448,
      "grad_norm": 1.3659427165985107,
      "learning_rate": 0.001711229946524064,
      "loss": 25.0459,
      "step": 140
    },
    {
      "epoch": 0.4544,
      "grad_norm": 1.124814748764038,
      "learning_rate": 0.0017069518716577542,
      "loss": 25.1648,
      "step": 142
    },
    {
      "epoch": 0.4608,
      "grad_norm": 1.0325303077697754,
      "learning_rate": 0.0017026737967914438,
      "loss": 25.4647,
      "step": 144
    },
    {
      "epoch": 0.4672,
      "grad_norm": 1.2640491724014282,
      "learning_rate": 0.0016983957219251337,
      "loss": 25.0149,
      "step": 146
    },
    {
      "epoch": 0.4736,
      "grad_norm": 1.0942180156707764,
      "learning_rate": 0.0016941176470588236,
      "loss": 25.0814,
      "step": 148
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.2712141275405884,
      "learning_rate": 0.0016898395721925135,
      "loss": 25.0341,
      "step": 150
    },
    {
      "epoch": 0.4864,
      "grad_norm": 1.2034132480621338,
      "learning_rate": 0.0016855614973262031,
      "loss": 25.0647,
      "step": 152
    },
    {
      "epoch": 0.4928,
      "grad_norm": 1.2348299026489258,
      "learning_rate": 0.0016812834224598932,
      "loss": 24.7844,
      "step": 154
    },
    {
      "epoch": 0.4992,
      "grad_norm": 1.691644549369812,
      "learning_rate": 0.001677005347593583,
      "loss": 24.7689,
      "step": 156
    },
    {
      "epoch": 0.5056,
      "grad_norm": 1.4753830432891846,
      "learning_rate": 0.0016727272727272726,
      "loss": 24.941,
      "step": 158
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.9928715229034424,
      "learning_rate": 0.0016684491978609627,
      "loss": 25.0275,
      "step": 160
    },
    {
      "epoch": 0.5184,
      "grad_norm": 1.1669100522994995,
      "learning_rate": 0.0016641711229946523,
      "loss": 24.9014,
      "step": 162
    },
    {
      "epoch": 0.5248,
      "grad_norm": 1.39118230342865,
      "learning_rate": 0.0016598930481283422,
      "loss": 24.6339,
      "step": 164
    },
    {
      "epoch": 0.5312,
      "grad_norm": 1.2533951997756958,
      "learning_rate": 0.001655614973262032,
      "loss": 24.6086,
      "step": 166
    },
    {
      "epoch": 0.5376,
      "grad_norm": 0.9174675941467285,
      "learning_rate": 0.001651336898395722,
      "loss": 25.0262,
      "step": 168
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.9981193542480469,
      "learning_rate": 0.0016470588235294116,
      "loss": 24.6074,
      "step": 170
    },
    {
      "epoch": 0.5504,
      "grad_norm": 1.4740686416625977,
      "learning_rate": 0.0016427807486631017,
      "loss": 24.5649,
      "step": 172
    },
    {
      "epoch": 0.5568,
      "grad_norm": 1.0988154411315918,
      "learning_rate": 0.0016385026737967914,
      "loss": 24.2709,
      "step": 174
    },
    {
      "epoch": 0.5632,
      "grad_norm": 1.1431447267532349,
      "learning_rate": 0.0016342245989304813,
      "loss": 24.7044,
      "step": 176
    },
    {
      "epoch": 0.5696,
      "grad_norm": 1.2967267036437988,
      "learning_rate": 0.0016299465240641711,
      "loss": 24.803,
      "step": 178
    },
    {
      "epoch": 0.576,
      "grad_norm": 1.1269863843917847,
      "learning_rate": 0.001625668449197861,
      "loss": 24.4945,
      "step": 180
    },
    {
      "epoch": 0.5824,
      "grad_norm": 0.9064795970916748,
      "learning_rate": 0.0016213903743315507,
      "loss": 24.5421,
      "step": 182
    },
    {
      "epoch": 0.5888,
      "grad_norm": 0.9420700669288635,
      "learning_rate": 0.0016171122994652408,
      "loss": 24.494,
      "step": 184
    },
    {
      "epoch": 0.5952,
      "grad_norm": 1.1436915397644043,
      "learning_rate": 0.0016128342245989304,
      "loss": 24.6417,
      "step": 186
    },
    {
      "epoch": 0.6016,
      "grad_norm": 0.9322279691696167,
      "learning_rate": 0.0016085561497326205,
      "loss": 24.6154,
      "step": 188
    },
    {
      "epoch": 0.608,
      "grad_norm": 1.0101568698883057,
      "learning_rate": 0.0016042780748663102,
      "loss": 24.4768,
      "step": 190
    },
    {
      "epoch": 0.6144,
      "grad_norm": 1.3874669075012207,
      "learning_rate": 0.0016,
      "loss": 24.6752,
      "step": 192
    },
    {
      "epoch": 0.6208,
      "grad_norm": 1.0779438018798828,
      "learning_rate": 0.00159572192513369,
      "loss": 24.7572,
      "step": 194
    },
    {
      "epoch": 0.6272,
      "grad_norm": 1.3621560335159302,
      "learning_rate": 0.0015914438502673798,
      "loss": 24.5775,
      "step": 196
    },
    {
      "epoch": 0.6336,
      "grad_norm": 1.291939616203308,
      "learning_rate": 0.0015871657754010695,
      "loss": 24.4171,
      "step": 198
    },
    {
      "epoch": 0.64,
      "grad_norm": 19.172954559326172,
      "learning_rate": 0.0015828877005347596,
      "loss": 30.6558,
      "step": 200
    },
    {
      "epoch": 0.6464,
      "grad_norm": 4.52994441986084,
      "learning_rate": 0.0015786096256684493,
      "loss": 25.1351,
      "step": 202
    },
    {
      "epoch": 0.6528,
      "grad_norm": 3.1266472339630127,
      "learning_rate": 0.001574331550802139,
      "loss": 24.8543,
      "step": 204
    },
    {
      "epoch": 0.6592,
      "grad_norm": 2.4678471088409424,
      "learning_rate": 0.001570053475935829,
      "loss": 24.716,
      "step": 206
    },
    {
      "epoch": 0.6656,
      "grad_norm": 3.5065557956695557,
      "learning_rate": 0.0015657754010695187,
      "loss": 24.6532,
      "step": 208
    },
    {
      "epoch": 0.672,
      "grad_norm": 2.501286745071411,
      "learning_rate": 0.0015614973262032086,
      "loss": 24.7766,
      "step": 210
    },
    {
      "epoch": 0.6784,
      "grad_norm": 1.9249541759490967,
      "learning_rate": 0.0015572192513368984,
      "loss": 24.2869,
      "step": 212
    },
    {
      "epoch": 0.6848,
      "grad_norm": 3.193232297897339,
      "learning_rate": 0.0015529411764705883,
      "loss": 24.7769,
      "step": 214
    },
    {
      "epoch": 0.6912,
      "grad_norm": 2.367741823196411,
      "learning_rate": 0.001548663101604278,
      "loss": 24.7073,
      "step": 216
    },
    {
      "epoch": 0.6976,
      "grad_norm": 1.5215927362442017,
      "learning_rate": 0.001544385026737968,
      "loss": 24.8652,
      "step": 218
    },
    {
      "epoch": 0.704,
      "grad_norm": 1.8906418085098267,
      "learning_rate": 0.0015401069518716577,
      "loss": 24.7487,
      "step": 220
    },
    {
      "epoch": 0.7104,
      "grad_norm": 1.5913176536560059,
      "learning_rate": 0.0015358288770053476,
      "loss": 25.017,
      "step": 222
    },
    {
      "epoch": 0.7168,
      "grad_norm": 1.6209039688110352,
      "learning_rate": 0.0015315508021390375,
      "loss": 24.339,
      "step": 224
    },
    {
      "epoch": 0.7232,
      "grad_norm": 1.5647910833358765,
      "learning_rate": 0.0015272727272727274,
      "loss": 24.1431,
      "step": 226
    },
    {
      "epoch": 0.7296,
      "grad_norm": 1.111008882522583,
      "learning_rate": 0.001522994652406417,
      "loss": 24.3126,
      "step": 228
    },
    {
      "epoch": 0.736,
      "grad_norm": 1.259020209312439,
      "learning_rate": 0.0015187165775401071,
      "loss": 24.1569,
      "step": 230
    },
    {
      "epoch": 0.7424,
      "grad_norm": 1.2711575031280518,
      "learning_rate": 0.0015144385026737968,
      "loss": 24.4551,
      "step": 232
    },
    {
      "epoch": 0.7488,
      "grad_norm": 1.1747275590896606,
      "learning_rate": 0.0015101604278074867,
      "loss": 24.2954,
      "step": 234
    },
    {
      "epoch": 0.7552,
      "grad_norm": 1.3041143417358398,
      "learning_rate": 0.0015058823529411766,
      "loss": 24.3246,
      "step": 236
    },
    {
      "epoch": 0.7616,
      "grad_norm": 1.2481331825256348,
      "learning_rate": 0.0015016042780748664,
      "loss": 24.3238,
      "step": 238
    },
    {
      "epoch": 0.768,
      "grad_norm": 1.1186060905456543,
      "learning_rate": 0.001497326203208556,
      "loss": 24.5642,
      "step": 240
    },
    {
      "epoch": 0.7744,
      "grad_norm": 1.3123741149902344,
