{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1956,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005112474437627812,
      "grad_norm": 3.3048973083496094,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 1.1371,
      "step": 1
    },
    {
      "epoch": 0.0010224948875255625,
      "grad_norm": 3.735586404800415,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.1928,
      "step": 2
    },
    {
      "epoch": 0.0015337423312883436,
      "grad_norm": 3.930663585662842,
      "learning_rate": 3e-06,
      "loss": 1.3484,
      "step": 3
    },
    {
      "epoch": 0.002044989775051125,
      "grad_norm": 4.375387668609619,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.2683,
      "step": 4
    },
    {
      "epoch": 0.002556237218813906,
      "grad_norm": 5.1499857902526855,
      "learning_rate": 5e-06,
      "loss": 1.1936,
      "step": 5
    },
    {
      "epoch": 0.003067484662576687,
      "grad_norm": 5.033441066741943,
      "learning_rate": 6e-06,
      "loss": 1.3051,
      "step": 6
    },
    {
      "epoch": 0.0035787321063394683,
      "grad_norm": 5.029082775115967,
      "learning_rate": 7.000000000000001e-06,
      "loss": 1.1804,
      "step": 7
    },
    {
      "epoch": 0.00408997955010225,
      "grad_norm": 4.745914459228516,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.2379,
      "step": 8
    },
    {
      "epoch": 0.004601226993865031,
      "grad_norm": 4.976576805114746,
      "learning_rate": 9e-06,
      "loss": 1.282,
      "step": 9
    },
    {
      "epoch": 0.005112474437627812,
      "grad_norm": 4.350550174713135,
      "learning_rate": 1e-05,
      "loss": 1.1822,
      "step": 10
    },
    {
      "epoch": 0.005623721881390593,
      "grad_norm": 4.9506940841674805,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 1.2666,
      "step": 11
    },
    {
      "epoch": 0.006134969325153374,
      "grad_norm": 5.068627834320068,
      "learning_rate": 1.2e-05,
      "loss": 1.2411,
      "step": 12
    },
    {
      "epoch": 0.0066462167689161555,
      "grad_norm": 5.628175735473633,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 1.2263,
      "step": 13
    },
    {
      "epoch": 0.007157464212678937,
      "grad_norm": 5.16189432144165,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 1.2809,
      "step": 14
    },
    {
      "epoch": 0.007668711656441718,
      "grad_norm": 4.856331825256348,
      "learning_rate": 1.5e-05,
      "loss": 1.1944,
      "step": 15
    },
    {
      "epoch": 0.0081799591002045,
      "grad_norm": 4.980719089508057,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.2265,
      "step": 16
    },
    {
      "epoch": 0.008691206543967281,
      "grad_norm": 5.2572832107543945,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 1.2408,
      "step": 17
    },
    {
      "epoch": 0.009202453987730062,
      "grad_norm": 5.128628730773926,
      "learning_rate": 1.8e-05,
      "loss": 1.2272,
      "step": 18
    },
    {
      "epoch": 0.009713701431492843,
      "grad_norm": 5.2293829917907715,
      "learning_rate": 1.9e-05,
      "loss": 1.195,
      "step": 19
    },
    {
      "epoch": 0.010224948875255624,
      "grad_norm": 4.794898986816406,
      "learning_rate": 2e-05,
      "loss": 1.2411,
      "step": 20
    },
    {
      "epoch": 0.010736196319018405,
      "grad_norm": 5.007977485656738,
      "learning_rate": 2.1e-05,
      "loss": 1.1821,
      "step": 21
    },
    {
      "epoch": 0.011247443762781187,
      "grad_norm": 4.727319240570068,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 1.188,
      "step": 22
    },
    {
      "epoch": 0.011758691206543968,
      "grad_norm": 4.606495380401611,
      "learning_rate": 2.3000000000000003e-05,
      "loss": 1.1372,
      "step": 23
    },
    {
      "epoch": 0.012269938650306749,
      "grad_norm": 4.6991753578186035,
      "learning_rate": 2.4e-05,
      "loss": 1.1395,
      "step": 24
    },
    {
      "epoch": 0.01278118609406953,
      "grad_norm": 4.635598659515381,
      "learning_rate": 2.5e-05,
      "loss": 1.1782,
      "step": 25
    },
    {
      "epoch": 0.013292433537832311,
      "grad_norm": 4.882657051086426,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 1.1939,
      "step": 26
    },
    {
      "epoch": 0.013803680981595092,
      "grad_norm": 5.13659143447876,
      "learning_rate": 2.7000000000000002e-05,
      "loss": 1.1567,
      "step": 27
    },
    {
      "epoch": 0.014314928425357873,
      "grad_norm": 5.102069854736328,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 1.2254,
      "step": 28
    },
    {
      "epoch": 0.014826175869120654,
      "grad_norm": 5.0762505531311035,
      "learning_rate": 2.9e-05,
      "loss": 1.1643,
      "step": 29
    },
    {
      "epoch": 0.015337423312883436,
      "grad_norm": 4.819282054901123,
      "learning_rate": 3e-05,
      "loss": 1.0718,
      "step": 30
    },
    {
      "epoch": 0.015848670756646217,
      "grad_norm": 4.949357032775879,
      "learning_rate": 3.1e-05,
      "loss": 1.091,
      "step": 31
    },
    {
      "epoch": 0.016359918200409,
      "grad_norm": 5.103873252868652,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 1.0602,
      "step": 32
    },
    {
      "epoch": 0.01687116564417178,
      "grad_norm": 4.818650722503662,
      "learning_rate": 3.3e-05,
      "loss": 0.9318,
      "step": 33
    },
    {
      "epoch": 0.017382413087934562,
      "grad_norm": 4.793953895568848,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 1.0556,
      "step": 34
    },
    {
      "epoch": 0.01789366053169734,
      "grad_norm": 5.540240287780762,
      "learning_rate": 3.5e-05,
      "loss": 1.1009,
      "step": 35
    },
    {
      "epoch": 0.018404907975460124,
      "grad_norm": 5.293509483337402,
      "learning_rate": 3.6e-05,
      "loss": 1.1285,
      "step": 36
    },
    {
      "epoch": 0.018916155419222903,
      "grad_norm": 5.887397289276123,
      "learning_rate": 3.7e-05,
      "loss": 1.0327,
      "step": 37
    },
    {
      "epoch": 0.019427402862985686,
      "grad_norm": 4.989732265472412,
      "learning_rate": 3.8e-05,
      "loss": 0.9327,
      "step": 38
    },
    {
      "epoch": 0.019938650306748466,
      "grad_norm": 5.251559257507324,
      "learning_rate": 3.9000000000000006e-05,
      "loss": 1.114,
      "step": 39
    },
    {
      "epoch": 0.02044989775051125,
      "grad_norm": 6.304392337799072,
      "learning_rate": 4e-05,
      "loss": 1.0943,
      "step": 40
    },
    {
      "epoch": 0.020961145194274028,
      "grad_norm": 5.590821266174316,
      "learning_rate": 4.1e-05,
      "loss": 1.0598,
      "step": 41
    },
    {
      "epoch": 0.02147239263803681,
      "grad_norm": 5.784496784210205,
      "learning_rate": 4.2e-05,
      "loss": 1.021,
      "step": 42
    },
    {
      "epoch": 0.02198364008179959,
      "grad_norm": 5.714466094970703,
      "learning_rate": 4.3e-05,
      "loss": 0.929,
      "step": 43
    },
    {
      "epoch": 0.022494887525562373,
      "grad_norm": 6.554811000823975,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 1.1903,
      "step": 44
    },
    {
      "epoch": 0.023006134969325152,
      "grad_norm": 6.44047212600708,
      "learning_rate": 4.5e-05,
      "loss": 0.8508,
      "step": 45
    },
    {
      "epoch": 0.023517382413087935,
      "grad_norm": 6.655950546264648,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.8196,
      "step": 46
    },
    {
      "epoch": 0.024028629856850715,
      "grad_norm": 6.728236198425293,
      "learning_rate": 4.7e-05,
      "loss": 0.9124,
      "step": 47
    },
    {
      "epoch": 0.024539877300613498,
      "grad_norm": 7.260775566101074,
      "learning_rate": 4.8e-05,
      "loss": 0.7433,
      "step": 48
    },
    {
      "epoch": 0.025051124744376277,
      "grad_norm": 8.426919937133789,
      "learning_rate": 4.9e-05,
      "loss": 0.902,
      "step": 49
    },
    {
      "epoch": 0.02556237218813906,
      "grad_norm": 14.32536506652832,
      "learning_rate": 5e-05,
      "loss": 0.8194,
      "step": 50
    },
    {
      "epoch": 0.02607361963190184,
      "grad_norm": 4.535055637359619,
      "learning_rate": 5.1000000000000006e-05,
      "loss": 1.0298,
      "step": 51
    },
    {
      "epoch": 0.026584867075664622,
      "grad_norm": 5.038204193115234,
      "learning_rate": 5.2000000000000004e-05,
      "loss": 1.0745,
      "step": 52
    },
    {
      "epoch": 0.0270961145194274,
      "grad_norm": 6.9380927085876465,
      "learning_rate": 5.300000000000001e-05,
      "loss": 1.2208,
      "step": 53
    },
    {
      "epoch": 0.027607361963190184,
      "grad_norm": 4.157510280609131,
      "learning_rate": 5.4000000000000005e-05,
      "loss": 1.135,
      "step": 54
    },
    {
      "epoch": 0.028118609406952964,
      "grad_norm": 3.202308416366577,
      "learning_rate": 5.500000000000001e-05,
      "loss": 1.0679,
      "step": 55
    },
    {
      "epoch": 0.028629856850715747,
      "grad_norm": 3.3872716426849365,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 1.0871,
      "step": 56
    },
    {
      "epoch": 0.029141104294478526,
      "grad_norm": 3.4161555767059326,
      "learning_rate": 5.6999999999999996e-05,
      "loss": 1.149,
      "step": 57
    },
    {
      "epoch": 0.02965235173824131,
      "grad_norm": 4.0091962814331055,
      "learning_rate": 5.8e-05,
      "loss": 1.096,
      "step": 58
    },
    {
      "epoch": 0.03016359918200409,
      "grad_norm": 3.4014410972595215,
      "learning_rate": 5.9e-05,
      "loss": 0.9183,
      "step": 59
    },
    {
      "epoch": 0.03067484662576687,
      "grad_norm": 3.9149327278137207,
      "learning_rate": 6e-05,
      "loss": 1.0704,
      "step": 60
    },
    {
      "epoch": 0.031186094069529654,
      "grad_norm": 4.084487438201904,
      "learning_rate": 6.1e-05,
      "loss": 1.0907,
      "step": 61
    },
    {
      "epoch": 0.03169734151329243,
      "grad_norm": 3.6023764610290527,
      "learning_rate": 6.2e-05,
      "loss": 0.9582,
      "step": 62
    },
    {
      "epoch": 0.032208588957055216,
      "grad_norm": 3.806703805923462,
      "learning_rate": 6.3e-05,
      "loss": 1.0072,
      "step": 63
    },
    {
      "epoch": 0.032719836400818,
      "grad_norm": 3.710463523864746,
      "learning_rate": 6.400000000000001e-05,
      "loss": 0.9751,
      "step": 64
    },
    {
      "epoch": 0.033231083844580775,
      "grad_norm": 3.8242218494415283,
      "learning_rate": 6.500000000000001e-05,
      "loss": 1.0477,
      "step": 65
    },
    {
      "epoch": 0.03374233128834356,
      "grad_norm": 4.025355815887451,
      "learning_rate": 6.6e-05,
      "loss": 1.0147,
      "step": 66
    },
    {
      "epoch": 0.03425357873210634,
      "grad_norm": 3.9239964485168457,
      "learning_rate": 6.7e-05,
      "loss": 1.1412,
      "step": 67
    },
    {
      "epoch": 0.034764826175869123,
      "grad_norm": 4.182487964630127,
      "learning_rate": 6.800000000000001e-05,
      "loss": 1.012,
      "step": 68
    },
    {
      "epoch": 0.0352760736196319,
      "grad_norm": 4.380066871643066,
      "learning_rate": 6.9e-05,
      "loss": 1.1207,
      "step": 69
    },
    {
      "epoch": 0.03578732106339468,
      "grad_norm": 4.1866230964660645,
      "learning_rate": 7e-05,
      "loss": 1.0963,
      "step": 70
    },
    {
      "epoch": 0.036298568507157465,
      "grad_norm": 4.3079833984375,
      "learning_rate": 7.1e-05,
      "loss": 0.9799,
      "step": 71
    },
    {
      "epoch": 0.03680981595092025,
      "grad_norm": 4.587730407714844,
      "learning_rate": 7.2e-05,
      "loss": 1.0129,
      "step": 72
    },
    {
      "epoch": 0.037321063394683024,
      "grad_norm": 4.465561389923096,
      "learning_rate": 7.3e-05,
      "loss": 1.0579,
      "step": 73
    },
    {
      "epoch": 0.03783231083844581,
      "grad_norm": 4.654197692871094,
      "learning_rate": 7.4e-05,
      "loss": 1.0536,
      "step": 74
    },
    {
      "epoch": 0.03834355828220859,
      "grad_norm": 4.685794353485107,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.0567,
      "step": 75
    },
    {
      "epoch": 0.03885480572597137,
      "grad_norm": 4.281529426574707,
      "learning_rate": 7.6e-05,
      "loss": 0.9689,
      "step": 76
    },
    {
      "epoch": 0.03936605316973415,
      "grad_norm": 4.096267223358154,
      "learning_rate": 7.7e-05,
      "loss": 1.0279,
      "step": 77
    },
    {
      "epoch": 0.03987730061349693,
      "grad_norm": 4.406935214996338,
      "learning_rate": 7.800000000000001e-05,
      "loss": 1.0197,
      "step": 78
    },
    {
      "epoch": 0.040388548057259714,
      "grad_norm": 4.344164848327637,
      "learning_rate": 7.900000000000001e-05,
      "loss": 0.9634,
      "step": 79
    },
    {
      "epoch": 0.0408997955010225,
      "grad_norm": 4.969428539276123,
      "learning_rate": 8e-05,
      "loss": 1.0437,
      "step": 80
    },
    {
      "epoch": 0.04141104294478527,
      "grad_norm": 4.627592086791992,
      "learning_rate": 8.1e-05,
      "loss": 1.0735,
      "step": 81
    },
    {
      "epoch": 0.041922290388548056,
      "grad_norm": 4.50595760345459,
      "learning_rate": 8.2e-05,
      "loss": 0.9614,
      "step": 82
    },
    {
      "epoch": 0.04243353783231084,
      "grad_norm": 4.57030725479126,
      "learning_rate": 8.3e-05,
      "loss": 0.9109,
      "step": 83
    },
    {
      "epoch": 0.04294478527607362,
      "grad_norm": 5.381258487701416,
      "learning_rate": 8.4e-05,
      "loss": 1.1194,
      "step": 84
    },
    {
      "epoch": 0.0434560327198364,
      "grad_norm": 5.224101543426514,
      "learning_rate": 8.5e-05,
      "loss": 1.0022,
      "step": 85
    },
    {
      "epoch": 0.04396728016359918,
      "grad_norm": 4.670103073120117,
      "learning_rate": 8.6e-05,
      "loss": 0.9712,
      "step": 86
    },
    {
      "epoch": 0.04447852760736196,
      "grad_norm": 5.264283657073975,
      "learning_rate": 8.7e-05,
      "loss": 0.9767,
      "step": 87
    },
    {
      "epoch": 0.044989775051124746,
      "grad_norm": 4.800637245178223,
      "learning_rate": 8.800000000000001e-05,
      "loss": 0.8713,
      "step": 88
    },
    {
      "epoch": 0.04550102249488753,
      "grad_norm": 5.133697986602783,
      "learning_rate": 8.900000000000001e-05,
      "loss": 0.9959,
      "step": 89
    },
    {
      "epoch": 0.046012269938650305,
      "grad_norm": 5.831608295440674,
      "learning_rate": 9e-05,
      "loss": 0.9995,
      "step": 90
    },
    {
      "epoch": 0.04652351738241309,
      "grad_norm": 5.324717998504639,
      "learning_rate": 9.1e-05,
      "loss": 0.9143,
      "step": 91
    },
    {
      "epoch": 0.04703476482617587,
      "grad_norm": 6.4028401374816895,
      "learning_rate": 9.200000000000001e-05,
      "loss": 1.1914,
      "step": 92
    },
    {
      "epoch": 0.04754601226993865,
      "grad_norm": 6.07197904586792,
      "learning_rate": 9.300000000000001e-05,
      "loss": 0.9809,
      "step": 93
    },
    {
      "epoch": 0.04805725971370143,
      "grad_norm": 5.957538604736328,
      "learning_rate": 9.4e-05,
      "loss": 1.0473,
      "step": 94
    },
    {
      "epoch": 0.04856850715746421,
      "grad_norm": 5.8357343673706055,
      "learning_rate": 9.5e-05,
      "loss": 0.9465,
      "step": 95
    },
    {
      "epoch": 0.049079754601226995,
      "grad_norm": 5.610650062561035,
      "learning_rate": 9.6e-05,
      "loss": 0.7326,
      "step": 96
    },
    {
      "epoch": 0.04959100204498978,
      "grad_norm": 6.265199661254883,
      "learning_rate": 9.7e-05,
      "loss": 0.9174,
      "step": 97
    },
    {
      "epoch": 0.050102249488752554,
      "grad_norm": 7.403947353363037,
      "learning_rate": 9.8e-05,
      "loss": 0.8325,
      "step": 98
    },
    {
      "epoch": 0.05061349693251534,
      "grad_norm": 7.57144021987915,
      "learning_rate": 9.900000000000001e-05,
      "loss": 0.6444,
      "step": 99
    },
    {
      "epoch": 0.05112474437627812,
      "grad_norm": 10.456438064575195,
      "learning_rate": 0.0001,
      "loss": 0.6351,
      "step": 100
    },
    {
      "epoch": 0.0516359918200409,
      "grad_norm": 3.450286865234375,
      "learning_rate": 9.999992837185459e-05,
      "loss": 1.1081,
      "step": 101
    },
    {
      "epoch": 0.05214723926380368,
      "grad_norm": 3.6296398639678955,
      "learning_rate": 9.999971348762359e-05,
      "loss": 1.1465,
      "step": 102
    },
    {
      "epoch": 0.05265848670756646,
      "grad_norm": 3.699775457382202,
      "learning_rate": 9.999935534792264e-05,
      "loss": 1.19,
      "step": 103
    },
    {
      "epoch": 0.053169734151329244,
      "grad_norm": 3.4316041469573975,
      "learning_rate": 9.999885395377788e-05,
      "loss": 1.1055,
      "step": 104
    },
    {
      "epoch": 0.05368098159509203,
      "grad_norm": 3.6994450092315674,
      "learning_rate": 9.999820930662585e-05,
      "loss": 1.1867,
      "step": 105
    },
    {
      "epoch": 0.0541922290388548,
      "grad_norm": 3.6903419494628906,
      "learning_rate": 9.999742140831357e-05,
      "loss": 1.077,
      "step": 106
    },
    {
      "epoch": 0.054703476482617586,
      "grad_norm": 4.079493522644043,
      "learning_rate": 9.999649026109845e-05,
      "loss": 1.1008,
      "step": 107
    },
    {
      "epoch": 0.05521472392638037,
      "grad_norm": 3.9423539638519287,
      "learning_rate": 9.999541586764836e-05,
      "loss": 1.2268,
      "step": 108
    },
    {
      "epoch": 0.05572597137014315,
      "grad_norm": 4.003759860992432,
      "learning_rate": 9.999419823104155e-05,
      "loss": 1.0735,
      "step": 109
    },
    {
      "epoch": 0.05623721881390593,
      "grad_norm": 3.989661693572998,
      "learning_rate": 9.999283735476672e-05,
      "loss": 1.0735,
      "step": 110
    },
    {
      "epoch": 0.05674846625766871,
      "grad_norm": 3.9586427211761475,
      "learning_rate": 9.999133324272294e-05,
      "loss": 1.0708,
      "step": 111
    },
    {
      "epoch": 0.05725971370143149,
      "grad_norm": 4.020519733428955,
      "learning_rate": 9.998968589921969e-05,
      "loss": 1.0245,
      "step": 112
    },
    {
      "epoch": 0.057770961145194276,
      "grad_norm": 4.426997184753418,
      "learning_rate": 9.998789532897681e-05,
      "loss": 1.0745,
      "step": 113
    },
    {
      "epoch": 0.05828220858895705,
      "grad_norm": 3.993692636489868,
      "learning_rate": 9.998596153712451e-05,
      "loss": 1.0767,
      "step": 114
    },
    {
      "epoch": 0.058793456032719835,
      "grad_norm": 4.429073333740234,
      "learning_rate": 9.998388452920334e-05,
      "loss": 1.017,
      "step": 115
    },
    {
      "epoch": 0.05930470347648262,
      "grad_norm": 3.983215093612671,
      "learning_rate": 9.99816643111642e-05,
      "loss": 1.021,
      "step": 116
    },
    {
      "epoch": 0.0598159509202454,
      "grad_norm": 4.212836742401123,
      "learning_rate": 9.997930088936828e-05,
      "loss": 0.982,
      "step": 117
    },
    {
      "epoch": 0.06032719836400818,
      "grad_norm": 4.086669921875,
      "learning_rate": 9.997679427058713e-05,
      "loss": 1.0085,
      "step": 118
    },
    {
      "epoch": 0.06083844580777096,
      "grad_norm": 3.8549580574035645,
      "learning_rate": 9.997414446200246e-05,
      "loss": 0.9703,
      "step": 119
    },
    {
      "epoch": 0.06134969325153374,
      "grad_norm": 4.432587623596191,
      "learning_rate": 9.997135147120633e-05,
      "loss": 1.0989,
      "step": 120
    },
    {
      "epoch": 0.061860940695296525,
      "grad_norm": 4.559906959533691,
      "learning_rate": 9.996841530620103e-05,
      "loss": 0.9276,
      "step": 121
    },
    {
      "epoch": 0.06237218813905931,
      "grad_norm": 4.4931721687316895,
      "learning_rate": 9.996533597539901e-05,
      "loss": 0.9899,
      "step": 122
    },
    {
      "epoch": 0.06288343558282208,
      "grad_norm": 4.057201862335205,
      "learning_rate": 9.996211348762296e-05,
      "loss": 1.0011,
      "step": 123
    },
    {
      "epoch": 0.06339468302658487,
      "grad_norm": 4.463088512420654,
      "learning_rate": 9.995874785210573e-05,
      "loss": 0.9554,
      "step": 124
    },
    {
      "epoch": 0.06390593047034765,
      "grad_norm": 4.215217590332031,
      "learning_rate": 9.995523907849024e-05,
      "loss": 0.9568,
      "step": 125
    },
    {
      "epoch": 0.06441717791411043,
      "grad_norm": 4.900539875030518,
      "learning_rate": 9.995158717682963e-05,
      "loss": 0.9864,
      "step": 126
    },
    {
      "epoch": 0.06492842535787322,
      "grad_norm": 4.836815357208252,
      "learning_rate": 9.9947792157587e-05,
      "loss": 0.9749,
      "step": 127
    },
    {
      "epoch": 0.065439672801636,
      "grad_norm": 4.402955055236816,
      "learning_rate": 9.99438540316356e-05,
      "loss": 0.9626,
      "step": 128
    },
    {
      "epoch": 0.06595092024539877,
      "grad_norm": 4.344229698181152,
      "learning_rate": 9.993977281025862e-05,
      "loss": 0.9272,
      "step": 129
    },
    {
      "epoch": 0.06646216768916155,
      "grad_norm": 4.652369976043701,
      "learning_rate": 9.99355485051493e-05,
      "loss": 1.0139,
      "step": 130
    },
    {
      "epoch": 0.06697341513292433,
      "grad_norm": 4.622213840484619,
      "learning_rate": 9.99311811284108e-05,
      "loss": 0.9265,
      "step": 131
    },
    {
      "epoch": 0.06748466257668712,
      "grad_norm": 5.247522830963135,
      "learning_rate": 9.992667069255619e-05,
      "loss": 1.0134,
      "step": 132
    },
    {
      "epoch": 0.0679959100204499,
      "grad_norm": 4.564031600952148,
      "learning_rate": 9.992201721050847e-05,
      "loss": 0.9342,
      "step": 133
    },
    {
      "epoch": 0.06850715746421268,
      "grad_norm": 4.3959245681762695,
      "learning_rate": 9.991722069560041e-05,
      "loss": 0.9452,
      "step": 134
    },
    {
      "epoch": 0.06901840490797546,
      "grad_norm": 4.695987701416016,
      "learning_rate": 9.991228116157466e-05,
      "loss": 0.9721,
      "step": 135
    },
    {
      "epoch": 0.06952965235173825,
      "grad_norm": 4.595483303070068,
      "learning_rate": 9.990719862258358e-05,
      "loss": 0.9288,
      "step": 136
    },
    {
      "epoch": 0.07004089979550102,
      "grad_norm": 4.6917009353637695,
      "learning_rate": 9.99019730931893e-05,
      "loss": 0.8785,
      "step": 137
    },
    {
      "epoch": 0.0705521472392638,
      "grad_norm": 6.002026081085205,
      "learning_rate": 9.98966045883636e-05,
      "loss": 0.9841,
      "step": 138
    },
    {
      "epoch": 0.07106339468302658,
      "grad_norm": 5.35076379776001,
      "learning_rate": 9.989109312348796e-05,
      "loss": 1.0745,
      "step": 139
    },
    {
      "epoch": 0.07157464212678936,
      "grad_norm": 5.220756530761719,
      "learning_rate": 9.98854387143534e-05,
      "loss": 0.9308,
      "step": 140
    },
    {
      "epoch": 0.07208588957055215,
      "grad_norm": 5.255858421325684,
      "learning_rate": 9.98796413771605e-05,
      "loss": 0.9596,
      "step": 141
    },
    {
      "epoch": 0.07259713701431493,
      "grad_norm": 5.472203254699707,
      "learning_rate": 9.987370112851939e-05,
      "loss": 0.9245,
      "step": 142
    },
    {
      "epoch": 0.07310838445807771,
      "grad_norm": 5.4891252517700195,
      "learning_rate": 9.98676179854496e-05,
      "loss": 0.9549,
      "step": 143
    },
    {
      "epoch": 0.0736196319018405,
      "grad_norm": 5.602303504943848,
      "learning_rate": 9.986139196538011e-05,
      "loss": 0.9336,
      "step": 144
    },
    {
      "epoch": 0.07413087934560328,
      "grad_norm": 6.35874080657959,
      "learning_rate": 9.985502308614927e-05,
      "loss": 1.0135,
      "step": 145
    },
    {
      "epoch": 0.07464212678936605,
      "grad_norm": 6.352203369140625,
      "learning_rate": 9.984851136600469e-05,
      "loss": 0.8946,
      "step": 146
    },
    {
      "epoch": 0.07515337423312883,
      "grad_norm": 6.650132179260254,
      "learning_rate": 9.984185682360327e-05,
      "loss": 0.943,
      "step": 147
    },
    {
      "epoch": 0.07566462167689161,
      "grad_norm": 7.255720138549805,
      "learning_rate": 9.983505947801115e-05,
      "loss": 0.8747,
      "step": 148
    },
    {
      "epoch": 0.0761758691206544,
      "grad_norm": 7.385144233703613,
      "learning_rate": 9.982811934870353e-05,
      "loss": 0.7153,
      "step": 149
    },
    {
      "epoch": 0.07668711656441718,
      "grad_norm": 11.432644844055176,
      "learning_rate": 9.982103645556478e-05,
      "loss": 1.1386,
      "step": 150
    },
    {
      "epoch": 0.07719836400817996,
      "grad_norm": 3.1650919914245605,
      "learning_rate": 9.981381081888828e-05,
      "loss": 1.0661,
      "step": 151
    },
    {
      "epoch": 0.07770961145194274,
      "grad_norm": 3.7646584510803223,
      "learning_rate": 9.980644245937639e-05,
      "loss": 1.1944,
      "step": 152
    },
    {
      "epoch": 0.07822085889570553,
      "grad_norm": 3.43169903755188,
      "learning_rate": 9.979893139814039e-05,
      "loss": 1.1582,
      "step": 153
    },
    {
      "epoch": 0.0787321063394683,
      "grad_norm": 3.5647740364074707,
      "learning_rate": 9.979127765670039e-05,
      "loss": 1.1398,
      "step": 154
    },
    {
      "epoch": 0.07924335378323108,
      "grad_norm": 3.8447012901306152,
      "learning_rate": 9.978348125698534e-05,
      "loss": 1.1126,
      "step": 155
    },
    {
      "epoch": 0.07975460122699386,
      "grad_norm": 3.9558143615722656,
      "learning_rate": 9.977554222133292e-05,
      "loss": 1.1155,
      "step": 156
    },
    {
      "epoch": 0.08026584867075665,
      "grad_norm": 3.363036870956421,
      "learning_rate": 9.976746057248943e-05,
      "loss": 0.9753,
      "step": 157
    },
    {
      "epoch": 0.08077709611451943,
      "grad_norm": 4.080865383148193,
      "learning_rate": 9.975923633360985e-05,
      "loss": 1.1136,
      "step": 158
    },
    {
      "epoch": 0.08128834355828221,
      "grad_norm": 3.7571935653686523,
      "learning_rate": 9.975086952825762e-05,
      "loss": 1.1015,
      "step": 159
    },
    {
      "epoch": 0.081799591002045,
      "grad_norm": 3.6946327686309814,
      "learning_rate": 9.974236018040474e-05,
      "loss": 1.0684,
      "step": 160
    },
    {
      "epoch": 0.08231083844580778,
      "grad_norm": 4.085092544555664,
      "learning_rate": 9.973370831443152e-05,
      "loss": 1.082,
      "step": 161
    },
    {
      "epoch": 0.08282208588957055,
      "grad_norm": 4.023382186889648,
      "learning_rate": 9.972491395512666e-05,
      "loss": 1.059,
      "step": 162
    },
    {
      "epoch": 0.08333333333333333,
      "grad_norm": 4.119484901428223,
      "learning_rate": 9.97159771276871e-05,
      "loss": 0.996,
      "step": 163
    },
    {
      "epoch": 0.08384458077709611,
      "grad_norm": 3.6977996826171875,
      "learning_rate": 9.970689785771798e-05,
      "loss": 1.0283,
      "step": 164
    },
    {
      "epoch": 0.0843558282208589,
      "grad_norm": 3.9231314659118652,
      "learning_rate": 9.969767617123256e-05,
      "loss": 1.0912,
      "step": 165
    },
    {
      "epoch": 0.08486707566462168,
      "grad_norm": 3.965254306793213,
      "learning_rate": 9.968831209465211e-05,
      "loss": 1.0856,
      "step": 166
    },
    {
      "epoch": 0.08537832310838446,
      "grad_norm": 3.6064834594726562,
      "learning_rate": 9.967880565480589e-05,
      "loss": 0.9545,
      "step": 167
    },
    {
      "epoch": 0.08588957055214724,
      "grad_norm": 4.335506439208984,
      "learning_rate": 9.966915687893108e-05,
      "loss": 1.0064,
      "step": 168
    },
    {
      "epoch": 0.08640081799591003,
      "grad_norm": 4.090417385101318,
      "learning_rate": 9.96593657946726e-05,
      "loss": 0.9032,
      "step": 169
    },
    {
      "epoch": 0.0869120654396728,
      "grad_norm": 4.244670391082764,
      "learning_rate": 9.964943243008315e-05,
      "loss": 0.9544,
      "step": 170
    },
    {
      "epoch": 0.08742331288343558,
      "grad_norm": 4.027634143829346,
      "learning_rate": 9.963935681362308e-05,
      "loss": 1.0019,
      "step": 171
    },
    {
      "epoch": 0.08793456032719836,
      "grad_norm": 4.118989944458008,
      "learning_rate": 9.962913897416028e-05,
      "loss": 0.9132,
      "step": 172
    },
    {
      "epoch": 0.08844580777096114,
      "grad_norm": 4.275842666625977,
      "learning_rate": 9.961877894097016e-05,
      "loss": 0.9875,
      "step": 173
    },
    {
      "epoch": 0.08895705521472393,
      "grad_norm": 4.40177059173584,
      "learning_rate": 9.96082767437355e-05,
      "loss": 1.0284,
      "step": 174
    },
    {
      "epoch": 0.08946830265848671,
      "grad_norm": 4.55718994140625,
      "learning_rate": 9.959763241254645e-05,
      "loss": 1.0491,
      "step": 175
    },
    {
      "epoch": 0.08997955010224949,
      "grad_norm": 3.950263261795044,
      "learning_rate": 9.958684597790031e-05,
      "loss": 0.8668,
      "step": 176
    },
    {
      "epoch": 0.09049079754601227,
      "grad_norm": 4.344690322875977,
      "learning_rate": 9.957591747070163e-05,
      "loss": 0.9561,
      "step": 177
    },
    {
      "epoch": 0.09100204498977506,
      "grad_norm": 4.7496232986450195,
      "learning_rate": 9.95648469222619e-05,
      "loss": 1.2058,
      "step": 178
    },
    {
      "epoch": 0.09151329243353783,
      "grad_norm": 4.409940242767334,
      "learning_rate": 9.955363436429968e-05,
      "loss": 1.0548,
      "step": 179
    },
    {
      "epoch": 0.09202453987730061,
      "grad_norm": 4.629715919494629,
      "learning_rate": 9.954227982894034e-05,
      "loss": 0.9482,
      "step": 180
    },
    {
      "epoch": 0.09253578732106339,
      "grad_norm": 4.308309078216553,
      "learning_rate": 9.953078334871606e-05,
      "loss": 0.9435,
      "step": 181
    },
    {
      "epoch": 0.09304703476482618,
      "grad_norm": 4.249781131744385,
      "learning_rate": 9.951914495656569e-05,
      "loss": 0.9578,
      "step": 182
    },
    {
      "epoch": 0.09355828220858896,
      "grad_norm": 4.906545162200928,
      "learning_rate": 9.950736468583468e-05,
      "loss": 1.0188,
      "step": 183
    },
    {
      "epoch": 0.09406952965235174,
      "grad_norm": 5.413058280944824,
      "learning_rate": 9.949544257027502e-05,
      "loss": 1.1337,
      "step": 184
    },
    {
      "epoch": 0.09458077709611452,
      "grad_norm": 4.498751640319824,
      "learning_rate": 9.948337864404504e-05,
      "loss": 0.9442,
      "step": 185
    },
    {
      "epoch": 0.0950920245398773,
      "grad_norm": 4.727104187011719,
      "learning_rate": 9.947117294170944e-05,
      "loss": 1.0067,
      "step": 186
    },
    {
      "epoch": 0.09560327198364008,
      "grad_norm": 4.974936008453369,
      "learning_rate": 9.945882549823906e-05,
      "loss": 1.0294,
      "step": 187
    },
    {
      "epoch": 0.09611451942740286,
      "grad_norm": 5.312014102935791,
      "learning_rate": 9.944633634901088e-05,
      "loss": 0.9887,
      "step": 188
    },
    {
      "epoch": 0.09662576687116564,
      "grad_norm": 5.3448405265808105,
      "learning_rate": 9.943370552980791e-05,
      "loss": 0.9342,
      "step": 189
    },
    {
      "epoch": 0.09713701431492842,
      "grad_norm": 5.098217964172363,
      "learning_rate": 9.942093307681901e-05,
      "loss": 0.9,
      "step": 190
    },
    {
      "epoch": 0.09764826175869121,
      "grad_norm": 5.733004570007324,
      "learning_rate": 9.940801902663889e-05,
      "loss": 1.0274,
      "step": 191
    },