      "learning_rate": 0.0014930481283422462,
      "loss": 24.5782,
      "step": 242
    },
    {
      "epoch": 0.7808,
      "grad_norm": 1.0362588167190552,
      "learning_rate": 0.0014887700534759359,
      "loss": 24.519,
      "step": 244
    },
    {
      "epoch": 0.7872,
      "grad_norm": 0.9986498355865479,
      "learning_rate": 0.0014844919786096255,
      "loss": 24.6729,
      "step": 246
    },
    {
      "epoch": 0.7936,
      "grad_norm": 1.0757042169570923,
      "learning_rate": 0.0014802139037433156,
      "loss": 24.2834,
      "step": 248
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.1017972230911255,
      "learning_rate": 0.0014759358288770053,
      "loss": 23.9501,
      "step": 250
    },
    {
      "epoch": 0.8064,
      "grad_norm": 1.1957108974456787,
      "learning_rate": 0.0014716577540106952,
      "loss": 24.0922,
      "step": 252
    },
    {
      "epoch": 0.8128,
      "grad_norm": 1.1655299663543701,
      "learning_rate": 0.001467379679144385,
      "loss": 24.3735,
      "step": 254
    },
    {
      "epoch": 0.8192,
      "grad_norm": 0.9843050837516785,
      "learning_rate": 0.001463101604278075,
      "loss": 24.1656,
      "step": 256
    },
    {
      "epoch": 0.8256,
      "grad_norm": 1.0374311208724976,
      "learning_rate": 0.0014588235294117646,
      "loss": 24.2207,
      "step": 258
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.9728087186813354,
      "learning_rate": 0.0014545454545454547,
      "loss": 24.1451,
      "step": 260
    },
    {
      "epoch": 0.8384,
      "grad_norm": 0.9153401255607605,
      "learning_rate": 0.0014502673796791443,
      "loss": 24.0903,
      "step": 262
    },
    {
      "epoch": 0.8448,
      "grad_norm": 1.1150263547897339,
      "learning_rate": 0.0014459893048128342,
      "loss": 24.375,
      "step": 264
    },
    {
      "epoch": 0.8512,
      "grad_norm": 1.2261894941329956,
      "learning_rate": 0.001441711229946524,
      "loss": 24.5365,
      "step": 266
    },
    {
      "epoch": 0.8576,
      "grad_norm": 0.8708057403564453,
      "learning_rate": 0.001437433155080214,
      "loss": 24.0552,
      "step": 268
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.8445040583610535,
      "learning_rate": 0.0014331550802139036,
      "loss": 24.2762,
      "step": 270
    },
    {
      "epoch": 0.8704,
      "grad_norm": 1.1743762493133545,
      "learning_rate": 0.0014288770053475937,
      "loss": 24.4677,
      "step": 272
    },
    {
      "epoch": 0.8768,
      "grad_norm": 1.5417506694793701,
      "learning_rate": 0.0014245989304812834,
      "loss": 24.3304,
      "step": 274
    },
    {
      "epoch": 0.8832,
      "grad_norm": 1.1955397129058838,
      "learning_rate": 0.0014203208556149733,
      "loss": 24.2557,
      "step": 276
    },
    {
      "epoch": 0.8896,
      "grad_norm": 0.8730282187461853,
      "learning_rate": 0.0014160427807486632,
      "loss": 24.2498,
      "step": 278
    },
    {
      "epoch": 0.896,
      "grad_norm": 1.2659244537353516,
      "learning_rate": 0.001411764705882353,
      "loss": 24.1083,
      "step": 280
    },
    {
      "epoch": 0.9024,
      "grad_norm": 1.3372520208358765,
      "learning_rate": 0.0014074866310160427,
      "loss": 24.3094,
      "step": 282
    },
    {
      "epoch": 0.9088,
      "grad_norm": 0.9774726033210754,
      "learning_rate": 0.0014032085561497328,
      "loss": 24.2246,
      "step": 284
    },
    {
      "epoch": 0.9152,
      "grad_norm": 1.2357187271118164,
      "learning_rate": 0.0013989304812834225,
      "loss": 23.8482,
      "step": 286
    },
    {
      "epoch": 0.9216,
      "grad_norm": 0.8730471730232239,
      "learning_rate": 0.0013946524064171121,
      "loss": 24.1488,
      "step": 288
    },
    {
      "epoch": 0.928,
      "grad_norm": 1.1759275197982788,
      "learning_rate": 0.0013903743315508022,
      "loss": 24.0689,
      "step": 290
    },
    {
      "epoch": 0.9344,
      "grad_norm": 1.0625526905059814,
      "learning_rate": 0.0013860962566844919,
      "loss": 23.9011,
      "step": 292
    },
    {
      "epoch": 0.9408,
      "grad_norm": 0.9794639945030212,
      "learning_rate": 0.0013818181818181818,
      "loss": 24.1742,
      "step": 294
    },
    {
      "epoch": 0.9472,
      "grad_norm": 0.8726012110710144,
      "learning_rate": 0.0013775401069518716,
      "loss": 24.2234,
      "step": 296
    },
    {
      "epoch": 0.9536,
      "grad_norm": 0.9825778007507324,
      "learning_rate": 0.0013732620320855615,
      "loss": 23.993,
      "step": 298
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.0842067003250122,
      "learning_rate": 0.0013689839572192512,
      "loss": 24.1026,
      "step": 300
    },
    {
      "epoch": 0.9664,
      "grad_norm": 0.9663395881652832,
      "learning_rate": 0.0013647058823529413,
      "loss": 24.1273,
      "step": 302
    },
    {
      "epoch": 0.9728,
      "grad_norm": 1.0321509838104248,
      "learning_rate": 0.001360427807486631,
      "loss": 24.0919,
      "step": 304
    },
    {
      "epoch": 0.9792,
      "grad_norm": 1.0301271677017212,
      "learning_rate": 0.001356149732620321,
      "loss": 23.9162,
      "step": 306
    },
    {
      "epoch": 0.9856,
      "grad_norm": 1.1476960182189941,
      "learning_rate": 0.0013518716577540107,
      "loss": 23.9016,
      "step": 308
    },
    {
      "epoch": 0.992,
      "grad_norm": 1.044398307800293,
      "learning_rate": 0.0013475935828877006,
      "loss": 24.11,
      "step": 310
    },
    {
      "epoch": 0.9984,
      "grad_norm": 1.3517370223999023,
      "learning_rate": 0.0013433155080213905,
      "loss": 23.9528,
      "step": 312
    },
    {
      "epoch": 1.0032,
      "grad_norm": 1.1732239723205566,
      "learning_rate": 0.0013390374331550803,
      "loss": 18.1742,
      "step": 314
    },
    {
      "epoch": 1.0096,
      "grad_norm": 1.0016353130340576,
      "learning_rate": 0.00133475935828877,
      "loss": 23.7102,
      "step": 316
    },
    {
      "epoch": 1.016,
      "grad_norm": 1.3539528846740723,
      "learning_rate": 0.00133048128342246,
      "loss": 23.8078,
      "step": 318
    },
    {
      "epoch": 1.0224,
      "grad_norm": 1.2378896474838257,
      "learning_rate": 0.0013262032085561498,
      "loss": 24.2912,
      "step": 320
    },
    {
      "epoch": 1.0288,
      "grad_norm": 1.118072748184204,
      "learning_rate": 0.0013219251336898396,
      "loss": 23.9082,
      "step": 322
    },
    {
      "epoch": 1.0352,
      "grad_norm": 1.4039278030395508,
      "learning_rate": 0.0013176470588235295,
      "loss": 24.1193,
      "step": 324
    },
    {
      "epoch": 1.0416,
      "grad_norm": 0.9324209094047546,
      "learning_rate": 0.0013133689839572194,
      "loss": 23.9758,
      "step": 326
    },
    {
      "epoch": 1.048,
      "grad_norm": 1.21006178855896,
      "learning_rate": 0.001309090909090909,
      "loss": 24.0922,
      "step": 328
    },
    {
      "epoch": 1.0544,
      "grad_norm": 1.083890438079834,
      "learning_rate": 0.0013048128342245992,
      "loss": 23.9699,
      "step": 330
    },
    {
      "epoch": 1.0608,
      "grad_norm": 0.8704414367675781,
      "learning_rate": 0.0013005347593582888,
      "loss": 23.6364,
      "step": 332
    },
    {
      "epoch": 1.0672,
      "grad_norm": 0.9826658964157104,
      "learning_rate": 0.0012962566844919785,
      "loss": 24.1282,
      "step": 334
    },
    {
      "epoch": 1.0735999999999999,
      "grad_norm": 1.005998969078064,
      "learning_rate": 0.0012919786096256686,
      "loss": 24.0106,
      "step": 336
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.9576805233955383,
      "learning_rate": 0.0012877005347593582,
      "loss": 23.7264,
      "step": 338
    },
    {
      "epoch": 1.0864,
      "grad_norm": 0.9244794845581055,
      "learning_rate": 0.0012834224598930481,
      "loss": 23.7184,
      "step": 340
    },
    {
      "epoch": 1.0928,
      "grad_norm": 0.8726948499679565,
      "learning_rate": 0.001279144385026738,
      "loss": 23.8847,
      "step": 342
    },
    {
      "epoch": 1.0992,
      "grad_norm": 1.184464931488037,
      "learning_rate": 0.0012748663101604279,
      "loss": 23.9564,
      "step": 344
    },
    {
      "epoch": 1.1056,
      "grad_norm": 1.0483590364456177,
      "learning_rate": 0.0012705882352941175,
      "loss": 24.0259,
      "step": 346
    },
    {
      "epoch": 1.112,
      "grad_norm": 0.85725998878479,
      "learning_rate": 0.0012663101604278076,
      "loss": 23.9331,
      "step": 348
    },
    {
      "epoch": 1.1184,
      "grad_norm": 0.9352357387542725,
      "learning_rate": 0.0012620320855614973,
      "loss": 23.9268,
      "step": 350
    },
    {
      "epoch": 1.1248,
      "grad_norm": 0.9234675765037537,
      "learning_rate": 0.0012577540106951872,
      "loss": 24.0202,
      "step": 352
    },
    {
      "epoch": 1.1312,
      "grad_norm": 0.8931582570075989,
      "learning_rate": 0.001253475935828877,
      "loss": 23.7643,
      "step": 354
    },
    {
      "epoch": 1.1376,
      "grad_norm": 0.9177488684654236,
      "learning_rate": 0.001249197860962567,
      "loss": 23.7432,
      "step": 356
    },
    {
      "epoch": 1.144,
      "grad_norm": 0.8955037593841553,
      "learning_rate": 0.0012449197860962566,
      "loss": 23.7637,
      "step": 358
    },
    {
      "epoch": 1.1504,
      "grad_norm": 1.0113520622253418,
      "learning_rate": 0.0012406417112299467,
      "loss": 23.931,
      "step": 360
    },
    {
      "epoch": 1.1568,
      "grad_norm": 0.8994941711425781,
      "learning_rate": 0.0012363636363636364,
      "loss": 23.7517,
      "step": 362
    },
    {
      "epoch": 1.1632,
      "grad_norm": 0.9318895936012268,
      "learning_rate": 0.0012320855614973262,
      "loss": 23.8983,
      "step": 364
    },
    {
      "epoch": 1.1696,
      "grad_norm": 1.0647019147872925,
      "learning_rate": 0.0012278074866310161,
      "loss": 23.8213,
      "step": 366
    },
    {
      "epoch": 1.176,
      "grad_norm": 1.132948637008667,
      "learning_rate": 0.001223529411764706,
      "loss": 23.6347,
      "step": 368
    },
    {
      "epoch": 1.1824,
      "grad_norm": 0.9466244578361511,
      "learning_rate": 0.0012192513368983957,
      "loss": 23.9028,
      "step": 370
    },
    {
      "epoch": 1.1888,
      "grad_norm": 0.9593258500099182,
      "learning_rate": 0.0012149732620320858,
      "loss": 23.98,
      "step": 372
    },
    {
      "epoch": 1.1952,
      "grad_norm": 1.0898252725601196,
      "learning_rate": 0.0012106951871657754,
      "loss": 23.5301,
      "step": 374
    },
    {
      "epoch": 1.2016,
      "grad_norm": 1.0035079717636108,
      "learning_rate": 0.001206417112299465,
      "loss": 23.6481,
      "step": 376
    },
    {
      "epoch": 1.208,
      "grad_norm": 0.890784502029419,
      "learning_rate": 0.0012021390374331552,
      "loss": 23.638,
      "step": 378
    },
    {
      "epoch": 1.2144,
      "grad_norm": 1.0010117292404175,
      "learning_rate": 0.0011978609625668448,
      "loss": 23.7085,
      "step": 380
    },
    {
      "epoch": 1.2208,
      "grad_norm": 0.92851322889328,
      "learning_rate": 0.0011935828877005347,
      "loss": 23.9561,
      "step": 382
    },
    {
      "epoch": 1.2272,
      "grad_norm": 0.966708242893219,
      "learning_rate": 0.0011893048128342246,
      "loss": 23.5308,
      "step": 384
    },
    {
      "epoch": 1.2336,
      "grad_norm": 0.9790504574775696,
      "learning_rate": 0.0011850267379679145,
      "loss": 23.8811,
      "step": 386
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.9370564818382263,
      "learning_rate": 0.0011807486631016041,
      "loss": 23.6945,
      "step": 388
    },
    {
      "epoch": 1.2464,
      "grad_norm": 1.0068321228027344,
      "learning_rate": 0.0011764705882352942,
      "loss": 23.7471,
      "step": 390
    },
    {
      "epoch": 1.2528000000000001,
      "grad_norm": 0.880582332611084,
      "learning_rate": 0.001172192513368984,
      "loss": 23.4096,
      "step": 392
    },
    {
      "epoch": 1.2591999999999999,
      "grad_norm": 0.8774146437644958,
      "learning_rate": 0.0011679144385026738,
      "loss": 23.7207,
      "step": 394
    },
    {
      "epoch": 1.2656,
      "grad_norm": 1.070694923400879,
      "learning_rate": 0.0011636363636363637,
      "loss": 23.6461,
      "step": 396
    },
    {
      "epoch": 1.272,
      "grad_norm": 1.0852477550506592,
      "learning_rate": 0.0011593582887700535,
      "loss": 23.8393,
      "step": 398
    },
    {
      "epoch": 1.2784,
      "grad_norm": 0.9843823313713074,
      "learning_rate": 0.0011550802139037432,
      "loss": 23.798,
      "step": 400
    },
    {
      "epoch": 1.2784,
      "eval_loss": 1.4875859022140503,
      "eval_runtime": 1157.3489,
      "eval_samples_per_second": 8.64,
      "eval_steps_per_second": 1.08,
      "step": 400
    },
    {
      "epoch": 1.2848,
      "grad_norm": 0.9516158103942871,
      "learning_rate": 0.0011508021390374333,
      "loss": 23.8632,
      "step": 402
    },
    {
      "epoch": 1.2912,
      "grad_norm": 0.9639537930488586,
      "learning_rate": 0.001146524064171123,
      "loss": 23.7238,
      "step": 404
    },
    {
      "epoch": 1.2976,
      "grad_norm": 1.018441081047058,
      "learning_rate": 0.0011422459893048128,
      "loss": 23.5542,
      "step": 406
    },
    {
      "epoch": 1.304,
      "grad_norm": 0.9791119694709778,
      "learning_rate": 0.0011379679144385027,
      "loss": 23.5262,
      "step": 408
    },
    {
      "epoch": 1.3104,
      "grad_norm": 0.9863831996917725,
      "learning_rate": 0.0011336898395721926,
      "loss": 23.763,
      "step": 410
    },
    {
      "epoch": 1.3168,
      "grad_norm": 0.9786751866340637,
      "learning_rate": 0.0011294117647058823,
      "loss": 23.6861,
      "step": 412
    },
    {
      "epoch": 1.3232,
      "grad_norm": 0.9191985130310059,
      "learning_rate": 0.0011251336898395724,
      "loss": 23.5714,
      "step": 414
    },
    {
      "epoch": 1.3296000000000001,
      "grad_norm": 1.1038320064544678,
      "learning_rate": 0.001120855614973262,
      "loss": 23.6797,
      "step": 416
    },
    {
      "epoch": 1.336,
      "grad_norm": 0.9039956331253052,
      "learning_rate": 0.0011165775401069521,
      "loss": 23.5835,
      "step": 418
    },
    {
      "epoch": 1.3424,
      "grad_norm": 0.9752773642539978,
      "learning_rate": 0.0011122994652406418,
      "loss": 23.8994,
      "step": 420
    },
    {
      "epoch": 1.3488,
      "grad_norm": 1.058692455291748,
      "learning_rate": 0.0011080213903743314,
      "loss": 23.7038,
      "step": 422
    },
    {
      "epoch": 1.3552,
      "grad_norm": 0.89438396692276,
      "learning_rate": 0.0011037433155080215,
      "loss": 23.6139,
      "step": 424
    },
    {
      "epoch": 1.3616,
      "grad_norm": 0.9645833373069763,
      "learning_rate": 0.0010994652406417112,
      "loss": 23.603,
      "step": 426
    },
    {
      "epoch": 1.3679999999999999,
      "grad_norm": 0.9102054834365845,
      "learning_rate": 0.001095187165775401,
      "loss": 23.6805,
      "step": 428
    },
    {
      "epoch": 1.3744,
      "grad_norm": 1.066123366355896,
      "learning_rate": 0.001090909090909091,
      "loss": 23.623,
      "step": 430
    },
    {
      "epoch": 1.3808,
      "grad_norm": 0.9730815291404724,
      "learning_rate": 0.0010866310160427808,
      "loss": 23.9918,
      "step": 432
    },
    {
      "epoch": 1.3872,
      "grad_norm": 0.953152596950531,
      "learning_rate": 0.0010823529411764705,
      "loss": 23.5223,
      "step": 434
    },
    {
      "epoch": 1.3936,
      "grad_norm": 0.9269129633903503,
      "learning_rate": 0.0010780748663101606,
      "loss": 23.5794,
      "step": 436
    },
    {
      "epoch": 1.4,
      "grad_norm": 1.0842361450195312,
      "learning_rate": 0.0010737967914438503,
      "loss": 23.0374,
      "step": 438
    },
    {
      "epoch": 1.4064,
      "grad_norm": 0.9895943999290466,
      "learning_rate": 0.0010695187165775401,
      "loss": 23.6018,
      "step": 440
    },
    {
      "epoch": 1.4128,
      "grad_norm": 0.9779914617538452,
      "learning_rate": 0.00106524064171123,
      "loss": 23.4881,
      "step": 442
    },
    {
      "epoch": 1.4192,