    {
      "epoch": 0.09815950920245399,
      "grad_norm": 5.353964328765869,
      "learning_rate": 9.939496341626791e-05,
      "loss": 1.0018,
      "step": 192
    },
    {
      "epoch": 0.09867075664621677,
      "grad_norm": 5.414928913116455,
      "learning_rate": 9.938176628311204e-05,
      "loss": 0.8237,
      "step": 193
    },
    {
      "epoch": 0.09918200408997956,
      "grad_norm": 5.456855297088623,
      "learning_rate": 9.936842766498273e-05,
      "loss": 0.9474,
      "step": 194
    },
    {
      "epoch": 0.09969325153374232,
      "grad_norm": 6.7383856773376465,
      "learning_rate": 9.935494760009678e-05,
      "loss": 1.0278,
      "step": 195
    },
    {
      "epoch": 0.10020449897750511,
      "grad_norm": 6.588730812072754,
      "learning_rate": 9.934132612707632e-05,
      "loss": 1.0217,
      "step": 196
    },
    {
      "epoch": 0.10071574642126789,
      "grad_norm": 7.4416093826293945,
      "learning_rate": 9.932756328494852e-05,
      "loss": 0.9317,
      "step": 197
    },
    {
      "epoch": 0.10122699386503067,
      "grad_norm": 7.225059986114502,
      "learning_rate": 9.931365911314572e-05,
      "loss": 0.8769,
      "step": 198
    },
    {
      "epoch": 0.10173824130879346,
      "grad_norm": 7.279494285583496,
      "learning_rate": 9.929961365150506e-05,
      "loss": 0.6817,
      "step": 199
    },
    {
      "epoch": 0.10224948875255624,
      "grad_norm": 10.894880294799805,
      "learning_rate": 9.928542694026862e-05,
      "loss": 0.9749,
      "step": 200
    },
    {
      "epoch": 0.10276073619631902,
      "grad_norm": 3.4081413745880127,
      "learning_rate": 9.927109902008303e-05,
      "loss": 0.9225,
      "step": 201
    },
    {
      "epoch": 0.1032719836400818,
      "grad_norm": 4.234488010406494,
      "learning_rate": 9.925662993199967e-05,
      "loss": 1.1298,
      "step": 202
    },
    {
      "epoch": 0.10378323108384459,
      "grad_norm": 3.798875093460083,
      "learning_rate": 9.924201971747423e-05,
      "loss": 1.0496,
      "step": 203
    },
    {
      "epoch": 0.10429447852760736,
      "grad_norm": 3.6677727699279785,
      "learning_rate": 9.922726841836684e-05,
      "loss": 1.1259,
      "step": 204
    },
    {
      "epoch": 0.10480572597137014,
      "grad_norm": 3.7715201377868652,
      "learning_rate": 9.921237607694184e-05,
      "loss": 1.105,
      "step": 205
    },
    {
      "epoch": 0.10531697341513292,
      "grad_norm": 3.534221649169922,
      "learning_rate": 9.919734273586767e-05,
      "loss": 1.0223,
      "step": 206
    },
    {
      "epoch": 0.1058282208588957,
      "grad_norm": 3.9099631309509277,
      "learning_rate": 9.91821684382167e-05,
      "loss": 1.0874,
      "step": 207
    },
    {
      "epoch": 0.10633946830265849,
      "grad_norm": 3.318946361541748,
      "learning_rate": 9.916685322746524e-05,
      "loss": 0.9967,
      "step": 208
    },
    {
      "epoch": 0.10685071574642127,
      "grad_norm": 4.089531421661377,
      "learning_rate": 9.915139714749328e-05,
      "loss": 0.973,
      "step": 209
    },
    {
      "epoch": 0.10736196319018405,
      "grad_norm": 3.753154993057251,
      "learning_rate": 9.913580024258442e-05,
      "loss": 1.0184,
      "step": 210
    },
    {
      "epoch": 0.10787321063394684,
      "grad_norm": 4.053317070007324,
      "learning_rate": 9.912006255742579e-05,
      "loss": 1.1166,
      "step": 211
    },
    {
      "epoch": 0.1083844580777096,
      "grad_norm": 3.554300308227539,
      "learning_rate": 9.91041841371078e-05,
      "loss": 0.9794,
      "step": 212
    },
    {
      "epoch": 0.10889570552147239,
      "grad_norm": 3.9923880100250244,
      "learning_rate": 9.908816502712415e-05,
      "loss": 1.1504,
      "step": 213
    },
    {
      "epoch": 0.10940695296523517,
      "grad_norm": 3.900874137878418,
      "learning_rate": 9.90720052733716e-05,
      "loss": 1.0554,
      "step": 214
    },
    {
      "epoch": 0.10991820040899795,
      "grad_norm": 3.914747714996338,
      "learning_rate": 9.905570492214987e-05,
      "loss": 1.0801,
      "step": 215
    },
    {
      "epoch": 0.11042944785276074,
      "grad_norm": 4.028392791748047,
      "learning_rate": 9.903926402016153e-05,
      "loss": 1.0332,
      "step": 216
    },
    {
      "epoch": 0.11094069529652352,
      "grad_norm": 4.589059829711914,
      "learning_rate": 9.902268261451182e-05,
      "loss": 1.0179,
      "step": 217
    },
    {
      "epoch": 0.1114519427402863,
      "grad_norm": 4.278644561767578,
      "learning_rate": 9.900596075270856e-05,
      "loss": 1.0743,
      "step": 218
    },
    {
      "epoch": 0.11196319018404909,
      "grad_norm": 4.3040008544921875,
      "learning_rate": 9.898909848266196e-05,
      "loss": 0.9975,
      "step": 219
    },
    {
      "epoch": 0.11247443762781185,
      "grad_norm": 3.8854451179504395,
      "learning_rate": 9.897209585268458e-05,
      "loss": 1.003,
      "step": 220
    },
    {
      "epoch": 0.11298568507157464,
      "grad_norm": 4.615149021148682,
      "learning_rate": 9.89549529114911e-05,
      "loss": 0.9241,
      "step": 221
    },
    {
      "epoch": 0.11349693251533742,
      "grad_norm": 4.247766494750977,
      "learning_rate": 9.893766970819819e-05,
      "loss": 0.9801,
      "step": 222
    },
    {
      "epoch": 0.1140081799591002,
      "grad_norm": 4.650023460388184,
      "learning_rate": 9.89202462923244e-05,
      "loss": 1.0031,
      "step": 223
    },
    {
      "epoch": 0.11451942740286299,
      "grad_norm": 4.242456912994385,
      "learning_rate": 9.890268271379e-05,
      "loss": 1.0527,
      "step": 224
    },
    {
      "epoch": 0.11503067484662577,
      "grad_norm": 4.012096881866455,
      "learning_rate": 9.888497902291685e-05,
      "loss": 0.9761,
      "step": 225
    },
    {
      "epoch": 0.11554192229038855,
      "grad_norm": 4.403133392333984,
      "learning_rate": 9.886713527042828e-05,
      "loss": 0.9944,
      "step": 226
    },
    {
      "epoch": 0.11605316973415133,
      "grad_norm": 4.339402675628662,
      "learning_rate": 9.884915150744885e-05,
      "loss": 1.0336,
      "step": 227
    },
    {
      "epoch": 0.1165644171779141,
      "grad_norm": 5.434019088745117,
      "learning_rate": 9.883102778550434e-05,
      "loss": 1.0191,
      "step": 228
    },
    {
      "epoch": 0.11707566462167689,
      "grad_norm": 4.682615280151367,
      "learning_rate": 9.881276415652148e-05,
      "loss": 0.9405,
      "step": 229
    },
    {
      "epoch": 0.11758691206543967,
      "grad_norm": 4.088139057159424,
      "learning_rate": 9.879436067282783e-05,
      "loss": 0.8859,
      "step": 230
    },
    {
      "epoch": 0.11809815950920245,
      "grad_norm": 4.25062370300293,
      "learning_rate": 9.877581738715174e-05,
      "loss": 0.9826,
      "step": 231
    },
    {
      "epoch": 0.11860940695296524,
      "grad_norm": 4.560945510864258,
      "learning_rate": 9.875713435262203e-05,
      "loss": 0.8834,
      "step": 232
    },
    {
      "epoch": 0.11912065439672802,
      "grad_norm": 4.465799331665039,
      "learning_rate": 9.873831162276796e-05,
      "loss": 0.9343,
      "step": 233
    },
    {
      "epoch": 0.1196319018404908,
      "grad_norm": 4.841387748718262,
      "learning_rate": 9.871934925151898e-05,
      "loss": 0.9551,
      "step": 234
    },
    {
      "epoch": 0.12014314928425358,
      "grad_norm": 4.5967326164245605,
      "learning_rate": 9.870024729320471e-05,
      "loss": 0.9566,
      "step": 235
    },
    {
      "epoch": 0.12065439672801637,
      "grad_norm": 5.284646987915039,
      "learning_rate": 9.868100580255466e-05,
      "loss": 0.9958,
      "step": 236
    },
    {
      "epoch": 0.12116564417177914,
      "grad_norm": 5.193093299865723,
      "learning_rate": 9.86616248346981e-05,
      "loss": 1.0898,
      "step": 237
    },
    {
      "epoch": 0.12167689161554192,
      "grad_norm": 5.222631931304932,
      "learning_rate": 9.864210444516395e-05,
      "loss": 0.9452,
      "step": 238
    },
    {
      "epoch": 0.1221881390593047,
      "grad_norm": 5.9728569984436035,
      "learning_rate": 9.86224446898806e-05,
      "loss": 1.0229,
      "step": 239
    },
    {
      "epoch": 0.12269938650306748,
      "grad_norm": 4.982256889343262,
      "learning_rate": 9.86026456251757e-05,
      "loss": 1.0025,
      "step": 240
    },
    {
      "epoch": 0.12321063394683027,
      "grad_norm": 5.724108695983887,
      "learning_rate": 9.858270730777608e-05,
      "loss": 1.0477,
      "step": 241
    },
    {
      "epoch": 0.12372188139059305,
      "grad_norm": 5.632500648498535,
      "learning_rate": 9.85626297948075e-05,
      "loss": 0.927,
      "step": 242
    },
    {
      "epoch": 0.12423312883435583,
      "grad_norm": 5.947004795074463,
      "learning_rate": 9.854241314379457e-05,
      "loss": 1.0072,
      "step": 243
    },
    {
      "epoch": 0.12474437627811862,
      "grad_norm": 5.381914138793945,
      "learning_rate": 9.852205741266058e-05,
      "loss": 0.8549,
      "step": 244
    },
    {
      "epoch": 0.1252556237218814,
      "grad_norm": 6.360842227935791,
      "learning_rate": 9.850156265972721e-05,
      "loss": 0.911,
      "step": 245
    },
    {
      "epoch": 0.12576687116564417,
      "grad_norm": 6.210946083068848,
      "learning_rate": 9.848092894371452e-05,
      "loss": 0.8747,
      "step": 246
    },
    {
      "epoch": 0.12627811860940696,
      "grad_norm": 7.40360689163208,
      "learning_rate": 9.84601563237407e-05,
      "loss": 1.0297,
      "step": 247
    },
    {
      "epoch": 0.12678936605316973,
      "grad_norm": 6.7166032791137695,
      "learning_rate": 9.843924485932194e-05,
      "loss": 0.791,
      "step": 248
    },
    {
      "epoch": 0.1273006134969325,
      "grad_norm": 8.442200660705566,
      "learning_rate": 9.841819461037219e-05,
      "loss": 0.9056,
      "step": 249
    },
    {
      "epoch": 0.1278118609406953,
      "grad_norm": 10.764420509338379,
      "learning_rate": 9.839700563720309e-05,
      "loss": 0.9064,
      "step": 250
    },
    {
      "epoch": 0.12832310838445807,
      "grad_norm": 2.9699931144714355,
      "learning_rate": 9.837567800052368e-05,
      "loss": 0.909,
      "step": 251
    },
    {
      "epoch": 0.12883435582822086,
      "grad_norm": 3.270521640777588,
      "learning_rate": 9.835421176144035e-05,
      "loss": 1.0767,
      "step": 252
    },
    {
      "epoch": 0.12934560327198363,
      "grad_norm": 3.5505306720733643,
      "learning_rate": 9.833260698145656e-05,
      "loss": 1.102,
      "step": 253
    },
    {
      "epoch": 0.12985685071574643,
      "grad_norm": 3.706258535385132,
      "learning_rate": 9.831086372247272e-05,
      "loss": 1.0359,
      "step": 254
    },
    {
      "epoch": 0.1303680981595092,
      "grad_norm": 3.501894950866699,
      "learning_rate": 9.828898204678603e-05,
      "loss": 1.0805,
      "step": 255
    },
    {
      "epoch": 0.130879345603272,
      "grad_norm": 3.5987539291381836,
      "learning_rate": 9.826696201709021e-05,
      "loss": 1.0488,
      "step": 256
    },
    {
      "epoch": 0.13139059304703476,
      "grad_norm": 3.721916437149048,
      "learning_rate": 9.824480369647544e-05,
      "loss": 1.0055,
      "step": 257
    },
    {
      "epoch": 0.13190184049079753,
      "grad_norm": 3.430462121963501,
      "learning_rate": 9.82225071484281e-05,
      "loss": 0.9892,
      "step": 258
    },
    {
      "epoch": 0.13241308793456033,
      "grad_norm": 3.6975953578948975,
      "learning_rate": 9.820007243683057e-05,
      "loss": 1.0917,
      "step": 259
    },
    {
      "epoch": 0.1329243353783231,
      "grad_norm": 3.9236834049224854,
      "learning_rate": 9.817749962596115e-05,
      "loss": 1.0534,
      "step": 260
    },
    {
      "epoch": 0.1334355828220859,
      "grad_norm": 3.7525970935821533,
      "learning_rate": 9.815478878049378e-05,
      "loss": 0.9877,
      "step": 261
    },
    {
      "epoch": 0.13394683026584867,
      "grad_norm": 3.929924488067627,
      "learning_rate": 9.813193996549789e-05,
      "loss": 1.0387,
      "step": 262
    },
    {
      "epoch": 0.13445807770961146,
      "grad_norm": 3.305387496948242,
      "learning_rate": 9.810895324643821e-05,
      "loss": 0.8865,
      "step": 263
    },
    {
      "epoch": 0.13496932515337423,
      "grad_norm": 4.399075984954834,
      "learning_rate": 9.808582868917458e-05,
      "loss": 1.048,
      "step": 264
    },
    {
      "epoch": 0.13548057259713703,
      "grad_norm": 3.7269322872161865,
      "learning_rate": 9.806256635996175e-05,
      "loss": 1.0156,
      "step": 265
    },
    {
      "epoch": 0.1359918200408998,
      "grad_norm": 3.96358323097229,
      "learning_rate": 9.803916632544924e-05,
      "loss": 1.0946,
      "step": 266
    },
    {
      "epoch": 0.13650306748466257,
      "grad_norm": 3.7565560340881348,
      "learning_rate": 9.801562865268109e-05,
      "loss": 0.8618,
      "step": 267
    },
    {
      "epoch": 0.13701431492842536,
      "grad_norm": 3.869610548019409,
      "learning_rate": 9.79919534090957e-05,
      "loss": 1.0215,
      "step": 268
    },
    {
      "epoch": 0.13752556237218813,
      "grad_norm": 4.0557403564453125,
      "learning_rate": 9.79681406625256e-05,
      "loss": 1.019,
      "step": 269
    },
    {
      "epoch": 0.13803680981595093,
      "grad_norm": 4.523059368133545,
      "learning_rate": 9.794419048119733e-05,
      "loss": 0.9854,
      "step": 270
    },
    {
      "epoch": 0.1385480572597137,
      "grad_norm": 4.085573673248291,
      "learning_rate": 9.792010293373116e-05,
      "loss": 0.9711,
      "step": 271
    },
    {
      "epoch": 0.1390593047034765,
      "grad_norm": 4.123945713043213,
      "learning_rate": 9.789587808914093e-05,
      "loss": 0.9732,
      "step": 272
    },
    {
      "epoch": 0.13957055214723926,
      "grad_norm": 4.804892063140869,
      "learning_rate": 9.78715160168339e-05,
      "loss": 0.9582,
      "step": 273
    },
    {
      "epoch": 0.14008179959100203,
      "grad_norm": 4.866881370544434,
      "learning_rate": 9.784701678661045e-05,
      "loss": 0.9736,
      "step": 274
    },
    {
      "epoch": 0.14059304703476483,
      "grad_norm": 4.524308681488037,
      "learning_rate": 9.782238046866397e-05,
      "loss": 0.9397,
      "step": 275
    },
    {
      "epoch": 0.1411042944785276,
      "grad_norm": 4.348950386047363,
      "learning_rate": 9.779760713358059e-05,
      "loss": 1.0025,
      "step": 276
    },
    {
      "epoch": 0.1416155419222904,
      "grad_norm": 4.245871543884277,
      "learning_rate": 9.777269685233906e-05,
      "loss": 1.002,
      "step": 277
    },
    {
      "epoch": 0.14212678936605316,
      "grad_norm": 5.086630344390869,
      "learning_rate": 9.774764969631043e-05,
      "loss": 0.9445,
      "step": 278
    },
    {
      "epoch": 0.14263803680981596,
      "grad_norm": 4.560818195343018,
      "learning_rate": 9.772246573725799e-05,
      "loss": 0.9957,
      "step": 279
    },
    {
      "epoch": 0.14314928425357873,
      "grad_norm": 5.210485458374023,
      "learning_rate": 9.769714504733694e-05,
      "loss": 0.903,
      "step": 280
    },
    {
      "epoch": 0.14366053169734153,
      "grad_norm": 4.558165550231934,
      "learning_rate": 9.767168769909425e-05,
      "loss": 1.0033,
      "step": 281
    },
    {
      "epoch": 0.1441717791411043,
      "grad_norm": 4.681556701660156,
      "learning_rate": 9.76460937654684e-05,
      "loss": 0.9035,
      "step": 282
    },
    {
      "epoch": 0.14468302658486706,
      "grad_norm": 4.702577114105225,
      "learning_rate": 9.762036331978927e-05,
      "loss": 1.0228,
      "step": 283
    },
    {
      "epoch": 0.14519427402862986,
      "grad_norm": 5.256160259246826,
      "learning_rate": 9.759449643577778e-05,
      "loss": 1.0509,
      "step": 284
    },
    {
      "epoch": 0.14570552147239263,
      "grad_norm": 4.62410831451416,
      "learning_rate": 9.756849318754584e-05,
      "loss": 0.9495,
      "step": 285
    },
    {
      "epoch": 0.14621676891615543,
      "grad_norm": 4.917038917541504,
      "learning_rate": 9.754235364959601e-05,
      "loss": 1.0343,
      "step": 286
    },
    {
      "epoch": 0.1467280163599182,
      "grad_norm": 5.558690547943115,
      "learning_rate": 9.751607789682138e-05,
      "loss": 0.9006,
      "step": 287
    },
    {
      "epoch": 0.147239263803681,
      "grad_norm": 5.684017658233643,
      "learning_rate": 9.748966600450525e-05,
      "loss": 0.955,
      "step": 288
    },
    {
      "epoch": 0.14775051124744376,
      "grad_norm": 5.3489508628845215,
      "learning_rate": 9.746311804832105e-05,
      "loss": 0.8096,
      "step": 289
    },
    {
      "epoch": 0.14826175869120656,
      "grad_norm": 5.76107931137085,
      "learning_rate": 9.743643410433197e-05,
      "loss": 1.0569,
      "step": 290
    },
    {
      "epoch": 0.14877300613496933,
      "grad_norm": 5.042691707611084,
      "learning_rate": 9.740961424899092e-05,
      "loss": 0.8595,
      "step": 291
    },
    {
      "epoch": 0.1492842535787321,
      "grad_norm": 5.10003137588501,
      "learning_rate": 9.738265855914013e-05,
      "loss": 0.8284,
      "step": 292
    },
    {
      "epoch": 0.1497955010224949,
      "grad_norm": 5.545564651489258,
      "learning_rate": 9.735556711201103e-05,
      "loss": 0.9223,
      "step": 293
    },
    {
      "epoch": 0.15030674846625766,
      "grad_norm": 5.7029337882995605,
      "learning_rate": 9.732833998522407e-05,
      "loss": 0.8667,
      "step": 294
    },
    {
      "epoch": 0.15081799591002046,
      "grad_norm": 6.165698528289795,
      "learning_rate": 9.730097725678834e-05,
      "loss": 0.912,
      "step": 295
    },
    {
      "epoch": 0.15132924335378323,
      "grad_norm": 6.472256183624268,
      "learning_rate": 9.727347900510155e-05,
      "loss": 0.9143,
      "step": 296
    },
    {
      "epoch": 0.15184049079754602,
      "grad_norm": 6.160712718963623,
      "learning_rate": 9.724584530894962e-05,
      "loss": 0.8261,
      "step": 297
    },
    {
      "epoch": 0.1523517382413088,
      "grad_norm": 7.447754859924316,
      "learning_rate": 9.721807624750658e-05,
      "loss": 0.996,
      "step": 298
    },
    {
      "epoch": 0.15286298568507156,
      "grad_norm": 7.772396087646484,
      "learning_rate": 9.719017190033425e-05,
      "loss": 0.7686,
      "step": 299
    },
    {
      "epoch": 0.15337423312883436,
      "grad_norm": 8.158390045166016,
      "learning_rate": 9.716213234738215e-05,
      "loss": 0.5601,
      "step": 300
    },
    {
      "epoch": 0.15388548057259713,
      "grad_norm": 3.221916437149048,
      "learning_rate": 9.71339576689871e-05,
      "loss": 0.8234,
      "step": 301
    },
    {
      "epoch": 0.15439672801635992,
      "grad_norm": 3.221381187438965,
      "learning_rate": 9.710564794587309e-05,
      "loss": 1.1071,
      "step": 302
    },
    {
      "epoch": 0.1549079754601227,
      "grad_norm": 3.519752025604248,
      "learning_rate": 9.707720325915104e-05,
      "loss": 1.0924,
      "step": 303
    },
    {
      "epoch": 0.1554192229038855,
      "grad_norm": 3.5323832035064697,
      "learning_rate": 9.704862369031857e-05,
      "loss": 1.034,
      "step": 304
    },
    {
      "epoch": 0.15593047034764826,
      "grad_norm": 3.57119083404541,
      "learning_rate": 9.701990932125976e-05,
      "loss": 1.1247,
      "step": 305
    },
    {
      "epoch": 0.15644171779141106,
      "grad_norm": 3.4789652824401855,
      "learning_rate": 9.699106023424482e-05,
      "loss": 1.1318,
      "step": 306
    },
    {
      "epoch": 0.15695296523517382,
      "grad_norm": 3.551086664199829,
      "learning_rate": 9.696207651193008e-05,
      "loss": 1.0059,
      "step": 307
    },
    {
      "epoch": 0.1574642126789366,
      "grad_norm": 3.465805768966675,
      "learning_rate": 9.693295823735753e-05,
      "loss": 1.1119,
      "step": 308
    },
    {
      "epoch": 0.1579754601226994,
      "grad_norm": 4.0047807693481445,
      "learning_rate": 9.690370549395468e-05,
      "loss": 0.9925,
      "step": 309
    },
    {
      "epoch": 0.15848670756646216,
      "grad_norm": 3.724668502807617,
      "learning_rate": 9.687431836553434e-05,
      "loss": 1.0408,
      "step": 310
    },
    {
      "epoch": 0.15899795501022496,
      "grad_norm": 4.307526111602783,
      "learning_rate": 9.684479693629432e-05,
      "loss": 1.162,
      "step": 311
    },
    {
      "epoch": 0.15950920245398773,
      "grad_norm": 4.771853446960449,
      "learning_rate": 9.681514129081724e-05,
      "loss": 1.0714,
      "step": 312
    },
    {
      "epoch": 0.16002044989775052,
      "grad_norm": 3.7539167404174805,
      "learning_rate": 9.678535151407023e-05,
      "loss": 0.9453,
      "step": 313
    },
    {
      "epoch": 0.1605316973415133,
      "grad_norm": 4.115946292877197,
      "learning_rate": 9.675542769140476e-05,
      "loss": 1.1075,
      "step": 314
    },
    {
      "epoch": 0.16104294478527606,
      "grad_norm": 3.835968255996704,
      "learning_rate": 9.672536990855635e-05,
      "loss": 0.93,
      "step": 315
    },
    {
      "epoch": 0.16155419222903886,
      "grad_norm": 4.084163665771484,
      "learning_rate": 9.669517825164434e-05,
      "loss": 1.0297,
      "step": 316
    },
    {
      "epoch": 0.16206543967280163,
      "grad_norm": 5.519273281097412,
      "learning_rate": 9.666485280717161e-05,
      "loss": 0.9544,
      "step": 317
    },
    {
      "epoch": 0.16257668711656442,
      "grad_norm": 4.394032955169678,
      "learning_rate": 9.663439366202438e-05,
      "loss": 0.9425,
      "step": 318
    },
    {
      "epoch": 0.1630879345603272,
      "grad_norm": 4.335214138031006,
      "learning_rate": 9.660380090347192e-05,
      "loss": 0.9202,
      "step": 319
    },
    {
      "epoch": 0.16359918200409,
      "grad_norm": 3.6991333961486816,
      "learning_rate": 9.657307461916635e-05,
      "loss": 0.9667,
      "step": 320
    },
    {
      "epoch": 0.16411042944785276,
      "grad_norm": 4.654758453369141,
      "learning_rate": 9.654221489714234e-05,
      "loss": 1.0158,
      "step": 321
    },
    {
      "epoch": 0.16462167689161555,
      "grad_norm": 4.374093055725098,
      "learning_rate": 9.651122182581688e-05,
      "loss": 1.0641,
      "step": 322
    },
    {
      "epoch": 0.16513292433537832,
      "grad_norm": 4.553246974945068,
      "learning_rate": 9.6480095493989e-05,
      "loss": 0.9888,
      "step": 323
    },
    {
      "epoch": 0.1656441717791411,
      "grad_norm": 3.8664913177490234,
      "learning_rate": 9.644883599083958e-05,
      "loss": 0.89,
      "step": 324
    },
    {
      "epoch": 0.1661554192229039,
      "grad_norm": 4.2256760597229,
      "learning_rate": 9.6417443405931e-05,
      "loss": 0.9653,
      "step": 325
    },
    {
      "epoch": 0.16666666666666666,
      "grad_norm": 4.243971347808838,
      "learning_rate": 9.638591782920698e-05,
      "loss": 0.9588,
      "step": 326
    },
    {
      "epoch": 0.16717791411042945,
      "grad_norm": 4.24473762512207,
      "learning_rate": 9.635425935099228e-05,
      "loss": 0.9181,
      "step": 327
    },
    {
      "epoch": 0.16768916155419222,
      "grad_norm": 4.342863082885742,
      "learning_rate": 9.632246806199241e-05,
      "loss": 0.9603,
      "step": 328
    },
    {
      "epoch": 0.16820040899795502,
      "grad_norm": 4.400531768798828,
      "learning_rate": 9.62905440532934e-05,
      "loss": 0.9983,
      "step": 329
    },
    {
      "epoch": 0.1687116564417178,
      "grad_norm": 4.428309440612793,
      "learning_rate": 9.625848741636157e-05,
      "loss": 0.8637,
      "step": 330
    },
    {
      "epoch": 0.16922290388548059,
      "grad_norm": 4.119474411010742,
      "learning_rate": 9.62262982430432e-05,
      "loss": 0.8998,
      "step": 331
    },
    {
      "epoch": 0.16973415132924335,
      "grad_norm": 4.993500709533691,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.9449,
      "step": 332
    },
    {
      "epoch": 0.17024539877300612,
      "grad_norm": 4.711348056793213,
      "learning_rate": 9.616152265653048e-05,
      "loss": 1.0019,
      "step": 333
    },
    {
      "epoch": 0.17075664621676892,
      "grad_norm": 4.277347564697266,
      "learning_rate": 9.612893642892634e-05,
      "loss": 0.9081,
      "step": 334
    },
    {
      "epoch": 0.1712678936605317,
      "grad_norm": 4.902201175689697,
      "learning_rate": 9.609621803611555e-05,
      "loss": 0.9769,
      "step": 335
    },
    {
      "epoch": 0.17177914110429449,
      "grad_norm": 4.935635566711426,
      "learning_rate": 9.60633675718404e-05,
      "loss": 1.0006,
      "step": 336
    },
    {
      "epoch": 0.17229038854805726,
      "grad_norm": 4.689595699310303,
      "learning_rate": 9.603038513022165e-05,
      "loss": 0.9397,
      "step": 337
    },
    {
      "epoch": 0.17280163599182005,
      "grad_norm": 5.0966362953186035,
      "learning_rate": 9.59972708057581e-05,
      "loss": 0.9297,
      "step": 338
    },
    {
      "epoch": 0.17331288343558282,
      "grad_norm": 4.845507621765137,
      "learning_rate": 9.596402469332648e-05,
      "loss": 0.9348,
      "step": 339
    },
    {
      "epoch": 0.1738241308793456,
      "grad_norm": 4.724996089935303,
      "learning_rate": 9.59306468881811e-05,
      "loss": 0.9311,
      "step": 340
    },
    {
      "epoch": 0.1743353783231084,
      "grad_norm": 4.883242130279541,
      "learning_rate": 9.589713748595352e-05,
      "loss": 0.8652,
      "step": 341
    },
    {
      "epoch": 0.17484662576687116,
      "grad_norm": 5.192939758300781,
      "learning_rate": 9.586349658265245e-05,
      "loss": 0.8496,
      "step": 342
    },
    {
      "epoch": 0.17535787321063395,
      "grad_norm": 4.756380558013916,
      "learning_rate": 9.582972427466328e-05,
      "loss": 0.8408,
      "step": 343
    },
    {
      "epoch": 0.17586912065439672,
      "grad_norm": 5.135041236877441,
      "learning_rate": 9.579582065874793e-05,
      "loss": 0.8428,
      "step": 344
    },
    {
      "epoch": 0.17638036809815952,
      "grad_norm": 6.151747703552246,
      "learning_rate": 9.576178583204453e-05,
      "loss": 0.9985,
      "step": 345
    },
    {
      "epoch": 0.1768916155419223,
      "grad_norm": 6.850955486297607,
      "learning_rate": 9.572761989206712e-05,
      "loss": 0.824,
      "step": 346
    },
    {
      "epoch": 0.17740286298568508,
      "grad_norm": 6.425426959991455,
      "learning_rate": 9.569332293670543e-05,
      "loss": 0.8296,
      "step": 347
    },
    {
      "epoch": 0.17791411042944785,
      "grad_norm": 6.790580749511719,
      "learning_rate": 9.565889506422456e-05,
      "loss": 0.9905,
      "step": 348
    },
    {
      "epoch": 0.17842535787321062,
      "grad_norm": 6.514560699462891,
      "learning_rate": 9.562433637326468e-05,
      "loss": 0.6521,
      "step": 349
    },
    {
      "epoch": 0.17893660531697342,
      "grad_norm": 9.105517387390137,
      "learning_rate": 9.558964696284081e-05,
      "loss": 0.7948,
      "step": 350
    },
    {
      "epoch": 0.1794478527607362,
      "grad_norm": 3.3637101650238037,
      "learning_rate": 9.555482693234245e-05,
      "loss": 1.011,
      "step": 351
    },
    {
      "epoch": 0.17995910020449898,
      "grad_norm": 2.648695230484009,
      "learning_rate": 9.551987638153339e-05,
      "loss": 0.9487,
      "step": 352
    },
    {
      "epoch": 0.18047034764826175,
      "grad_norm": 3.222560405731201,
      "learning_rate": 9.548479541055133e-05,
      "loss": 1.0157,
      "step": 353
    },
    {
      "epoch": 0.18098159509202455,
      "grad_norm": 3.2273850440979004,
      "learning_rate": 9.54495841199077e-05,
      "loss": 1.116,
      "step": 354
    },
    {
      "epoch": 0.18149284253578732,
      "grad_norm": 3.239750862121582,
      "learning_rate": 9.541424261048725e-05,
      "loss": 0.9958,
      "step": 355
    },
    {
      "epoch": 0.18200408997955012,
      "grad_norm": 3.241530418395996,
      "learning_rate": 9.537877098354786e-05,
      "loss": 1.0277,
      "step": 356
    },
    {
      "epoch": 0.18251533742331288,
      "grad_norm": 3.041856050491333,
      "learning_rate": 9.534316934072021e-05,
      "loss": 1.0157,
      "step": 357
    },
    {
      "epoch": 0.18302658486707565,
      "grad_norm": 3.3174169063568115,
      "learning_rate": 9.530743778400746e-05,
      "loss": 1.0286,
      "step": 358
    },
    {
      "epoch": 0.18353783231083845,
      "grad_norm": 3.524456262588501,
      "learning_rate": 9.527157641578506e-05,
      "loss": 1.0808,
      "step": 359
    },
    {
      "epoch": 0.18404907975460122,
      "grad_norm": 3.1490230560302734,
      "learning_rate": 9.52355853388003e-05,
      "loss": 1.0235,
      "step": 360
    },
    {
      "epoch": 0.18456032719836402,
      "grad_norm": 3.3029637336730957,
      "learning_rate": 9.519946465617218e-05,
      "loss": 1.0764,
      "step": 361
    },
    {
      "epoch": 0.18507157464212678,
      "grad_norm": 3.6158411502838135,
      "learning_rate": 9.516321447139096e-05,
      "loss": 1.0236,
      "step": 362
    },
    {
      "epoch": 0.18558282208588958,
      "grad_norm": 3.52386736869812,
      "learning_rate": 9.512683488831802e-05,
      "loss": 1.0615,
      "step": 363
    },
    {
      "epoch": 0.18609406952965235,
      "grad_norm": 4.17709493637085,
      "learning_rate": 9.509032601118541e-05,
      "loss": 0.9873,
      "step": 364
    },
    {
      "epoch": 0.18660531697341512,
      "grad_norm": 3.8697471618652344,
      "learning_rate": 9.505368794459568e-05,
      "loss": 1.0644,
      "step": 365
    },
    {
      "epoch": 0.18711656441717792,
      "grad_norm": 3.6970717906951904,
      "learning_rate": 9.50169207935215e-05,
      "loss": 1.0334,
      "step": 366
    },
    {
      "epoch": 0.18762781186094069,
      "grad_norm": 3.9770524501800537,
      "learning_rate": 9.498002466330535e-05,
      "loss": 0.9283,
      "step": 367
    },
    {
      "epoch": 0.18813905930470348,
      "grad_norm": 3.809328317642212,
      "learning_rate": 9.494299965965933e-05,
      "loss": 0.9207,
      "step": 368
    },
    {
      "epoch": 0.18865030674846625,
      "grad_norm": 4.586056232452393,
      "learning_rate": 9.490584588866471e-05,
      "loss": 0.9965,
      "step": 369
    },
    {
      "epoch": 0.18916155419222905,
      "grad_norm": 4.179121494293213,
      "learning_rate": 9.486856345677173e-05,
      "loss": 0.921,
      "step": 370
    },
    {
      "epoch": 0.18967280163599182,
      "grad_norm": 4.585204124450684,
      "learning_rate": 9.483115247079924e-05,
      "loss": 0.95,
      "step": 371
    },
    {
      "epoch": 0.1901840490797546,
      "grad_norm": 4.396795749664307,
      "learning_rate": 9.47936130379344e-05,
      "loss": 1.0279,
      "step": 372
    },
    {
      "epoch": 0.19069529652351738,
      "grad_norm": 3.7172300815582275,
      "learning_rate": 9.475594526573245e-05,
      "loss": 1.0017,
      "step": 373
    },
    {
      "epoch": 0.19120654396728015,
      "grad_norm": 3.854496717453003,
      "learning_rate": 9.471814926211627e-05,
      "loss": 0.9455,
      "step": 374
    },
    {
      "epoch": 0.19171779141104295,
      "grad_norm": 4.243702411651611,
      "learning_rate": 9.468022513537617e-05,
      "loss": 0.9785,
      "step": 375
    },
    {
      "epoch": 0.19222903885480572,
      "grad_norm": 4.117717266082764,
      "learning_rate": 9.464217299416956e-05,
      "loss": 0.8867,
      "step": 376
    },
    {
      "epoch": 0.19274028629856851,
      "grad_norm": 4.680307865142822,
      "learning_rate": 9.460399294752061e-05,
      "loss": 1.0332,
      "step": 377
    },
    {
      "epoch": 0.19325153374233128,
      "grad_norm": 3.853069543838501,
      "learning_rate": 9.456568510481992e-05,
      "loss": 0.8719,
      "step": 378
    },
    {
      "epoch": 0.19376278118609408,
      "grad_norm": 4.585060119628906,
      "learning_rate": 9.452724957582433e-05,
      "loss": 0.8617,
      "step": 379
    },
    {
      "epoch": 0.19427402862985685,
      "grad_norm": 4.1220598220825195,
      "learning_rate": 9.448868647065642e-05,
      "loss": 0.9412,
      "step": 380
    },
    {
      "epoch": 0.19478527607361965,
      "grad_norm": 4.529792785644531,
      "learning_rate": 9.444999589980437e-05,
      "loss": 0.8709,
      "step": 381
    },
    {
      "epoch": 0.19529652351738241,
      "grad_norm": 4.1829447746276855,
      "learning_rate": 9.441117797412154e-05,
      "loss": 0.9142,
      "step": 382
    },
    {
      "epoch": 0.19580777096114518,
      "grad_norm": 4.374628067016602,
      "learning_rate": 9.437223280482613e-05,
      "loss": 1.0274,
      "step": 383
    },
    {
      "epoch": 0.19631901840490798,
      "grad_norm": 4.125802516937256,
      "learning_rate": 9.433316050350099e-05,
      "loss": 0.937,
      "step": 384
    },
    {
      "epoch": 0.19683026584867075,
      "grad_norm": 4.6389360427856445,
      "learning_rate": 9.429396118209316e-05,
      "loss": 1.1074,
      "step": 385
    },
    {
      "epoch": 0.19734151329243355,
      "grad_norm": 5.501649379730225,
      "learning_rate": 9.425463495291363e-05,
      "loss": 0.9531,
      "step": 386
    },
    {
      "epoch": 0.19785276073619631,
      "grad_norm": 4.594613552093506,
      "learning_rate": 9.421518192863701e-05,
      "loss": 0.8191,
      "step": 387
    },
    {
      "epoch": 0.1983640081799591,
      "grad_norm": 4.314398288726807,
      "learning_rate": 9.417560222230115e-05,
      "loss": 0.8347,
      "step": 388
    },
    {
      "epoch": 0.19887525562372188,
      "grad_norm": 4.536204814910889,
      "learning_rate": 9.413589594730692e-05,
      "loss": 0.9885,
      "step": 389
    },
    {
      "epoch": 0.19938650306748465,
      "grad_norm": 4.814162254333496,
      "learning_rate": 9.409606321741775e-05,
      "loss": 0.8417,
      "step": 390
    },
    {
      "epoch": 0.19989775051124745,
      "grad_norm": 5.067505359649658,
      "learning_rate": 9.405610414675948e-05,
      "loss": 0.9362,
      "step": 391
    },
    {
      "epoch": 0.20040899795501022,
      "grad_norm": 5.531371593475342,
      "learning_rate": 9.401601884981983e-05,
      "loss": 0.8284,
      "step": 392
    },
    {
      "epoch": 0.200920245398773,
      "grad_norm": 5.133208751678467,
      "learning_rate": 9.397580744144822e-05,
      "loss": 1.0076,
      "step": 393
    },
    {
      "epoch": 0.20143149284253578,
      "grad_norm": 5.163256645202637,
      "learning_rate": 9.393547003685543e-05,
      "loss": 0.8889,
      "step": 394
    },
    {
      "epoch": 0.20194274028629858,
      "grad_norm": 4.979630947113037,
      "learning_rate": 9.389500675161318e-05,
      "loss": 0.8315,
      "step": 395
    },
    {
      "epoch": 0.20245398773006135,
      "grad_norm": 5.355342864990234,
      "learning_rate": 9.385441770165385e-05,
      "loss": 0.822,
      "step": 396
    },
    {
      "epoch": 0.20296523517382414,
      "grad_norm": 5.970470428466797,
      "learning_rate": 9.381370300327021e-05,
      "loss": 0.8511,
      "step": 397
    },
    {
      "epoch": 0.2034764826175869,
      "grad_norm": 6.616512775421143,
      "learning_rate": 9.377286277311496e-05,
      "loss": 0.8692,
      "step": 398
    },
    {
      "epoch": 0.20398773006134968,
      "grad_norm": 6.648594856262207,
      "learning_rate": 9.373189712820055e-05,
      "loss": 0.8196,
      "step": 399
    },
    {
      "epoch": 0.20449897750511248,
      "grad_norm": 9.843427658081055,
      "learning_rate": 9.369080618589864e-05,
      "loss": 1.005,
      "step": 400
    },
    {
      "epoch": 0.20501022494887525,
      "grad_norm": 3.1759207248687744,
      "learning_rate": 9.364959006394002e-05,
      "loss": 0.9657,
      "step": 401
    },
    {
      "epoch": 0.20552147239263804,
      "grad_norm": 3.081962823867798,
      "learning_rate": 9.3608248880414e-05,
      "loss": 1.008,
      "step": 402
    },
    {
      "epoch": 0.2060327198364008,
      "grad_norm": 3.0837666988372803,
      "learning_rate": 9.356678275376832e-05,
      "loss": 1.0316,
      "step": 403
    },
    {
      "epoch": 0.2065439672801636,
      "grad_norm": 3.0529189109802246,
      "learning_rate": 9.35251918028086e-05,
      "loss": 1.086,
      "step": 404
    },
    {
      "epoch": 0.20705521472392638,
      "grad_norm": 3.318873405456543,
      "learning_rate": 9.34834761466982e-05,
      "loss": 1.0928,
      "step": 405
    },
    {
      "epoch": 0.20756646216768918,
      "grad_norm": 3.416926860809326,
      "learning_rate": 9.344163590495771e-05,
      "loss": 1.2316,
      "step": 406
    },
    {
      "epoch": 0.20807770961145194,
      "grad_norm": 3.2749698162078857,
      "learning_rate": 9.339967119746465e-05,
      "loss": 1.0556,
      "step": 407
    },
    {
      "epoch": 0.2085889570552147,
      "grad_norm": 3.5399835109710693,
      "learning_rate": 9.335758214445324e-05,
      "loss": 1.0681,
      "step": 408
    },
    {
      "epoch": 0.2091002044989775,
      "grad_norm": 3.271688222885132,
      "learning_rate": 9.331536886651387e-05,
      "loss": 0.903,
      "step": 409
    },
    {
      "epoch": 0.20961145194274028,
      "grad_norm": 3.6455349922180176,
      "learning_rate": 9.327303148459293e-05,
      "loss": 1.0631,
      "step": 410
    },
    {
      "epoch": 0.21012269938650308,
      "grad_norm": 3.230680227279663,
      "learning_rate": 9.32305701199923e-05,
      "loss": 1.0304,
      "step": 411
    },
    {
      "epoch": 0.21063394683026584,
      "grad_norm": 3.522308349609375,
      "learning_rate": 9.318798489436917e-05,
      "loss": 1.0796,
      "step": 412
    },
    {
      "epoch": 0.21114519427402864,
      "grad_norm": 3.4290144443511963,
      "learning_rate": 9.314527592973555e-05,
      "loss": 1.0312,
      "step": 413
    },
    {
      "epoch": 0.2116564417177914,
      "grad_norm": 3.476480007171631,
      "learning_rate": 9.310244334845801e-05,
      "loss": 0.9853,
      "step": 414
    },
    {
      "epoch": 0.21216768916155418,
      "grad_norm": 3.495586633682251,
      "learning_rate": 9.305948727325728e-05,
      "loss": 0.9321,
      "step": 415
    },
    {
      "epoch": 0.21267893660531698,
      "grad_norm": 3.8010475635528564,
      "learning_rate": 9.301640782720792e-05,
      "loss": 0.9269,
      "step": 416
    },
    {
      "epoch": 0.21319018404907975,
      "grad_norm": 3.931910276412964,
      "learning_rate": 9.297320513373795e-05,
      "loss": 1.0371,
      "step": 417
    },
    {
      "epoch": 0.21370143149284254,
      "grad_norm": 3.3684935569763184,
      "learning_rate": 9.292987931662855e-05,
      "loss": 0.9407,
      "step": 418
    },
    {
      "epoch": 0.2142126789366053,
      "grad_norm": 3.5472538471221924,
      "learning_rate": 9.288643050001361e-05,
      "loss": 0.9555,
      "step": 419
    },
    {
      "epoch": 0.2147239263803681,
      "grad_norm": 4.427033424377441,
      "learning_rate": 9.284285880837946e-05,
      "loss": 0.9743,
      "step": 420
    },
    {
      "epoch": 0.21523517382413088,
      "grad_norm": 3.681061267852783,
      "learning_rate": 9.279916436656451e-05,
      "loss": 0.8528,
      "step": 421
    },
    {
      "epoch": 0.21574642126789367,
      "grad_norm": 3.8058080673217773,
      "learning_rate": 9.275534729975879e-05,
      "loss": 0.9228,
      "step": 422
    },
    {
      "epoch": 0.21625766871165644,
      "grad_norm": 3.4031405448913574,
      "learning_rate": 9.271140773350373e-05,
      "loss": 0.8928,
      "step": 423
    },
    {
      "epoch": 0.2167689161554192,
      "grad_norm": 3.6969170570373535,
      "learning_rate": 9.266734579369172e-05,
      "loss": 0.9488,
      "step": 424
    },
    {
      "epoch": 0.217280163599182,
      "grad_norm": 3.792942762374878,
      "learning_rate": 9.262316160656575e-05,
      "loss": 0.9386,
      "step": 425
    },
    {
      "epoch": 0.21779141104294478,
      "grad_norm": 4.491650581359863,
      "learning_rate": 9.257885529871908e-05,
      "loss": 0.9395,
      "step": 426
    },
    {
      "epoch": 0.21830265848670757,
      "grad_norm": 3.8580541610717773,
      "learning_rate": 9.253442699709485e-05,
      "loss": 0.9405,
      "step": 427
    },
    {
      "epoch": 0.21881390593047034,
      "grad_norm": 4.1208696365356445,
      "learning_rate": 9.248987682898575e-05,
      "loss": 0.9959,
      "step": 428
    },
    {
      "epoch": 0.21932515337423314,
      "grad_norm": 4.131077766418457,
      "learning_rate": 9.24452049220336e-05,
      "loss": 1.0929,
      "step": 429
    },
    {
      "epoch": 0.2198364008179959,
      "grad_norm": 4.098513603210449,
      "learning_rate": 9.240041140422905e-05,
      "loss": 0.8605,
      "step": 430
    },
    {
      "epoch": 0.22034764826175868,
      "grad_norm": 3.7456324100494385,
      "learning_rate": 9.235549640391115e-05,
      "loss": 0.88,
      "step": 431
    },
    {
      "epoch": 0.22085889570552147,
      "grad_norm": 4.278169631958008,
      "learning_rate": 9.231046004976704e-05,
      "loss": 1.0368,
      "step": 432
    },
    {
      "epoch": 0.22137014314928424,
      "grad_norm": 4.015793800354004,
      "learning_rate": 9.226530247083153e-05,
      "loss": 1.0141,
      "step": 433
    },
    {
      "epoch": 0.22188139059304704,
      "grad_norm": 5.274085998535156,
      "learning_rate": 9.222002379648675e-05,
      "loss": 0.8493,
      "step": 434
    },
    {
      "epoch": 0.2223926380368098,
      "grad_norm": 4.8212432861328125,
      "learning_rate": 9.217462415646185e-05,
      "loss": 0.9706,
      "step": 435
    },
    {
      "epoch": 0.2229038854805726,
      "grad_norm": 3.999441385269165,
      "learning_rate": 9.212910368083245e-05,
      "loss": 0.8746,
      "step": 436
    },
    {
      "epoch": 0.22341513292433537,
      "grad_norm": 3.9513213634490967,
      "learning_rate": 9.208346250002049e-05,
      "loss": 0.8408,
      "step": 437
    },
    {
      "epoch": 0.22392638036809817,
      "grad_norm": 4.810235977172852,
      "learning_rate": 9.203770074479367e-05,
      "loss": 0.9253,
      "step": 438
    },
    {
      "epoch": 0.22443762781186094,
      "grad_norm": 5.90993070602417,
      "learning_rate": 9.199181854626517e-05,
      "loss": 0.9008,
      "step": 439
    },
    {
      "epoch": 0.2249488752556237,
      "grad_norm": 5.908392906188965,
      "learning_rate": 9.194581603589328e-05,
      "loss": 0.9596,
      "step": 440
    },
    {
      "epoch": 0.2254601226993865,
      "grad_norm": 4.973458766937256,
      "learning_rate": 9.189969334548096e-05,
      "loss": 1.0126,
      "step": 441
    },
    {
      "epoch": 0.22597137014314927,
      "grad_norm": 5.1067962646484375,
      "learning_rate": 9.185345060717554e-05,
      "loss": 0.8453,
      "step": 442
    },
    {
      "epoch": 0.22648261758691207,
      "grad_norm": 5.444033622741699,
      "learning_rate": 9.180708795346829e-05,
      "loss": 0.7763,
      "step": 443
    },
    {
      "epoch": 0.22699386503067484,
      "grad_norm": 5.796935558319092,
      "learning_rate": 9.1760605517194e-05,
      "loss": 1.0804,
      "step": 444
    },
    {
      "epoch": 0.22750511247443764,
      "grad_norm": 5.549903869628906,
      "learning_rate": 9.171400343153076e-05,
      "loss": 1.0069,
      "step": 445
    },
    {
      "epoch": 0.2280163599182004,
      "grad_norm": 5.64030122756958,
      "learning_rate": 9.166728182999936e-05,
      "loss": 0.8366,
      "step": 446
    },
    {
      "epoch": 0.2285276073619632,
      "grad_norm": 6.40301513671875,
      "learning_rate": 9.16204408464631e-05,
      "loss": 0.9099,
      "step": 447
    },
    {
      "epoch": 0.22903885480572597,
      "grad_norm": 6.116087913513184,
      "learning_rate": 9.157348061512727e-05,
      "loss": 0.7831,
      "step": 448
    },
    {
      "epoch": 0.22955010224948874,
      "grad_norm": 7.226142406463623,
      "learning_rate": 9.152640127053884e-05,
      "loss": 0.8518,
      "step": 449
    },
    {
      "epoch": 0.23006134969325154,
      "grad_norm": 8.906120300292969,
      "learning_rate": 9.147920294758607e-05,
      "loss": 0.6824,
      "step": 450
    },
    {
      "epoch": 0.2305725971370143,
      "grad_norm": 2.9229636192321777,
      "learning_rate": 9.143188578149809e-05,
      "loss": 1.0234,
      "step": 451
    },
    {
      "epoch": 0.2310838445807771,
      "grad_norm": 3.875776529312134,
      "learning_rate": 9.138444990784453e-05,
      "loss": 1.0901,
      "step": 452
    },
    {
      "epoch": 0.23159509202453987,
      "grad_norm": 3.7210988998413086,
      "learning_rate": 9.133689546253514e-05,
      "loss": 1.0057,
      "step": 453
    },
    {
      "epoch": 0.23210633946830267,
      "grad_norm": 3.645733118057251,
      "learning_rate": 9.128922258181938e-05,
      "loss": 0.9967,
      "step": 454
    },
    {
      "epoch": 0.23261758691206544,
      "grad_norm": 3.363543748855591,
      "learning_rate": 9.124143140228608e-05,
      "loss": 1.0951,
      "step": 455
    },
    {
      "epoch": 0.2331288343558282,
      "grad_norm": 3.1317014694213867,
      "learning_rate": 9.119352206086293e-05,
      "loss": 1.0259,
      "step": 456
    },
    {
      "epoch": 0.233640081799591,
      "grad_norm": 3.3655569553375244,
      "learning_rate": 9.114549469481626e-05,
      "loss": 1.119,
      "step": 457
    },
    {
      "epoch": 0.23415132924335377,
      "grad_norm": 3.224952220916748,
      "learning_rate": 9.10973494417505e-05,
      "loss": 1.0946,
      "step": 458
    },
    {
      "epoch": 0.23466257668711657,
      "grad_norm": 3.7233197689056396,
      "learning_rate": 9.104908643960787e-05,
      "loss": 0.9717,
      "step": 459
    },
    {
      "epoch": 0.23517382413087934,
      "grad_norm": 3.78554630279541,
      "learning_rate": 9.100070582666795e-05,
      "loss": 1.049,
      "step": 460
    },
    {
      "epoch": 0.23568507157464214,
      "grad_norm": 3.5677905082702637,
      "learning_rate": 9.095220774154726e-05,
      "loss": 0.9873,
      "step": 461
    },
    {
      "epoch": 0.2361963190184049,
      "grad_norm": 3.1176674365997314,
      "learning_rate": 9.090359232319893e-05,
      "loss": 0.8864,
      "step": 462
    },
    {
      "epoch": 0.2367075664621677,
      "grad_norm": 3.3484392166137695,
      "learning_rate": 9.085485971091225e-05,
      "loss": 0.9754,
      "step": 463
    },
    {
      "epoch": 0.23721881390593047,
      "grad_norm": 3.456024408340454,
      "learning_rate": 9.080601004431229e-05,
      "loss": 0.8267,
      "step": 464
    },
    {
      "epoch": 0.23773006134969324,
      "grad_norm": 3.6923391819000244,
      "learning_rate": 9.075704346335947e-05,
      "loss": 0.916,
      "step": 465
    },
    {
      "epoch": 0.23824130879345604,
      "grad_norm": 3.570155143737793,
      "learning_rate": 9.070796010834923e-05,
      "loss": 0.9735,
      "step": 466
    },
    {
      "epoch": 0.2387525562372188,
      "grad_norm": 4.115891456604004,
      "learning_rate": 9.065876011991155e-05,
      "loss": 1.0247,
      "step": 467
    },
    {
      "epoch": 0.2392638036809816,
      "grad_norm": 3.4448201656341553,
      "learning_rate": 9.060944363901056e-05,
      "loss": 0.9963,
      "step": 468
    },
    {
      "epoch": 0.23977505112474437,
      "grad_norm": 3.9490420818328857,
      "learning_rate": 9.056001080694423e-05,
      "loss": 0.9454,
      "step": 469
    },
    {
      "epoch": 0.24028629856850717,
      "grad_norm": 3.538975954055786,
      "learning_rate": 9.05104617653438e-05,
      "loss": 0.9077,
      "step": 470
    },
    {
      "epoch": 0.24079754601226994,
      "grad_norm": 3.586324453353882,
      "learning_rate": 9.046079665617354e-05,
      "loss": 0.9294,
      "step": 471
    },
    {
      "epoch": 0.24130879345603273,
      "grad_norm": 3.7943360805511475,
      "learning_rate": 9.041101562173023e-05,
      "loss": 0.9706,
      "step": 472
    },
    {
      "epoch": 0.2418200408997955,
      "grad_norm": 3.871798276901245,
      "learning_rate": 9.036111880464277e-05,
      "loss": 0.9572,
      "step": 473
    },
    {
      "epoch": 0.24233128834355827,
      "grad_norm": 3.678107976913452,
      "learning_rate": 9.031110634787186e-05,
      "loss": 0.9078,
      "step": 474
    },
    {
      "epoch": 0.24284253578732107,
      "grad_norm": 3.8214452266693115,
      "learning_rate": 9.026097839470947e-05,
      "loss": 0.899,
      "step": 475
    },
    {
      "epoch": 0.24335378323108384,
      "grad_norm": 4.086860656738281,
      "learning_rate": 9.021073508877845e-05,
      "loss": 0.8796,
      "step": 476
    },
    {
      "epoch": 0.24386503067484663,
      "grad_norm": 3.7152156829833984,
      "learning_rate": 9.016037657403224e-05,
      "loss": 0.8959,
      "step": 477
    },
    {
      "epoch": 0.2443762781186094,
      "grad_norm": 4.333080291748047,
      "learning_rate": 9.010990299475433e-05,
      "loss": 0.8454,
      "step": 478
    },
    {
      "epoch": 0.2448875255623722,
      "grad_norm": 3.956611156463623,
      "learning_rate": 9.005931449555782e-05,
      "loss": 1.0178,
      "step": 479
    },
    {
      "epoch": 0.24539877300613497,
      "grad_norm": 3.650317430496216,
      "learning_rate": 9.000861122138517e-05,
      "loss": 0.8329,
      "step": 480
    },
    {
      "epoch": 0.24591002044989774,
      "grad_norm": 4.062465190887451,
      "learning_rate": 8.995779331750764e-05,
      "loss": 0.9728,
      "step": 481
    },
    {
      "epoch": 0.24642126789366053,
      "grad_norm": 4.255726337432861,
      "learning_rate": 8.99068609295249e-05,
      "loss": 0.9331,
      "step": 482
    },
    {
      "epoch": 0.2469325153374233,
      "grad_norm": 4.312367916107178,
      "learning_rate": 8.985581420336465e-05,
      "loss": 0.9291,
      "step": 483
    },
    {
      "epoch": 0.2474437627811861,
      "grad_norm": 4.059680461883545,
      "learning_rate": 8.980465328528219e-05,
      "loss": 0.9143,
      "step": 484
    },
    {
      "epoch": 0.24795501022494887,
      "grad_norm": 4.124285697937012,
      "learning_rate": 8.975337832186e-05,
      "loss": 0.8942,
      "step": 485
    },
    {
      "epoch": 0.24846625766871167,
      "grad_norm": 4.120577335357666,
      "learning_rate": 8.970198946000727e-05,
      "loss": 0.8887,
      "step": 486
    },
    {
      "epoch": 0.24897750511247443,
      "grad_norm": 4.375182628631592,
      "learning_rate": 8.965048684695955e-05,
      "loss": 0.9409,
      "step": 487
    },
    {
      "epoch": 0.24948875255623723,
      "grad_norm": 4.595582962036133,
      "learning_rate": 8.959887063027837e-05,
      "loss": 0.8955,
      "step": 488
    },
    {
      "epoch": 0.25,
      "grad_norm": 4.253859519958496,
      "learning_rate": 8.954714095785062e-05,
      "loss": 0.8379,
      "step": 489
    },
    {
      "epoch": 0.2505112474437628,
      "grad_norm": 4.908374309539795,
      "learning_rate": 8.949529797788836e-05,
      "loss": 0.9577,
      "step": 490
    },
    {
      "epoch": 0.25102249488752554,
      "grad_norm": 4.503420829772949,
      "learning_rate": 8.944334183892822e-05,
      "loss": 0.8906,
      "step": 491
    },
    {
      "epoch": 0.25153374233128833,
      "grad_norm": 6.037133693695068,
      "learning_rate": 8.939127268983108e-05,
      "loss": 0.95,
      "step": 492
    },
    {
      "epoch": 0.25204498977505113,
      "grad_norm": 4.602914333343506,
      "learning_rate": 8.933909067978163e-05,
      "loss": 0.8129,
      "step": 493
    },
    {
      "epoch": 0.25255623721881393,
      "grad_norm": 5.256838321685791,
      "learning_rate": 8.928679595828786e-05,
      "loss": 0.8297,
      "step": 494
    },
    {
      "epoch": 0.25306748466257667,
      "grad_norm": 5.184623718261719,
      "learning_rate": 8.923438867518075e-05,
      "loss": 0.9879,
      "step": 495
    },
    {
      "epoch": 0.25357873210633947,
      "grad_norm": 5.045862674713135,
      "learning_rate": 8.918186898061376e-05,
      "loss": 0.911,
      "step": 496
    },
    {
      "epoch": 0.25408997955010226,
      "grad_norm": 5.845618724822998,
      "learning_rate": 8.912923702506241e-05,
      "loss": 0.8597,
      "step": 497
    },
    {
      "epoch": 0.254601226993865,
      "grad_norm": 6.090303421020508,
      "learning_rate": 8.907649295932387e-05,
      "loss": 0.8402,
      "step": 498
    },
    {
      "epoch": 0.2551124744376278,
      "grad_norm": 6.933328628540039,
      "learning_rate": 8.902363693451654e-05,
      "loss": 0.92,
      "step": 499
    },
    {
      "epoch": 0.2556237218813906,
      "grad_norm": 8.922635078430176,
      "learning_rate": 8.897066910207958e-05,
      "loss": 0.9011,
      "step": 500
    },
    {
      "epoch": 0.2561349693251534,
      "grad_norm": 2.395228147506714,
      "learning_rate": 8.89175896137725e-05,
      "loss": 0.8274,
      "step": 501
    },
    {
      "epoch": 0.25664621676891614,
      "grad_norm": 2.550231695175171,
      "learning_rate": 8.886439862167469e-05,
      "loss": 0.9732,
      "step": 502
    },
    {
      "epoch": 0.25715746421267893,
      "grad_norm": 3.1747238636016846,
      "learning_rate": 8.881109627818504e-05,
      "loss": 1.1108,
      "step": 503
    },
    {
      "epoch": 0.25766871165644173,
      "grad_norm": 3.0045149326324463,
      "learning_rate": 8.875768273602148e-05,
      "loss": 1.0678,
      "step": 504
    },
    {
      "epoch": 0.2581799591002045,
      "grad_norm": 2.9451730251312256,
      "learning_rate": 8.870415814822052e-05,
      "loss": 1.025,
      "step": 505
    },
    {
      "epoch": 0.25869120654396727,
      "grad_norm": 3.0222766399383545,
      "learning_rate": 8.865052266813685e-05,
      "loss": 1.1336,
      "step": 506
    },
    {
      "epoch": 0.25920245398773006,
      "grad_norm": 3.1302473545074463,
      "learning_rate": 8.859677644944287e-05,
      "loss": 1.088,
      "step": 507
    },
    {
      "epoch": 0.25971370143149286,
      "grad_norm": 3.14927339553833,
      "learning_rate": 8.854291964612825e-05,
      "loss": 1.0584,
      "step": 508
    },
    {
      "epoch": 0.2602249488752556,
      "grad_norm": 3.135263204574585,
      "learning_rate": 8.848895241249949e-05,
      "loss": 1.0483,
      "step": 509
    },
    {
      "epoch": 0.2607361963190184,
      "grad_norm": 3.1471927165985107,
      "learning_rate": 8.843487490317954e-05,
      "loss": 1.0553,
      "step": 510
    },
    {
      "epoch": 0.2612474437627812,
      "grad_norm": 2.991029739379883,
      "learning_rate": 8.838068727310724e-05,
      "loss": 0.9881,
      "step": 511
    },
    {
      "epoch": 0.261758691206544,
      "grad_norm": 3.197028398513794,
      "learning_rate": 8.832638967753699e-05,
      "loss": 0.9918,
      "step": 512
    },
    {
      "epoch": 0.26226993865030673,
      "grad_norm": 3.349247694015503,
      "learning_rate": 8.827198227203821e-05,
      "loss": 0.9742,
      "step": 513
    },
    {
      "epoch": 0.26278118609406953,
      "grad_norm": 3.336989164352417,
      "learning_rate": 8.821746521249499e-05,
      "loss": 0.9569,
      "step": 514
    },
    {
      "epoch": 0.2632924335378323,
      "grad_norm": 3.321845531463623,
      "learning_rate": 8.816283865510554e-05,
      "loss": 0.9315,
      "step": 515
    },
    {
      "epoch": 0.26380368098159507,
      "grad_norm": 3.3562281131744385,
      "learning_rate": 8.810810275638183e-05,
      "loss": 1.0072,
      "step": 516
    },
    {
      "epoch": 0.26431492842535786,
      "grad_norm": 3.2323803901672363,
      "learning_rate": 8.805325767314909e-05,
      "loss": 0.9317,
      "step": 517
    },
    {
      "epoch": 0.26482617586912066,
      "grad_norm": 3.820136547088623,
      "learning_rate": 8.799830356254539e-05,
      "loss": 0.9336,
      "step": 518
    },
    {
      "epoch": 0.26533742331288346,
      "grad_norm": 3.5097525119781494,
      "learning_rate": 8.794324058202116e-05,
      "loss": 0.9786,
      "step": 519
    },
    {
      "epoch": 0.2658486707566462,
      "grad_norm": 3.6844258308410645,
      "learning_rate": 8.788806888933881e-05,
      "loss": 0.975,
      "step": 520
    },
    {
      "epoch": 0.266359918200409,
      "grad_norm": 3.6815667152404785,
      "learning_rate": 8.783278864257211e-05,
      "loss": 0.9411,
      "step": 521
    },
    {
      "epoch": 0.2668711656441718,
      "grad_norm": 3.773531675338745,
      "learning_rate": 8.777740000010599e-05,
      "loss": 0.9664,
      "step": 522
    },
    {
      "epoch": 0.26738241308793453,
      "grad_norm": 3.854552745819092,
      "learning_rate": 8.772190312063583e-05,
      "loss": 0.9537,
      "step": 523
    },
    {
      "epoch": 0.26789366053169733,
      "grad_norm": 4.011033058166504,
      "learning_rate": 8.766629816316721e-05,
      "loss": 0.9038,
      "step": 524
    },
    {
      "epoch": 0.2684049079754601,
      "grad_norm": 3.57070255279541,
      "learning_rate": 8.761058528701529e-05,
      "loss": 0.8889,
      "step": 525
    },
    {
      "epoch": 0.2689161554192229,
      "grad_norm": 3.7002241611480713,
      "learning_rate": 8.75547646518045e-05,
      "loss": 0.9296,
      "step": 526
    },
    {
      "epoch": 0.26942740286298567,
      "grad_norm": 3.872769594192505,
      "learning_rate": 8.749883641746798e-05,
      "loss": 0.9211,
      "step": 527
    },
    {
      "epoch": 0.26993865030674846,
      "grad_norm": 4.013867378234863,
      "learning_rate": 8.744280074424713e-05,
      "loss": 1.0296,
      "step": 528
    },
    {
      "epoch": 0.27044989775051126,
      "grad_norm": 3.432490587234497,
      "learning_rate": 8.738665779269125e-05,
      "loss": 0.7884,
      "step": 529
    },
    {
      "epoch": 0.27096114519427406,
      "grad_norm": 3.7517478466033936,
      "learning_rate": 8.733040772365693e-05,
      "loss": 0.8918,
      "step": 530
    },
    {
      "epoch": 0.2714723926380368,
      "grad_norm": 3.8097121715545654,
      "learning_rate": 8.72740506983077e-05,
      "loss": 1.0147,
      "step": 531
    },
    {
      "epoch": 0.2719836400817996,
      "grad_norm": 4.307250022888184,
      "learning_rate": 8.721758687811352e-05,
      "loss": 0.9346,
      "step": 532
    },
    {
      "epoch": 0.2724948875255624,
      "grad_norm": 4.444828033447266,
      "learning_rate": 8.716101642485035e-05,
      "loss": 1.0747,
      "step": 533
    },
    {
      "epoch": 0.27300613496932513,
      "grad_norm": 4.189756393432617,
      "learning_rate": 8.710433950059967e-05,
      "loss": 0.9034,
      "step": 534
    },
    {
      "epoch": 0.27351738241308793,
      "grad_norm": 4.013859748840332,
      "learning_rate": 8.704755626774796e-05,
      "loss": 0.9403,
      "step": 535
    },
    {
      "epoch": 0.2740286298568507,
      "grad_norm": 4.528006553649902,
      "learning_rate": 8.699066688898635e-05,
      "loss": 0.9282,
      "step": 536
    },
    {
      "epoch": 0.2745398773006135,
      "grad_norm": 4.314260482788086,
      "learning_rate": 8.69336715273101e-05,
      "loss": 0.9731,
      "step": 537
    },
    {
      "epoch": 0.27505112474437626,
      "grad_norm": 4.596576690673828,
      "learning_rate": 8.687657034601801e-05,
      "loss": 0.8781,
      "step": 538
    },
    {
      "epoch": 0.27556237218813906,
      "grad_norm": 4.438699245452881,
      "learning_rate": 8.681936350871222e-05,
      "loss": 0.8967,
      "step": 539
    },
    {
      "epoch": 0.27607361963190186,
      "grad_norm": 5.038392543792725,
      "learning_rate": 8.676205117929752e-05,
      "loss": 1.0196,
      "step": 540
    },
    {
      "epoch": 0.2765848670756646,
      "grad_norm": 4.725849628448486,
      "learning_rate": 8.670463352198088e-05,
      "loss": 0.903,
      "step": 541
    },
    {
      "epoch": 0.2770961145194274,
      "grad_norm": 4.913168907165527,
      "learning_rate": 8.664711070127117e-05,
      "loss": 0.9673,
      "step": 542
    },
    {
      "epoch": 0.2776073619631902,
      "grad_norm": 4.891239166259766,
      "learning_rate": 8.658948288197848e-05,
      "loss": 0.9758,
      "step": 543
    },
    {
      "epoch": 0.278118609406953,
      "grad_norm": 4.979687213897705,
      "learning_rate": 8.65317502292138e-05,
      "loss": 0.9457,
      "step": 544
    },
    {
      "epoch": 0.27862985685071573,
      "grad_norm": 5.6750359535217285,
      "learning_rate": 8.647391290838838e-05,
      "loss": 0.8713,
      "step": 545
    },
    {
      "epoch": 0.2791411042944785,
      "grad_norm": 5.564166069030762,
      "learning_rate": 8.641597108521348e-05,
      "loss": 1.0865,
      "step": 546
    },
    {
      "epoch": 0.2796523517382413,
      "grad_norm": 5.280520439147949,
      "learning_rate": 8.635792492569967e-05,
      "loss": 0.8019,
      "step": 547
    },
    {
      "epoch": 0.28016359918200406,
      "grad_norm": 6.287930965423584,
      "learning_rate": 8.629977459615655e-05,
      "loss": 0.8003,
      "step": 548
    },
    {
      "epoch": 0.28067484662576686,
      "grad_norm": 6.9441914558410645,
      "learning_rate": 8.624152026319208e-05,
      "loss": 0.8961,
      "step": 549
    },
    {
      "epoch": 0.28118609406952966,
      "grad_norm": 8.341976165771484,
      "learning_rate": 8.618316209371228e-05,
      "loss": 0.6323,
      "step": 550
    },
    {
      "epoch": 0.28169734151329245,
      "grad_norm": 2.6429460048675537,
      "learning_rate": 8.612470025492065e-05,
      "loss": 1.0268,
      "step": 551
    },
    {
      "epoch": 0.2822085889570552,
      "grad_norm": 2.9558145999908447,
      "learning_rate": 8.60661349143177e-05,
      "loss": 0.9619,
      "step": 552
    },
    {
      "epoch": 0.282719836400818,
      "grad_norm": 2.9567391872406006,
      "learning_rate": 8.600746623970051e-05,
      "loss": 0.9782,
      "step": 553
    },
    {
      "epoch": 0.2832310838445808,
      "grad_norm": 2.8197598457336426,
      "learning_rate": 8.59486943991622e-05,
      "loss": 0.9863,
      "step": 554
    },
    {
      "epoch": 0.2837423312883436,
      "grad_norm": 2.895512819290161,
      "learning_rate": 8.588981956109151e-05,
      "loss": 0.9764,
      "step": 555
    },
    {
      "epoch": 0.2842535787321063,
      "grad_norm": 2.976269245147705,
      "learning_rate": 8.583084189417224e-05,
      "loss": 1.1054,
      "step": 556
    },
    {
      "epoch": 0.2847648261758691,
      "grad_norm": 2.900879144668579,
      "learning_rate": 8.577176156738282e-05,
      "loss": 1.0042,
      "step": 557
    },
    {
      "epoch": 0.2852760736196319,
      "grad_norm": 3.3032069206237793,
      "learning_rate": 8.571257874999585e-05,
      "loss": 1.0139,
      "step": 558
    },
    {
      "epoch": 0.28578732106339466,
      "grad_norm": 3.3368518352508545,
      "learning_rate": 8.565329361157752e-05,
      "loss": 1.037,
      "step": 559
    },
    {
      "epoch": 0.28629856850715746,
      "grad_norm": 2.9615988731384277,
      "learning_rate": 8.559390632198723e-05,
      "loss": 0.916,
      "step": 560
    },
    {
      "epoch": 0.28680981595092025,
      "grad_norm": 3.3652186393737793,
      "learning_rate": 8.553441705137702e-05,
      "loss": 1.0821,
      "step": 561
    },
    {
      "epoch": 0.28732106339468305,
      "grad_norm": 3.1181633472442627,
      "learning_rate": 8.547482597019114e-05,
      "loss": 0.9685,
      "step": 562
    },
    {
      "epoch": 0.2878323108384458,