      "grad_norm": 0.9720066785812378,
      "learning_rate": 0.0010609625668449199,
      "loss": 23.7365,
      "step": 444
    },
    {
      "epoch": 1.4256,
      "grad_norm": 0.9314095973968506,
      "learning_rate": 0.0010566844919786096,
      "loss": 23.6696,
      "step": 446
    },
    {
      "epoch": 1.432,
      "grad_norm": 0.9558563232421875,
      "learning_rate": 0.0010524064171122996,
      "loss": 23.5277,
      "step": 448
    },
    {
      "epoch": 1.4384000000000001,
      "grad_norm": 0.9274935722351074,
      "learning_rate": 0.0010481283422459893,
      "loss": 23.8089,
      "step": 450
    },
    {
      "epoch": 1.4447999999999999,
      "grad_norm": 0.8989819884300232,
      "learning_rate": 0.0010438502673796792,
      "loss": 23.4861,
      "step": 452
    },
    {
      "epoch": 1.4512,
      "grad_norm": 0.8949527144432068,
      "learning_rate": 0.001039572192513369,
      "loss": 23.8258,
      "step": 454
    },
    {
      "epoch": 1.4576,
      "grad_norm": 1.0032631158828735,
      "learning_rate": 0.001035294117647059,
      "loss": 23.6332,
      "step": 456
    },
    {
      "epoch": 1.464,
      "grad_norm": 0.8726269006729126,
      "learning_rate": 0.0010310160427807486,
      "loss": 23.852,
      "step": 458
    },
    {
      "epoch": 1.4704,
      "grad_norm": 1.013187289237976,
      "learning_rate": 0.0010267379679144387,
      "loss": 23.6731,
      "step": 460
    },
    {
      "epoch": 1.4768,
      "grad_norm": 0.9792904853820801,
      "learning_rate": 0.0010224598930481284,
      "loss": 23.4724,
      "step": 462
    },
    {
      "epoch": 1.4832,
      "grad_norm": 0.9881017804145813,
      "learning_rate": 0.001018181818181818,
      "loss": 23.6523,
      "step": 464
    },
    {
      "epoch": 1.4896,
      "grad_norm": 1.0167256593704224,
      "learning_rate": 0.0010139037433155081,
      "loss": 23.7077,
      "step": 466
    },
    {
      "epoch": 1.496,
      "grad_norm": 1.028531551361084,
      "learning_rate": 0.0010096256684491978,
      "loss": 24.1109,
      "step": 468
    },
    {
      "epoch": 1.5024,
      "grad_norm": 0.9525850415229797,
      "learning_rate": 0.0010053475935828877,
      "loss": 23.9823,
      "step": 470
    },
    {
      "epoch": 1.5088,
      "grad_norm": 1.1063846349716187,
      "learning_rate": 0.0010010695187165776,
      "loss": 23.6939,
      "step": 472
    },
    {
      "epoch": 1.5152,
      "grad_norm": 1.023152232170105,
      "learning_rate": 0.0009967914438502674,
      "loss": 23.5762,
      "step": 474
    },
    {
      "epoch": 1.5215999999999998,
      "grad_norm": 0.9533156752586365,
      "learning_rate": 0.0009925133689839573,
      "loss": 23.8931,
      "step": 476
    },
    {
      "epoch": 1.528,
      "grad_norm": 1.01291823387146,
      "learning_rate": 0.0009882352941176472,
      "loss": 23.7593,
      "step": 478
    },
    {
      "epoch": 1.5344,
      "grad_norm": 0.967445969581604,
      "learning_rate": 0.000983957219251337,
      "loss": 23.5105,
      "step": 480
    },
    {
      "epoch": 1.5408,
      "grad_norm": 0.8999999165534973,
      "learning_rate": 0.0009796791443850267,
      "loss": 23.9054,
      "step": 482
    },
    {
      "epoch": 1.5472000000000001,
      "grad_norm": 1.0055955648422241,
      "learning_rate": 0.0009754010695187165,
      "loss": 23.6609,
      "step": 484
    },
    {
      "epoch": 1.5535999999999999,
      "grad_norm": 0.9917962551116943,
      "learning_rate": 0.0009711229946524064,
      "loss": 23.664,
      "step": 486
    },
    {
      "epoch": 1.56,
      "grad_norm": 1.0509769916534424,
      "learning_rate": 0.0009668449197860963,
      "loss": 23.5889,
      "step": 488
    },
    {
      "epoch": 1.5664,
      "grad_norm": 0.9780173301696777,
      "learning_rate": 0.000962566844919786,
      "loss": 23.3563,
      "step": 490
    },
    {
      "epoch": 1.5728,
      "grad_norm": 1.0623761415481567,
      "learning_rate": 0.0009582887700534759,
      "loss": 23.7324,
      "step": 492
    },
    {
      "epoch": 1.5792000000000002,
      "grad_norm": 0.9415885806083679,
      "learning_rate": 0.0009540106951871658,
      "loss": 23.6075,
      "step": 494
    },
    {
      "epoch": 1.5856,
      "grad_norm": 1.0558156967163086,
      "learning_rate": 0.0009497326203208556,
      "loss": 23.9617,
      "step": 496
    },
    {
      "epoch": 1.592,
      "grad_norm": 1.2639505863189697,
      "learning_rate": 0.0009454545454545454,
      "loss": 23.7581,
      "step": 498
    },
    {
      "epoch": 1.5984,
      "grad_norm": 0.9966816902160645,
      "learning_rate": 0.0009411764705882353,
      "loss": 23.4159,
      "step": 500
    },
    {
      "epoch": 1.6048,
      "grad_norm": 1.1803079843521118,
      "learning_rate": 0.0009368983957219251,
      "loss": 23.6882,
      "step": 502
    },
    {
      "epoch": 1.6112,
      "grad_norm": 1.068477749824524,
      "learning_rate": 0.000932620320855615,
      "loss": 23.3086,
      "step": 504
    },
    {
      "epoch": 1.6176,
      "grad_norm": 1.0496388673782349,
      "learning_rate": 0.0009283422459893048,
      "loss": 23.5714,
      "step": 506
    },
    {
      "epoch": 1.624,
      "grad_norm": 0.9868847727775574,
      "learning_rate": 0.0009240641711229947,
      "loss": 23.3231,
      "step": 508
    },
    {
      "epoch": 1.6303999999999998,
      "grad_norm": 1.0146859884262085,
      "learning_rate": 0.0009197860962566845,
      "loss": 23.474,
      "step": 510
    },
    {
      "epoch": 1.6368,
      "grad_norm": 1.3640587329864502,
      "learning_rate": 0.0009155080213903744,
      "loss": 23.728,
      "step": 512
    },
    {
      "epoch": 1.6432,
      "grad_norm": 1.0222820043563843,
      "learning_rate": 0.0009112299465240643,
      "loss": 23.8034,
      "step": 514
    },
    {
      "epoch": 1.6496,
      "grad_norm": 0.9659361243247986,
      "learning_rate": 0.000906951871657754,
      "loss": 23.454,
      "step": 516
    },
    {
      "epoch": 1.6560000000000001,
      "grad_norm": 1.0694700479507446,
      "learning_rate": 0.0009026737967914439,
      "loss": 23.7977,
      "step": 518
    },
    {
      "epoch": 1.6623999999999999,
      "grad_norm": 0.9259259700775146,
      "learning_rate": 0.0008983957219251338,
      "loss": 23.5328,
      "step": 520
    },
    {
      "epoch": 1.6688,
      "grad_norm": 1.1431095600128174,
      "learning_rate": 0.0008941176470588236,
      "loss": 23.5061,
      "step": 522
    },
    {
      "epoch": 1.6752,
      "grad_norm": 1.0091369152069092,
      "learning_rate": 0.0008898395721925134,
      "loss": 23.414,
      "step": 524
    },
    {
      "epoch": 1.6816,
      "grad_norm": 1.0047186613082886,
      "learning_rate": 0.0008855614973262033,
      "loss": 23.7231,
      "step": 526
    },
    {
      "epoch": 1.688,
      "grad_norm": 1.020326018333435,
      "learning_rate": 0.000881283422459893,
      "loss": 23.5656,
      "step": 528
    },
    {
      "epoch": 1.6944,
      "grad_norm": 1.2622140645980835,
      "learning_rate": 0.0008770053475935829,
      "loss": 23.1935,
      "step": 530
    },
    {
      "epoch": 1.7008,
      "grad_norm": 0.9633255004882812,
      "learning_rate": 0.0008727272727272727,
      "loss": 23.7261,
      "step": 532
    },
    {
      "epoch": 1.7072,
      "grad_norm": 0.9747726321220398,
      "learning_rate": 0.0008684491978609625,
      "loss": 23.3123,
      "step": 534
    },
    {
      "epoch": 1.7136,
      "grad_norm": 0.8444503545761108,
      "learning_rate": 0.0008641711229946524,
      "loss": 23.703,
      "step": 536
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.9087849855422974,
      "learning_rate": 0.0008598930481283423,
      "loss": 23.3917,
      "step": 538
    },
    {
      "epoch": 1.7264,
      "grad_norm": 0.9356536269187927,
      "learning_rate": 0.000855614973262032,
      "loss": 23.6751,
      "step": 540
    },
    {
      "epoch": 1.7328000000000001,
      "grad_norm": 0.9630663990974426,
      "learning_rate": 0.0008513368983957219,
      "loss": 23.4119,
      "step": 542
    },
    {
      "epoch": 1.7391999999999999,
      "grad_norm": 1.0205367803573608,