      "grad_norm": 3.476203203201294,
      "learning_rate": 8.541513324916555e-05,
      "loss": 0.9567,
      "step": 563
    },
    {
      "epoch": 0.2883435582822086,
      "grad_norm": 3.4987919330596924,
      "learning_rate": 8.535533905932738e-05,
      "loss": 1.0481,
      "step": 564
    },
    {
      "epoch": 0.2888548057259714,
      "grad_norm": 3.677459478378296,
      "learning_rate": 8.529544357199453e-05,
      "loss": 0.9656,
      "step": 565
    },
    {
      "epoch": 0.2893660531697341,
      "grad_norm": 3.3879263401031494,
      "learning_rate": 8.523544695877508e-05,
      "loss": 0.9833,
      "step": 566
    },
    {
      "epoch": 0.2898773006134969,
      "grad_norm": 3.864903688430786,
      "learning_rate": 8.51753493915669e-05,
      "loss": 0.9329,
      "step": 567
    },
    {
      "epoch": 0.2903885480572597,
      "grad_norm": 4.15344762802124,
      "learning_rate": 8.51151510425571e-05,
      "loss": 1.049,
      "step": 568
    },
    {
      "epoch": 0.2908997955010225,
      "grad_norm": 3.7756378650665283,
      "learning_rate": 8.505485208422146e-05,
      "loss": 0.9141,
      "step": 569
    },
    {
      "epoch": 0.29141104294478526,
      "grad_norm": 3.6224896907806396,
      "learning_rate": 8.499445268932413e-05,
      "loss": 0.9635,
      "step": 570
    },
    {
      "epoch": 0.29192229038854806,
      "grad_norm": 3.796037197113037,
      "learning_rate": 8.493395303091697e-05,
      "loss": 1.0782,
      "step": 571
    },
    {
      "epoch": 0.29243353783231085,
      "grad_norm": 3.62247371673584,
      "learning_rate": 8.487335328233912e-05,
      "loss": 0.9458,
      "step": 572
    },
    {
      "epoch": 0.2929447852760736,
      "grad_norm": 3.6064841747283936,
      "learning_rate": 8.481265361721645e-05,
      "loss": 0.877,
      "step": 573
    },
    {
      "epoch": 0.2934560327198364,
      "grad_norm": 3.84397292137146,
      "learning_rate": 8.475185420946118e-05,
      "loss": 1.0055,
      "step": 574
    },
    {
      "epoch": 0.2939672801635992,
      "grad_norm": 3.980118751525879,
      "learning_rate": 8.469095523327124e-05,
      "loss": 0.9517,
      "step": 575
    },
    {
      "epoch": 0.294478527607362,
      "grad_norm": 3.9962069988250732,
      "learning_rate": 8.462995686312985e-05,
      "loss": 0.9557,
      "step": 576
    },
    {
      "epoch": 0.2949897750511247,
      "grad_norm": 3.9836580753326416,
      "learning_rate": 8.456885927380504e-05,
      "loss": 1.0057,
      "step": 577
    },
    {
      "epoch": 0.2955010224948875,
      "grad_norm": 3.9834954738616943,
      "learning_rate": 8.450766264034907e-05,
      "loss": 0.9853,
      "step": 578
    },
    {
      "epoch": 0.2960122699386503,
      "grad_norm": 3.7003509998321533,
      "learning_rate": 8.444636713809801e-05,
      "loss": 0.8674,
      "step": 579
    },
    {
      "epoch": 0.2965235173824131,
      "grad_norm": 3.7649359703063965,
      "learning_rate": 8.438497294267117e-05,
      "loss": 0.9366,
      "step": 580
    },
    {
      "epoch": 0.29703476482617586,
      "grad_norm": 4.7126994132995605,
      "learning_rate": 8.432348022997066e-05,
      "loss": 0.911,
      "step": 581
    },
    {
      "epoch": 0.29754601226993865,
      "grad_norm": 3.622835159301758,
      "learning_rate": 8.426188917618083e-05,
      "loss": 0.8138,
      "step": 582
    },
    {
      "epoch": 0.29805725971370145,
      "grad_norm": 3.986794948577881,
      "learning_rate": 8.420019995776779e-05,
      "loss": 0.978,
      "step": 583
    },
    {
      "epoch": 0.2985685071574642,
      "grad_norm": 4.0647478103637695,
      "learning_rate": 8.413841275147892e-05,
      "loss": 0.9557,
      "step": 584
    },
    {
      "epoch": 0.299079754601227,
      "grad_norm": 4.3159637451171875,
      "learning_rate": 8.407652773434236e-05,
      "loss": 0.8454,
      "step": 585
    },
    {
      "epoch": 0.2995910020449898,
      "grad_norm": 4.256467342376709,
      "learning_rate": 8.401454508366643e-05,
      "loss": 0.9827,
      "step": 586
    },
    {
      "epoch": 0.3001022494887526,
      "grad_norm": 4.750210762023926,
      "learning_rate": 8.395246497703925e-05,
      "loss": 0.8911,
      "step": 587
    },
    {
      "epoch": 0.3006134969325153,
      "grad_norm": 4.348392963409424,
      "learning_rate": 8.389028759232815e-05,
      "loss": 0.8436,
      "step": 588
    },
    {
      "epoch": 0.3011247443762781,
      "grad_norm": 4.388984680175781,
      "learning_rate": 8.382801310767912e-05,
      "loss": 0.9827,
      "step": 589
    },
    {
      "epoch": 0.3016359918200409,
      "grad_norm": 4.420820236206055,
      "learning_rate": 8.376564170151642e-05,
      "loss": 0.896,
      "step": 590
    },
    {
      "epoch": 0.30214723926380366,
      "grad_norm": 4.915138244628906,
      "learning_rate": 8.370317355254197e-05,
      "loss": 0.8738,
      "step": 591
    },
    {
      "epoch": 0.30265848670756645,
      "grad_norm": 5.0341644287109375,
      "learning_rate": 8.364060883973489e-05,
      "loss": 0.8735,
      "step": 592
    },
    {
      "epoch": 0.30316973415132925,
      "grad_norm": 4.664186000823975,
      "learning_rate": 8.357794774235092e-05,
      "loss": 0.82,
      "step": 593
    },
    {
      "epoch": 0.30368098159509205,
      "grad_norm": 5.188631057739258,
      "learning_rate": 8.351519043992201e-05,
      "loss": 0.8746,
      "step": 594
    },
    {
      "epoch": 0.3041922290388548,
      "grad_norm": 5.177921295166016,
      "learning_rate": 8.345233711225574e-05,
      "loss": 0.8297,
      "step": 595
    },
    {
      "epoch": 0.3047034764826176,
      "grad_norm": 5.583620548248291,
      "learning_rate": 8.338938793943478e-05,
      "loss": 1.0842,
      "step": 596
    },
    {
      "epoch": 0.3052147239263804,
      "grad_norm": 6.216187953948975,
      "learning_rate": 8.332634310181644e-05,
      "loss": 0.9047,
      "step": 597
    },
    {
      "epoch": 0.3057259713701431,
      "grad_norm": 5.41096305847168,
      "learning_rate": 8.326320278003211e-05,
      "loss": 0.6631,
      "step": 598
    },
    {
      "epoch": 0.3062372188139059,
      "grad_norm": 7.021457672119141,
      "learning_rate": 8.319996715498675e-05,
      "loss": 0.8118,
      "step": 599
    },
    {
      "epoch": 0.3067484662576687,
      "grad_norm": 9.457502365112305,
      "learning_rate": 8.313663640785839e-05,
      "loss": 0.5443,
      "step": 600
    },
    {
      "epoch": 0.3072597137014315,
      "grad_norm": 2.781672239303589,
      "learning_rate": 8.307321072009759e-05,
      "loss": 1.0299,
      "step": 601
    },
    {
      "epoch": 0.30777096114519426,
      "grad_norm": 3.157245635986328,
      "learning_rate": 8.300969027342692e-05,
      "loss": 1.0549,
      "step": 602
    },
    {
      "epoch": 0.30828220858895705,
      "grad_norm": 3.0656280517578125,
      "learning_rate": 8.294607524984045e-05,
      "loss": 1.1016,
      "step": 603
    },
    {
      "epoch": 0.30879345603271985,
      "grad_norm": 3.2387661933898926,
      "learning_rate": 8.288236583160322e-05,
      "loss": 1.0825,
      "step": 604
    },
    {
      "epoch": 0.30930470347648265,
      "grad_norm": 3.1593637466430664,
      "learning_rate": 8.281856220125076e-05,
      "loss": 1.0234,
      "step": 605
    },
    {
      "epoch": 0.3098159509202454,
      "grad_norm": 3.0978777408599854,
      "learning_rate": 8.275466454158847e-05,
      "loss": 1.0093,
      "step": 606
    },
    {
      "epoch": 0.3103271983640082,
      "grad_norm": 2.8354928493499756,
      "learning_rate": 8.269067303569118e-05,
      "loss": 1.0444,
      "step": 607
    },
    {
      "epoch": 0.310838445807771,
      "grad_norm": 2.848942756652832,
      "learning_rate": 8.262658786690262e-05,
      "loss": 0.9818,
      "step": 608
    },
    {
      "epoch": 0.3113496932515337,
      "grad_norm": 3.241898536682129,
      "learning_rate": 8.256240921883487e-05,
      "loss": 1.0262,
      "step": 609
    },
    {
      "epoch": 0.3118609406952965,
      "grad_norm": 3.1023952960968018,
      "learning_rate": 8.249813727536781e-05,
      "loss": 1.1397,
      "step": 610
    },
    {
      "epoch": 0.3123721881390593,
      "grad_norm": 3.007742166519165,
      "learning_rate": 8.243377222064865e-05,
      "loss": 0.9883,
      "step": 611
    },
    {
      "epoch": 0.3128834355828221,
      "grad_norm": 2.9802401065826416,
      "learning_rate": 8.236931423909138e-05,
      "loss": 0.9806,
      "step": 612
    },
    {
      "epoch": 0.31339468302658485,
      "grad_norm": 3.3660130500793457,
      "learning_rate": 8.230476351537624e-05,
      "loss": 0.9672,
      "step": 613
    },
    {
      "epoch": 0.31390593047034765,
      "grad_norm": 3.3183295726776123,
      "learning_rate": 8.224012023444913e-05,
      "loss": 0.9066,
      "step": 614
    },
    {
      "epoch": 0.31441717791411045,
      "grad_norm": 3.352731466293335,
      "learning_rate": 8.217538458152122e-05,
      "loss": 0.9771,
      "step": 615
    },
    {
      "epoch": 0.3149284253578732,
      "grad_norm": 3.439236640930176,
      "learning_rate": 8.211055674206828e-05,
      "loss": 1.0707,
      "step": 616
    },
    {
      "epoch": 0.315439672801636,
      "grad_norm": 3.3290817737579346,
      "learning_rate": 8.204563690183024e-05,
      "loss": 1.0231,
      "step": 617
    },
    {
      "epoch": 0.3159509202453988,
      "grad_norm": 3.7434792518615723,
      "learning_rate": 8.198062524681061e-05,
      "loss": 0.9347,
      "step": 618
    },
    {
      "epoch": 0.3164621676891616,
      "grad_norm": 3.3235063552856445,
      "learning_rate": 8.191552196327596e-05,
      "loss": 0.8501,
      "step": 619
    },
    {
      "epoch": 0.3169734151329243,
      "grad_norm": 3.5552737712860107,
      "learning_rate": 8.185032723775539e-05,
      "loss": 0.971,
      "step": 620
    },
    {
      "epoch": 0.3174846625766871,
      "grad_norm": 3.5360236167907715,
      "learning_rate": 8.178504125703997e-05,
      "loss": 1.0067,
      "step": 621
    },
    {
      "epoch": 0.3179959100204499,
      "grad_norm": 3.9374284744262695,
      "learning_rate": 8.171966420818228e-05,
      "loss": 0.9743,
      "step": 622
    },
    {
      "epoch": 0.31850715746421265,
      "grad_norm": 3.668915271759033,
      "learning_rate": 8.165419627849578e-05,
      "loss": 0.9575,
      "step": 623
    },
    {
      "epoch": 0.31901840490797545,
      "grad_norm": 3.8559343814849854,
      "learning_rate": 8.15886376555543e-05,
      "loss": 0.9918,
      "step": 624
    },
    {
      "epoch": 0.31952965235173825,
      "grad_norm": 3.806727409362793,
      "learning_rate": 8.152298852719159e-05,
      "loss": 0.9559,
      "step": 625
    },
    {
      "epoch": 0.32004089979550104,
      "grad_norm": 3.8618004322052,
      "learning_rate": 8.145724908150063e-05,
      "loss": 0.9213,
      "step": 626
    },
    {
      "epoch": 0.3205521472392638,
      "grad_norm": 3.607926607131958,
      "learning_rate": 8.13914195068332e-05,
      "loss": 0.8309,
      "step": 627
    },
    {
      "epoch": 0.3210633946830266,
      "grad_norm": 3.718229055404663,
      "learning_rate": 8.132549999179933e-05,
      "loss": 0.8815,
      "step": 628
    },
    {
      "epoch": 0.3215746421267894,
      "grad_norm": 4.1860785484313965,
      "learning_rate": 8.125949072526673e-05,
      "loss": 0.9077,
      "step": 629
    },
    {
      "epoch": 0.3220858895705521,
      "grad_norm": 3.8918352127075195,
      "learning_rate": 8.119339189636023e-05,
      "loss": 0.9646,
      "step": 630
    },
    {
      "epoch": 0.3225971370143149,
      "grad_norm": 3.9071269035339355,
      "learning_rate": 8.11272036944613e-05,
      "loss": 0.8756,
      "step": 631
    },
    {
      "epoch": 0.3231083844580777,
      "grad_norm": 3.965230941772461,
      "learning_rate": 8.106092630920749e-05,
      "loss": 0.8488,
      "step": 632
    },
    {
      "epoch": 0.3236196319018405,
      "grad_norm": 4.22208309173584,
      "learning_rate": 8.099455993049181e-05,
      "loss": 0.9365,
      "step": 633
    },
    {
      "epoch": 0.32413087934560325,
      "grad_norm": 4.312554359436035,
      "learning_rate": 8.092810474846231e-05,
      "loss": 0.9539,
      "step": 634
    },
    {
      "epoch": 0.32464212678936605,
      "grad_norm": 4.0963592529296875,
      "learning_rate": 8.086156095352144e-05,
      "loss": 0.9983,
      "step": 635
    },
    {
      "epoch": 0.32515337423312884,
      "grad_norm": 4.197260856628418,
      "learning_rate": 8.079492873632554e-05,
      "loss": 1.0168,
      "step": 636
    },
    {
      "epoch": 0.32566462167689164,
      "grad_norm": 4.2196173667907715,
      "learning_rate": 8.07282082877843e-05,
      "loss": 0.9127,
      "step": 637
    },
    {
      "epoch": 0.3261758691206544,
      "grad_norm": 4.162968635559082,
      "learning_rate": 8.06613997990602e-05,
      "loss": 0.9794,
      "step": 638
    },
    {
      "epoch": 0.3266871165644172,
      "grad_norm": 4.384978294372559,
      "learning_rate": 8.059450346156796e-05,
      "loss": 0.9692,
      "step": 639
    },
    {
      "epoch": 0.32719836400818,
      "grad_norm": 4.631701469421387,
      "learning_rate": 8.052751946697403e-05,
      "loss": 0.9136,
      "step": 640
    },
    {
      "epoch": 0.3277096114519427,
      "grad_norm": 5.009700298309326,
      "learning_rate": 8.046044800719594e-05,
      "loss": 0.8666,
      "step": 641
    },
    {
      "epoch": 0.3282208588957055,
      "grad_norm": 5.072287082672119,
      "learning_rate": 8.039328927440188e-05,
      "loss": 0.9186,
      "step": 642
    },
    {
      "epoch": 0.3287321063394683,
      "grad_norm": 5.042779445648193,
      "learning_rate": 8.032604346101009e-05,
      "loss": 0.843,
      "step": 643
    },
    {
      "epoch": 0.3292433537832311,
      "grad_norm": 4.215601921081543,
      "learning_rate": 8.025871075968828e-05,
      "loss": 0.8339,
      "step": 644
    },
    {
      "epoch": 0.32975460122699385,
      "grad_norm": 4.962833881378174,
      "learning_rate": 8.019129136335306e-05,
      "loss": 0.8041,
      "step": 645
    },
    {
      "epoch": 0.33026584867075665,
      "grad_norm": 5.858077526092529,
      "learning_rate": 8.012378546516954e-05,
      "loss": 0.763,
      "step": 646
    },
    {
      "epoch": 0.33077709611451944,
      "grad_norm": 6.057731628417969,
      "learning_rate": 8.00561932585506e-05,
      "loss": 0.8779,
      "step": 647
    },
    {
      "epoch": 0.3312883435582822,
      "grad_norm": 6.040656089782715,
      "learning_rate": 7.99885149371564e-05,
      "loss": 0.7984,
      "step": 648
    },
    {
      "epoch": 0.331799591002045,
      "grad_norm": 4.905963897705078,
      "learning_rate": 7.992075069489387e-05,
      "loss": 0.5278,
      "step": 649
    },
    {
      "epoch": 0.3323108384458078,
      "grad_norm": 7.595052242279053,
      "learning_rate": 7.985290072591605e-05,
      "loss": 0.5099,
      "step": 650
    },
    {
      "epoch": 0.3328220858895706,
      "grad_norm": 2.33105206489563,
      "learning_rate": 7.978496522462167e-05,
      "loss": 0.8827,
      "step": 651
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 2.849806070327759,
      "learning_rate": 7.97169443856545e-05,
      "loss": 0.9505,
      "step": 652
    },
    {
      "epoch": 0.3338445807770961,
      "grad_norm": 3.0055220127105713,
      "learning_rate": 7.964883840390276e-05,
      "loss": 1.0614,
      "step": 653
    },
    {
      "epoch": 0.3343558282208589,
      "grad_norm": 3.2699313163757324,
      "learning_rate": 7.958064747449869e-05,
      "loss": 1.0308,
      "step": 654
    },
    {
      "epoch": 0.33486707566462165,
      "grad_norm": 2.836352586746216,
      "learning_rate": 7.951237179281788e-05,
      "loss": 1.0131,
      "step": 655
    },
    {
      "epoch": 0.33537832310838445,
      "grad_norm": 2.913180351257324,
      "learning_rate": 7.944401155447871e-05,
      "loss": 0.9972,
      "step": 656
    },
    {
      "epoch": 0.33588957055214724,
      "grad_norm": 3.03965163230896,
      "learning_rate": 7.937556695534193e-05,
      "loss": 0.968,
      "step": 657
    },
    {
      "epoch": 0.33640081799591004,
      "grad_norm": 2.8682761192321777,
      "learning_rate": 7.930703819150987e-05,
      "loss": 0.9992,
      "step": 658
    },
    {
      "epoch": 0.3369120654396728,
      "grad_norm": 2.7956974506378174,
      "learning_rate": 7.92384254593261e-05,
      "loss": 1.0082,
      "step": 659
    },
    {
      "epoch": 0.3374233128834356,
      "grad_norm": 3.3379359245300293,
      "learning_rate": 7.916972895537471e-05,
      "loss": 1.0037,
      "step": 660
    },
    {
      "epoch": 0.3379345603271984,
      "grad_norm": 3.0874998569488525,
      "learning_rate": 7.91009488764798e-05,
      "loss": 1.0145,
      "step": 661
    },
    {
      "epoch": 0.33844580777096117,
      "grad_norm": 3.038001537322998,
      "learning_rate": 7.903208541970501e-05,
      "loss": 0.9002,
      "step": 662
    },
    {
      "epoch": 0.3389570552147239,
      "grad_norm": 3.2250359058380127,
      "learning_rate": 7.896313878235278e-05,
      "loss": 0.9536,
      "step": 663
    },
    {
      "epoch": 0.3394683026584867,
      "grad_norm": 3.5659546852111816,
      "learning_rate": 7.889410916196389e-05,
      "loss": 0.9676,
      "step": 664
    },
    {
      "epoch": 0.3399795501022495,
      "grad_norm": 3.521686553955078,
      "learning_rate": 7.882499675631689e-05,
      "loss": 0.9694,
      "step": 665
    },
    {
      "epoch": 0.34049079754601225,
      "grad_norm": 3.544236660003662,
      "learning_rate": 7.875580176342753e-05,
      "loss": 0.8514,
      "step": 666
    },
    {
      "epoch": 0.34100204498977504,
      "grad_norm": 3.529057025909424,
      "learning_rate": 7.868652438154815e-05,
      "loss": 1.0767,
      "step": 667
    },
    {
      "epoch": 0.34151329243353784,
      "grad_norm": 3.835726261138916,
      "learning_rate": 7.86171648091672e-05,
      "loss": 0.9648,
      "step": 668
    },
    {
      "epoch": 0.34202453987730064,
      "grad_norm": 3.570894479751587,
      "learning_rate": 7.854772324500854e-05,
      "loss": 0.9455,
      "step": 669
    },
    {
      "epoch": 0.3425357873210634,
      "grad_norm": 3.518613815307617,
      "learning_rate": 7.847819988803102e-05,
      "loss": 1.0018,
      "step": 670
    },
    {
      "epoch": 0.3430470347648262,
      "grad_norm": 3.49399733543396,
      "learning_rate": 7.84085949374278e-05,
      "loss": 0.8864,
      "step": 671
    },
    {
      "epoch": 0.34355828220858897,
      "grad_norm": 3.5140910148620605,
      "learning_rate": 7.833890859262579e-05,
      "loss": 1.067,
      "step": 672
    },
    {
      "epoch": 0.3440695296523517,
      "grad_norm": 3.35764479637146,
      "learning_rate": 7.826914105328519e-05,
      "loss": 0.9411,
      "step": 673
    },
    {
      "epoch": 0.3445807770961145,
      "grad_norm": 3.6672184467315674,
      "learning_rate": 7.819929251929873e-05,
      "loss": 0.9742,
      "step": 674
    },
    {
      "epoch": 0.3450920245398773,
      "grad_norm": 3.7109756469726562,
      "learning_rate": 7.812936319079127e-05,
      "loss": 0.9916,
      "step": 675
    },
    {
      "epoch": 0.3456032719836401,
      "grad_norm": 3.371408224105835,
      "learning_rate": 7.805935326811912e-05,
      "loss": 0.9044,
      "step": 676
    },
    {
      "epoch": 0.34611451942740284,
      "grad_norm": 3.713839054107666,
      "learning_rate": 7.798926295186955e-05,
      "loss": 0.9528,
      "step": 677
    },
    {
      "epoch": 0.34662576687116564,
      "grad_norm": 3.572603702545166,
      "learning_rate": 7.791909244286009e-05,
      "loss": 0.8672,
      "step": 678
    },
    {
      "epoch": 0.34713701431492844,
      "grad_norm": 3.5241007804870605,
      "learning_rate": 7.784884194213811e-05,
      "loss": 0.862,
      "step": 679
    },
    {
      "epoch": 0.3476482617586912,
      "grad_norm": 4.026426792144775,
      "learning_rate": 7.777851165098012e-05,
      "loss": 0.96,
      "step": 680
    },
    {
      "epoch": 0.348159509202454,
      "grad_norm": 4.031967639923096,
      "learning_rate": 7.770810177089126e-05,
      "loss": 1.0187,
      "step": 681
    },
    {
      "epoch": 0.3486707566462168,
      "grad_norm": 3.622410297393799,
      "learning_rate": 7.763761250360467e-05,
      "loss": 0.8931,
      "step": 682
    },
    {
      "epoch": 0.34918200408997957,
      "grad_norm": 3.932042121887207,
      "learning_rate": 7.756704405108102e-05,
      "loss": 0.8535,
      "step": 683
    },
    {
      "epoch": 0.3496932515337423,
      "grad_norm": 3.7137181758880615,
      "learning_rate": 7.749639661550775e-05,
      "loss": 0.808,
      "step": 684
    },
    {
      "epoch": 0.3502044989775051,
      "grad_norm": 3.7734367847442627,
      "learning_rate": 7.74256703992987e-05,
      "loss": 0.8274,
      "step": 685
    },
    {
      "epoch": 0.3507157464212679,
      "grad_norm": 4.37022590637207,
      "learning_rate": 7.735486560509332e-05,
      "loss": 0.913,
      "step": 686
    },
    {
      "epoch": 0.3512269938650307,
      "grad_norm": 3.8798720836639404,
      "learning_rate": 7.728398243575631e-05,
      "loss": 0.8297,
      "step": 687
    },
    {
      "epoch": 0.35173824130879344,
      "grad_norm": 4.2243266105651855,
      "learning_rate": 7.721302109437685e-05,
      "loss": 0.9338,
      "step": 688
    },
    {
      "epoch": 0.35224948875255624,
      "grad_norm": 4.538561820983887,
      "learning_rate": 7.71419817842681e-05,
      "loss": 0.917,
      "step": 689
    },
    {
      "epoch": 0.35276073619631904,
      "grad_norm": 5.12889289855957,
      "learning_rate": 7.707086470896663e-05,
      "loss": 0.8964,
      "step": 690
    },
    {
      "epoch": 0.3532719836400818,
      "grad_norm": 4.37795877456665,
      "learning_rate": 7.699967007223182e-05,
      "loss": 0.8095,
      "step": 691
    },
    {
      "epoch": 0.3537832310838446,
      "grad_norm": 4.92649507522583,
      "learning_rate": 7.692839807804521e-05,
      "loss": 0.9034,
      "step": 692
    },
    {
      "epoch": 0.35429447852760737,
      "grad_norm": 4.200683116912842,
      "learning_rate": 7.68570489306101e-05,
      "loss": 0.8081,
      "step": 693
    },
    {
      "epoch": 0.35480572597137017,
      "grad_norm": 5.237700462341309,
      "learning_rate": 7.678562283435076e-05,
      "loss": 1.0181,
      "step": 694
    },
    {
      "epoch": 0.3553169734151329,
      "grad_norm": 5.287177562713623,
      "learning_rate": 7.67141199939119e-05,
      "loss": 1.0058,
      "step": 695
    },
    {
      "epoch": 0.3558282208588957,
      "grad_norm": 5.213104724884033,
      "learning_rate": 7.664254061415818e-05,
      "loss": 0.881,
      "step": 696
    },
    {
      "epoch": 0.3563394683026585,
      "grad_norm": 5.322814464569092,
      "learning_rate": 7.657088490017354e-05,
      "loss": 0.9247,
      "step": 697
    },
    {
      "epoch": 0.35685071574642124,
      "grad_norm": 5.443381309509277,
      "learning_rate": 7.64991530572606e-05,
      "loss": 0.789,
      "step": 698
    },
    {
      "epoch": 0.35736196319018404,
      "grad_norm": 6.3556060791015625,
      "learning_rate": 7.642734529094012e-05,
      "loss": 0.8338,
      "step": 699
    },
    {
      "epoch": 0.35787321063394684,
      "grad_norm": 8.609705924987793,
      "learning_rate": 7.635546180695038e-05,
      "loss": 0.9239,
      "step": 700
    },
    {
      "epoch": 0.35838445807770963,
      "grad_norm": 2.269519329071045,
      "learning_rate": 7.628350281124663e-05,
      "loss": 0.8918,
      "step": 701
    },
    {
      "epoch": 0.3588957055214724,
      "grad_norm": 2.6576449871063232,
      "learning_rate": 7.621146851000043e-05,
      "loss": 1.0788,
      "step": 702
    },
    {
      "epoch": 0.35940695296523517,
      "grad_norm": 2.7672042846679688,
      "learning_rate": 7.61393591095991e-05,
      "loss": 1.0559,
      "step": 703
    },
    {
      "epoch": 0.35991820040899797,
      "grad_norm": 2.5997817516326904,
      "learning_rate": 7.606717481664514e-05,
      "loss": 1.0025,
      "step": 704
    },
    {
      "epoch": 0.3604294478527607,
      "grad_norm": 2.5708072185516357,
      "learning_rate": 7.599491583795569e-05,
      "loss": 0.9982,
      "step": 705
    },
    {
      "epoch": 0.3609406952965235,
      "grad_norm": 2.8804173469543457,
      "learning_rate": 7.592258238056174e-05,
      "loss": 1.0188,
      "step": 706
    },
    {
      "epoch": 0.3614519427402863,
      "grad_norm": 2.8373563289642334,
      "learning_rate": 7.58501746517078e-05,
      "loss": 0.921,
      "step": 707
    },
    {
      "epoch": 0.3619631901840491,
      "grad_norm": 2.876730442047119,
      "learning_rate": 7.577769285885109e-05,
      "loss": 0.9169,
      "step": 708
    },
    {
      "epoch": 0.36247443762781184,
      "grad_norm": 2.9293148517608643,
      "learning_rate": 7.570513720966108e-05,
      "loss": 1.01,
      "step": 709
    },
    {
      "epoch": 0.36298568507157464,
      "grad_norm": 3.1806929111480713,
      "learning_rate": 7.563250791201886e-05,
      "loss": 0.9419,
      "step": 710
    },
    {
      "epoch": 0.36349693251533743,
      "grad_norm": 3.0284066200256348,
      "learning_rate": 7.555980517401645e-05,
      "loss": 1.011,
      "step": 711
    },
    {
      "epoch": 0.36400817995910023,
      "grad_norm": 3.130155086517334,
      "learning_rate": 7.548702920395638e-05,
      "loss": 0.9283,
      "step": 712
    },
    {
      "epoch": 0.36451942740286297,
      "grad_norm": 3.167332172393799,
      "learning_rate": 7.541418021035098e-05,
      "loss": 0.9085,
      "step": 713
    },
    {
      "epoch": 0.36503067484662577,
      "grad_norm": 3.1641294956207275,
      "learning_rate": 7.534125840192172e-05,
      "loss": 0.9266,
      "step": 714
    },
    {
      "epoch": 0.36554192229038857,
      "grad_norm": 3.073336601257324,
      "learning_rate": 7.526826398759881e-05,
      "loss": 0.8859,
      "step": 715
    },
    {
      "epoch": 0.3660531697341513,
      "grad_norm": 2.870695114135742,
      "learning_rate": 7.519519717652039e-05,
      "loss": 0.7983,
      "step": 716
    },
    {
      "epoch": 0.3665644171779141,
      "grad_norm": 3.243685483932495,
      "learning_rate": 7.512205817803211e-05,
      "loss": 0.9613,
      "step": 717
    },
    {
      "epoch": 0.3670756646216769,
      "grad_norm": 3.117615222930908,
      "learning_rate": 7.504884720168637e-05,
      "loss": 0.859,
      "step": 718
    },
    {
      "epoch": 0.3675869120654397,
      "grad_norm": 3.478621244430542,
      "learning_rate": 7.497556445724183e-05,
      "loss": 0.9985,
      "step": 719
    },
    {
      "epoch": 0.36809815950920244,
      "grad_norm": 3.571955680847168,
      "learning_rate": 7.490221015466279e-05,
      "loss": 0.9798,
      "step": 720
    },
    {
      "epoch": 0.36860940695296524,
      "grad_norm": 3.45243239402771,
      "learning_rate": 7.482878450411854e-05,
      "loss": 0.97,
      "step": 721
    },
    {
      "epoch": 0.36912065439672803,
      "grad_norm": 3.4372360706329346,
      "learning_rate": 7.475528771598283e-05,
      "loss": 1.0453,
      "step": 722
    },
    {
      "epoch": 0.3696319018404908,
      "grad_norm": 3.8749423027038574,
      "learning_rate": 7.468172000083319e-05,
      "loss": 0.9453,
      "step": 723
    },
    {
      "epoch": 0.37014314928425357,
      "grad_norm": 3.3690943717956543,
      "learning_rate": 7.460808156945036e-05,
      "loss": 0.8492,
      "step": 724
    },
    {
      "epoch": 0.37065439672801637,
      "grad_norm": 3.636127471923828,
      "learning_rate": 7.453437263281777e-05,
      "loss": 0.9653,
      "step": 725
    },
    {
      "epoch": 0.37116564417177916,
      "grad_norm": 3.700061082839966,
      "learning_rate": 7.446059340212072e-05,
      "loss": 0.9498,
      "step": 726
    },
    {
      "epoch": 0.3716768916155419,
      "grad_norm": 3.9007370471954346,
      "learning_rate": 7.438674408874604e-05,
      "loss": 0.8941,
      "step": 727
    },
    {
      "epoch": 0.3721881390593047,
      "grad_norm": 3.541825771331787,
      "learning_rate": 7.431282490428129e-05,
      "loss": 0.8871,
      "step": 728
    },
    {
      "epoch": 0.3726993865030675,
      "grad_norm": 3.9885671138763428,
      "learning_rate": 7.423883606051424e-05,
      "loss": 1.0297,
      "step": 729
    },
    {
      "epoch": 0.37321063394683024,
      "grad_norm": 3.7179479598999023,
      "learning_rate": 7.416477776943222e-05,
      "loss": 0.8756,
      "step": 730
    },
    {
      "epoch": 0.37372188139059304,
      "grad_norm": 4.2167229652404785,
      "learning_rate": 7.409065024322158e-05,
      "loss": 0.9019,
      "step": 731
    },
    {
      "epoch": 0.37423312883435583,
      "grad_norm": 3.5955021381378174,
      "learning_rate": 7.401645369426697e-05,
      "loss": 0.8644,
      "step": 732
    },
    {
      "epoch": 0.37474437627811863,
      "grad_norm": 3.7380001544952393,
      "learning_rate": 7.394218833515088e-05,
      "loss": 0.8517,
      "step": 733
    },
    {
      "epoch": 0.37525562372188137,
      "grad_norm": 3.396381378173828,
      "learning_rate": 7.386785437865287e-05,
      "loss": 0.7941,
      "step": 734
    },
    {
      "epoch": 0.37576687116564417,
      "grad_norm": 3.9136502742767334,
      "learning_rate": 7.37934520377491e-05,
      "loss": 0.8351,
      "step": 735
    },
    {
      "epoch": 0.37627811860940696,
      "grad_norm": 3.892547130584717,
      "learning_rate": 7.371898152561166e-05,
      "loss": 0.9343,
      "step": 736
    },
    {
      "epoch": 0.37678936605316976,
      "grad_norm": 3.785649061203003,
      "learning_rate": 7.364444305560787e-05,
      "loss": 0.8726,
      "step": 737
    },
    {
      "epoch": 0.3773006134969325,
      "grad_norm": 4.7827372550964355,
      "learning_rate": 7.35698368412999e-05,
      "loss": 0.9685,
      "step": 738
    },
    {
      "epoch": 0.3778118609406953,
      "grad_norm": 3.921762466430664,
      "learning_rate": 7.349516309644388e-05,
      "loss": 0.8303,
      "step": 739
    },
    {
      "epoch": 0.3783231083844581,
      "grad_norm": 4.252273082733154,
      "learning_rate": 7.342042203498951e-05,
      "loss": 0.8746,
      "step": 740
    },
    {
      "epoch": 0.37883435582822084,
      "grad_norm": 4.284645080566406,
      "learning_rate": 7.334561387107935e-05,
      "loss": 0.8135,
      "step": 741
    },
    {
      "epoch": 0.37934560327198363,
      "grad_norm": 4.3504638671875,
      "learning_rate": 7.327073881904819e-05,
      "loss": 0.8595,
      "step": 742
    },
    {
      "epoch": 0.37985685071574643,
      "grad_norm": 4.201150417327881,
      "learning_rate": 7.319579709342244e-05,
      "loss": 0.7961,
      "step": 743
    },
    {
      "epoch": 0.3803680981595092,
      "grad_norm": 4.760327339172363,
      "learning_rate": 7.312078890891963e-05,
      "loss": 0.7495,
      "step": 744
    },
    {
      "epoch": 0.38087934560327197,
      "grad_norm": 5.05194091796875,
      "learning_rate": 7.304571448044759e-05,
      "loss": 0.9073,
      "step": 745
    },
    {
      "epoch": 0.38139059304703476,
      "grad_norm": 5.017317295074463,
      "learning_rate": 7.297057402310406e-05,
      "loss": 0.7862,
      "step": 746
    },
    {
      "epoch": 0.38190184049079756,
      "grad_norm": 5.787736892700195,
      "learning_rate": 7.289536775217587e-05,
      "loss": 0.8488,
      "step": 747
    },
    {
      "epoch": 0.3824130879345603,
      "grad_norm": 5.555422306060791,
      "learning_rate": 7.282009588313845e-05,
      "loss": 0.7949,
      "step": 748
    },
    {
      "epoch": 0.3829243353783231,
      "grad_norm": 6.039085865020752,
      "learning_rate": 7.274475863165518e-05,
      "loss": 0.5657,
      "step": 749
    },
    {
      "epoch": 0.3834355828220859,
      "grad_norm": 7.193869590759277,
      "learning_rate": 7.266935621357677e-05,
      "loss": 0.6427,
      "step": 750
    },
    {
      "epoch": 0.3839468302658487,
      "grad_norm": 1.978843331336975,
      "learning_rate": 7.259388884494064e-05,
      "loss": 0.853,
      "step": 751
    },
    {
      "epoch": 0.38445807770961143,
      "grad_norm": 2.8742833137512207,
      "learning_rate": 7.251835674197029e-05,
      "loss": 1.0315,
      "step": 752
    },
    {
      "epoch": 0.38496932515337423,
      "grad_norm": 2.5204415321350098,
      "learning_rate": 7.244276012107467e-05,
      "loss": 1.0057,
      "step": 753
    },
    {
      "epoch": 0.38548057259713703,
      "grad_norm": 2.7091565132141113,
      "learning_rate": 7.236709919884763e-05,
      "loss": 1.066,
      "step": 754
    },
    {
      "epoch": 0.38599182004089977,
      "grad_norm": 2.9578049182891846,
      "learning_rate": 7.229137419206727e-05,
      "loss": 1.0613,
      "step": 755
    },
    {
      "epoch": 0.38650306748466257,
      "grad_norm": 2.7542765140533447,
      "learning_rate": 7.221558531769519e-05,
      "loss": 0.9996,
      "step": 756
    },
    {
      "epoch": 0.38701431492842536,
      "grad_norm": 2.6492252349853516,
      "learning_rate": 7.21397327928761e-05,
      "loss": 0.9659,
      "step": 757
    },
    {
      "epoch": 0.38752556237218816,
      "grad_norm": 2.741001605987549,
      "learning_rate": 7.206381683493702e-05,
      "loss": 0.9089,
      "step": 758
    },
    {
      "epoch": 0.3880368098159509,
      "grad_norm": 2.586782693862915,
      "learning_rate": 7.19878376613867e-05,
      "loss": 0.9266,
      "step": 759
    },
    {
      "epoch": 0.3885480572597137,
      "grad_norm": 2.960686206817627,
      "learning_rate": 7.191179548991507e-05,
      "loss": 0.9666,
      "step": 760
    },
    {
      "epoch": 0.3890593047034765,
      "grad_norm": 3.008049249649048,
      "learning_rate": 7.183569053839248e-05,
      "loss": 1.0174,
      "step": 761
    },
    {
      "epoch": 0.3895705521472393,
      "grad_norm": 2.975431203842163,
      "learning_rate": 7.17595230248692e-05,
      "loss": 0.9742,
      "step": 762
    },
    {
      "epoch": 0.39008179959100203,
      "grad_norm": 3.148494243621826,
      "learning_rate": 7.168329316757475e-05,
      "loss": 0.9141,
      "step": 763
    },
    {
      "epoch": 0.39059304703476483,
      "grad_norm": 3.210510730743408,
      "learning_rate": 7.160700118491728e-05,
      "loss": 0.9639,
      "step": 764
    },
    {
      "epoch": 0.3911042944785276,
      "grad_norm": 3.1042428016662598,
      "learning_rate": 7.15306472954829e-05,
      "loss": 0.9609,
      "step": 765
    },
    {
      "epoch": 0.39161554192229037,
      "grad_norm": 3.2748963832855225,
      "learning_rate": 7.14542317180351e-05,
      "loss": 0.9741,
      "step": 766
    },
    {
      "epoch": 0.39212678936605316,
      "grad_norm": 3.0430991649627686,
      "learning_rate": 7.137775467151411e-05,
      "loss": 0.9359,
      "step": 767
    },
    {
      "epoch": 0.39263803680981596,
      "grad_norm": 3.2744600772857666,
      "learning_rate": 7.130121637503632e-05,
      "loss": 0.8993,
      "step": 768
    },
    {
      "epoch": 0.39314928425357876,
      "grad_norm": 3.287580728530884,
      "learning_rate": 7.122461704789358e-05,
      "loss": 0.8918,
      "step": 769
    },
    {
      "epoch": 0.3936605316973415,
      "grad_norm": 3.8326618671417236,
      "learning_rate": 7.11479569095526e-05,
      "loss": 0.8956,
      "step": 770
    },
    {
      "epoch": 0.3941717791411043,
      "grad_norm": 3.353121519088745,
      "learning_rate": 7.10712361796543e-05,
      "loss": 0.9102,
      "step": 771
    },
    {
      "epoch": 0.3946830265848671,
      "grad_norm": 3.537923574447632,
      "learning_rate": 7.099445507801323e-05,
      "loss": 0.9339,
      "step": 772
    },
    {
      "epoch": 0.39519427402862983,
      "grad_norm": 3.2914555072784424,
      "learning_rate": 7.091761382461692e-05,
      "loss": 0.8794,
      "step": 773
    },
    {
      "epoch": 0.39570552147239263,
      "grad_norm": 3.461108922958374,
      "learning_rate": 7.084071263962521e-05,
      "loss": 0.9089,
      "step": 774
    },
    {
      "epoch": 0.3962167689161554,
      "grad_norm": 3.281498908996582,
      "learning_rate": 7.076375174336972e-05,
      "loss": 0.8186,
      "step": 775
    },
    {
      "epoch": 0.3967280163599182,
      "grad_norm": 3.6270294189453125,
      "learning_rate": 7.068673135635302e-05,
      "loss": 0.8756,
      "step": 776
    },
    {
      "epoch": 0.39723926380368096,
      "grad_norm": 3.7898643016815186,
      "learning_rate": 7.060965169924828e-05,
      "loss": 0.9115,
      "step": 777
    },
    {
      "epoch": 0.39775051124744376,
      "grad_norm": 3.4987165927886963,
      "learning_rate": 7.053251299289837e-05,
      "loss": 0.8832,
      "step": 778
    },
    {
      "epoch": 0.39826175869120656,
      "grad_norm": 3.6572365760803223,
      "learning_rate": 7.045531545831541e-05,
      "loss": 0.9144,
      "step": 779
    },
    {
      "epoch": 0.3987730061349693,
      "grad_norm": 4.028952598571777,
      "learning_rate": 7.037805931668005e-05,
      "loss": 0.9033,
      "step": 780
    },
    {
      "epoch": 0.3992842535787321,
      "grad_norm": 3.5009331703186035,
      "learning_rate": 7.030074478934085e-05,
      "loss": 0.8903,
      "step": 781
    },
    {
      "epoch": 0.3997955010224949,
      "grad_norm": 3.7041122913360596,
      "learning_rate": 7.022337209781367e-05,
      "loss": 0.8992,
      "step": 782
    },
    {
      "epoch": 0.4003067484662577,
      "grad_norm": 3.8405399322509766,
      "learning_rate": 7.014594146378097e-05,
      "loss": 0.8085,
      "step": 783
    },
    {
      "epoch": 0.40081799591002043,
      "grad_norm": 3.704244375228882,
      "learning_rate": 7.006845310909131e-05,
      "loss": 0.8948,
      "step": 784
    },
    {
      "epoch": 0.4013292433537832,
      "grad_norm": 3.5414698123931885,
      "learning_rate": 6.999090725575854e-05,
      "loss": 0.7511,
      "step": 785
    },
    {
      "epoch": 0.401840490797546,
      "grad_norm": 4.033692836761475,
      "learning_rate": 6.99133041259613e-05,
      "loss": 0.8177,
      "step": 786
    },
    {
      "epoch": 0.4023517382413088,
      "grad_norm": 4.002652168273926,
      "learning_rate": 6.98356439420423e-05,
      "loss": 0.8903,
      "step": 787
    },
    {
      "epoch": 0.40286298568507156,
      "grad_norm": 4.07108736038208,
      "learning_rate": 6.975792692650777e-05,
      "loss": 0.972,
      "step": 788
    },
    {
      "epoch": 0.40337423312883436,
      "grad_norm": 3.756382942199707,
      "learning_rate": 6.968015330202672e-05,
      "loss": 0.8346,
      "step": 789
    },
    {
      "epoch": 0.40388548057259716,
      "grad_norm": 4.131988525390625,
      "learning_rate": 6.960232329143039e-05,
      "loss": 0.9014,
      "step": 790
    },
    {
      "epoch": 0.4043967280163599,
      "grad_norm": 4.3247833251953125,
      "learning_rate": 6.952443711771151e-05,
      "loss": 0.8945,
      "step": 791
    },
    {
      "epoch": 0.4049079754601227,
      "grad_norm": 4.037295818328857,
      "learning_rate": 6.94464950040238e-05,
      "loss": 0.9311,
      "step": 792
    },
    {
      "epoch": 0.4054192229038855,
      "grad_norm": 4.475683689117432,
      "learning_rate": 6.936849717368122e-05,
      "loss": 0.8492,
      "step": 793
    },
    {
      "epoch": 0.4059304703476483,
      "grad_norm": 4.772801399230957,
      "learning_rate": 6.929044385015735e-05,
      "loss": 0.9542,
      "step": 794
    },
    {
      "epoch": 0.40644171779141103,
      "grad_norm": 4.975774765014648,
      "learning_rate": 6.921233525708479e-05,
      "loss": 0.9035,
      "step": 795
    },
    {
      "epoch": 0.4069529652351738,
      "grad_norm": 4.816725254058838,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.9328,
      "step": 796
    },
    {
      "epoch": 0.4074642126789366,
      "grad_norm": 5.29307222366333,
      "learning_rate": 6.905595315761511e-05,
      "loss": 0.6613,
      "step": 797
    },
    {
      "epoch": 0.40797546012269936,
      "grad_norm": 5.284379959106445,
      "learning_rate": 6.897768009927239e-05,
      "loss": 0.6603,
      "step": 798
    },
    {
      "epoch": 0.40848670756646216,
      "grad_norm": 6.468302249908447,
      "learning_rate": 6.889935266748848e-05,
      "loss": 0.5906,
      "step": 799
    },
    {
      "epoch": 0.40899795501022496,
      "grad_norm": 7.637302398681641,
      "learning_rate": 6.882097108668132e-05,
      "loss": 0.8293,
      "step": 800
    },
    {
      "epoch": 0.40950920245398775,
      "grad_norm": 2.510533094406128,
      "learning_rate": 6.8742535581424e-05,
      "loss": 0.8645,
      "step": 801
    },
    {
      "epoch": 0.4100204498977505,
      "grad_norm": 2.7541604042053223,
      "learning_rate": 6.866404637644414e-05,
      "loss": 1.0526,
      "step": 802
    },
    {
      "epoch": 0.4105316973415133,
      "grad_norm": 2.601616382598877,
      "learning_rate": 6.858550369662316e-05,
      "loss": 0.9845,
      "step": 803
    },
    {
      "epoch": 0.4110429447852761,
      "grad_norm": 3.236582040786743,
      "learning_rate": 6.850690776699573e-05,
      "loss": 1.0338,
      "step": 804
    },
    {
      "epoch": 0.41155419222903883,
      "grad_norm": 3.11091685295105,
      "learning_rate": 6.842825881274905e-05,
      "loss": 0.9695,
      "step": 805
    },
    {
      "epoch": 0.4120654396728016,
      "grad_norm": 3.053615093231201,
      "learning_rate": 6.834955705922232e-05,
      "loss": 0.9887,
      "step": 806
    },
    {
      "epoch": 0.4125766871165644,
      "grad_norm": 2.8809502124786377,
      "learning_rate": 6.827080273190593e-05,
      "loss": 0.9319,
      "step": 807
    },
    {
      "epoch": 0.4130879345603272,
      "grad_norm": 2.974292039871216,
      "learning_rate": 6.819199605644094e-05,
      "loss": 0.9261,
      "step": 808
    },
    {
      "epoch": 0.41359918200408996,
      "grad_norm": 2.8838043212890625,
      "learning_rate": 6.81131372586184e-05,
      "loss": 0.9214,
      "step": 809
    },
    {
      "epoch": 0.41411042944785276,
      "grad_norm": 3.148404598236084,
      "learning_rate": 6.803422656437867e-05,
      "loss": 1.0052,
      "step": 810
    },
    {
      "epoch": 0.41462167689161555,
      "grad_norm": 4.05100679397583,
      "learning_rate": 6.795526419981084e-05,
      "loss": 1.0043,
      "step": 811
    },
    {
      "epoch": 0.41513292433537835,
      "grad_norm": 3.163527727127075,
      "learning_rate": 6.7876250391152e-05,
      "loss": 1.018,
      "step": 812
    },
    {
      "epoch": 0.4156441717791411,
      "grad_norm": 3.0442557334899902,
      "learning_rate": 6.779718536478667e-05,
      "loss": 1.0564,
      "step": 813
    },
    {
      "epoch": 0.4161554192229039,
      "grad_norm": 3.1469643115997314,
      "learning_rate": 6.771806934724609e-05,
      "loss": 0.9531,
      "step": 814
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 2.9278955459594727,
      "learning_rate": 6.763890256520761e-05,
      "loss": 0.8475,
      "step": 815
    },
    {
      "epoch": 0.4171779141104294,
      "grad_norm": 2.9922728538513184,
      "learning_rate": 6.755968524549402e-05,
      "loss": 0.9396,
      "step": 816
    },
    {
      "epoch": 0.4176891615541922,
      "grad_norm": 3.1033873558044434,
      "learning_rate": 6.748041761507289e-05,
      "loss": 0.9287,
      "step": 817
    },
    {
      "epoch": 0.418200408997955,
      "grad_norm": 3.116170644760132,
      "learning_rate": 6.740109990105599e-05,
      "loss": 0.9613,
      "step": 818
    },
    {
      "epoch": 0.4187116564417178,
      "grad_norm": 3.148620367050171,
      "learning_rate": 6.73217323306985e-05,
      "loss": 0.8181,
      "step": 819
    },
    {
      "epoch": 0.41922290388548056,
      "grad_norm": 3.3708975315093994,
      "learning_rate": 6.724231513139852e-05,
      "loss": 0.932,
      "step": 820
    },
    {
      "epoch": 0.41973415132924335,
      "grad_norm": 3.3369412422180176,
      "learning_rate": 6.716284853069634e-05,
      "loss": 0.8951,
      "step": 821
    },
    {
      "epoch": 0.42024539877300615,
      "grad_norm": 3.5555686950683594,
      "learning_rate": 6.708333275627374e-05,
      "loss": 0.9422,
      "step": 822
    },
    {
      "epoch": 0.4207566462167689,
      "grad_norm": 3.544792890548706,
      "learning_rate": 6.700376803595343e-05,
      "loss": 0.9328,
      "step": 823
    },
    {
      "epoch": 0.4212678936605317,
      "grad_norm": 3.218123197555542,
      "learning_rate": 6.692415459769836e-05,
      "loss": 0.9012,
      "step": 824
    },
    {
      "epoch": 0.4217791411042945,
      "grad_norm": 3.3866398334503174,
      "learning_rate": 6.6844492669611e-05,
      "loss": 0.8799,
      "step": 825
    },
    {
      "epoch": 0.4222903885480573,
      "grad_norm": 4.010083198547363,
      "learning_rate": 6.676478247993284e-05,
      "loss": 0.9741,
      "step": 826
    },
    {
      "epoch": 0.42280163599182,
      "grad_norm": 3.7181904315948486,
      "learning_rate": 6.668502425704359e-05,
      "loss": 0.9802,
      "step": 827
    },
    {
      "epoch": 0.4233128834355828,
      "grad_norm": 3.6271464824676514,
      "learning_rate": 6.66052182294606e-05,
      "loss": 0.9424,
      "step": 828
    },
    {
      "epoch": 0.4238241308793456,
      "grad_norm": 3.841322898864746,
      "learning_rate": 6.652536462583817e-05,
      "loss": 0.9182,
      "step": 829
    },
    {
      "epoch": 0.42433537832310836,
      "grad_norm": 4.184422492980957,
      "learning_rate": 6.644546367496692e-05,
      "loss": 0.8529,
      "step": 830
    },
    {
      "epoch": 0.42484662576687116,
      "grad_norm": 3.9780843257904053,
      "learning_rate": 6.636551560577312e-05,
      "loss": 0.881,
      "step": 831
    },
    {
      "epoch": 0.42535787321063395,
      "grad_norm": 3.784928798675537,
      "learning_rate": 6.628552064731807e-05,
      "loss": 0.9653,
      "step": 832
    },
    {
      "epoch": 0.42586912065439675,
      "grad_norm": 3.955132484436035,
      "learning_rate": 6.620547902879738e-05,
      "loss": 0.9304,
      "step": 833
    },
    {
      "epoch": 0.4263803680981595,
      "grad_norm": 3.502516031265259,
      "learning_rate": 6.612539097954035e-05,
      "loss": 0.7941,
      "step": 834
    },
    {
      "epoch": 0.4268916155419223,
      "grad_norm": 4.0014824867248535,
      "learning_rate": 6.604525672900932e-05,
      "loss": 0.9795,
      "step": 835
    },
    {
      "epoch": 0.4274028629856851,
      "grad_norm": 4.320385932922363,
      "learning_rate": 6.5965076506799e-05,
      "loss": 1.14,
      "step": 836
    },
    {
      "epoch": 0.4279141104294479,
      "grad_norm": 3.5574934482574463,
      "learning_rate": 6.588485054263583e-05,
      "loss": 0.8958,
      "step": 837
    },
    {
      "epoch": 0.4284253578732106,
      "grad_norm": 4.321437358856201,
      "learning_rate": 6.580457906637727e-05,
      "loss": 0.9631,
      "step": 838
    },
    {
      "epoch": 0.4289366053169734,
      "grad_norm": 3.9037375450134277,
      "learning_rate": 6.57242623080112e-05,
      "loss": 0.8151,
      "step": 839
    },
    {
      "epoch": 0.4294478527607362,
      "grad_norm": 4.19363260269165,
      "learning_rate": 6.564390049765528e-05,
      "loss": 0.9471,
      "step": 840
    },
    {
      "epoch": 0.42995910020449896,
      "grad_norm": 3.846937894821167,
      "learning_rate": 6.556349386555615e-05,
      "loss": 0.761,
      "step": 841
    },
    {
      "epoch": 0.43047034764826175,
      "grad_norm": 3.813044786453247,
      "learning_rate": 6.548304264208894e-05,
      "loss": 0.7439,
      "step": 842
    },
    {
      "epoch": 0.43098159509202455,
      "grad_norm": 4.229809284210205,
      "learning_rate": 6.540254705775657e-05,
      "loss": 0.8742,
      "step": 843
    },
    {
      "epoch": 0.43149284253578735,
      "grad_norm": 4.918692588806152,
      "learning_rate": 6.532200734318896e-05,
      "loss": 0.9837,
      "step": 844
    },
    {
      "epoch": 0.4320040899795501,
      "grad_norm": 4.5872955322265625,
      "learning_rate": 6.524142372914255e-05,
      "loss": 0.9378,
      "step": 845
    },
    {
      "epoch": 0.4325153374233129,
      "grad_norm": 4.543177604675293,
      "learning_rate": 6.516079644649954e-05,
      "loss": 0.7143,
      "step": 846
    },
    {
      "epoch": 0.4330265848670757,
      "grad_norm": 5.153186321258545,
      "learning_rate": 6.508012572626724e-05,
      "loss": 0.743,
      "step": 847
    },
    {
      "epoch": 0.4335378323108384,
      "grad_norm": 4.661349296569824,
      "learning_rate": 6.49994117995774e-05,
      "loss": 0.7493,
      "step": 848
    },
    {
      "epoch": 0.4340490797546012,
      "grad_norm": 5.541446685791016,
      "learning_rate": 6.491865489768556e-05,
      "loss": 0.6782,
      "step": 849
    },
    {
      "epoch": 0.434560327198364,
      "grad_norm": 7.626821041107178,
      "learning_rate": 6.483785525197044e-05,
      "loss": 0.7414,
      "step": 850
    },
    {
      "epoch": 0.4350715746421268,
      "grad_norm": 2.0656769275665283,
      "learning_rate": 6.475701309393318e-05,
      "loss": 0.898,
      "step": 851
    },
    {
      "epoch": 0.43558282208588955,
      "grad_norm": 2.457845449447632,
      "learning_rate": 6.467612865519674e-05,
      "loss": 0.9019,
      "step": 852
    },
    {
      "epoch": 0.43609406952965235,
      "grad_norm": 2.4107584953308105,
      "learning_rate": 6.45952021675052e-05,
      "loss": 0.9884,
      "step": 853
    },
    {
      "epoch": 0.43660531697341515,
      "grad_norm": 2.524989128112793,
      "learning_rate": 6.451423386272312e-05,
      "loss": 1.0709,
      "step": 854
    },
    {
      "epoch": 0.4371165644171779,
      "grad_norm": 2.8318159580230713,
      "learning_rate": 6.44332239728349e-05,
      "loss": 0.9647,
      "step": 855
    },
    {
      "epoch": 0.4376278118609407,
      "grad_norm": 2.6179370880126953,
      "learning_rate": 6.435217272994406e-05,
      "loss": 1.0725,
      "step": 856
    },
    {
      "epoch": 0.4381390593047035,
      "grad_norm": 2.7386882305145264,
      "learning_rate": 6.427108036627262e-05,
      "loss": 0.9827,
      "step": 857
    },
    {
      "epoch": 0.4386503067484663,
      "grad_norm": 2.72357439994812,
      "learning_rate": 6.418994711416038e-05,
      "loss": 0.945,
      "step": 858
    },
    {
      "epoch": 0.439161554192229,
      "grad_norm": 2.9313924312591553,
      "learning_rate": 6.410877320606432e-05,
      "loss": 1.0524,
      "step": 859
    },
    {
      "epoch": 0.4396728016359918,
      "grad_norm": 3.1169660091400146,
      "learning_rate": 6.402755887455792e-05,
      "loss": 1.0112,
      "step": 860
    },
    {
      "epoch": 0.4401840490797546,
      "grad_norm": 3.151090621948242,
      "learning_rate": 6.394630435233045e-05,
      "loss": 1.1376,
      "step": 861
    },
    {
      "epoch": 0.44069529652351735,
      "grad_norm": 3.0045671463012695,
      "learning_rate": 6.386500987218633e-05,
      "loss": 0.8827,
      "step": 862
    },
    {
      "epoch": 0.44120654396728015,
      "grad_norm": 3.427995443344116,
      "learning_rate": 6.378367566704449e-05,
      "loss": 0.9608,
      "step": 863
    },
    {
      "epoch": 0.44171779141104295,
      "grad_norm": 3.3372836112976074,
      "learning_rate": 6.370230196993763e-05,
      "loss": 1.0425,
      "step": 864
    },
    {
      "epoch": 0.44222903885480574,
      "grad_norm": 2.8694262504577637,
      "learning_rate": 6.362088901401165e-05,
      "loss": 0.9048,
      "step": 865
    },
    {
      "epoch": 0.4427402862985685,
      "grad_norm": 3.1645376682281494,
      "learning_rate": 6.353943703252493e-05,
      "loss": 1.0502,
      "step": 866
    },
    {
      "epoch": 0.4432515337423313,
      "grad_norm": 3.32002592086792,
      "learning_rate": 6.345794625884762e-05,
      "loss": 0.9785,
      "step": 867
    },
    {
      "epoch": 0.4437627811860941,
      "grad_norm": 3.2119669914245605,
      "learning_rate": 6.337641692646106e-05,
      "loss": 0.9322,
      "step": 868
    },
    {
      "epoch": 0.4442740286298569,
      "grad_norm": 3.1519265174865723,
      "learning_rate": 6.3294849268957e-05,
      "loss": 0.9326,
      "step": 869
    },
    {
      "epoch": 0.4447852760736196,
      "grad_norm": 3.1073238849639893,
      "learning_rate": 6.321324352003711e-05,
      "loss": 0.915,
      "step": 870
    },
    {
      "epoch": 0.4452965235173824,
      "grad_norm": 3.0250244140625,
      "learning_rate": 6.313159991351206e-05,
      "loss": 0.8389,
      "step": 871
    },
    {
      "epoch": 0.4458077709611452,
      "grad_norm": 3.6045401096343994,
      "learning_rate": 6.30499186833011e-05,
      "loss": 0.996,
      "step": 872
    },
    {
      "epoch": 0.44631901840490795,
      "grad_norm": 3.9004733562469482,
      "learning_rate": 6.296820006343122e-05,
      "loss": 0.993,
      "step": 873
    },
    {
      "epoch": 0.44683026584867075,
      "grad_norm": 3.555574417114258,
      "learning_rate": 6.288644428803653e-05,
      "loss": 0.918,
      "step": 874
    },
    {
      "epoch": 0.44734151329243355,
      "grad_norm": 3.5566821098327637,
      "learning_rate": 6.280465159135763e-05,
      "loss": 0.8514,
      "step": 875
    },
    {
      "epoch": 0.44785276073619634,
      "grad_norm": 3.650169849395752,
      "learning_rate": 6.272282220774091e-05,
      "loss": 0.9625,
      "step": 876
    },
    {
      "epoch": 0.4483640081799591,
      "grad_norm": 3.5784823894500732,
      "learning_rate": 6.264095637163779e-05,
      "loss": 0.9592,
      "step": 877
    },
    {
      "epoch": 0.4488752556237219,
      "grad_norm": 3.529385566711426,
      "learning_rate": 6.255905431760424e-05,
      "loss": 0.9422,
      "step": 878
    },
    {
      "epoch": 0.4493865030674847,
      "grad_norm": 3.6039819717407227,
      "learning_rate": 6.247711628029993e-05,
      "loss": 0.8743,
      "step": 879
    },
    {
      "epoch": 0.4498977505112474,
      "grad_norm": 3.4549014568328857,
      "learning_rate": 6.239514249448767e-05,
      "loss": 0.889,
      "step": 880
    },
    {
      "epoch": 0.4504089979550102,
      "grad_norm": 3.8468194007873535,
      "learning_rate": 6.231313319503264e-05,
      "loss": 1.023,
      "step": 881
    },
    {
      "epoch": 0.450920245398773,
      "grad_norm": 4.1849365234375,
      "learning_rate": 6.22310886169018e-05,
      "loss": 0.9639,
      "step": 882
    },
    {
      "epoch": 0.4514314928425358,
      "grad_norm": 3.9194602966308594,
      "learning_rate": 6.21490089951632e-05,
      "loss": 0.9048,
      "step": 883
    },
    {
      "epoch": 0.45194274028629855,
      "grad_norm": 3.7644968032836914,
      "learning_rate": 6.206689456498529e-05,
      "loss": 0.8741,
      "step": 884
    },
    {
      "epoch": 0.45245398773006135,
      "grad_norm": 3.7964026927948,
      "learning_rate": 6.198474556163623e-05,
      "loss": 0.8582,
      "step": 885
    },
    {
      "epoch": 0.45296523517382414,
      "grad_norm": 4.092611312866211,
      "learning_rate": 6.190256222048327e-05,
      "loss": 0.8069,
      "step": 886
    },
    {
      "epoch": 0.4534764826175869,
      "grad_norm": 3.79891300201416,
      "learning_rate": 6.1820344776992e-05,
      "loss": 0.887,
      "step": 887
    },
    {
      "epoch": 0.4539877300613497,
      "grad_norm": 4.135311126708984,
      "learning_rate": 6.173809346672574e-05,
      "loss": 0.8481,
      "step": 888
    },
    {
      "epoch": 0.4544989775051125,
      "grad_norm": 4.441306114196777,
      "learning_rate": 6.165580852534487e-05,
      "loss": 0.8705,
      "step": 889
    },
    {
      "epoch": 0.4550102249488753,
      "grad_norm": 4.326587677001953,
      "learning_rate": 6.157349018860607e-05,
      "loss": 0.9348,
      "step": 890
    },
    {
      "epoch": 0.455521472392638,
      "grad_norm": 5.497438907623291,
      "learning_rate": 6.149113869236175e-05,
      "loss": 0.7662,
      "step": 891
    },
    {
      "epoch": 0.4560327198364008,
      "grad_norm": 4.413537979125977,
      "learning_rate": 6.14087542725593e-05,
      "loss": 0.9143,
      "step": 892
    },
    {
      "epoch": 0.4565439672801636,
      "grad_norm": 4.53402853012085,
      "learning_rate": 6.132633716524046e-05,
      "loss": 0.789,
      "step": 893
    },
    {
      "epoch": 0.4570552147239264,
      "grad_norm": 4.856945037841797,
      "learning_rate": 6.124388760654059e-05,
      "loss": 0.9029,
      "step": 894
    },
    {
      "epoch": 0.45756646216768915,
      "grad_norm": 5.3796610832214355,
      "learning_rate": 6.116140583268806e-05,
      "loss": 0.7887,
      "step": 895
    },
    {
      "epoch": 0.45807770961145194,
      "grad_norm": 4.99921178817749,
      "learning_rate": 6.107889208000354e-05,
      "loss": 0.8267,
      "step": 896
    },
    {
      "epoch": 0.45858895705521474,
      "grad_norm": 5.182647705078125,
      "learning_rate": 6.0996346584899286e-05,
      "loss": 0.8042,
      "step": 897
    },
    {
      "epoch": 0.4591002044989775,
      "grad_norm": 4.6235270500183105,
      "learning_rate": 6.0913769583878555e-05,
      "loss": 0.6185,
      "step": 898
    },
    {
      "epoch": 0.4596114519427403,
      "grad_norm": 5.750977039337158,
      "learning_rate": 6.083116131353484e-05,
      "loss": 0.7183,
      "step": 899
    },
    {
      "epoch": 0.4601226993865031,
      "grad_norm": 7.206790447235107,
      "learning_rate": 6.0748522010551215e-05,
      "loss": 0.4708,
      "step": 900
    },
    {
      "epoch": 0.4606339468302659,
      "grad_norm": 2.1739280223846436,
      "learning_rate": 6.066585191169969e-05,
      "loss": 0.8614,
      "step": 901
    },
    {
      "epoch": 0.4611451942740286,
      "grad_norm": 2.5511627197265625,
      "learning_rate": 6.0583151253840486e-05,
      "loss": 0.9832,
      "step": 902
    },
    {
      "epoch": 0.4616564417177914,
      "grad_norm": 2.4750165939331055,
      "learning_rate": 6.050042027392142e-05,
      "loss": 0.9799,
      "step": 903
    },
    {
      "epoch": 0.4621676891615542,
      "grad_norm": 2.503610849380493,
      "learning_rate": 6.0417659208977127e-05,
      "loss": 0.8716,
      "step": 904
    },
    {
      "epoch": 0.46267893660531695,
      "grad_norm": 2.991971492767334,
      "learning_rate": 6.033486829612851e-05,
      "loss": 0.9855,
      "step": 905
    },
    {
      "epoch": 0.46319018404907975,
      "grad_norm": 2.931039571762085,
      "learning_rate": 6.025204777258191e-05,
      "loss": 1.0956,
      "step": 906
    },
    {
      "epoch": 0.46370143149284254,
      "grad_norm": 2.9422197341918945,
      "learning_rate": 6.016919787562858e-05,
      "loss": 0.9248,
      "step": 907
    },
    {
      "epoch": 0.46421267893660534,
      "grad_norm": 2.9751358032226562,
      "learning_rate": 6.008631884264388e-05,
      "loss": 0.9444,
      "step": 908
    },
    {
      "epoch": 0.4647239263803681,
      "grad_norm": 2.9331753253936768,
      "learning_rate": 6.000341091108665e-05,
      "loss": 1.0345,
      "step": 909
    },
    {
      "epoch": 0.4652351738241309,
      "grad_norm": 2.805023670196533,
      "learning_rate": 5.99204743184986e-05,
      "loss": 0.9525,
      "step": 910
    },
    {
      "epoch": 0.4657464212678937,
      "grad_norm": 2.883270740509033,
      "learning_rate": 5.983750930250345e-05,
      "loss": 0.9199,
      "step": 911
    },
    {
      "epoch": 0.4662576687116564,
      "grad_norm": 3.328139066696167,
      "learning_rate": 5.9754516100806423e-05,
      "loss": 1.0521,
      "step": 912
    },
    {
      "epoch": 0.4667689161554192,
      "grad_norm": 2.935974359512329,
      "learning_rate": 5.9671494951193486e-05,
      "loss": 0.9543,
      "step": 913
    },
    {
      "epoch": 0.467280163599182,
      "grad_norm": 2.9338274002075195,
      "learning_rate": 5.958844609153068e-05,
      "loss": 0.8694,
      "step": 914
    },
    {
      "epoch": 0.4677914110429448,
      "grad_norm": 3.335564374923706,
      "learning_rate": 5.950536975976345e-05,
      "loss": 0.9262,
      "step": 915
    },
    {
      "epoch": 0.46830265848670755,
      "grad_norm": 3.262054681777954,
      "learning_rate": 5.9422266193915924e-05,
      "loss": 0.9122,
      "step": 916
    },
    {
      "epoch": 0.46881390593047034,
      "grad_norm": 3.0267891883850098,
      "learning_rate": 5.933913563209026e-05,
      "loss": 0.8729,
      "step": 917
    },
    {
      "epoch": 0.46932515337423314,
      "grad_norm": 3.102147102355957,
      "learning_rate": 5.925597831246601e-05,
      "loss": 0.9029,
      "step": 918
    },
    {
      "epoch": 0.46983640081799594,
      "grad_norm": 3.2279419898986816,
      "learning_rate": 5.917279447329933e-05,
      "loss": 0.9477,
      "step": 919
    },
    {
      "epoch": 0.4703476482617587,
      "grad_norm": 3.404975652694702,
      "learning_rate": 5.908958435292241e-05,
      "loss": 0.9897,
      "step": 920
    },
    {
      "epoch": 0.4708588957055215,
      "grad_norm": 3.5861120223999023,
      "learning_rate": 5.900634818974269e-05,
      "loss": 0.915,
      "step": 921
    },
    {
      "epoch": 0.47137014314928427,
      "grad_norm": 3.6053731441497803,
      "learning_rate": 5.8923086222242255e-05,
      "loss": 0.8687,
      "step": 922
    },
    {
      "epoch": 0.471881390593047,
      "grad_norm": 3.653125286102295,
      "learning_rate": 5.883979868897712e-05,
      "loss": 0.8448,
      "step": 923
    },
    {
      "epoch": 0.4723926380368098,
      "grad_norm": 3.7469069957733154,
      "learning_rate": 5.8756485828576544e-05,
      "loss": 0.971,
      "step": 924
    },
    {
      "epoch": 0.4729038854805726,
      "grad_norm": 3.449720859527588,
      "learning_rate": 5.867314787974237e-05,
      "loss": 0.877,
      "step": 925
    },
    {
      "epoch": 0.4734151329243354,
      "grad_norm": 3.7333271503448486,
      "learning_rate": 5.858978508124829e-05,
      "loss": 0.9484,
      "step": 926
    },
    {
      "epoch": 0.47392638036809814,
      "grad_norm": 3.670445442199707,
      "learning_rate": 5.8506397671939214e-05,
      "loss": 0.9182,
      "step": 927
    },
    {
      "epoch": 0.47443762781186094,
      "grad_norm": 4.089524745941162,
      "learning_rate": 5.8422985890730576e-05,
      "loss": 0.9254,
      "step": 928
    },
    {
      "epoch": 0.47494887525562374,
      "grad_norm": 3.6270928382873535,
      "learning_rate": 5.833954997660761e-05,
      "loss": 0.9148,
      "step": 929
    },
    {
      "epoch": 0.4754601226993865,
      "grad_norm": 3.7784345149993896,
      "learning_rate": 5.825609016862469e-05,
      "loss": 0.8091,
      "step": 930
    },
    {
      "epoch": 0.4759713701431493,
      "grad_norm": 4.248725414276123,
      "learning_rate": 5.81726067059047e-05,
      "loss": 1.0091,
      "step": 931
    },
    {
      "epoch": 0.47648261758691207,
      "grad_norm": 3.9264259338378906,
      "learning_rate": 5.808909982763825e-05,
      "loss": 0.9502,
      "step": 932
    },
    {
      "epoch": 0.47699386503067487,
      "grad_norm": 3.8590493202209473,
      "learning_rate": 5.8005569773083035e-05,
      "loss": 0.8454,
      "step": 933
    },
    {
      "epoch": 0.4775051124744376,
      "grad_norm": 4.2589802742004395,
      "learning_rate": 5.7922016781563205e-05,
      "loss": 1.0871,