      "learning_rate": 0.0008470588235294118,
      "loss": 23.2226,
      "step": 544
    },
    {
      "epoch": 1.7456,
      "grad_norm": 1.0030864477157593,
      "learning_rate": 0.0008427807486631016,
      "loss": 23.6982,
      "step": 546
    },
    {
      "epoch": 1.752,
      "grad_norm": 0.9470692276954651,
      "learning_rate": 0.0008385026737967914,
      "loss": 23.6646,
      "step": 548
    },
    {
      "epoch": 1.7584,
      "grad_norm": 0.9366554617881775,
      "learning_rate": 0.0008342245989304813,
      "loss": 23.7817,
      "step": 550
    },
    {
      "epoch": 1.7648000000000001,
      "grad_norm": 0.9047606587409973,
      "learning_rate": 0.0008299465240641711,
      "loss": 23.5856,
      "step": 552
    },
    {
      "epoch": 1.7711999999999999,
      "grad_norm": 0.9083183407783508,
      "learning_rate": 0.000825668449197861,
      "loss": 23.694,
      "step": 554
    },
    {
      "epoch": 1.7776,
      "grad_norm": 0.9529908895492554,
      "learning_rate": 0.0008213903743315509,
      "loss": 23.9012,
      "step": 556
    },
    {
      "epoch": 1.784,
      "grad_norm": 0.9241888523101807,
      "learning_rate": 0.0008171122994652406,
      "loss": 23.6046,
      "step": 558
    },
    {
      "epoch": 1.7904,
      "grad_norm": 0.869388997554779,
      "learning_rate": 0.0008128342245989305,
      "loss": 23.4944,
      "step": 560
    },
    {
      "epoch": 1.7968,
      "grad_norm": 0.8648447394371033,
      "learning_rate": 0.0008085561497326204,
      "loss": 23.2361,
      "step": 562
    },
    {
      "epoch": 1.8032,
      "grad_norm": 1.0052869319915771,
      "learning_rate": 0.0008042780748663103,
      "loss": 23.2241,
      "step": 564
    },
    {
      "epoch": 1.8096,
      "grad_norm": 0.9710979461669922,
      "learning_rate": 0.0008,
      "loss": 23.3083,
      "step": 566
    },
    {
      "epoch": 1.8159999999999998,
      "grad_norm": 0.9305456280708313,
      "learning_rate": 0.0007957219251336899,
      "loss": 23.5668,
      "step": 568
    },
    {
      "epoch": 1.8224,
      "grad_norm": 0.9786150455474854,
      "learning_rate": 0.0007914438502673798,
      "loss": 23.576,
      "step": 570
    },
    {
      "epoch": 1.8288,
      "grad_norm": 0.9177144169807434,
      "learning_rate": 0.0007871657754010695,
      "loss": 23.8692,
      "step": 572
    },
    {
      "epoch": 1.8352,
      "grad_norm": 0.8715292811393738,
      "learning_rate": 0.0007828877005347593,
      "loss": 23.5924,
      "step": 574
    },
    {
      "epoch": 1.8416000000000001,
      "grad_norm": 0.9923094511032104,
      "learning_rate": 0.0007786096256684492,
      "loss": 23.4065,
      "step": 576
    },
    {
      "epoch": 1.8479999999999999,
      "grad_norm": 0.931535542011261,
      "learning_rate": 0.000774331550802139,
      "loss": 23.6056,
      "step": 578
    },
    {
      "epoch": 1.8544,
      "grad_norm": 0.8508500456809998,
      "learning_rate": 0.0007700534759358289,
      "loss": 23.4128,
      "step": 580
    },
    {
      "epoch": 1.8608,
      "grad_norm": 0.916963517665863,
      "learning_rate": 0.0007657754010695187,
      "loss": 23.6378,
      "step": 582
    },
    {
      "epoch": 1.8672,
      "grad_norm": 0.8805645108222961,
      "learning_rate": 0.0007614973262032085,
      "loss": 23.2751,
      "step": 584
    },
    {
      "epoch": 1.8736000000000002,
      "grad_norm": 0.9679768085479736,
      "learning_rate": 0.0007572192513368984,
      "loss": 23.496,
      "step": 586
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.9578665494918823,
      "learning_rate": 0.0007529411764705883,
      "loss": 23.5754,
      "step": 588
    },
    {
      "epoch": 1.8864,
      "grad_norm": 0.9930956959724426,
      "learning_rate": 0.000748663101604278,
      "loss": 23.3563,
      "step": 590
    },
    {
      "epoch": 1.8928,
      "grad_norm": 0.8266828656196594,
      "learning_rate": 0.0007443850267379679,
      "loss": 23.7507,
      "step": 592
    },
    {
      "epoch": 1.8992,
      "grad_norm": 0.9037677049636841,
      "learning_rate": 0.0007401069518716578,
      "loss": 23.3083,
      "step": 594
    },
    {
      "epoch": 1.9056,
      "grad_norm": 0.9325649738311768,
      "learning_rate": 0.0007358288770053476,
      "loss": 23.4873,
      "step": 596
    },
    {
      "epoch": 1.912,
      "grad_norm": 0.8425143957138062,
      "learning_rate": 0.0007315508021390375,
      "loss": 23.2706,
      "step": 598
    },
    {
      "epoch": 1.9184,
      "grad_norm": 0.9997607469558716,
      "learning_rate": 0.0007272727272727273,
      "loss": 23.4578,
      "step": 600
    },
    {
      "epoch": 1.9247999999999998,
      "grad_norm": 0.9266635775566101,
      "learning_rate": 0.0007229946524064171,
      "loss": 23.3936,
      "step": 602
    },
    {
      "epoch": 1.9312,
      "grad_norm": 0.9372814893722534,
      "learning_rate": 0.000718716577540107,
      "loss": 23.7373,
      "step": 604
    },
    {
      "epoch": 1.9376,
      "grad_norm": 1.0302927494049072,
      "learning_rate": 0.0007144385026737969,
      "loss": 23.3622,
      "step": 606
    },
    {
      "epoch": 1.944,
      "grad_norm": 0.8410359025001526,
      "learning_rate": 0.0007101604278074866,
      "loss": 23.6175,
      "step": 608
    },
    {
      "epoch": 1.9504000000000001,
      "grad_norm": 1.002364993095398,
      "learning_rate": 0.0007058823529411765,
      "loss": 23.7065,
      "step": 610
    },
    {
      "epoch": 1.9567999999999999,
      "grad_norm": 0.8989892601966858,
      "learning_rate": 0.0007016042780748664,
      "loss": 23.1151,
      "step": 612
    },
    {
      "epoch": 1.9632,
      "grad_norm": 0.9592454433441162,
      "learning_rate": 0.0006973262032085561,
      "loss": 23.6266,
      "step": 614
    },
    {
      "epoch": 1.9696,
      "grad_norm": 0.9210177063941956,
      "learning_rate": 0.0006930481283422459,
      "loss": 23.2886,
      "step": 616
    },
    {
      "epoch": 1.976,
      "grad_norm": 1.1872823238372803,
      "learning_rate": 0.0006887700534759358,
      "loss": 23.495,
      "step": 618
    },
    {
      "epoch": 1.9824000000000002,
      "grad_norm": 0.9874787330627441,
      "learning_rate": 0.0006844919786096256,
      "loss": 23.6608,
      "step": 620
    },
    {
      "epoch": 1.9888,
      "grad_norm": 1.1812684535980225,
      "learning_rate": 0.0006802139037433155,
      "loss": 23.5925,
      "step": 622
    },
    {
      "epoch": 1.9952,
      "grad_norm": 1.0033948421478271,
      "learning_rate": 0.0006759358288770053,
      "loss": 23.4883,
      "step": 624
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.7030152082443237,
      "learning_rate": 0.0006716577540106952,
      "loss": 17.5334,
      "step": 626
    },
    {
      "epoch": 2.0064,
      "grad_norm": 1.017948865890503,
      "learning_rate": 0.000667379679144385,
      "loss": 23.2155,
      "step": 628
    },
    {
      "epoch": 2.0128,
      "grad_norm": 0.9184821248054504,
      "learning_rate": 0.0006631016042780749,
      "loss": 23.2131,
      "step": 630
    },
    {
      "epoch": 2.0192,
      "grad_norm": 1.0140761137008667,
      "learning_rate": 0.0006588235294117648,
      "loss": 23.0792,
      "step": 632
    },
    {
      "epoch": 2.0256,
      "grad_norm": 0.9820951223373413,
      "learning_rate": 0.0006545454545454545,
      "loss": 23.2678,
      "step": 634
    },
    {
      "epoch": 2.032,
      "grad_norm": 0.9167227745056152,
      "learning_rate": 0.0006502673796791444,
      "loss": 23.7916,
      "step": 636
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.935867428779602,
      "learning_rate": 0.0006459893048128343,
      "loss": 23.5265,
      "step": 638
    },
    {
      "epoch": 2.0448,
      "grad_norm": 0.871694803237915,
      "learning_rate": 0.0006417112299465241,
      "loss": 23.2606,
      "step": 640
    },
    {
      "epoch": 2.0512,
      "grad_norm": 0.9383721947669983,
      "learning_rate": 0.0006374331550802139,
      "loss": 23.4792,
      "step": 642
    },
    {
      "epoch": 2.0576,
      "grad_norm": 0.9656952023506165,
      "learning_rate": 0.0006331550802139038,
      "loss": 23.1145,
      "step": 644
    },