      "step": 934
    },
    {
      "epoch": 0.4780163599182004,
      "grad_norm": 3.9979145526885986,
      "learning_rate": 5.7838441092468565e-05,
      "loss": 0.8526,
      "step": 935
    },
    {
      "epoch": 0.4785276073619632,
      "grad_norm": 4.0638041496276855,
      "learning_rate": 5.775484294525399e-05,
      "loss": 0.9021,
      "step": 936
    },
    {
      "epoch": 0.47903885480572594,
      "grad_norm": 3.9989681243896484,
      "learning_rate": 5.767122257943869e-05,
      "loss": 0.8017,
      "step": 937
    },
    {
      "epoch": 0.47955010224948874,
      "grad_norm": 3.931239366531372,
      "learning_rate": 5.758758023460553e-05,
      "loss": 0.8762,
      "step": 938
    },
    {
      "epoch": 0.48006134969325154,
      "grad_norm": 4.070004463195801,
      "learning_rate": 5.750391615040037e-05,
      "loss": 0.7725,
      "step": 939
    },
    {
      "epoch": 0.48057259713701433,
      "grad_norm": 3.9469223022460938,
      "learning_rate": 5.742023056653131e-05,
      "loss": 0.7972,
      "step": 940
    },
    {
      "epoch": 0.4810838445807771,
      "grad_norm": 4.728452682495117,
      "learning_rate": 5.733652372276809e-05,
      "loss": 0.9215,
      "step": 941
    },
    {
      "epoch": 0.4815950920245399,
      "grad_norm": 4.572932243347168,
      "learning_rate": 5.7252795858941344e-05,
      "loss": 0.8456,
      "step": 942
    },
    {
      "epoch": 0.48210633946830267,
      "grad_norm": 5.1015520095825195,
      "learning_rate": 5.716904721494195e-05,
      "loss": 0.9307,
      "step": 943
    },
    {
      "epoch": 0.48261758691206547,
      "grad_norm": 4.975411415100098,
      "learning_rate": 5.70852780307203e-05,
      "loss": 0.7936,
      "step": 944
    },
    {
      "epoch": 0.4831288343558282,
      "grad_norm": 4.388181209564209,
      "learning_rate": 5.7001488546285666e-05,
      "loss": 0.7024,
      "step": 945
    },
    {
      "epoch": 0.483640081799591,
      "grad_norm": 5.069786548614502,
      "learning_rate": 5.691767900170542e-05,
      "loss": 0.6689,
      "step": 946
    },
    {
      "epoch": 0.4841513292433538,
      "grad_norm": 5.24841833114624,
      "learning_rate": 5.6833849637104476e-05,
      "loss": 0.8018,
      "step": 947
    },
    {
      "epoch": 0.48466257668711654,
      "grad_norm": 5.333656311035156,
      "learning_rate": 5.675000069266451e-05,
      "loss": 0.7003,
      "step": 948
    },
    {
      "epoch": 0.48517382413087934,
      "grad_norm": 6.533324718475342,
      "learning_rate": 5.666613240862331e-05,
      "loss": 0.6977,
      "step": 949
    },
    {
      "epoch": 0.48568507157464214,
      "grad_norm": 8.306028366088867,
      "learning_rate": 5.658224502527404e-05,
      "loss": 1.1272,
      "step": 950
    },
    {
      "epoch": 0.48619631901840493,
      "grad_norm": 2.5492875576019287,
      "learning_rate": 5.649833878296462e-05,
      "loss": 0.8884,
      "step": 951
    },
    {
      "epoch": 0.4867075664621677,
      "grad_norm": 2.5944671630859375,
      "learning_rate": 5.641441392209699e-05,
      "loss": 0.8951,
      "step": 952
    },
    {
      "epoch": 0.48721881390593047,
      "grad_norm": 2.6134727001190186,
      "learning_rate": 5.6330470683126424e-05,
      "loss": 0.9544,
      "step": 953
    },
    {
      "epoch": 0.48773006134969327,
      "grad_norm": 2.720135450363159,
      "learning_rate": 5.624650930656089e-05,
      "loss": 0.9283,
      "step": 954
    },
    {
      "epoch": 0.488241308793456,
      "grad_norm": 2.600879669189453,
      "learning_rate": 5.616253003296027e-05,
      "loss": 0.9107,
      "step": 955
    },
    {
      "epoch": 0.4887525562372188,
      "grad_norm": 2.7407121658325195,
      "learning_rate": 5.6078533102935745e-05,
      "loss": 0.95,
      "step": 956
    },
    {
      "epoch": 0.4892638036809816,
      "grad_norm": 3.158928155899048,
      "learning_rate": 5.599451875714913e-05,
      "loss": 0.9942,
      "step": 957
    },
    {
      "epoch": 0.4897750511247444,
      "grad_norm": 2.9920670986175537,
      "learning_rate": 5.5910487236312045e-05,
      "loss": 1.0038,
      "step": 958
    },
    {
      "epoch": 0.49028629856850714,
      "grad_norm": 3.3155324459075928,
      "learning_rate": 5.582643878118541e-05,
      "loss": 0.9121,
      "step": 959
    },
    {
      "epoch": 0.49079754601226994,
      "grad_norm": 3.5027592182159424,
      "learning_rate": 5.574237363257858e-05,
      "loss": 0.9685,
      "step": 960
    },
    {
      "epoch": 0.49130879345603273,
      "grad_norm": 3.1491141319274902,
      "learning_rate": 5.565829203134881e-05,
      "loss": 1.1047,
      "step": 961
    },
    {
      "epoch": 0.4918200408997955,
      "grad_norm": 3.131809949874878,
      "learning_rate": 5.557419421840048e-05,
      "loss": 0.9373,
      "step": 962
    },
    {
      "epoch": 0.49233128834355827,
      "grad_norm": 3.011101245880127,
      "learning_rate": 5.549008043468439e-05,
      "loss": 0.9795,
      "step": 963
    },
    {
      "epoch": 0.49284253578732107,
      "grad_norm": 3.0677943229675293,
      "learning_rate": 5.540595092119709e-05,
      "loss": 0.9323,
      "step": 964
    },
    {
      "epoch": 0.49335378323108386,
      "grad_norm": 2.9634270668029785,
      "learning_rate": 5.532180591898026e-05,
      "loss": 0.917,
      "step": 965
    },
    {
      "epoch": 0.4938650306748466,
      "grad_norm": 3.0387773513793945,
      "learning_rate": 5.5237645669119895e-05,
      "loss": 0.8734,
      "step": 966
    },
    {
      "epoch": 0.4943762781186094,
      "grad_norm": 3.3674721717834473,
      "learning_rate": 5.51534704127457e-05,
      "loss": 0.9791,
      "step": 967
    },
    {
      "epoch": 0.4948875255623722,
      "grad_norm": 3.1492297649383545,
      "learning_rate": 5.50692803910304e-05,
      "loss": 0.9631,
      "step": 968
    },
    {
      "epoch": 0.495398773006135,
      "grad_norm": 3.6630218029022217,
      "learning_rate": 5.498507584518896e-05,
      "loss": 0.9708,
      "step": 969
    },
    {
      "epoch": 0.49591002044989774,
      "grad_norm": 3.3851146697998047,
      "learning_rate": 5.490085701647805e-05,
      "loss": 0.8997,
      "step": 970
    },
    {
      "epoch": 0.49642126789366053,
      "grad_norm": 3.156371831893921,
      "learning_rate": 5.481662414619515e-05,
      "loss": 0.9712,
      "step": 971
    },
    {
      "epoch": 0.49693251533742333,
      "grad_norm": 3.431939125061035,
      "learning_rate": 5.473237747567805e-05,
      "loss": 0.9149,
      "step": 972
    },
    {
      "epoch": 0.49744376278118607,
      "grad_norm": 3.7921383380889893,
      "learning_rate": 5.464811724630411e-05,
      "loss": 0.8872,
      "step": 973
    },
    {
      "epoch": 0.49795501022494887,
      "grad_norm": 3.2609188556671143,
      "learning_rate": 5.456384369948943e-05,
      "loss": 0.8625,
      "step": 974
    },
    {
      "epoch": 0.49846625766871167,
      "grad_norm": 3.48323392868042,
      "learning_rate": 5.4479557076688325e-05,
      "loss": 0.9007,
      "step": 975
    },
    {
      "epoch": 0.49897750511247446,
      "grad_norm": 3.2786710262298584,
      "learning_rate": 5.439525761939261e-05,
      "loss": 0.8592,
      "step": 976
    },
    {
      "epoch": 0.4994887525562372,
      "grad_norm": 3.686288833618164,
      "learning_rate": 5.431094556913082e-05,
      "loss": 0.8544,
      "step": 977
    },
    {
      "epoch": 0.5,
      "grad_norm": 4.029841899871826,
      "learning_rate": 5.422662116746759e-05,
      "loss": 0.9432,
      "step": 978
    },
    {
      "epoch": 0.5005112474437627,
      "grad_norm": 3.703411102294922,
      "learning_rate": 5.414228465600293e-05,
      "loss": 0.9156,
      "step": 979
    },
    {
      "epoch": 0.5010224948875256,
      "grad_norm": 4.18709659576416,
      "learning_rate": 5.4057936276371565e-05,
      "loss": 0.9549,
      "step": 980
    },
    {
      "epoch": 0.5015337423312883,
      "grad_norm": 3.750430107116699,
      "learning_rate": 5.397357627024221e-05,
      "loss": 0.8906,
      "step": 981
    },
    {
      "epoch": 0.5020449897750511,
      "grad_norm": 3.5980660915374756,
      "learning_rate": 5.3889204879316913e-05,
      "loss": 0.8738,
      "step": 982
    },
    {
      "epoch": 0.5025562372188139,
      "grad_norm": 4.250522136688232,
      "learning_rate": 5.3804822345330295e-05,
      "loss": 0.9729,
      "step": 983
    },
    {
      "epoch": 0.5030674846625767,
      "grad_norm": 4.09278678894043,
      "learning_rate": 5.372042891004896e-05,
      "loss": 0.8706,
      "step": 984
    },
    {
      "epoch": 0.5035787321063395,
      "grad_norm": 4.25390100479126,
      "learning_rate": 5.36360248152707e-05,
      "loss": 0.9537,
      "step": 985
    },
    {
      "epoch": 0.5040899795501023,
      "grad_norm": 3.719179153442383,
      "learning_rate": 5.355161030282387e-05,
      "loss": 0.8335,
      "step": 986
    },
    {
      "epoch": 0.504601226993865,
      "grad_norm": 3.710343837738037,
      "learning_rate": 5.346718561456669e-05,
      "loss": 0.8313,
      "step": 987
    },
    {
      "epoch": 0.5051124744376279,
      "grad_norm": 4.005296230316162,
      "learning_rate": 5.338275099238647e-05,
      "loss": 0.9525,
      "step": 988
    },
    {
      "epoch": 0.5056237218813906,
      "grad_norm": 4.096909523010254,
      "learning_rate": 5.329830667819905e-05,
      "loss": 0.8994,
      "step": 989
    },
    {
      "epoch": 0.5061349693251533,
      "grad_norm": 4.005760669708252,
      "learning_rate": 5.3213852913948026e-05,
      "loss": 0.894,
      "step": 990
    },
    {
      "epoch": 0.5066462167689162,
      "grad_norm": 4.101000785827637,
      "learning_rate": 5.312938994160404e-05,
      "loss": 0.872,
      "step": 991
    },
    {
      "epoch": 0.5071574642126789,
      "grad_norm": 4.393942356109619,
      "learning_rate": 5.3044918003164156e-05,
      "loss": 0.8976,
      "step": 992
    },
    {
      "epoch": 0.5076687116564417,
      "grad_norm": 4.222607135772705,
      "learning_rate": 5.296043734065108e-05,
      "loss": 0.7865,
      "step": 993
    },
    {
      "epoch": 0.5081799591002045,
      "grad_norm": 4.471764087677002,
      "learning_rate": 5.287594819611256e-05,
      "loss": 0.8599,
      "step": 994
    },
    {
      "epoch": 0.5086912065439673,
      "grad_norm": 4.455383777618408,
      "learning_rate": 5.2791450811620615e-05,
      "loss": 0.839,
      "step": 995
    },
    {
      "epoch": 0.50920245398773,
      "grad_norm": 4.264684200286865,
      "learning_rate": 5.270694542927088e-05,
      "loss": 0.7249,
      "step": 996
    },
    {
      "epoch": 0.5097137014314929,
      "grad_norm": 4.615438938140869,
      "learning_rate": 5.262243229118192e-05,
      "loss": 0.8047,
      "step": 997
    },
    {
      "epoch": 0.5102249488752556,
      "grad_norm": 5.186718940734863,
      "learning_rate": 5.2537911639494494e-05,
      "loss": 0.773,
      "step": 998
    },
    {
      "epoch": 0.5107361963190185,
      "grad_norm": 5.383667945861816,
      "learning_rate": 5.245338371637091e-05,
      "loss": 0.6268,
      "step": 999
    },
    {
      "epoch": 0.5112474437627812,
      "grad_norm": 9.21727466583252,
      "learning_rate": 5.236884876399429e-05,
      "loss": 0.9014,
      "step": 1000
    },
    {
      "epoch": 0.5117586912065439,
      "grad_norm": 2.15889573097229,
      "learning_rate": 5.2284307024567936e-05,
      "loss": 0.8665,
      "step": 1001
    },
    {
      "epoch": 0.5122699386503068,
      "grad_norm": 2.28460693359375,
      "learning_rate": 5.219975874031456e-05,
      "loss": 0.9948,
      "step": 1002
    },
    {
      "epoch": 0.5127811860940695,
      "grad_norm": 2.767993211746216,
      "learning_rate": 5.211520415347562e-05,
      "loss": 1.0241,
      "step": 1003
    },
    {
      "epoch": 0.5132924335378323,
      "grad_norm": 2.5393383502960205,
      "learning_rate": 5.203064350631064e-05,
      "loss": 0.8823,
      "step": 1004
    },
    {
      "epoch": 0.5138036809815951,
      "grad_norm": 2.8708670139312744,
      "learning_rate": 5.194607704109653e-05,
      "loss": 1.1419,
      "step": 1005
    },
    {
      "epoch": 0.5143149284253579,
      "grad_norm": 2.744375228881836,
      "learning_rate": 5.186150500012685e-05,
      "loss": 1.0083,
      "step": 1006
    },
    {
      "epoch": 0.5148261758691206,
      "grad_norm": 2.742351770401001,
      "learning_rate": 5.1776927625711156e-05,
      "loss": 0.9516,
      "step": 1007
    },
    {
      "epoch": 0.5153374233128835,
      "grad_norm": 3.1860601902008057,
      "learning_rate": 5.1692345160174225e-05,
      "loss": 0.9949,
      "step": 1008
    },
    {
      "epoch": 0.5158486707566462,
      "grad_norm": 2.853633403778076,
      "learning_rate": 5.160775784585551e-05,
      "loss": 0.9623,
      "step": 1009
    },
    {
      "epoch": 0.516359918200409,
      "grad_norm": 3.070064067840576,
      "learning_rate": 5.152316592510826e-05,
      "loss": 0.9656,
      "step": 1010
    },
    {
      "epoch": 0.5168711656441718,
      "grad_norm": 3.3969080448150635,
      "learning_rate": 5.1438569640299006e-05,
      "loss": 0.9152,
      "step": 1011
    },
    {
      "epoch": 0.5173824130879345,
      "grad_norm": 3.090078830718994,
      "learning_rate": 5.135396923380673e-05,
      "loss": 1.0513,
      "step": 1012
    },
    {
      "epoch": 0.5178936605316974,
      "grad_norm": 3.320444107055664,
      "learning_rate": 5.1269364948022255e-05,
      "loss": 0.9246,
      "step": 1013
    },
    {
      "epoch": 0.5184049079754601,
      "grad_norm": 3.0707502365112305,
      "learning_rate": 5.118475702534748e-05,
      "loss": 0.8337,
      "step": 1014
    },
    {
      "epoch": 0.5189161554192229,
      "grad_norm": 3.0075268745422363,
      "learning_rate": 5.110014570819478e-05,
      "loss": 0.9142,
      "step": 1015
    },
    {
      "epoch": 0.5194274028629857,
      "grad_norm": 2.8207309246063232,
      "learning_rate": 5.101553123898622e-05,
      "loss": 0.8493,
      "step": 1016
    },
    {
      "epoch": 0.5199386503067485,
      "grad_norm": 3.2407922744750977,
      "learning_rate": 5.093091386015287e-05,
      "loss": 0.8856,
      "step": 1017
    },
    {
      "epoch": 0.5204498977505112,
      "grad_norm": 3.477283239364624,
      "learning_rate": 5.08462938141342e-05,
      "loss": 0.989,
      "step": 1018
    },
    {
      "epoch": 0.5209611451942741,
      "grad_norm": 3.3261115550994873,
      "learning_rate": 5.0761671343377256e-05,
      "loss": 0.9585,
      "step": 1019
    },
    {
      "epoch": 0.5214723926380368,
      "grad_norm": 3.273200273513794,
      "learning_rate": 5.0677046690336096e-05,
      "loss": 0.914,
      "step": 1020
    },
    {
      "epoch": 0.5219836400817995,
      "grad_norm": 3.3354554176330566,
      "learning_rate": 5.0592420097471e-05,
      "loss": 0.9123,
      "step": 1021
    },
    {
      "epoch": 0.5224948875255624,
      "grad_norm": 3.29411244392395,
      "learning_rate": 5.050779180724776e-05,
      "loss": 0.865,
      "step": 1022
    },
    {
      "epoch": 0.5230061349693251,
      "grad_norm": 3.2440457344055176,
      "learning_rate": 5.042316206213712e-05,
      "loss": 0.7901,
      "step": 1023
    },
    {
      "epoch": 0.523517382413088,
      "grad_norm": 3.6321732997894287,
      "learning_rate": 5.0338531104613926e-05,
      "loss": 0.9108,
      "step": 1024
    },
    {
      "epoch": 0.5240286298568507,
      "grad_norm": 3.4314496517181396,
      "learning_rate": 5.025389917715653e-05,
      "loss": 0.8733,
      "step": 1025
    },
    {
      "epoch": 0.5245398773006135,
      "grad_norm": 3.502465009689331,
      "learning_rate": 5.016926652224604e-05,
      "loss": 1.011,
      "step": 1026
    },
    {
      "epoch": 0.5250511247443763,
      "grad_norm": 3.4616212844848633,
      "learning_rate": 5.008463338236566e-05,
      "loss": 0.8113,
      "step": 1027
    },
    {
      "epoch": 0.5255623721881391,
      "grad_norm": 3.4691240787506104,
      "learning_rate": 5e-05,
      "loss": 0.8563,
      "step": 1028
    },
    {
      "epoch": 0.5260736196319018,
      "grad_norm": 3.7538325786590576,
      "learning_rate": 4.991536661763434e-05,
      "loss": 0.9536,
      "step": 1029
    },
    {
      "epoch": 0.5265848670756647,
      "grad_norm": 3.647031307220459,
      "learning_rate": 4.983073347775397e-05,
      "loss": 0.9124,
      "step": 1030
    },
    {
      "epoch": 0.5270961145194274,
      "grad_norm": 3.6543843746185303,
      "learning_rate": 4.9746100822843477e-05,
      "loss": 0.9421,
      "step": 1031
    },
    {
      "epoch": 0.5276073619631901,
      "grad_norm": 3.728757619857788,
      "learning_rate": 4.966146889538608e-05,
      "loss": 0.8373,
      "step": 1032
    },
    {
      "epoch": 0.528118609406953,
      "grad_norm": 3.5960378646850586,
      "learning_rate": 4.95768379378629e-05,
      "loss": 0.9218,
      "step": 1033
    },
    {
      "epoch": 0.5286298568507157,
      "grad_norm": 3.6316988468170166,
      "learning_rate": 4.9492208192752256e-05,
      "loss": 0.9065,
      "step": 1034
    },
    {
      "epoch": 0.5291411042944786,
      "grad_norm": 4.244678020477295,
      "learning_rate": 4.940757990252902e-05,
      "loss": 0.8062,
      "step": 1035
    },
    {
      "epoch": 0.5296523517382413,
      "grad_norm": 3.584723711013794,
      "learning_rate": 4.9322953309663916e-05,
      "loss": 0.8503,
      "step": 1036
    },
    {
      "epoch": 0.5301635991820041,
      "grad_norm": 4.005549430847168,
      "learning_rate": 4.923832865662275e-05,
      "loss": 0.895,
      "step": 1037
    },
    {
      "epoch": 0.5306748466257669,
      "grad_norm": 4.480841636657715,
      "learning_rate": 4.9153706185865815e-05,
      "loss": 1.0696,
      "step": 1038
    },
    {
      "epoch": 0.5311860940695297,
      "grad_norm": 4.540712356567383,
      "learning_rate": 4.9069086139847134e-05,
      "loss": 1.0161,
      "step": 1039
    },
    {
      "epoch": 0.5316973415132924,
      "grad_norm": 4.2374372482299805,
      "learning_rate": 4.898446876101379e-05,
      "loss": 0.9821,
      "step": 1040
    },
    {
      "epoch": 0.5322085889570553,
      "grad_norm": 4.287907600402832,
      "learning_rate": 4.889985429180521e-05,
      "loss": 0.7927,
      "step": 1041
    },
    {
      "epoch": 0.532719836400818,
      "grad_norm": 4.230398654937744,
      "learning_rate": 4.881524297465251e-05,
      "loss": 0.8622,
      "step": 1042
    },
    {
      "epoch": 0.5332310838445807,
      "grad_norm": 4.255669593811035,
      "learning_rate": 4.873063505197777e-05,
      "loss": 0.7223,
      "step": 1043
    },
    {
      "epoch": 0.5337423312883436,
      "grad_norm": 4.257006645202637,
      "learning_rate": 4.8646030766193285e-05,
      "loss": 0.7052,
      "step": 1044
    },
    {
      "epoch": 0.5342535787321063,
      "grad_norm": 4.510742664337158,
      "learning_rate": 4.856143035970101e-05,
      "loss": 0.7036,
      "step": 1045
    },
    {
      "epoch": 0.5347648261758691,
      "grad_norm": 5.051736831665039,
      "learning_rate": 4.847683407489175e-05,
      "loss": 0.8619,
      "step": 1046
    },
    {
      "epoch": 0.5352760736196319,
      "grad_norm": 4.3064093589782715,
      "learning_rate": 4.83922421541445e-05,
      "loss": 0.6374,
      "step": 1047
    },
    {
      "epoch": 0.5357873210633947,
      "grad_norm": 4.907425403594971,
      "learning_rate": 4.830765483982578e-05,
      "loss": 0.6855,
      "step": 1048
    },
    {
      "epoch": 0.5362985685071575,
      "grad_norm": 6.278970241546631,
      "learning_rate": 4.822307237428885e-05,
      "loss": 0.7517,
      "step": 1049
    },
    {
      "epoch": 0.5368098159509203,
      "grad_norm": 7.862913608551025,
      "learning_rate": 4.813849499987314e-05,
      "loss": 0.7461,
      "step": 1050
    },
    {
      "epoch": 0.537321063394683,
      "grad_norm": 2.1761088371276855,
      "learning_rate": 4.8053922958903466e-05,
      "loss": 1.0014,
      "step": 1051
    },
    {
      "epoch": 0.5378323108384458,
      "grad_norm": 2.4289817810058594,
      "learning_rate": 4.796935649368935e-05,
      "loss": 1.0023,
      "step": 1052
    },
    {
      "epoch": 0.5383435582822086,
      "grad_norm": 2.4582669734954834,
      "learning_rate": 4.788479584652441e-05,
      "loss": 1.0581,
      "step": 1053
    },
    {
      "epoch": 0.5388548057259713,
      "grad_norm": 2.5545613765716553,
      "learning_rate": 4.7800241259685464e-05,
      "loss": 1.0115,
      "step": 1054
    },
    {
      "epoch": 0.5393660531697342,
      "grad_norm": 2.4820990562438965,
      "learning_rate": 4.7715692975432076e-05,
      "loss": 0.941,
      "step": 1055
    },
    {
      "epoch": 0.5398773006134969,
      "grad_norm": 2.6315999031066895,
      "learning_rate": 4.763115123600571e-05,
      "loss": 0.9529,
      "step": 1056
    },
    {
      "epoch": 0.5403885480572597,
      "grad_norm": 2.7553720474243164,
      "learning_rate": 4.7546616283629105e-05,
      "loss": 0.9689,
      "step": 1057
    },
    {
      "epoch": 0.5408997955010225,
      "grad_norm": 2.7258827686309814,
      "learning_rate": 4.746208836050551e-05,
      "loss": 0.9044,
      "step": 1058
    },
    {
      "epoch": 0.5414110429447853,
      "grad_norm": 2.9439940452575684,
      "learning_rate": 4.737756770881809e-05,
      "loss": 0.9474,
      "step": 1059
    },
    {
      "epoch": 0.5419222903885481,
      "grad_norm": 3.022825241088867,
      "learning_rate": 4.729305457072913e-05,
      "loss": 1.0523,
      "step": 1060
    },
    {
      "epoch": 0.5424335378323109,
      "grad_norm": 3.053812265396118,
      "learning_rate": 4.720854918837939e-05,
      "loss": 0.8946,
      "step": 1061
    },
    {
      "epoch": 0.5429447852760736,
      "grad_norm": 3.0392403602600098,
      "learning_rate": 4.712405180388745e-05,
      "loss": 0.9809,
      "step": 1062
    },
    {
      "epoch": 0.5434560327198364,
      "grad_norm": 2.9778902530670166,
      "learning_rate": 4.703956265934894e-05,
      "loss": 0.883,
      "step": 1063
    },
    {
      "epoch": 0.5439672801635992,
      "grad_norm": 3.0558431148529053,
      "learning_rate": 4.695508199683586e-05,
      "loss": 0.8867,
      "step": 1064
    },
    {
      "epoch": 0.5444785276073619,
      "grad_norm": 3.2434241771698,
      "learning_rate": 4.687061005839597e-05,
      "loss": 0.9211,
      "step": 1065
    },
    {
      "epoch": 0.5449897750511248,
      "grad_norm": 3.273984909057617,
      "learning_rate": 4.678614708605199e-05,
      "loss": 0.9619,
      "step": 1066
    },
    {
      "epoch": 0.5455010224948875,
      "grad_norm": 3.257916212081909,
      "learning_rate": 4.670169332180096e-05,
      "loss": 0.9638,
      "step": 1067
    },
    {
      "epoch": 0.5460122699386503,
      "grad_norm": 3.309859275817871,
      "learning_rate": 4.6617249007613544e-05,
      "loss": 1.0217,
      "step": 1068
    },
    {
      "epoch": 0.5465235173824131,
      "grad_norm": 3.2615549564361572,
      "learning_rate": 4.653281438543333e-05,
      "loss": 0.9133,
      "step": 1069
    },
    {
      "epoch": 0.5470347648261759,
      "grad_norm": 3.446503162384033,
      "learning_rate": 4.644838969717613e-05,
      "loss": 0.9331,
      "step": 1070
    },
    {
      "epoch": 0.5475460122699386,
      "grad_norm": 3.353502035140991,
      "learning_rate": 4.636397518472931e-05,
      "loss": 0.9512,
      "step": 1071
    },
    {
      "epoch": 0.5480572597137015,
      "grad_norm": 3.027780055999756,
      "learning_rate": 4.6279571089951054e-05,
      "loss": 0.8765,
      "step": 1072
    },
    {
      "epoch": 0.5485685071574642,
      "grad_norm": 3.0944087505340576,
      "learning_rate": 4.619517765466972e-05,
      "loss": 0.9157,
      "step": 1073
    },
    {
      "epoch": 0.549079754601227,
      "grad_norm": 3.496917486190796,
      "learning_rate": 4.61107951206831e-05,
      "loss": 0.9094,
      "step": 1074
    },
    {
      "epoch": 0.5495910020449898,
      "grad_norm": 3.2800345420837402,
      "learning_rate": 4.60264237297578e-05,
      "loss": 0.9757,
      "step": 1075
    },
    {
      "epoch": 0.5501022494887525,
      "grad_norm": 3.8397467136383057,
      "learning_rate": 4.594206372362845e-05,
      "loss": 0.9101,
      "step": 1076
    },
    {
      "epoch": 0.5506134969325154,
      "grad_norm": 3.393571376800537,
      "learning_rate": 4.5857715343997076e-05,
      "loss": 0.8935,
      "step": 1077
    },
    {
      "epoch": 0.5511247443762781,
      "grad_norm": 3.46854829788208,
      "learning_rate": 4.577337883253242e-05,
      "loss": 0.9132,
      "step": 1078
    },
    {
      "epoch": 0.5516359918200409,
      "grad_norm": 3.91577410697937,
      "learning_rate": 4.5689054430869184e-05,
      "loss": 0.8591,
      "step": 1079
    },
    {
      "epoch": 0.5521472392638037,
      "grad_norm": 3.561288356781006,
      "learning_rate": 4.560474238060739e-05,
      "loss": 0.8733,
      "step": 1080
    },
    {
      "epoch": 0.5526584867075665,
      "grad_norm": 3.7657833099365234,
      "learning_rate": 4.5520442923311666e-05,
      "loss": 0.8395,
      "step": 1081
    },
    {
      "epoch": 0.5531697341513292,
      "grad_norm": 3.7186317443847656,
      "learning_rate": 4.54361563005106e-05,
      "loss": 0.9149,
      "step": 1082
    },
    {
      "epoch": 0.553680981595092,
      "grad_norm": 4.215698719024658,
      "learning_rate": 4.535188275369592e-05,
      "loss": 0.8546,
      "step": 1083
    },
    {
      "epoch": 0.5541922290388548,
      "grad_norm": 4.031459808349609,
      "learning_rate": 4.526762252432195e-05,
      "loss": 0.8887,
      "step": 1084
    },
    {
      "epoch": 0.5547034764826176,
      "grad_norm": 3.89286208152771,
      "learning_rate": 4.5183375853804864e-05,
      "loss": 0.8501,
      "step": 1085
    },
    {
      "epoch": 0.5552147239263804,
      "grad_norm": 4.312307357788086,
      "learning_rate": 4.509914298352197e-05,
      "loss": 0.8708,
      "step": 1086
    },
    {
      "epoch": 0.5557259713701431,
      "grad_norm": 4.412914752960205,
      "learning_rate": 4.501492415481105e-05,
      "loss": 0.923,
      "step": 1087
    },
    {
      "epoch": 0.556237218813906,
      "grad_norm": 3.436657667160034,
      "learning_rate": 4.493071960896961e-05,
      "loss": 0.7221,
      "step": 1088
    },
    {
      "epoch": 0.5567484662576687,
      "grad_norm": 4.254183292388916,
      "learning_rate": 4.4846529587254296e-05,
      "loss": 0.8008,
      "step": 1089
    },
    {
      "epoch": 0.5572597137014315,
      "grad_norm": 4.078035354614258,
      "learning_rate": 4.476235433088011e-05,
      "loss": 0.7969,
      "step": 1090
    },
    {
      "epoch": 0.5577709611451943,
      "grad_norm": 4.366103649139404,
      "learning_rate": 4.4678194081019764e-05,
      "loss": 1.0288,
      "step": 1091
    },
    {
      "epoch": 0.558282208588957,
      "grad_norm": 4.415543079376221,
      "learning_rate": 4.4594049078802925e-05,
      "loss": 0.8554,
      "step": 1092
    },
    {
      "epoch": 0.5587934560327198,
      "grad_norm": 4.634888648986816,
      "learning_rate": 4.450991956531564e-05,
      "loss": 0.82,
      "step": 1093
    },
    {
      "epoch": 0.5593047034764826,
      "grad_norm": 5.047021865844727,
      "learning_rate": 4.4425805781599535e-05,
      "loss": 0.9254,
      "step": 1094
    },
    {
      "epoch": 0.5598159509202454,
      "grad_norm": 4.863358497619629,
      "learning_rate": 4.434170796865119e-05,
      "loss": 0.7494,
      "step": 1095
    },
    {
      "epoch": 0.5603271983640081,
      "grad_norm": 4.6440863609313965,
      "learning_rate": 4.425762636742143e-05,
      "loss": 0.7075,
      "step": 1096
    },
    {
      "epoch": 0.560838445807771,
      "grad_norm": 5.290436744689941,
      "learning_rate": 4.417356121881461e-05,
      "loss": 0.6747,
      "step": 1097
    },
    {
      "epoch": 0.5613496932515337,
      "grad_norm": 5.783541202545166,
      "learning_rate": 4.4089512763687966e-05,
      "loss": 0.764,
      "step": 1098
    },
    {
      "epoch": 0.5618609406952966,
      "grad_norm": 5.853088855743408,
      "learning_rate": 4.4005481242850884e-05,
      "loss": 0.5864,
      "step": 1099
    },
    {
      "epoch": 0.5623721881390593,
      "grad_norm": 7.8992791175842285,
      "learning_rate": 4.392146689706425e-05,
      "loss": 0.712,
      "step": 1100
    },
    {
      "epoch": 0.5628834355828221,
      "grad_norm": 1.9316519498825073,
      "learning_rate": 4.3837469967039745e-05,
      "loss": 0.7383,
      "step": 1101
    },
    {
      "epoch": 0.5633946830265849,
      "grad_norm": 2.5184829235076904,
      "learning_rate": 4.3753490693439135e-05,
      "loss": 0.9712,
      "step": 1102
    },
    {
      "epoch": 0.5639059304703476,
      "grad_norm": 2.3476760387420654,
      "learning_rate": 4.3669529316873594e-05,
      "loss": 0.9155,
      "step": 1103
    },
    {
      "epoch": 0.5644171779141104,
      "grad_norm": 2.6786763668060303,
      "learning_rate": 4.358558607790303e-05,
      "loss": 0.9806,
      "step": 1104
    },
    {
      "epoch": 0.5649284253578732,
      "grad_norm": 2.709429979324341,
      "learning_rate": 4.350166121703539e-05,
      "loss": 0.9601,
      "step": 1105
    },
    {
      "epoch": 0.565439672801636,
      "grad_norm": 2.7100303173065186,
      "learning_rate": 4.341775497472597e-05,
      "loss": 0.9591,
      "step": 1106
    },
    {
      "epoch": 0.5659509202453987,
      "grad_norm": 3.0667576789855957,
      "learning_rate": 4.3333867591376704e-05,
      "loss": 0.9735,
      "step": 1107
    },
    {
      "epoch": 0.5664621676891616,
      "grad_norm": 2.943694591522217,
      "learning_rate": 4.3249999307335495e-05,
      "loss": 0.9863,
      "step": 1108
    },
    {
      "epoch": 0.5669734151329243,
      "grad_norm": 3.2859537601470947,
      "learning_rate": 4.316615036289553e-05,
      "loss": 0.9712,
      "step": 1109
    },
    {
      "epoch": 0.5674846625766872,
      "grad_norm": 2.9579572677612305,
      "learning_rate": 4.3082320998294594e-05,
      "loss": 0.9693,
      "step": 1110
    },
    {
      "epoch": 0.5679959100204499,
      "grad_norm": 3.6050679683685303,
      "learning_rate": 4.2998511453714366e-05,
      "loss": 1.0155,
      "step": 1111
    },
    {
      "epoch": 0.5685071574642127,
      "grad_norm": 3.0228002071380615,
      "learning_rate": 4.2914721969279705e-05,
      "loss": 0.9553,
      "step": 1112
    },
    {
      "epoch": 0.5690184049079755,
      "grad_norm": 3.024217128753662,
      "learning_rate": 4.283095278505806e-05,
      "loss": 1.0179,
      "step": 1113
    },
    {
      "epoch": 0.5695296523517382,
      "grad_norm": 2.834343433380127,
      "learning_rate": 4.274720414105866e-05,
      "loss": 0.8619,
      "step": 1114
    },
    {
      "epoch": 0.570040899795501,
      "grad_norm": 2.9928078651428223,
      "learning_rate": 4.2663476277231915e-05,
      "loss": 0.9013,
      "step": 1115
    },
    {
      "epoch": 0.5705521472392638,
      "grad_norm": 2.8394672870635986,
      "learning_rate": 4.2579769433468694e-05,
      "loss": 0.8572,
      "step": 1116
    },
    {
      "epoch": 0.5710633946830266,
      "grad_norm": 3.117053508758545,
      "learning_rate": 4.249608384959964e-05,
      "loss": 0.9714,
      "step": 1117
    },
    {
      "epoch": 0.5715746421267893,
      "grad_norm": 3.2112934589385986,
      "learning_rate": 4.241241976539447e-05,
      "loss": 0.9014,
      "step": 1118
    },
    {
      "epoch": 0.5720858895705522,
      "grad_norm": 3.148987293243408,
      "learning_rate": 4.232877742056131e-05,
      "loss": 0.8606,
      "step": 1119
    },
    {
      "epoch": 0.5725971370143149,
      "grad_norm": 3.3100132942199707,
      "learning_rate": 4.224515705474603e-05,
      "loss": 0.9315,
      "step": 1120
    },
    {
      "epoch": 0.5731083844580777,
      "grad_norm": 3.2287204265594482,
      "learning_rate": 4.2161558907531454e-05,
      "loss": 0.8845,
      "step": 1121
    },
    {
      "epoch": 0.5736196319018405,
      "grad_norm": 3.292661428451538,
      "learning_rate": 4.207798321843681e-05,
      "loss": 0.9407,
      "step": 1122
    },
    {
      "epoch": 0.5741308793456033,
      "grad_norm": 3.141087055206299,
      "learning_rate": 4.199443022691698e-05,
      "loss": 0.8379,
      "step": 1123
    },
    {
      "epoch": 0.5746421267893661,
      "grad_norm": 3.3399198055267334,
      "learning_rate": 4.1910900172361764e-05,
      "loss": 0.8819,
      "step": 1124
    },
    {
      "epoch": 0.5751533742331288,
      "grad_norm": 3.5327308177948,
      "learning_rate": 4.182739329409532e-05,
      "loss": 0.9396,
      "step": 1125
    },
    {
      "epoch": 0.5756646216768916,
      "grad_norm": 3.4489853382110596,
      "learning_rate": 4.1743909831375316e-05,
      "loss": 0.8109,
      "step": 1126
    },
    {
      "epoch": 0.5761758691206544,
      "grad_norm": 3.4719979763031006,
      "learning_rate": 4.16604500233924e-05,
      "loss": 0.9068,
      "step": 1127
    },
    {
      "epoch": 0.5766871165644172,
      "grad_norm": 3.4496169090270996,
      "learning_rate": 4.157701410926943e-05,
      "loss": 0.898,
      "step": 1128
    },
    {
      "epoch": 0.5771983640081799,
      "grad_norm": 3.652449131011963,
      "learning_rate": 4.149360232806078e-05,
      "loss": 0.794,
      "step": 1129
    },
    {
      "epoch": 0.5777096114519428,
      "grad_norm": 4.12328577041626,
      "learning_rate": 4.141021491875172e-05,
      "loss": 0.9067,
      "step": 1130
    },
    {
      "epoch": 0.5782208588957055,
      "grad_norm": 3.9710137844085693,
      "learning_rate": 4.132685212025764e-05,
      "loss": 0.985,
      "step": 1131
    },
    {
      "epoch": 0.5787321063394683,
      "grad_norm": 3.8590469360351562,
      "learning_rate": 4.124351417142347e-05,
      "loss": 0.9224,
      "step": 1132
    },
    {
      "epoch": 0.5792433537832311,
      "grad_norm": 3.658128023147583,
      "learning_rate": 4.11602013110229e-05,
      "loss": 0.9181,
      "step": 1133
    },
    {
      "epoch": 0.5797546012269938,
      "grad_norm": 3.682197332382202,
      "learning_rate": 4.1076913777757757e-05,
      "loss": 0.8932,
      "step": 1134
    },
    {
      "epoch": 0.5802658486707567,
      "grad_norm": 3.68892765045166,
      "learning_rate": 4.099365181025733e-05,
      "loss": 0.8478,
      "step": 1135
    },
    {
      "epoch": 0.5807770961145194,
      "grad_norm": 3.860687255859375,
      "learning_rate": 4.09104156470776e-05,
      "loss": 0.7584,
      "step": 1136
    },
    {
      "epoch": 0.5812883435582822,
      "grad_norm": 3.890394926071167,
      "learning_rate": 4.0827205526700673e-05,
      "loss": 0.8957,
      "step": 1137
    },
    {
      "epoch": 0.581799591002045,
      "grad_norm": 4.4587321281433105,
      "learning_rate": 4.074402168753399e-05,
      "loss": 0.9948,
      "step": 1138
    },
    {
      "epoch": 0.5823108384458078,
      "grad_norm": 4.47606897354126,
      "learning_rate": 4.0660864367909734e-05,
      "loss": 0.8756,
      "step": 1139
    },
    {
      "epoch": 0.5828220858895705,
      "grad_norm": 4.071111679077148,
      "learning_rate": 4.057773380608411e-05,
      "loss": 0.8526,
      "step": 1140
    },
    {
      "epoch": 0.5833333333333334,
      "grad_norm": 4.525618553161621,
      "learning_rate": 4.049463024023657e-05,
      "loss": 1.0175,
      "step": 1141
    },
    {
      "epoch": 0.5838445807770961,
      "grad_norm": 4.352678298950195,
      "learning_rate": 4.041155390846933e-05,
      "loss": 0.8302,
      "step": 1142
    },
    {
      "epoch": 0.5843558282208589,
      "grad_norm": 4.422893047332764,
      "learning_rate": 4.0328505048806525e-05,
      "loss": 0.871,
      "step": 1143
    },
    {
      "epoch": 0.5848670756646217,
      "grad_norm": 4.7299275398254395,
      "learning_rate": 4.0245483899193595e-05,
      "loss": 0.7768,
      "step": 1144
    },
    {
      "epoch": 0.5853783231083844,
      "grad_norm": 4.770334720611572,
      "learning_rate": 4.016249069749657e-05,
      "loss": 0.8351,
      "step": 1145
    },
    {
      "epoch": 0.5858895705521472,
      "grad_norm": 5.195528984069824,
      "learning_rate": 4.0079525681501415e-05,
      "loss": 0.6847,
      "step": 1146
    },
    {
      "epoch": 0.58640081799591,
      "grad_norm": 4.96982479095459,
      "learning_rate": 3.999658908891334e-05,
      "loss": 0.709,
      "step": 1147
    },
    {
      "epoch": 0.5869120654396728,
      "grad_norm": 4.830311298370361,
      "learning_rate": 3.991368115735612e-05,
      "loss": 0.5307,
      "step": 1148
    },
    {
      "epoch": 0.5874233128834356,
      "grad_norm": 5.926014423370361,
      "learning_rate": 3.983080212437144e-05,
      "loss": 0.706,
      "step": 1149
    },
    {
      "epoch": 0.5879345603271984,
      "grad_norm": 6.610885143280029,
      "learning_rate": 3.9747952227418096e-05,
      "loss": 0.5711,
      "step": 1150
    },
    {
      "epoch": 0.5884458077709611,
      "grad_norm": 1.9169435501098633,
      "learning_rate": 3.96651317038715e-05,
      "loss": 0.923,
      "step": 1151
    },
    {
      "epoch": 0.588957055214724,
      "grad_norm": 2.1922850608825684,
      "learning_rate": 3.958234079102288e-05,
      "loss": 0.9157,
      "step": 1152
    },
    {
      "epoch": 0.5894683026584867,
      "grad_norm": 2.3200840950012207,
      "learning_rate": 3.9499579726078594e-05,
      "loss": 0.8759,
      "step": 1153
    },
    {
      "epoch": 0.5899795501022495,
      "grad_norm": 2.6276164054870605,
      "learning_rate": 3.941684874615952e-05,
      "loss": 0.9767,
      "step": 1154
    },
    {
      "epoch": 0.5904907975460123,
      "grad_norm": 2.754929304122925,
      "learning_rate": 3.933414808830033e-05,
      "loss": 0.9734,
      "step": 1155
    },
    {
      "epoch": 0.591002044989775,
      "grad_norm": 2.80434513092041,
      "learning_rate": 3.92514779894488e-05,
      "loss": 1.0555,
      "step": 1156
    },
    {
      "epoch": 0.5915132924335378,
      "grad_norm": 2.974435806274414,
      "learning_rate": 3.916883868646517e-05,
      "loss": 1.018,
      "step": 1157
    },
    {
      "epoch": 0.5920245398773006,
      "grad_norm": 2.803382396697998,
      "learning_rate": 3.908623041612144e-05,
      "loss": 0.9716,
      "step": 1158
    },
    {
      "epoch": 0.5925357873210634,
      "grad_norm": 2.8717260360717773,
      "learning_rate": 3.900365341510073e-05,
      "loss": 0.9645,
      "step": 1159
    },
    {
      "epoch": 0.5930470347648262,
      "grad_norm": 3.0683510303497314,
      "learning_rate": 3.892110791999649e-05,
      "loss": 0.8748,
      "step": 1160
    },
    {
      "epoch": 0.593558282208589,
      "grad_norm": 2.787853240966797,
      "learning_rate": 3.883859416731195e-05,
      "loss": 0.9037,
      "step": 1161
    },
    {
      "epoch": 0.5940695296523517,
      "grad_norm": 2.759490728378296,
      "learning_rate": 3.875611239345942e-05,
      "loss": 0.8391,
      "step": 1162
    },
    {
      "epoch": 0.5945807770961146,
      "grad_norm": 2.975919008255005,
      "learning_rate": 3.867366283475955e-05,
      "loss": 0.89,
      "step": 1163
    },
    {
      "epoch": 0.5950920245398773,
      "grad_norm": 2.9492640495300293,
      "learning_rate": 3.859124572744071e-05,
      "loss": 0.8719,
      "step": 1164
    },
    {
      "epoch": 0.59560327198364,
      "grad_norm": 3.2285163402557373,
      "learning_rate": 3.850886130763825e-05,
      "loss": 0.9296,
      "step": 1165
    },
    {
      "epoch": 0.5961145194274029,
      "grad_norm": 3.5377800464630127,
      "learning_rate": 3.842650981139393e-05,
      "loss": 0.9344,
      "step": 1166
    },
    {
      "epoch": 0.5966257668711656,
      "grad_norm": 3.2627103328704834,
      "learning_rate": 3.834419147465513e-05,
      "loss": 0.9323,
      "step": 1167
    },
    {
      "epoch": 0.5971370143149284,
      "grad_norm": 3.227404832839966,
      "learning_rate": 3.8261906533274254e-05,
      "loss": 0.8478,
      "step": 1168
    },
    {
      "epoch": 0.5976482617586912,
      "grad_norm": 3.3979251384735107,
      "learning_rate": 3.817965522300803e-05,
      "loss": 0.8301,
      "step": 1169
    },
    {
      "epoch": 0.598159509202454,
      "grad_norm": 2.9473681449890137,
      "learning_rate": 3.809743777951675e-05,
      "loss": 0.881,
      "step": 1170
    },
    {
      "epoch": 0.5986707566462167,
      "grad_norm": 3.4166736602783203,
      "learning_rate": 3.801525443836378e-05,
      "loss": 0.9838,
      "step": 1171
    },
    {
      "epoch": 0.5991820040899796,
      "grad_norm": 3.37152099609375,
      "learning_rate": 3.793310543501473e-05,
      "loss": 0.9654,
      "step": 1172
    },
    {
      "epoch": 0.5996932515337423,
      "grad_norm": 3.3099162578582764,
      "learning_rate": 3.785099100483681e-05,
      "loss": 0.9081,
      "step": 1173
    },
    {
      "epoch": 0.6002044989775052,
      "grad_norm": 3.897919178009033,
      "learning_rate": 3.776891138309821e-05,
      "loss": 0.8691,
      "step": 1174
    },
    {
      "epoch": 0.6007157464212679,
      "grad_norm": 3.367326259613037,
      "learning_rate": 3.768686680496737e-05,
      "loss": 0.8366,
      "step": 1175
    },
    {
      "epoch": 0.6012269938650306,
      "grad_norm": 3.388650417327881,
      "learning_rate": 3.7604857505512345e-05,
      "loss": 0.8758,
      "step": 1176
    },
    {
      "epoch": 0.6017382413087935,
      "grad_norm": 3.35302996635437,
      "learning_rate": 3.752288371970006e-05,
      "loss": 0.9105,
      "step": 1177
    },
    {
      "epoch": 0.6022494887525562,
      "grad_norm": 3.6092076301574707,
      "learning_rate": 3.744094568239577e-05,
      "loss": 0.9705,
      "step": 1178
    },
    {
      "epoch": 0.602760736196319,
      "grad_norm": 3.6518003940582275,
      "learning_rate": 3.735904362836222e-05,
      "loss": 0.9318,
      "step": 1179
    },
    {
      "epoch": 0.6032719836400818,
      "grad_norm": 3.510641098022461,
      "learning_rate": 3.7277177792259114e-05,
      "loss": 0.8585,
      "step": 1180
    },
    {
      "epoch": 0.6037832310838446,
      "grad_norm": 3.7477800846099854,
      "learning_rate": 3.719534840864237e-05,
      "loss": 0.9278,
      "step": 1181
    },
    {
      "epoch": 0.6042944785276073,
      "grad_norm": 3.890324115753174,
      "learning_rate": 3.7113555711963474e-05,
      "loss": 0.9617,
      "step": 1182
    },
    {
      "epoch": 0.6048057259713702,
      "grad_norm": 3.7491960525512695,
      "learning_rate": 3.70317999365688e-05,
      "loss": 0.8225,
      "step": 1183
    },
    {
      "epoch": 0.6053169734151329,
      "grad_norm": 3.997459888458252,
      "learning_rate": 3.695008131669891e-05,
      "loss": 0.8853,
      "step": 1184
    },
    {
      "epoch": 0.6058282208588958,
      "grad_norm": 3.706289529800415,
      "learning_rate": 3.686840008648794e-05,
      "loss": 0.8468,
      "step": 1185
    },
    {
      "epoch": 0.6063394683026585,
      "grad_norm": 3.999207019805908,
      "learning_rate": 3.67867564799629e-05,
      "loss": 0.9332,
      "step": 1186
    },
    {
      "epoch": 0.6068507157464212,
      "grad_norm": 3.972015857696533,
      "learning_rate": 3.6705150731043005e-05,
      "loss": 0.9797,
      "step": 1187
    },
    {
      "epoch": 0.6073619631901841,
      "grad_norm": 4.267761707305908,
      "learning_rate": 3.6623583073538966e-05,
      "loss": 0.984,
      "step": 1188
    },
    {
      "epoch": 0.6078732106339468,
      "grad_norm": 4.182210445404053,
      "learning_rate": 3.6542053741152394e-05,
      "loss": 0.7828,
      "step": 1189
    },
    {
      "epoch": 0.6083844580777096,
      "grad_norm": 4.105729103088379,
      "learning_rate": 3.646056296747507e-05,
      "loss": 0.9388,
      "step": 1190
    },
    {
      "epoch": 0.6088957055214724,
      "grad_norm": 4.066761493682861,
      "learning_rate": 3.637911098598836e-05,
      "loss": 0.9364,
      "step": 1191
    },
    {
      "epoch": 0.6094069529652352,
      "grad_norm": 4.344300270080566,
      "learning_rate": 3.629769803006239e-05,
      "loss": 0.7844,
      "step": 1192
    },
    {
      "epoch": 0.6099182004089979,
      "grad_norm": 4.508508682250977,
      "learning_rate": 3.621632433295553e-05,
      "loss": 0.8825,
      "step": 1193
    },
    {
      "epoch": 0.6104294478527608,
      "grad_norm": 4.3253583908081055,
      "learning_rate": 3.613499012781368e-05,
      "loss": 0.8134,
      "step": 1194
    },
    {
      "epoch": 0.6109406952965235,
      "grad_norm": 4.819972515106201,
      "learning_rate": 3.605369564766956e-05,
      "loss": 0.8738,
      "step": 1195
    },
    {
      "epoch": 0.6114519427402862,
      "grad_norm": 5.187696933746338,
      "learning_rate": 3.597244112544208e-05,
      "loss": 0.8424,
      "step": 1196
    },
    {
      "epoch": 0.6119631901840491,
      "grad_norm": 5.0562968254089355,
      "learning_rate": 3.5891226793935675e-05,
      "loss": 0.7322,
      "step": 1197
    },
    {
      "epoch": 0.6124744376278118,
      "grad_norm": 5.8710832595825195,
      "learning_rate": 3.581005288583964e-05,
      "loss": 0.8522,
      "step": 1198
    },
    {
      "epoch": 0.6129856850715747,
      "grad_norm": 5.613986968994141,
      "learning_rate": 3.57289196337274e-05,
      "loss": 0.8068,
      "step": 1199
    },
    {
      "epoch": 0.6134969325153374,
      "grad_norm": 5.67006778717041,
      "learning_rate": 3.5647827270055945e-05,
      "loss": 0.3523,
      "step": 1200
    },
    {
      "epoch": 0.6140081799591002,
      "grad_norm": 2.078099489212036,
      "learning_rate": 3.5566776027165106e-05,
      "loss": 0.8653,
      "step": 1201
    },
    {
      "epoch": 0.614519427402863,
      "grad_norm": 2.6257896423339844,
      "learning_rate": 3.5485766137276894e-05,
      "loss": 1.0512,
      "step": 1202
    },
    {
      "epoch": 0.6150306748466258,
      "grad_norm": 2.1455605030059814,
      "learning_rate": 3.540479783249482e-05,
      "loss": 0.9154,
      "step": 1203
    },
    {
      "epoch": 0.6155419222903885,
      "grad_norm": 2.551408290863037,
      "learning_rate": 3.5323871344803263e-05,
      "loss": 0.9544,
      "step": 1204
    },
    {
      "epoch": 0.6160531697341514,
      "grad_norm": 2.475545883178711,
      "learning_rate": 3.524298690606681e-05,
      "loss": 0.902,
      "step": 1205
    },
    {
      "epoch": 0.6165644171779141,
      "grad_norm": 2.6632912158966064,
      "learning_rate": 3.5162144748029555e-05,
      "loss": 0.9852,
      "step": 1206
    },
    {
      "epoch": 0.6170756646216768,
      "grad_norm": 2.898348331451416,
      "learning_rate": 3.5081345102314454e-05,
      "loss": 0.9091,
      "step": 1207
    },
    {
      "epoch": 0.6175869120654397,
      "grad_norm": 2.6006739139556885,
      "learning_rate": 3.500058820042263e-05,
      "loss": 0.894,
      "step": 1208
    },
    {
      "epoch": 0.6180981595092024,
      "grad_norm": 2.914081573486328,
      "learning_rate": 3.4919874273732784e-05,
      "loss": 0.9868,
      "step": 1209
    },
    {
      "epoch": 0.6186094069529653,
      "grad_norm": 2.8525145053863525,
      "learning_rate": 3.4839203553500474e-05,
      "loss": 0.9638,
      "step": 1210
    },
    {
      "epoch": 0.619120654396728,
      "grad_norm": 2.761507034301758,
      "learning_rate": 3.475857627085746e-05,
      "loss": 0.9268,
      "step": 1211
    },
    {
      "epoch": 0.6196319018404908,
      "grad_norm": 2.962073802947998,
      "learning_rate": 3.467799265681105e-05,
      "loss": 0.9322,
      "step": 1212
    },
    {
      "epoch": 0.6201431492842536,
      "grad_norm": 3.011537551879883,
      "learning_rate": 3.4597452942243446e-05,
      "loss": 1.0676,
      "step": 1213
    },
    {
      "epoch": 0.6206543967280164,
      "grad_norm": 2.7569634914398193,
      "learning_rate": 3.451695735791106e-05,
      "loss": 0.8432,
      "step": 1214
    },
    {
      "epoch": 0.6211656441717791,
      "grad_norm": 3.254464864730835,
      "learning_rate": 3.4436506134443866e-05,
      "loss": 0.9145,
      "step": 1215
    },
    {
      "epoch": 0.621676891615542,
      "grad_norm": 3.0407156944274902,
      "learning_rate": 3.435609950234473e-05,
      "loss": 0.9686,
      "step": 1216
    },
    {
      "epoch": 0.6221881390593047,
      "grad_norm": 3.2604401111602783,
      "learning_rate": 3.427573769198879e-05,
      "loss": 0.8542,
      "step": 1217
    },
    {
      "epoch": 0.6226993865030674,
      "grad_norm": 3.351675271987915,
      "learning_rate": 3.419542093362274e-05,
      "loss": 0.873,
      "step": 1218
    },
    {
      "epoch": 0.6232106339468303,
      "grad_norm": 2.7927772998809814,
      "learning_rate": 3.411514945736419e-05,
      "loss": 0.7677,
      "step": 1219
    },
    {
      "epoch": 0.623721881390593,
      "grad_norm": 3.3357203006744385,
      "learning_rate": 3.403492349320101e-05,
      "loss": 0.9263,
      "step": 1220
    },
    {
      "epoch": 0.6242331288343558,
      "grad_norm": 3.1823763847351074,
      "learning_rate": 3.3954743270990695e-05,
      "loss": 0.8802,
      "step": 1221
    },
    {
      "epoch": 0.6247443762781186,
      "grad_norm": 3.5401298999786377,
      "learning_rate": 3.387460902045967e-05,
      "loss": 0.9269,
      "step": 1222
    },
    {
      "epoch": 0.6252556237218814,
      "grad_norm": 3.1181726455688477,
      "learning_rate": 3.3794520971202635e-05,
      "loss": 0.8232,
      "step": 1223
    },
    {
      "epoch": 0.6257668711656442,
      "grad_norm": 3.3269927501678467,
      "learning_rate": 3.371447935268194e-05,
      "loss": 0.9379,
      "step": 1224
    },
    {
      "epoch": 0.626278118609407,
      "grad_norm": 3.3684628009796143,
      "learning_rate": 3.363448439422688e-05,
      "loss": 0.7728,
      "step": 1225
    },
    {
      "epoch": 0.6267893660531697,
      "grad_norm": 3.450457811355591,
      "learning_rate": 3.3554536325033095e-05,
      "loss": 0.9889,
      "step": 1226
    },
    {
      "epoch": 0.6273006134969326,
      "grad_norm": 3.5818159580230713,
      "learning_rate": 3.3474635374161845e-05,
      "loss": 0.9753,
      "step": 1227
    },
    {
      "epoch": 0.6278118609406953,
      "grad_norm": 3.563751697540283,
      "learning_rate": 3.339478177053941e-05,
      "loss": 0.9556,
      "step": 1228
    },
    {
      "epoch": 0.628323108384458,
      "grad_norm": 3.62951922416687,
      "learning_rate": 3.3314975742956424e-05,
      "loss": 0.9656,
      "step": 1229
    },
    {
      "epoch": 0.6288343558282209,
      "grad_norm": 3.5825490951538086,
      "learning_rate": 3.323521752006716e-05,
      "loss": 0.8455,
      "step": 1230
    },
    {
      "epoch": 0.6293456032719836,
      "grad_norm": 3.4373691082000732,
      "learning_rate": 3.3155507330389e-05,
      "loss": 0.7958,
      "step": 1231
    },
    {
      "epoch": 0.6298568507157464,
      "grad_norm": 4.127106666564941,
      "learning_rate": 3.3075845402301655e-05,
      "loss": 0.893,
      "step": 1232
    },
    {
      "epoch": 0.6303680981595092,
      "grad_norm": 3.9865024089813232,
      "learning_rate": 3.299623196404657e-05,
      "loss": 0.8928,
      "step": 1233
    },
    {
      "epoch": 0.630879345603272,
      "grad_norm": 3.4670865535736084,
      "learning_rate": 3.291666724372626e-05,
      "loss": 0.8081,
      "step": 1234
    },
    {
      "epoch": 0.6313905930470347,
      "grad_norm": 3.6847434043884277,
      "learning_rate": 3.2837151469303664e-05,
      "loss": 0.744,
      "step": 1235
    },
    {
      "epoch": 0.6319018404907976,
      "grad_norm": 3.732963800430298,
      "learning_rate": 3.275768486860149e-05,
      "loss": 0.7568,
      "step": 1236
    },
    {
      "epoch": 0.6324130879345603,
      "grad_norm": 3.961927652359009,
      "learning_rate": 3.267826766930152e-05,
      "loss": 0.9264,
      "step": 1237
    },
    {
      "epoch": 0.6329243353783232,
      "grad_norm": 4.04052209854126,
      "learning_rate": 3.2598900098944043e-05,
      "loss": 0.9307,
      "step": 1238
    },
    {
      "epoch": 0.6334355828220859,
      "grad_norm": 4.026215076446533,
      "learning_rate": 3.251958238492711e-05,
      "loss": 0.8288,
      "step": 1239
    },
    {
      "epoch": 0.6339468302658486,
      "grad_norm": 4.72583532333374,
      "learning_rate": 3.244031475450599e-05,
      "loss": 0.8492,
      "step": 1240
    },
    {
      "epoch": 0.6344580777096115,
      "grad_norm": 4.102770805358887,
      "learning_rate": 3.2361097434792396e-05,
      "loss": 0.8145,
      "step": 1241
    },
    {
      "epoch": 0.6349693251533742,
      "grad_norm": 4.25362491607666,
      "learning_rate": 3.228193065275391e-05,
      "loss": 0.7773,
      "step": 1242
    },
    {
      "epoch": 0.635480572597137,
      "grad_norm": 5.0050225257873535,
      "learning_rate": 3.220281463521333e-05,
      "loss": 0.9292,
      "step": 1243
    },
    {
      "epoch": 0.6359918200408998,
      "grad_norm": 4.613053321838379,
      "learning_rate": 3.2123749608848e-05,
      "loss": 0.7985,
      "step": 1244
    },
    {
      "epoch": 0.6365030674846626,
      "grad_norm": 4.7116546630859375,
      "learning_rate": 3.204473580018916e-05,
      "loss": 0.858,
      "step": 1245
    },
    {
      "epoch": 0.6370143149284253,
      "grad_norm": 5.056521892547607,
      "learning_rate": 3.196577343562135e-05,
      "loss": 0.8168,
      "step": 1246
    },
    {
      "epoch": 0.6375255623721882,
      "grad_norm": 5.530933380126953,
      "learning_rate": 3.188686274138163e-05,
      "loss": 0.7587,
      "step": 1247
    },
    {
      "epoch": 0.6380368098159509,
      "grad_norm": 5.684156894683838,
      "learning_rate": 3.180800394355908e-05,
      "loss": 0.7681,
      "step": 1248
    },
    {
      "epoch": 0.6385480572597138,
      "grad_norm": 5.361364841461182,
      "learning_rate": 3.172919726809409e-05,
      "loss": 0.7175,
      "step": 1249
    },
    {
      "epoch": 0.6390593047034765,
      "grad_norm": 6.80435848236084,
      "learning_rate": 3.1650442940777695e-05,
      "loss": 0.5696,
      "step": 1250
    },
    {
      "epoch": 0.6395705521472392,
      "grad_norm": 1.996549367904663,
      "learning_rate": 3.1571741187250945e-05,
      "loss": 0.8583,
      "step": 1251
    },
    {
      "epoch": 0.6400817995910021,
      "grad_norm": 2.412594795227051,
      "learning_rate": 3.149309223300428e-05,
      "loss": 0.985,
      "step": 1252
    },
    {
      "epoch": 0.6405930470347648,
      "grad_norm": 2.6278765201568604,
      "learning_rate": 3.141449630337685e-05,
      "loss": 0.9277,
      "step": 1253
    },
    {
      "epoch": 0.6411042944785276,
      "grad_norm": 2.631162405014038,
      "learning_rate": 3.1335953623555865e-05,
      "loss": 0.9586,
      "step": 1254
    },
    {
      "epoch": 0.6416155419222904,
      "grad_norm": 2.5849802494049072,
      "learning_rate": 3.1257464418575986e-05,
      "loss": 0.8985,
      "step": 1255
    },
    {
      "epoch": 0.6421267893660532,
      "grad_norm": 2.6325793266296387,
      "learning_rate": 3.11790289133187e-05,
      "loss": 0.9677,
      "step": 1256
    },
    {
      "epoch": 0.6426380368098159,
      "grad_norm": 2.7737441062927246,
      "learning_rate": 3.110064733251154e-05,
      "loss": 0.9517,
      "step": 1257
    },
    {
      "epoch": 0.6431492842535788,
      "grad_norm": 2.836144208908081,
      "learning_rate": 3.102231990072763e-05,
      "loss": 1.0036,
      "step": 1258
    },
    {
      "epoch": 0.6436605316973415,
      "grad_norm": 3.148357391357422,
      "learning_rate": 3.094404684238489e-05,
      "loss": 0.9177,
      "step": 1259
    },
    {
      "epoch": 0.6441717791411042,
      "grad_norm": 3.0685768127441406,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.9073,
      "step": 1260
    },
    {
      "epoch": 0.6446830265848671,
      "grad_norm": 2.947803020477295,
      "learning_rate": 3.078766474291522e-05,
      "loss": 0.91,
      "step": 1261
    },
    {
      "epoch": 0.6451942740286298,
      "grad_norm": 3.012056350708008,
      "learning_rate": 3.0709556149842664e-05,
      "loss": 0.8971,
      "step": 1262
    },
    {
      "epoch": 0.6457055214723927,
      "grad_norm": 3.072740316390991,
      "learning_rate": 3.063150282631879e-05,
      "loss": 0.9495,
      "step": 1263
    },
    {
      "epoch": 0.6462167689161554,
      "grad_norm": 3.0233309268951416,
      "learning_rate": 3.05535049959762e-05,
      "loss": 0.9246,
      "step": 1264
    },
    {
      "epoch": 0.6467280163599182,
      "grad_norm": 3.3996291160583496,
      "learning_rate": 3.047556288228851e-05,
      "loss": 0.9428,
      "step": 1265
    },
    {
      "epoch": 0.647239263803681,
      "grad_norm": 3.211404323577881,
      "learning_rate": 3.0397676708569633e-05,
      "loss": 0.8686,
      "step": 1266
    },
    {
      "epoch": 0.6477505112474438,
      "grad_norm": 3.0419328212738037,
      "learning_rate": 3.031984669797328e-05,
      "loss": 0.891,
      "step": 1267
    },
    {
      "epoch": 0.6482617586912065,
      "grad_norm": 3.230339527130127,
      "learning_rate": 3.024207307349224e-05,
      "loss": 0.8741,
      "step": 1268
    },
    {
      "epoch": 0.6487730061349694,
      "grad_norm": 3.1209537982940674,
      "learning_rate": 3.0164356057957706e-05,
      "loss": 0.917,
      "step": 1269
    },
    {
      "epoch": 0.6492842535787321,
      "grad_norm": 3.3551604747772217,
      "learning_rate": 3.0086695874038717e-05,
      "loss": 0.9233,
      "step": 1270
    },
    {
      "epoch": 0.6497955010224948,
      "grad_norm": 3.477663516998291,
      "learning_rate": 3.0009092744241472e-05,
      "loss": 0.8906,
      "step": 1271
    },
    {
      "epoch": 0.6503067484662577,
      "grad_norm": 3.3048524856567383,
      "learning_rate": 2.9931546890908697e-05,
      "loss": 0.9094,
      "step": 1272
    },
    {
      "epoch": 0.6508179959100204,
      "grad_norm": 3.2910995483398438,
      "learning_rate": 2.985405853621902e-05,
      "loss": 0.8444,
      "step": 1273
    },
    {
      "epoch": 0.6513292433537833,
      "grad_norm": 3.342087507247925,
      "learning_rate": 2.9776627902186338e-05,
      "loss": 0.8766,
      "step": 1274
    },
    {
      "epoch": 0.651840490797546,
      "grad_norm": 3.8440937995910645,
      "learning_rate": 2.9699255210659166e-05,
      "loss": 0.9573,
      "step": 1275
    },
    {
      "epoch": 0.6523517382413088,
      "grad_norm": 3.666816473007202,
      "learning_rate": 2.962194068331996e-05,
      "loss": 0.9012,
      "step": 1276
    },
    {
      "epoch": 0.6528629856850716,
      "grad_norm": 3.627305746078491,
      "learning_rate": 2.9544684541684598e-05,
      "loss": 0.9463,
      "step": 1277
    },
    {
      "epoch": 0.6533742331288344,
      "grad_norm": 3.7801358699798584,
      "learning_rate": 2.9467487007101636e-05,
      "loss": 0.9248,
      "step": 1278
    },
    {
      "epoch": 0.6538854805725971,
      "grad_norm": 3.5962491035461426,
      "learning_rate": 2.939034830075173e-05,
      "loss": 0.9153,
      "step": 1279
    },
    {
      "epoch": 0.65439672801636,
      "grad_norm": 3.716352939605713,
      "learning_rate": 2.9313268643646986e-05,
      "loss": 0.8698,
      "step": 1280
    },
    {
      "epoch": 0.6549079754601227,
      "grad_norm": 3.7603189945220947,
      "learning_rate": 2.92362482566303e-05,
      "loss": 1.0128,
      "step": 1281
    },
    {
      "epoch": 0.6554192229038854,
      "grad_norm": 3.9676363468170166,
      "learning_rate": 2.915928736037478e-05,
      "loss": 0.8961,
      "step": 1282
    },
    {
      "epoch": 0.6559304703476483,
      "grad_norm": 4.314609527587891,
      "learning_rate": 2.908238617538307e-05,
      "loss": 0.8685,
      "step": 1283
    },
    {
      "epoch": 0.656441717791411,
      "grad_norm": 3.607283115386963,
      "learning_rate": 2.900554492198677e-05,
      "loss": 0.7664,
      "step": 1284
    },
    {
      "epoch": 0.6569529652351738,
      "grad_norm": 3.701615810394287,
      "learning_rate": 2.8928763820345716e-05,
      "loss": 0.8553,
      "step": 1285
    },
    {
      "epoch": 0.6574642126789366,
      "grad_norm": 4.535073757171631,
      "learning_rate": 2.8852043090447423e-05,
      "loss": 0.849,
      "step": 1286
    },
    {
      "epoch": 0.6579754601226994,
      "grad_norm": 3.8456053733825684,
      "learning_rate": 2.877538295210644e-05,
      "loss": 0.7781,
      "step": 1287
    },
    {
      "epoch": 0.6584867075664622,
      "grad_norm": 4.1655168533325195,
      "learning_rate": 2.869878362496368e-05,
      "loss": 0.8895,
      "step": 1288
    },
    {
      "epoch": 0.658997955010225,
      "grad_norm": 4.286038398742676,
      "learning_rate": 2.8622245328485907e-05,
      "loss": 0.8201,
      "step": 1289
    },
    {
      "epoch": 0.6595092024539877,
      "grad_norm": 4.441523551940918,
      "learning_rate": 2.8545768281964925e-05,
      "loss": 0.7817,
      "step": 1290
    },
    {
      "epoch": 0.6600204498977505,
      "grad_norm": 4.569100379943848,
      "learning_rate": 2.846935270451712e-05,
      "loss": 0.8183,
      "step": 1291
    },
    {
      "epoch": 0.6605316973415133,
      "grad_norm": 4.366547584533691,
      "learning_rate": 2.8392998815082717e-05,
      "loss": 0.8923,
      "step": 1292
    },
    {
      "epoch": 0.661042944785276,
      "grad_norm": 4.581328392028809,
      "learning_rate": 2.8316706832425243e-05,
      "loss": 0.7825,
      "step": 1293
    },
    {
      "epoch": 0.6615541922290389,
      "grad_norm": 5.016780376434326,
      "learning_rate": 2.8240476975130803e-05,
      "loss": 0.912,
      "step": 1294
    },
    {
      "epoch": 0.6620654396728016,
      "grad_norm": 4.488846302032471,
      "learning_rate": 2.8164309461607547e-05,
      "loss": 0.719,
      "step": 1295
    },
    {
      "epoch": 0.6625766871165644,
      "grad_norm": 5.28438138961792,
      "learning_rate": 2.808820451008495e-05,
      "loss": 0.8663,
      "step": 1296
    },
    {
      "epoch": 0.6630879345603272,
      "grad_norm": 5.2693071365356445,
      "learning_rate": 2.801216233861331e-05,
      "loss": 0.7197,
      "step": 1297
    },
    {
      "epoch": 0.66359918200409,
      "grad_norm": 5.987064838409424,
      "learning_rate": 2.793618316506299e-05,
      "loss": 0.6999,
      "step": 1298
    },
    {
      "epoch": 0.6641104294478528,
      "grad_norm": 5.712385654449463,
      "learning_rate": 2.7860267207123914e-05,
      "loss": 0.7776,
      "step": 1299
    },
    {
      "epoch": 0.6646216768916156,
      "grad_norm": 5.60577917098999,
      "learning_rate": 2.7784414682304832e-05,
      "loss": 0.3877,
      "step": 1300
    },
    {
      "epoch": 0.6651329243353783,
      "grad_norm": 1.8261439800262451,
      "learning_rate": 2.7708625807932754e-05,
      "loss": 0.8337,
      "step": 1301
    },
    {
      "epoch": 0.6656441717791411,
      "grad_norm": 2.3331668376922607,
      "learning_rate": 2.763290080115238e-05,
      "loss": 0.8949,
      "step": 1302
    },
    {
      "epoch": 0.6661554192229039,
      "grad_norm": 2.5809342861175537,
      "learning_rate": 2.7557239878925335e-05,
      "loss": 0.9388,
      "step": 1303
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 2.492849111557007,
      "learning_rate": 2.7481643258029748e-05,
      "loss": 0.9331,