    {
      "epoch": 2.064,
      "grad_norm": 0.8393946886062622,
      "learning_rate": 0.0006288770053475936,
      "loss": 23.2965,
      "step": 646
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.8161150217056274,
      "learning_rate": 0.0006245989304812835,
      "loss": 23.4888,
      "step": 648
    },
    {
      "epoch": 2.0768,
      "grad_norm": 0.8854630589485168,
      "learning_rate": 0.0006203208556149733,
      "loss": 23.4233,
      "step": 650
    },
    {
      "epoch": 2.0832,
      "grad_norm": 0.9034832119941711,
      "learning_rate": 0.0006160427807486631,
      "loss": 23.1208,
      "step": 652
    },
    {
      "epoch": 2.0896,
      "grad_norm": 0.9552251696586609,
      "learning_rate": 0.000611764705882353,
      "loss": 23.2775,
      "step": 654
    },
    {
      "epoch": 2.096,
      "grad_norm": 0.9383313059806824,
      "learning_rate": 0.0006074866310160429,
      "loss": 23.096,
      "step": 656
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.9303053021430969,
      "learning_rate": 0.0006032085561497325,
      "loss": 23.6542,
      "step": 658
    },
    {
      "epoch": 2.1088,
      "grad_norm": 0.9048749804496765,
      "learning_rate": 0.0005989304812834224,
      "loss": 23.3699,
      "step": 660
    },
    {
      "epoch": 2.1152,
      "grad_norm": 0.9803421497344971,
      "learning_rate": 0.0005946524064171123,
      "loss": 23.3341,
      "step": 662
    },
    {
      "epoch": 2.1216,
      "grad_norm": 0.8741451501846313,
      "learning_rate": 0.0005903743315508021,
      "loss": 23.2353,
      "step": 664
    },
    {
      "epoch": 2.128,
      "grad_norm": 0.9050042629241943,
      "learning_rate": 0.000586096256684492,
      "loss": 23.7765,
      "step": 666
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.8833932280540466,
      "learning_rate": 0.0005818181818181818,
      "loss": 23.154,
      "step": 668
    },
    {
      "epoch": 2.1408,
      "grad_norm": 0.872748851776123,
      "learning_rate": 0.0005775401069518716,
      "loss": 23.5524,
      "step": 670
    },
    {
      "epoch": 2.1471999999999998,
      "grad_norm": 0.9564759731292725,
      "learning_rate": 0.0005732620320855615,
      "loss": 23.3869,
      "step": 672
    },
    {
      "epoch": 2.1536,
      "grad_norm": 0.9444039463996887,
      "learning_rate": 0.0005689839572192514,
      "loss": 23.5905,
      "step": 674
    },
    {
      "epoch": 2.16,
      "grad_norm": 1.1383410692214966,
      "learning_rate": 0.0005647058823529411,
      "loss": 23.1071,
      "step": 676
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.8750917911529541,
      "learning_rate": 0.000560427807486631,
      "loss": 23.5244,
      "step": 678
    },
    {
      "epoch": 2.1728,
      "grad_norm": 0.8726853132247925,
      "learning_rate": 0.0005561497326203209,
      "loss": 23.0605,
      "step": 680
    },
    {
      "epoch": 2.1792,
      "grad_norm": 0.9074258208274841,
      "learning_rate": 0.0005518716577540108,
      "loss": 23.243,
      "step": 682
    },
    {
      "epoch": 2.1856,
      "grad_norm": 0.9091795086860657,
      "learning_rate": 0.0005475935828877005,
      "loss": 23.3617,
      "step": 684
    },
    {
      "epoch": 2.192,
      "grad_norm": 0.909899115562439,
      "learning_rate": 0.0005433155080213904,
      "loss": 23.1892,
      "step": 686
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.9476255774497986,
      "learning_rate": 0.0005390374331550803,
      "loss": 23.3369,
      "step": 688
    },
    {
      "epoch": 2.2048,
      "grad_norm": 0.9601420760154724,
      "learning_rate": 0.0005347593582887701,
      "loss": 23.5679,
      "step": 690
    },
    {
      "epoch": 2.2112,
      "grad_norm": 0.9360019564628601,
      "learning_rate": 0.0005304812834224599,
      "loss": 23.007,
      "step": 692
    },
    {
      "epoch": 2.2176,
      "grad_norm": 0.9346802830696106,
      "learning_rate": 0.0005262032085561498,
      "loss": 23.0159,
      "step": 694
    },
    {
      "epoch": 2.224,
      "grad_norm": 0.9711213111877441,
      "learning_rate": 0.0005219251336898396,
      "loss": 23.3363,
      "step": 696
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.9546957612037659,
      "learning_rate": 0.0005176470588235295,
      "loss": 23.1201,
      "step": 698
    },
    {
      "epoch": 2.2368,
      "grad_norm": 0.916655957698822,
      "learning_rate": 0.0005133689839572194,
      "loss": 23.2946,
      "step": 700
    },
    {
      "epoch": 2.2432,
      "grad_norm": 0.8734573125839233,
      "learning_rate": 0.000509090909090909,
      "loss": 23.2751,
      "step": 702
    },
    {
      "epoch": 2.2496,
      "grad_norm": 0.9505683779716492,
      "learning_rate": 0.0005048128342245989,
      "loss": 23.3028,
      "step": 704
    },
    {
      "epoch": 2.2560000000000002,
      "grad_norm": 0.934208333492279,
      "learning_rate": 0.0005005347593582888,
      "loss": 23.2606,
      "step": 706
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.8738580942153931,
      "learning_rate": 0.0004962566844919787,
      "loss": 23.0636,
      "step": 708
    },
    {
      "epoch": 2.2688,
      "grad_norm": 0.9497429132461548,
      "learning_rate": 0.0004919786096256685,
      "loss": 23.3791,
      "step": 710
    },
    {
      "epoch": 2.2752,
      "grad_norm": 0.9983781576156616,
      "learning_rate": 0.00048770053475935825,
      "loss": 23.4518,
      "step": 712
    },
    {
      "epoch": 2.2816,
      "grad_norm": 0.9503285884857178,
      "learning_rate": 0.00048342245989304813,
      "loss": 23.2358,
      "step": 714
    },
    {
      "epoch": 2.288,
      "grad_norm": 0.9116323590278625,
      "learning_rate": 0.00047914438502673796,
      "loss": 23.0996,
      "step": 716
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.8628079295158386,
      "learning_rate": 0.0004748663101604278,
      "loss": 23.5187,
      "step": 718
    },
    {
      "epoch": 2.3008,
      "grad_norm": 0.9411908984184265,
      "learning_rate": 0.00047058823529411766,
      "loss": 23.5873,
      "step": 720
    },
    {
      "epoch": 2.3072,
      "grad_norm": 0.9314679503440857,
      "learning_rate": 0.0004663101604278075,
      "loss": 23.3654,
      "step": 722
    },
    {
      "epoch": 2.3136,
      "grad_norm": 0.9462237358093262,
      "learning_rate": 0.00046203208556149736,
      "loss": 23.0101,
      "step": 724
    },
    {
      "epoch": 2.32,
      "grad_norm": 0.889033317565918,
      "learning_rate": 0.0004577540106951872,
      "loss": 23.4283,
      "step": 726
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.9167575836181641,
      "learning_rate": 0.000453475935828877,
      "loss": 23.2583,
      "step": 728
    },
    {
      "epoch": 2.3327999999999998,
      "grad_norm": 0.9349282383918762,
      "learning_rate": 0.0004491978609625669,
      "loss": 23.463,
      "step": 730
    },
    {
      "epoch": 2.3392,
      "grad_norm": 1.114784836769104,
      "learning_rate": 0.0004449197860962567,
      "loss": 23.2864,
      "step": 732
    },
    {
      "epoch": 2.3456,
      "grad_norm": 0.9500236511230469,
      "learning_rate": 0.0004406417112299465,
      "loss": 23.4031,
      "step": 734
    },
    {
      "epoch": 2.352,
      "grad_norm": 1.0214568376541138,
      "learning_rate": 0.00043636363636363637,
      "loss": 23.2233,
      "step": 736
    },
    {
      "epoch": 2.3584,
      "grad_norm": 1.118489146232605,
      "learning_rate": 0.0004320855614973262,
      "loss": 23.7254,
      "step": 738
    },
    {
      "epoch": 2.3648,
      "grad_norm": 0.9648391008377075,
      "learning_rate": 0.000427807486631016,
      "loss": 23.3774,
      "step": 740
    },
    {
      "epoch": 2.3712,
      "grad_norm": 0.9128082394599915,
      "learning_rate": 0.0004235294117647059,
      "loss": 23.5378,
      "step": 742
    },
    {
      "epoch": 2.3776,
      "grad_norm": 0.9346900582313538,
      "learning_rate": 0.0004192513368983957,
      "loss": 23.6871,
      "step": 744
    },
    {
      "epoch": 2.384,
      "grad_norm": 1.0307550430297852,
      "learning_rate": 0.00041497326203208555,
      "loss": 23.2258,
      "step": 746
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.9579919576644897,
      "learning_rate": 0.00041069518716577543,
      "loss": 23.269,
      "step": 748
    },
    {
      "epoch": 2.3968,
      "grad_norm": 0.9140896201133728,
      "learning_rate": 0.00040641711229946525,
      "loss": 23.3352,
      "step": 750
    },
    {
      "epoch": 2.4032,
      "grad_norm": 0.9879317879676819,
      "learning_rate": 0.00040213903743315513,
      "loss": 23.2264,
      "step": 752
    },
    {
      "epoch": 2.4096,
      "grad_norm": 1.1229231357574463,
      "learning_rate": 0.00039786096256684496,
      "loss": 23.5893,
      "step": 754
    },
    {
      "epoch": 2.416,
      "grad_norm": 0.8806203007698059,
      "learning_rate": 0.00039358288770053473,
      "loss": 23.3295,
      "step": 756
    },
    {
      "epoch": 2.4224,
      "grad_norm": 1.0155560970306396,
      "learning_rate": 0.0003893048128342246,
      "loss": 23.0444,
      "step": 758
    },
    {
      "epoch": 2.4288,
      "grad_norm": 0.9045442342758179,
      "learning_rate": 0.00038502673796791443,
      "loss": 22.9648,
      "step": 760
    },
    {
      "epoch": 2.4352,
      "grad_norm": 0.9359453320503235,
      "learning_rate": 0.00038074866310160426,
      "loss": 23.6534,
      "step": 762
    },
    {
      "epoch": 2.4416,
      "grad_norm": 0.9333987236022949,
      "learning_rate": 0.00037647058823529414,
      "loss": 23.3959,
      "step": 764
    },
    {
      "epoch": 2.448,
      "grad_norm": 0.8483988046646118,
      "learning_rate": 0.00037219251336898396,
      "loss": 23.8575,
      "step": 766
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.8853796720504761,
      "learning_rate": 0.0003679144385026738,
      "loss": 23.0032,
      "step": 768
    },
    {
      "epoch": 2.4608,
      "grad_norm": 0.9364005327224731,
      "learning_rate": 0.00036363636363636367,
      "loss": 23.1803,
      "step": 770
    },
    {
      "epoch": 2.4672,
      "grad_norm": 0.820959210395813,
      "learning_rate": 0.0003593582887700535,
      "loss": 23.4326,
      "step": 772
    },
    {
      "epoch": 2.4736000000000002,
      "grad_norm": 0.927909255027771,
      "learning_rate": 0.0003550802139037433,
      "loss": 22.8314,
      "step": 774
    },
    {
      "epoch": 2.48,
      "grad_norm": 0.9819096922874451,
      "learning_rate": 0.0003508021390374332,
      "loss": 23.3604,
      "step": 776
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.8942359089851379,
      "learning_rate": 0.00034652406417112297,
      "loss": 23.3988,
      "step": 778
    },
    {
      "epoch": 2.4928,
      "grad_norm": 0.9608676433563232,
      "learning_rate": 0.0003422459893048128,
      "loss": 23.2326,
      "step": 780
    },
    {
      "epoch": 2.4992,
      "grad_norm": 0.8843042254447937,
      "learning_rate": 0.0003379679144385027,
      "loss": 23.3506,
      "step": 782
    },
    {
      "epoch": 2.5056000000000003,
      "grad_norm": 0.8857446312904358,
      "learning_rate": 0.0003336898395721925,
      "loss": 23.1861,
      "step": 784
    },
    {
      "epoch": 2.512,
      "grad_norm": 0.8458705544471741,
      "learning_rate": 0.0003294117647058824,
      "loss": 23.0167,
      "step": 786
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.8760762214660645,
      "learning_rate": 0.0003251336898395722,
      "loss": 23.5173,
      "step": 788
    },
    {
      "epoch": 2.5248,
      "grad_norm": 0.8493122458457947,
      "learning_rate": 0.00032085561497326203,
      "loss": 23.257,
      "step": 790
    },
    {
      "epoch": 2.5312,
      "grad_norm": 0.8905952572822571,
      "learning_rate": 0.0003165775401069519,
      "loss": 23.1194,
      "step": 792
    },
    {
      "epoch": 2.5376,
      "grad_norm": 0.9125280976295471,
      "learning_rate": 0.00031229946524064173,
      "loss": 23.0237,
      "step": 794
    },
    {
      "epoch": 2.544,
      "grad_norm": 0.9156554341316223,
      "learning_rate": 0.00030802139037433156,
      "loss": 23.4626,
      "step": 796
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.8901218771934509,
      "learning_rate": 0.00030374331550802144,
      "loss": 22.9936,
      "step": 798
    },
    {
      "epoch": 2.5568,
      "grad_norm": 0.8714930415153503,
      "learning_rate": 0.0002994652406417112,
      "loss": 23.1742,
      "step": 800
    },
    {
      "epoch": 2.5568,
      "eval_loss": 1.4636129140853882,
      "eval_runtime": 1115.7802,
      "eval_samples_per_second": 8.962,
      "eval_steps_per_second": 1.12,
      "step": 800
    },
    {
      "epoch": 2.5632,
      "grad_norm": 0.8894703984260559,
      "learning_rate": 0.00029518716577540103,
      "loss": 23.6523,
      "step": 802
    },
    {
      "epoch": 2.5696,
      "grad_norm": 0.8964853286743164,
      "learning_rate": 0.0002909090909090909,
      "loss": 22.9383,
      "step": 804
    },
    {
      "epoch": 2.576,
      "grad_norm": 0.8526540398597717,
      "learning_rate": 0.00028663101604278074,
      "loss": 23.1684,
      "step": 806
    },
    {
      "epoch": 2.5824,
      "grad_norm": 1.0182241201400757,
      "learning_rate": 0.00028235294117647056,
      "loss": 23.3383,
      "step": 808
    },
    {
      "epoch": 2.5888,
      "grad_norm": 0.9509109854698181,
      "learning_rate": 0.00027807486631016044,
      "loss": 23.4199,
      "step": 810
    },
    {
      "epoch": 2.5952,
      "grad_norm": 0.8805685639381409,
      "learning_rate": 0.00027379679144385027,
      "loss": 23.3654,
      "step": 812
    },
    {
      "epoch": 2.6016,
      "grad_norm": 0.9017741680145264,
      "learning_rate": 0.00026951871657754015,
      "loss": 23.2814,
      "step": 814
    },
    {
      "epoch": 2.608,
      "grad_norm": 0.844363808631897,
      "learning_rate": 0.00026524064171122997,
      "loss": 23.3663,
      "step": 816
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.868501603603363,
      "learning_rate": 0.0002609625668449198,
      "loss": 23.3927,
      "step": 818
    },
    {
      "epoch": 2.6208,
      "grad_norm": 0.965802788734436,
      "learning_rate": 0.0002566844919786097,
      "loss": 23.3886,
      "step": 820
    },
    {
      "epoch": 2.6272,
      "grad_norm": 0.9137208461761475,
      "learning_rate": 0.00025240641711229945,
      "loss": 23.2161,
      "step": 822
    },
    {
      "epoch": 2.6336,
      "grad_norm": 0.8800885081291199,
      "learning_rate": 0.00024812834224598933,
      "loss": 23.4479,
      "step": 824
    },
    {
      "epoch": 2.64,
      "grad_norm": 0.8783839344978333,
      "learning_rate": 0.00024385026737967913,
      "loss": 23.1788,
      "step": 826
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.8877150416374207,
      "learning_rate": 0.00023957219251336898,
      "loss": 23.3931,
      "step": 828
    },
    {
      "epoch": 2.6528,
      "grad_norm": 0.8638187050819397,
      "learning_rate": 0.00023529411764705883,
      "loss": 23.1699,
      "step": 830
    },
    {
      "epoch": 2.6592000000000002,
      "grad_norm": 0.8768156170845032,
      "learning_rate": 0.00023101604278074868,
      "loss": 23.2747,
      "step": 832
    },
    {
      "epoch": 2.6656,
      "grad_norm": 0.9169402718544006,
      "learning_rate": 0.0002267379679144385,
      "loss": 23.281,
      "step": 834
    },
    {
      "epoch": 2.672,
      "grad_norm": 0.978992223739624,
      "learning_rate": 0.00022245989304812836,
      "loss": 23.0263,
      "step": 836
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.9082220792770386,
      "learning_rate": 0.00021818181818181818,
      "loss": 23.1189,
      "step": 838
    },
    {
      "epoch": 2.6848,
      "grad_norm": 0.9290631413459778,
      "learning_rate": 0.000213903743315508,
      "loss": 23.6318,
      "step": 840
    },
    {
      "epoch": 2.6912000000000003,
      "grad_norm": 0.9007918834686279,
      "learning_rate": 0.00020962566844919786,
      "loss": 23.0228,
      "step": 842
    },
    {
      "epoch": 2.6976,
      "grad_norm": 0.9045092463493347,
      "learning_rate": 0.00020534759358288771,
      "loss": 23.1173,
      "step": 844
    },
    {
      "epoch": 2.7039999999999997,
      "grad_norm": 0.8459358215332031,
      "learning_rate": 0.00020106951871657757,
      "loss": 23.0842,
      "step": 846
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.8688234686851501,
      "learning_rate": 0.00019679144385026736,
      "loss": 23.1473,
      "step": 848
    },
    {
      "epoch": 2.7168,
      "grad_norm": 0.8705434203147888,
      "learning_rate": 0.00019251336898395722,
      "loss": 23.2901,
      "step": 850
    },
    {
      "epoch": 2.7232,
      "grad_norm": 0.9049879908561707,
      "learning_rate": 0.00018823529411764707,
      "loss": 23.2844,
      "step": 852
    },
    {
      "epoch": 2.7296,
      "grad_norm": 0.8677183985710144,
      "learning_rate": 0.0001839572192513369,
      "loss": 23.441,
      "step": 854
    },
    {
      "epoch": 2.7359999999999998,
      "grad_norm": 0.8370922207832336,
      "learning_rate": 0.00017967914438502675,
      "loss": 23.3627,
      "step": 856
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.827491819858551,
      "learning_rate": 0.0001754010695187166,
      "loss": 23.3567,
      "step": 858
    },
    {
      "epoch": 2.7488,
      "grad_norm": 0.8676392436027527,
      "learning_rate": 0.0001711229946524064,
      "loss": 23.5228,
      "step": 860
    },
    {
      "epoch": 2.7552,
      "grad_norm": 0.916731059551239,
      "learning_rate": 0.00016684491978609625,
      "loss": 23.4274,
      "step": 862
    },
    {
      "epoch": 2.7616,
      "grad_norm": 0.8114993572235107,
      "learning_rate": 0.0001625668449197861,
      "loss": 23.0131,
      "step": 864
    },
    {
      "epoch": 2.768,
      "grad_norm": 0.821692705154419,
      "learning_rate": 0.00015828877005347595,
      "loss": 22.9853,
      "step": 866
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.8291884064674377,
      "learning_rate": 0.00015401069518716578,
      "loss": 22.6708,
      "step": 868
    },
    {
      "epoch": 2.7808,
      "grad_norm": 0.8806215524673462,
      "learning_rate": 0.0001497326203208556,
      "loss": 23.079,
      "step": 870
    },
    {
      "epoch": 2.7872,
      "grad_norm": 0.8456304669380188,
      "learning_rate": 0.00014545454545454546,
      "loss": 23.3899,
      "step": 872
    },
    {
      "epoch": 2.7936,
      "grad_norm": 0.8849140405654907,
      "learning_rate": 0.00014117647058823528,
      "loss": 23.3624,
      "step": 874
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.8567163944244385,
      "learning_rate": 0.00013689839572192513,
      "loss": 23.2244,
      "step": 876
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.886084258556366,
      "learning_rate": 0.00013262032085561499,
      "loss": 23.275,
      "step": 878
    },
    {
      "epoch": 2.8128,
      "grad_norm": 0.8626444339752197,
      "learning_rate": 0.00012834224598930484,
      "loss": 23.3534,
      "step": 880
    },
    {
      "epoch": 2.8192,
      "grad_norm": 0.8719081282615662,
      "learning_rate": 0.00012406417112299466,
      "loss": 23.7974,
      "step": 882
    },
    {
      "epoch": 2.8256,
      "grad_norm": 0.8361597061157227,
      "learning_rate": 0.00011978609625668449,
      "loss": 23.2917,
      "step": 884
    },
    {
      "epoch": 2.832,
      "grad_norm": 0.9440580606460571,
      "learning_rate": 0.00011550802139037434,
      "loss": 23.2307,
      "step": 886
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.84004145860672,
      "learning_rate": 0.00011122994652406418,
      "loss": 23.135,
      "step": 888
    },
    {
      "epoch": 2.8448,
      "grad_norm": 0.836331307888031,
      "learning_rate": 0.000106951871657754,
      "loss": 23.327,
      "step": 890
    },
    {
      "epoch": 2.8512,
      "grad_norm": 0.9749852418899536,
      "learning_rate": 0.00010267379679144386,
      "loss": 23.0129,
      "step": 892
    },
    {
      "epoch": 2.8576,
      "grad_norm": 0.902911365032196,
      "learning_rate": 9.839572192513368e-05,
      "loss": 23.4656,
      "step": 894
    },
    {
      "epoch": 2.864,
      "grad_norm": 0.8584335446357727,
      "learning_rate": 9.411764705882353e-05,
      "loss": 22.9288,
      "step": 896
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.8301668167114258,
      "learning_rate": 8.983957219251337e-05,
      "loss": 23.4383,
      "step": 898
    },
    {
      "epoch": 2.8768000000000002,
      "grad_norm": 0.8426039814949036,
      "learning_rate": 8.55614973262032e-05,
      "loss": 23.2636,
      "step": 900
    },
    {
      "epoch": 2.8832,
      "grad_norm": 0.8241990208625793,
      "learning_rate": 8.128342245989305e-05,
      "loss": 23.4808,
      "step": 902
    },
    {
      "epoch": 2.8895999999999997,
      "grad_norm": 0.8555430769920349,
      "learning_rate": 7.700534759358289e-05,
      "loss": 22.9983,
      "step": 904
    },
    {
      "epoch": 2.896,
      "grad_norm": 0.8600494265556335,
      "learning_rate": 7.272727272727273e-05,
      "loss": 23.2584,
      "step": 906
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.9131312966346741,
      "learning_rate": 6.844919786096257e-05,
      "loss": 23.4513,
      "step": 908
    },
    {
      "epoch": 2.9088000000000003,
      "grad_norm": 0.8168982267379761,
      "learning_rate": 6.417112299465242e-05,
      "loss": 23.4352,
      "step": 910
    },
    {
      "epoch": 2.9152,
      "grad_norm": 0.8447926640510559,
      "learning_rate": 5.9893048128342244e-05,
      "loss": 23.2975,
      "step": 912
    },
    {
      "epoch": 2.9215999999999998,
      "grad_norm": 0.8652685284614563,
      "learning_rate": 5.561497326203209e-05,
      "loss": 23.5172,
      "step": 914
    },
    {
      "epoch": 2.928,
      "grad_norm": 0.8298205733299255,
      "learning_rate": 5.133689839572193e-05,
      "loss": 23.5154,
      "step": 916
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.8782302141189575,
      "learning_rate": 4.705882352941177e-05,
      "loss": 23.3597,
      "step": 918
    },
    {
      "epoch": 2.9408,
      "grad_norm": 0.9018718004226685,
      "learning_rate": 4.27807486631016e-05,
      "loss": 23.601,
      "step": 920
    },
    {
      "epoch": 2.9472,
      "grad_norm": 0.912187933921814,
      "learning_rate": 3.8502673796791445e-05,
      "loss": 23.4442,
      "step": 922
    },
    {
      "epoch": 2.9536,
      "grad_norm": 0.8485115170478821,
      "learning_rate": 3.4224598930481284e-05,
      "loss": 23.0539,
      "step": 924
    },
    {
      "epoch": 2.96,
      "grad_norm": 0.880998432636261,
      "learning_rate": 2.9946524064171122e-05,
      "loss": 23.1452,
      "step": 926
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.8453803658485413,
      "learning_rate": 2.5668449197860964e-05,
      "loss": 22.8998,
      "step": 928
    },
    {
      "epoch": 2.9728,
      "grad_norm": 0.8399450182914734,
      "learning_rate": 2.13903743315508e-05,
      "loss": 23.0928,
      "step": 930
    },
    {
      "epoch": 2.9792,
      "grad_norm": 0.8890286684036255,
      "learning_rate": 1.7112299465240642e-05,
      "loss": 23.0015,
      "step": 932
    },
    {
      "epoch": 2.9856,
      "grad_norm": 0.8455818295478821,
      "learning_rate": 1.2834224598930482e-05,
      "loss": 23.4294,
      "step": 934
    },
    {
      "epoch": 2.992,
      "grad_norm": 0.8911629915237427,
      "learning_rate": 8.556149732620321e-06,
      "loss": 23.0343,
      "step": 936
    },
    {
      "epoch": 2.992,
      "step": 936,
      "total_flos": 1.015865179865088e+19,
      "train_loss": 28.25554455243624,
      "train_runtime": 30178.7669,
      "train_samples_per_second": 7.953,
      "train_steps_per_second": 0.031
    },
    {
      "epoch": 2.992,
      "eval_loss": 1.4616048336029053,
      "eval_runtime": 1115.0715,
      "eval_samples_per_second": 8.968,
      "eval_steps_per_second": 1.121,
      "step": 936
    },
    {
      "epoch": 2.992,
      "eval_loss": 1.4569849967956543,
      "eval_runtime": 1130.2457,
      "eval_samples_per_second": 8.848,
      "eval_steps_per_second": 1.106,
      "step": 936
    }
  ],
  "logging_steps": 2,
  "max_steps": 936,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.015865179865088e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}