      "step": 1304
    },
    {
      "epoch": 0.6671779141104295,
      "grad_norm": 2.6156399250030518,
      "learning_rate": 2.740611115505937e-05,
      "loss": 0.9623,
      "step": 1305
    },
    {
      "epoch": 0.6676891615541922,
      "grad_norm": 2.552900791168213,
      "learning_rate": 2.7330643786423237e-05,
      "loss": 0.8438,
      "step": 1306
    },
    {
      "epoch": 0.668200408997955,
      "grad_norm": 2.678558588027954,
      "learning_rate": 2.7255241368344815e-05,
      "loss": 0.8981,
      "step": 1307
    },
    {
      "epoch": 0.6687116564417178,
      "grad_norm": 2.8384506702423096,
      "learning_rate": 2.7179904116861556e-05,
      "loss": 1.0157,
      "step": 1308
    },
    {
      "epoch": 0.6692229038854806,
      "grad_norm": 3.0380685329437256,
      "learning_rate": 2.7104632247824126e-05,
      "loss": 0.8501,
      "step": 1309
    },
    {
      "epoch": 0.6697341513292433,
      "grad_norm": 2.7943999767303467,
      "learning_rate": 2.7029425976895943e-05,
      "loss": 0.8643,
      "step": 1310
    },
    {
      "epoch": 0.6702453987730062,
      "grad_norm": 2.9918973445892334,
      "learning_rate": 2.6954285519552415e-05,
      "loss": 0.8909,
      "step": 1311
    },
    {
      "epoch": 0.6707566462167689,
      "grad_norm": 2.9454028606414795,
      "learning_rate": 2.687921109108038e-05,
      "loss": 0.919,
      "step": 1312
    },
    {
      "epoch": 0.6712678936605317,
      "grad_norm": 3.3940649032592773,
      "learning_rate": 2.680420290657757e-05,
      "loss": 1.051,
      "step": 1313
    },
    {
      "epoch": 0.6717791411042945,
      "grad_norm": 3.3205361366271973,
      "learning_rate": 2.6729261180951847e-05,
      "loss": 0.8133,
      "step": 1314
    },
    {
      "epoch": 0.6722903885480572,
      "grad_norm": 3.0560505390167236,
      "learning_rate": 2.6654386128920683e-05,
      "loss": 0.8722,
      "step": 1315
    },
    {
      "epoch": 0.6728016359918201,
      "grad_norm": 3.47533917427063,
      "learning_rate": 2.65795779650105e-05,
      "loss": 0.8353,
      "step": 1316
    },
    {
      "epoch": 0.6733128834355828,
      "grad_norm": 3.5020523071289062,
      "learning_rate": 2.650483690355614e-05,
      "loss": 0.9415,
      "step": 1317
    },
    {
      "epoch": 0.6738241308793456,
      "grad_norm": 3.358147144317627,
      "learning_rate": 2.6430163158700115e-05,
      "loss": 0.8847,
      "step": 1318
    },
    {
      "epoch": 0.6743353783231084,
      "grad_norm": 3.2747859954833984,
      "learning_rate": 2.6355556944392136e-05,
      "loss": 0.893,
      "step": 1319
    },
    {
      "epoch": 0.6748466257668712,
      "grad_norm": 3.4800641536712646,
      "learning_rate": 2.628101847438835e-05,
      "loss": 0.8938,
      "step": 1320
    },
    {
      "epoch": 0.6753578732106339,
      "grad_norm": 3.310898780822754,
      "learning_rate": 2.6206547962250894e-05,
      "loss": 0.8795,
      "step": 1321
    },
    {
      "epoch": 0.6758691206543967,
      "grad_norm": 3.385996103286743,
      "learning_rate": 2.6132145621347114e-05,
      "loss": 0.8546,
      "step": 1322
    },
    {
      "epoch": 0.6763803680981595,
      "grad_norm": 3.280492067337036,
      "learning_rate": 2.605781166484914e-05,
      "loss": 0.914,
      "step": 1323
    },
    {
      "epoch": 0.6768916155419223,
      "grad_norm": 3.5544273853302,
      "learning_rate": 2.598354630573303e-05,
      "loss": 0.8673,
      "step": 1324
    },
    {
      "epoch": 0.6774028629856851,
      "grad_norm": 3.4663734436035156,
      "learning_rate": 2.5909349756778446e-05,
      "loss": 0.9238,
      "step": 1325
    },
    {
      "epoch": 0.6779141104294478,
      "grad_norm": 3.577603578567505,
      "learning_rate": 2.5835222230567803e-05,
      "loss": 0.9245,
      "step": 1326
    },
    {
      "epoch": 0.6784253578732107,
      "grad_norm": 3.5449306964874268,
      "learning_rate": 2.5761163939485772e-05,
      "loss": 0.9191,
      "step": 1327
    },
    {
      "epoch": 0.6789366053169734,
      "grad_norm": 3.638456344604492,
      "learning_rate": 2.5687175095718723e-05,
      "loss": 0.8672,
      "step": 1328
    },
    {
      "epoch": 0.6794478527607362,
      "grad_norm": 3.5813286304473877,
      "learning_rate": 2.5613255911253963e-05,
      "loss": 0.9806,
      "step": 1329
    },
    {
      "epoch": 0.679959100204499,
      "grad_norm": 3.563659191131592,
      "learning_rate": 2.553940659787929e-05,
      "loss": 0.8792,
      "step": 1330
    },
    {
      "epoch": 0.6804703476482618,
      "grad_norm": 3.872799873352051,
      "learning_rate": 2.5465627367182243e-05,
      "loss": 0.8539,
      "step": 1331
    },
    {
      "epoch": 0.6809815950920245,
      "grad_norm": 3.625586986541748,
      "learning_rate": 2.539191843054963e-05,
      "loss": 0.968,
      "step": 1332
    },
    {
      "epoch": 0.6814928425357873,
      "grad_norm": 3.7508835792541504,
      "learning_rate": 2.5318279999166828e-05,
      "loss": 0.7757,
      "step": 1333
    },
    {
      "epoch": 0.6820040899795501,
      "grad_norm": 3.7676475048065186,
      "learning_rate": 2.5244712284017186e-05,
      "loss": 0.8444,
      "step": 1334
    },
    {
      "epoch": 0.6825153374233128,
      "grad_norm": 3.5776097774505615,
      "learning_rate": 2.5171215495881458e-05,
      "loss": 0.8295,
      "step": 1335
    },
    {
      "epoch": 0.6830265848670757,
      "grad_norm": 4.256826877593994,
      "learning_rate": 2.5097789845337223e-05,
      "loss": 0.9222,
      "step": 1336
    },
    {
      "epoch": 0.6835378323108384,
      "grad_norm": 3.883566379547119,
      "learning_rate": 2.5024435542758173e-05,
      "loss": 0.8951,
      "step": 1337
    },
    {
      "epoch": 0.6840490797546013,
      "grad_norm": 4.0634284019470215,
      "learning_rate": 2.495115279831365e-05,
      "loss": 0.7853,
      "step": 1338
    },
    {
      "epoch": 0.684560327198364,
      "grad_norm": 4.287288665771484,
      "learning_rate": 2.4877941821967914e-05,
      "loss": 0.7984,
      "step": 1339
    },
    {
      "epoch": 0.6850715746421268,
      "grad_norm": 4.691751003265381,
      "learning_rate": 2.4804802823479613e-05,
      "loss": 0.8143,
      "step": 1340
    },
    {
      "epoch": 0.6855828220858896,
      "grad_norm": 4.170199394226074,
      "learning_rate": 2.4731736012401207e-05,
      "loss": 0.8166,
      "step": 1341
    },
    {
      "epoch": 0.6860940695296524,
      "grad_norm": 4.324507713317871,
      "learning_rate": 2.465874159807828e-05,
      "loss": 0.7594,
      "step": 1342
    },
    {
      "epoch": 0.6866053169734151,
      "grad_norm": 4.2558913230896,
      "learning_rate": 2.4585819789649057e-05,
      "loss": 0.7372,
      "step": 1343
    },
    {
      "epoch": 0.6871165644171779,
      "grad_norm": 4.4216461181640625,
      "learning_rate": 2.4512970796043616e-05,
      "loss": 0.6851,
      "step": 1344
    },
    {
      "epoch": 0.6876278118609407,
      "grad_norm": 4.879353046417236,
      "learning_rate": 2.444019482598356e-05,
      "loss": 0.6924,
      "step": 1345
    },
    {
      "epoch": 0.6881390593047034,
      "grad_norm": 5.454744338989258,
      "learning_rate": 2.4367492087981155e-05,
      "loss": 0.8629,
      "step": 1346
    },
    {
      "epoch": 0.6886503067484663,
      "grad_norm": 4.753787517547607,
      "learning_rate": 2.4294862790338917e-05,
      "loss": 0.7606,
      "step": 1347
    },
    {
      "epoch": 0.689161554192229,
      "grad_norm": 4.211785316467285,
      "learning_rate": 2.422230714114891e-05,
      "loss": 0.469,
      "step": 1348
    },
    {
      "epoch": 0.6896728016359919,
      "grad_norm": 6.177483558654785,
      "learning_rate": 2.4149825348292215e-05,
      "loss": 0.9659,
      "step": 1349
    },
    {
      "epoch": 0.6901840490797546,
      "grad_norm": 6.019729137420654,
      "learning_rate": 2.4077417619438275e-05,
      "loss": 0.4232,
      "step": 1350
    },
    {
      "epoch": 0.6906952965235174,
      "grad_norm": 1.807778239250183,
      "learning_rate": 2.400508416204433e-05,
      "loss": 0.7991,
      "step": 1351
    },
    {
      "epoch": 0.6912065439672802,
      "grad_norm": 2.253737688064575,
      "learning_rate": 2.393282518335486e-05,
      "loss": 0.9311,
      "step": 1352
    },
    {
      "epoch": 0.691717791411043,
      "grad_norm": 2.301241159439087,
      "learning_rate": 2.3860640890400925e-05,
      "loss": 0.9348,
      "step": 1353
    },
    {
      "epoch": 0.6922290388548057,
      "grad_norm": 2.6120917797088623,
      "learning_rate": 2.3788531489999604e-05,
      "loss": 1.0091,
      "step": 1354
    },
    {
      "epoch": 0.6927402862985685,
      "grad_norm": 2.5849595069885254,
      "learning_rate": 2.371649718875338e-05,
      "loss": 0.9379,
      "step": 1355
    },
    {
      "epoch": 0.6932515337423313,
      "grad_norm": 2.7695066928863525,
      "learning_rate": 2.3644538193049625e-05,
      "loss": 0.9027,
      "step": 1356
    },
    {
      "epoch": 0.693762781186094,
      "grad_norm": 2.7553393840789795,
      "learning_rate": 2.3572654709059882e-05,
      "loss": 0.8889,
      "step": 1357
    },
    {
      "epoch": 0.6942740286298569,
      "grad_norm": 2.850353717803955,
      "learning_rate": 2.3500846942739406e-05,
      "loss": 0.9374,
      "step": 1358
    },
    {
      "epoch": 0.6947852760736196,
      "grad_norm": 2.645514726638794,
      "learning_rate": 2.3429115099826458e-05,
      "loss": 0.888,
      "step": 1359
    },
    {
      "epoch": 0.6952965235173824,
      "grad_norm": 2.834381341934204,
      "learning_rate": 2.3357459385841823e-05,
      "loss": 0.8912,
      "step": 1360
    },
    {
      "epoch": 0.6958077709611452,
      "grad_norm": 2.820368766784668,
      "learning_rate": 2.328588000608812e-05,
      "loss": 0.8868,
      "step": 1361
    },
    {
      "epoch": 0.696319018404908,
      "grad_norm": 2.9356560707092285,
      "learning_rate": 2.321437716564927e-05,
      "loss": 0.9155,
      "step": 1362
    },
    {
      "epoch": 0.6968302658486708,
      "grad_norm": 2.9917140007019043,
      "learning_rate": 2.3142951069389902e-05,
      "loss": 0.9129,
      "step": 1363
    },
    {
      "epoch": 0.6973415132924335,
      "grad_norm": 2.977782726287842,
      "learning_rate": 2.3071601921954794e-05,
      "loss": 0.8904,
      "step": 1364
    },
    {
      "epoch": 0.6978527607361963,
      "grad_norm": 2.948878765106201,
      "learning_rate": 2.3000329927768214e-05,
      "loss": 0.8717,
      "step": 1365
    },
    {
      "epoch": 0.6983640081799591,
      "grad_norm": 3.2418344020843506,
      "learning_rate": 2.292913529103338e-05,
      "loss": 0.8991,
      "step": 1366
    },
    {
      "epoch": 0.6988752556237219,
      "grad_norm": 3.2583718299865723,
      "learning_rate": 2.285801821573192e-05,
      "loss": 0.8856,
      "step": 1367
    },
    {
      "epoch": 0.6993865030674846,
      "grad_norm": 3.554612398147583,
      "learning_rate": 2.278697890562316e-05,
      "loss": 0.9651,
      "step": 1368
    },
    {
      "epoch": 0.6998977505112475,
      "grad_norm": 3.229438543319702,
      "learning_rate": 2.2716017564243697e-05,
      "loss": 0.8992,
      "step": 1369
    },
    {
      "epoch": 0.7004089979550102,
      "grad_norm": 3.106508255004883,
      "learning_rate": 2.2645134394906676e-05,
      "loss": 0.9012,
      "step": 1370
    },
    {
      "epoch": 0.700920245398773,
      "grad_norm": 3.6647939682006836,
      "learning_rate": 2.257432960070132e-05,
      "loss": 0.9815,
      "step": 1371
    },
    {
      "epoch": 0.7014314928425358,
      "grad_norm": 3.2601795196533203,
      "learning_rate": 2.250360338449226e-05,
      "loss": 0.8928,
      "step": 1372
    },
    {
      "epoch": 0.7019427402862985,
      "grad_norm": 3.264993906021118,
      "learning_rate": 2.243295594891901e-05,
      "loss": 0.9133,
      "step": 1373
    },
    {
      "epoch": 0.7024539877300614,
      "grad_norm": 3.4938395023345947,
      "learning_rate": 2.2362387496395333e-05,
      "loss": 0.8657,
      "step": 1374
    },
    {
      "epoch": 0.7029652351738241,
      "grad_norm": 3.3451359272003174,
      "learning_rate": 2.229189822910876e-05,
      "loss": 0.8688,
      "step": 1375
    },
    {
      "epoch": 0.7034764826175869,
      "grad_norm": 3.640519857406616,
      "learning_rate": 2.2221488349019903e-05,
      "loss": 0.8744,
      "step": 1376
    },
    {
      "epoch": 0.7039877300613497,
      "grad_norm": 3.371683359146118,
      "learning_rate": 2.21511580578619e-05,
      "loss": 0.887,
      "step": 1377
    },
    {
      "epoch": 0.7044989775051125,
      "grad_norm": 3.3256847858428955,
      "learning_rate": 2.2080907557139922e-05,
      "loss": 0.8098,
      "step": 1378
    },
    {
      "epoch": 0.7050102249488752,
      "grad_norm": 3.6227996349334717,
      "learning_rate": 2.201073704813046e-05,
      "loss": 0.8696,
      "step": 1379
    },
    {
      "epoch": 0.7055214723926381,
      "grad_norm": 3.5112345218658447,
      "learning_rate": 2.194064673188089e-05,
      "loss": 0.8226,
      "step": 1380
    },
    {
      "epoch": 0.7060327198364008,
      "grad_norm": 3.5785815715789795,
      "learning_rate": 2.187063680920875e-05,
      "loss": 0.9444,
      "step": 1381
    },
    {
      "epoch": 0.7065439672801636,
      "grad_norm": 3.671478271484375,
      "learning_rate": 2.18007074807013e-05,
      "loss": 0.8786,
      "step": 1382
    },
    {
      "epoch": 0.7070552147239264,
      "grad_norm": 3.485398054122925,
      "learning_rate": 2.1730858946714828e-05,
      "loss": 0.7943,
      "step": 1383
    },
    {
      "epoch": 0.7075664621676891,
      "grad_norm": 3.856492280960083,
      "learning_rate": 2.1661091407374218e-05,
      "loss": 0.8619,
      "step": 1384
    },
    {
      "epoch": 0.7080777096114519,
      "grad_norm": 4.099680423736572,
      "learning_rate": 2.1591405062572213e-05,
      "loss": 0.8321,
      "step": 1385
    },
    {
      "epoch": 0.7085889570552147,
      "grad_norm": 4.101649761199951,
      "learning_rate": 2.1521800111968992e-05,
      "loss": 0.9423,
      "step": 1386
    },
    {
      "epoch": 0.7091002044989775,
      "grad_norm": 4.159656047821045,
      "learning_rate": 2.1452276754991456e-05,
      "loss": 0.8539,
      "step": 1387
    },
    {
      "epoch": 0.7096114519427403,
      "grad_norm": 4.193243026733398,
      "learning_rate": 2.1382835190832813e-05,
      "loss": 0.8896,
      "step": 1388
    },
    {
      "epoch": 0.7101226993865031,
      "grad_norm": 4.159454345703125,
      "learning_rate": 2.1313475618451857e-05,
      "loss": 0.8223,
      "step": 1389
    },
    {
      "epoch": 0.7106339468302658,
      "grad_norm": 4.094180107116699,
      "learning_rate": 2.1244198236572477e-05,
      "loss": 0.7764,
      "step": 1390
    },
    {
      "epoch": 0.7111451942740287,
      "grad_norm": 4.319839000701904,
      "learning_rate": 2.1175003243683117e-05,
      "loss": 0.7791,
      "step": 1391
    },
    {
      "epoch": 0.7116564417177914,
      "grad_norm": 4.31974983215332,
      "learning_rate": 2.110589083803613e-05,
      "loss": 0.831,
      "step": 1392
    },
    {
      "epoch": 0.7121676891615542,
      "grad_norm": 4.274682521820068,
      "learning_rate": 2.1036861217647242e-05,
      "loss": 0.8928,
      "step": 1393
    },
    {
      "epoch": 0.712678936605317,
      "grad_norm": 4.526281356811523,
      "learning_rate": 2.096791458029499e-05,
      "loss": 0.6852,
      "step": 1394
    },
    {
      "epoch": 0.7131901840490797,
      "grad_norm": 4.865596771240234,
      "learning_rate": 2.0899051123520196e-05,
      "loss": 0.8364,
      "step": 1395
    },
    {
      "epoch": 0.7137014314928425,
      "grad_norm": 4.703789710998535,
      "learning_rate": 2.08302710446253e-05,
      "loss": 0.7496,
      "step": 1396
    },
    {
      "epoch": 0.7142126789366053,
      "grad_norm": 5.552756309509277,
      "learning_rate": 2.076157454067391e-05,
      "loss": 0.9237,
      "step": 1397
    },
    {
      "epoch": 0.7147239263803681,
      "grad_norm": 4.742315769195557,
      "learning_rate": 2.069296180849012e-05,
      "loss": 0.6129,
      "step": 1398
    },
    {
      "epoch": 0.7152351738241309,
      "grad_norm": 6.7453413009643555,
      "learning_rate": 2.0624433044658076e-05,
      "loss": 0.8049,
      "step": 1399
    },
    {
      "epoch": 0.7157464212678937,
      "grad_norm": 5.575685977935791,
      "learning_rate": 2.055598844552129e-05,
      "loss": 0.3437,
      "step": 1400
    },
    {
      "epoch": 0.7162576687116564,
      "grad_norm": 1.7778542041778564,
      "learning_rate": 2.0487628207182148e-05,
      "loss": 0.7406,
      "step": 1401
    },
    {
      "epoch": 0.7167689161554193,
      "grad_norm": 2.223677158355713,
      "learning_rate": 2.0419352525501316e-05,
      "loss": 0.9807,
      "step": 1402
    },
    {
      "epoch": 0.717280163599182,
      "grad_norm": 2.3648645877838135,
      "learning_rate": 2.035116159609725e-05,
      "loss": 0.9588,
      "step": 1403
    },
    {
      "epoch": 0.7177914110429447,
      "grad_norm": 2.5223405361175537,
      "learning_rate": 2.0283055614345532e-05,
      "loss": 0.982,
      "step": 1404
    },
    {
      "epoch": 0.7183026584867076,
      "grad_norm": 2.8171517848968506,
      "learning_rate": 2.0215034775378332e-05,
      "loss": 0.9429,
      "step": 1405
    },
    {
      "epoch": 0.7188139059304703,
      "grad_norm": 2.740797281265259,
      "learning_rate": 2.014709927408397e-05,
      "loss": 0.9401,
      "step": 1406
    },
    {
      "epoch": 0.7193251533742331,
      "grad_norm": 2.6437759399414062,
      "learning_rate": 2.0079249305106146e-05,
      "loss": 0.8826,
      "step": 1407
    },
    {
      "epoch": 0.7198364008179959,
      "grad_norm": 3.0211195945739746,
      "learning_rate": 2.001148506284361e-05,
      "loss": 0.9285,
      "step": 1408
    },
    {
      "epoch": 0.7203476482617587,
      "grad_norm": 3.08404541015625,
      "learning_rate": 1.99438067414494e-05,
      "loss": 0.9163,
      "step": 1409
    },
    {
      "epoch": 0.7208588957055214,
      "grad_norm": 3.098724842071533,
      "learning_rate": 1.9876214534830477e-05,
      "loss": 0.9279,
      "step": 1410
    },
    {
      "epoch": 0.7213701431492843,
      "grad_norm": 2.9764244556427,
      "learning_rate": 1.980870863664695e-05,
      "loss": 0.8716,
      "step": 1411
    },
    {
      "epoch": 0.721881390593047,
      "grad_norm": 2.8523361682891846,
      "learning_rate": 1.9741289240311755e-05,
      "loss": 0.9273,
      "step": 1412
    },
    {
      "epoch": 0.7223926380368099,
      "grad_norm": 2.991556406021118,
      "learning_rate": 1.9673956538989912e-05,
      "loss": 0.928,
      "step": 1413
    },
    {
      "epoch": 0.7229038854805726,
      "grad_norm": 2.887350559234619,
      "learning_rate": 1.960671072559812e-05,
      "loss": 0.8584,
      "step": 1414
    },
    {
      "epoch": 0.7234151329243353,
      "grad_norm": 3.0709550380706787,
      "learning_rate": 1.953955199280408e-05,
      "loss": 0.8571,
      "step": 1415
    },
    {
      "epoch": 0.7239263803680982,
      "grad_norm": 3.103337287902832,
      "learning_rate": 1.9472480533025984e-05,
      "loss": 0.9068,
      "step": 1416
    },
    {
      "epoch": 0.7244376278118609,
      "grad_norm": 2.9760520458221436,
      "learning_rate": 1.9405496538432043e-05,
      "loss": 0.8228,
      "step": 1417
    },
    {
      "epoch": 0.7249488752556237,
      "grad_norm": 3.2029881477355957,
      "learning_rate": 1.9338600200939805e-05,
      "loss": 0.9329,
      "step": 1418
    },
    {
      "epoch": 0.7254601226993865,
      "grad_norm": 3.1520378589630127,
      "learning_rate": 1.927179171221571e-05,
      "loss": 0.9649,
      "step": 1419
    },
    {
      "epoch": 0.7259713701431493,
      "grad_norm": 3.432126045227051,
      "learning_rate": 1.920507126367448e-05,
      "loss": 0.8563,
      "step": 1420
    },
    {
      "epoch": 0.726482617586912,
      "grad_norm": 3.369917154312134,
      "learning_rate": 1.9138439046478585e-05,
      "loss": 0.9227,
      "step": 1421
    },
    {
      "epoch": 0.7269938650306749,
      "grad_norm": 3.478285551071167,
      "learning_rate": 1.9071895251537702e-05,
      "loss": 0.8868,
      "step": 1422
    },
    {
      "epoch": 0.7275051124744376,
      "grad_norm": 3.168760061264038,
      "learning_rate": 1.90054400695082e-05,
      "loss": 0.8772,
      "step": 1423
    },
    {
      "epoch": 0.7280163599182005,
      "grad_norm": 3.382513999938965,
      "learning_rate": 1.893907369079252e-05,
      "loss": 0.7854,
      "step": 1424
    },
    {
      "epoch": 0.7285276073619632,
      "grad_norm": 3.3271732330322266,
      "learning_rate": 1.8872796305538698e-05,
      "loss": 0.9083,
      "step": 1425
    },
    {
      "epoch": 0.7290388548057259,
      "grad_norm": 3.714656352996826,
      "learning_rate": 1.8806608103639766e-05,
      "loss": 0.9312,
      "step": 1426
    },
    {
      "epoch": 0.7295501022494888,
      "grad_norm": 3.502117872238159,
      "learning_rate": 1.8740509274733276e-05,
      "loss": 0.8154,
      "step": 1427
    },
    {
      "epoch": 0.7300613496932515,
      "grad_norm": 3.463674783706665,
      "learning_rate": 1.8674500008200674e-05,
      "loss": 0.8553,
      "step": 1428
    },
    {
      "epoch": 0.7305725971370143,
      "grad_norm": 3.8900396823883057,
      "learning_rate": 1.8608580493166795e-05,
      "loss": 0.972,
      "step": 1429
    },
    {
      "epoch": 0.7310838445807771,
      "grad_norm": 3.5256094932556152,
      "learning_rate": 1.8542750918499395e-05,
      "loss": 0.8967,
      "step": 1430
    },
    {
      "epoch": 0.7315950920245399,
      "grad_norm": 3.558422327041626,
      "learning_rate": 1.8477011472808424e-05,
      "loss": 0.9316,
      "step": 1431
    },
    {
      "epoch": 0.7321063394683026,
      "grad_norm": 3.8706467151641846,
      "learning_rate": 1.8411362344445708e-05,
      "loss": 0.856,
      "step": 1432
    },
    {
      "epoch": 0.7326175869120655,
      "grad_norm": 4.188419342041016,
      "learning_rate": 1.8345803721504234e-05,
      "loss": 0.9773,
      "step": 1433
    },
    {
      "epoch": 0.7331288343558282,
      "grad_norm": 3.983508825302124,
      "learning_rate": 1.8280335791817733e-05,
      "loss": 0.875,
      "step": 1434
    },
    {
      "epoch": 0.733640081799591,
      "grad_norm": 3.518186569213867,
      "learning_rate": 1.821495874296003e-05,
      "loss": 0.7896,
      "step": 1435
    },
    {
      "epoch": 0.7341513292433538,
      "grad_norm": 4.46856164932251,
      "learning_rate": 1.8149672762244624e-05,
      "loss": 1.0133,
      "step": 1436
    },
    {
      "epoch": 0.7346625766871165,
      "grad_norm": 3.835460662841797,
      "learning_rate": 1.808447803672404e-05,
      "loss": 0.837,
      "step": 1437
    },
    {
      "epoch": 0.7351738241308794,
      "grad_norm": 4.0243916511535645,
      "learning_rate": 1.801937475318939e-05,
      "loss": 0.8681,
      "step": 1438
    },
    {
      "epoch": 0.7356850715746421,
      "grad_norm": 4.4127960205078125,
      "learning_rate": 1.7954363098169768e-05,
      "loss": 0.9832,
      "step": 1439
    },
    {
      "epoch": 0.7361963190184049,
      "grad_norm": 4.0299248695373535,
      "learning_rate": 1.7889443257931737e-05,
      "loss": 0.8392,
      "step": 1440
    },
    {
      "epoch": 0.7367075664621677,
      "grad_norm": 4.006964206695557,
      "learning_rate": 1.782461541847879e-05,
      "loss": 0.7381,
      "step": 1441
    },
    {
      "epoch": 0.7372188139059305,
      "grad_norm": 4.23606014251709,
      "learning_rate": 1.7759879765550887e-05,
      "loss": 0.8984,
      "step": 1442
    },
    {
      "epoch": 0.7377300613496932,
      "grad_norm": 4.232422351837158,
      "learning_rate": 1.769523648462379e-05,
      "loss": 0.8403,
      "step": 1443
    },
    {
      "epoch": 0.7382413087934561,
      "grad_norm": 4.366641044616699,
      "learning_rate": 1.7630685760908622e-05,
      "loss": 0.8691,
      "step": 1444
    },
    {
      "epoch": 0.7387525562372188,
      "grad_norm": 4.925327777862549,
      "learning_rate": 1.7566227779351357e-05,
      "loss": 0.9101,
      "step": 1445
    },
    {
      "epoch": 0.7392638036809815,
      "grad_norm": 5.099564552307129,
      "learning_rate": 1.750186272463219e-05,
      "loss": 0.7557,
      "step": 1446
    },
    {
      "epoch": 0.7397750511247444,
      "grad_norm": 5.312896728515625,
      "learning_rate": 1.7437590781165138e-05,
      "loss": 0.6865,
      "step": 1447
    },
    {
      "epoch": 0.7402862985685071,
      "grad_norm": 5.26222038269043,
      "learning_rate": 1.7373412133097372e-05,
      "loss": 0.7639,
      "step": 1448
    },
    {
      "epoch": 0.74079754601227,
      "grad_norm": 4.816617012023926,
      "learning_rate": 1.7309326964308838e-05,
      "loss": 0.4676,
      "step": 1449
    },
    {
      "epoch": 0.7413087934560327,
      "grad_norm": 6.375493049621582,
      "learning_rate": 1.7245335458411542e-05,
      "loss": 0.4741,
      "step": 1450
    },
    {
      "epoch": 0.7418200408997955,
      "grad_norm": 1.9384807348251343,
      "learning_rate": 1.7181437798749256e-05,
      "loss": 0.8033,
      "step": 1451
    },
    {
      "epoch": 0.7423312883435583,
      "grad_norm": 2.3475661277770996,
      "learning_rate": 1.7117634168396774e-05,
      "loss": 0.9676,
      "step": 1452
    },
    {
      "epoch": 0.7428425357873211,
      "grad_norm": 2.614190101623535,
      "learning_rate": 1.705392475015956e-05,
      "loss": 1.0618,
      "step": 1453
    },
    {
      "epoch": 0.7433537832310838,
      "grad_norm": 2.739084243774414,
      "learning_rate": 1.6990309726573095e-05,
      "loss": 0.9619,
      "step": 1454
    },
    {
      "epoch": 0.7438650306748467,
      "grad_norm": 2.832260847091675,
      "learning_rate": 1.6926789279902412e-05,
      "loss": 0.9322,
      "step": 1455
    },
    {
      "epoch": 0.7443762781186094,
      "grad_norm": 2.8056933879852295,
      "learning_rate": 1.6863363592141618e-05,
      "loss": 0.9499,
      "step": 1456
    },
    {
      "epoch": 0.7448875255623721,
      "grad_norm": 2.8965537548065186,
      "learning_rate": 1.6800032845013247e-05,
      "loss": 0.907,
      "step": 1457
    },
    {
      "epoch": 0.745398773006135,
      "grad_norm": 3.008425235748291,
      "learning_rate": 1.673679721996789e-05,
      "loss": 0.9932,
      "step": 1458
    },
    {
      "epoch": 0.7459100204498977,
      "grad_norm": 2.81937837600708,
      "learning_rate": 1.6673656898183572e-05,
      "loss": 0.9586,
      "step": 1459
    },
    {
      "epoch": 0.7464212678936605,
      "grad_norm": 2.9769203662872314,
      "learning_rate": 1.6610612060565234e-05,
      "loss": 0.9148,
      "step": 1460
    },
    {
      "epoch": 0.7469325153374233,
      "grad_norm": 3.3971211910247803,
      "learning_rate": 1.6547662887744265e-05,
      "loss": 0.906,
      "step": 1461
    },
    {
      "epoch": 0.7474437627811861,
      "grad_norm": 2.9512698650360107,
      "learning_rate": 1.648480956007799e-05,
      "loss": 0.9082,
      "step": 1462
    },
    {
      "epoch": 0.7479550102249489,
      "grad_norm": 3.2894976139068604,
      "learning_rate": 1.6422052257649078e-05,
      "loss": 0.9537,
      "step": 1463
    },
    {
      "epoch": 0.7484662576687117,
      "grad_norm": 2.9925243854522705,
      "learning_rate": 1.6359391160265125e-05,
      "loss": 0.9018,
      "step": 1464
    },
    {
      "epoch": 0.7489775051124744,
      "grad_norm": 3.4137401580810547,
      "learning_rate": 1.629682644745802e-05,
      "loss": 0.9413,
      "step": 1465
    },
    {
      "epoch": 0.7494887525562373,
      "grad_norm": 3.152463436126709,
      "learning_rate": 1.6234358298483575e-05,
      "loss": 0.9389,
      "step": 1466
    },
    {
      "epoch": 0.75,
      "grad_norm": 3.088907480239868,
      "learning_rate": 1.6171986892320884e-05,
      "loss": 0.8818,
      "step": 1467
    },
    {
      "epoch": 0.7505112474437627,
      "grad_norm": 2.9647679328918457,
      "learning_rate": 1.6109712407671867e-05,
      "loss": 0.8208,
      "step": 1468
    },
    {
      "epoch": 0.7510224948875256,
      "grad_norm": 3.121256113052368,
      "learning_rate": 1.6047535022960757e-05,
      "loss": 0.8292,
      "step": 1469
    },
    {
      "epoch": 0.7515337423312883,
      "grad_norm": 3.0089125633239746,
      "learning_rate": 1.5985454916333577e-05,
      "loss": 0.8211,
      "step": 1470
    },
    {
      "epoch": 0.7520449897750511,
      "grad_norm": 3.4958393573760986,
      "learning_rate": 1.592347226565766e-05,
      "loss": 0.9042,
      "step": 1471
    },
    {
      "epoch": 0.7525562372188139,
      "grad_norm": 3.168567419052124,
      "learning_rate": 1.586158724852108e-05,
      "loss": 0.783,
      "step": 1472
    },
    {
      "epoch": 0.7530674846625767,
      "grad_norm": 3.506746530532837,
      "learning_rate": 1.579980004223222e-05,
      "loss": 0.8124,
      "step": 1473
    },
    {
      "epoch": 0.7535787321063395,
      "grad_norm": 3.678039789199829,
      "learning_rate": 1.573811082381918e-05,
      "loss": 0.9267,
      "step": 1474
    },
    {
      "epoch": 0.7540899795501023,
      "grad_norm": 3.4833285808563232,
      "learning_rate": 1.567651977002935e-05,
      "loss": 0.8201,
      "step": 1475
    },
    {
      "epoch": 0.754601226993865,
      "grad_norm": 3.767958879470825,
      "learning_rate": 1.561502705732883e-05,
      "loss": 0.9726,
      "step": 1476
    },
    {
      "epoch": 0.7551124744376279,
      "grad_norm": 3.3126537799835205,
      "learning_rate": 1.5553632861901995e-05,
      "loss": 0.7471,
      "step": 1477
    },
    {
      "epoch": 0.7556237218813906,
      "grad_norm": 3.590383529663086,
      "learning_rate": 1.5492337359650937e-05,
      "loss": 0.8543,
      "step": 1478
    },
    {
      "epoch": 0.7561349693251533,
      "grad_norm": 3.4559550285339355,
      "learning_rate": 1.5431140726194974e-05,
      "loss": 0.8802,
      "step": 1479
    },
    {
      "epoch": 0.7566462167689162,
      "grad_norm": 3.7260870933532715,
      "learning_rate": 1.5370043136870148e-05,
      "loss": 0.9178,
      "step": 1480
    },
    {
      "epoch": 0.7571574642126789,
      "grad_norm": 4.135682106018066,
      "learning_rate": 1.5309044766728775e-05,
      "loss": 0.8823,
      "step": 1481
    },
    {
      "epoch": 0.7576687116564417,
      "grad_norm": 3.654975175857544,
      "learning_rate": 1.5248145790538837e-05,
      "loss": 0.8367,
      "step": 1482
    },
    {
      "epoch": 0.7581799591002045,
      "grad_norm": 3.653458833694458,
      "learning_rate": 1.5187346382783552e-05,
      "loss": 0.8005,
      "step": 1483
    },
    {
      "epoch": 0.7586912065439673,
      "grad_norm": 3.8800647258758545,
      "learning_rate": 1.5126646717660897e-05,
      "loss": 0.9004,
      "step": 1484
    },
    {
      "epoch": 0.75920245398773,
      "grad_norm": 3.79221773147583,
      "learning_rate": 1.5066046969083026e-05,
      "loss": 0.7986,
      "step": 1485
    },
    {
      "epoch": 0.7597137014314929,
      "grad_norm": 4.0721025466918945,
      "learning_rate": 1.5005547310675872e-05,
      "loss": 0.9721,
      "step": 1486
    },
    {
      "epoch": 0.7602249488752556,
      "grad_norm": 3.911992311477661,
      "learning_rate": 1.4945147915778535e-05,
      "loss": 0.8429,
      "step": 1487
    },
    {
      "epoch": 0.7607361963190185,
      "grad_norm": 4.218154430389404,
      "learning_rate": 1.4884848957442931e-05,
      "loss": 0.7957,
      "step": 1488
    },
    {
      "epoch": 0.7612474437627812,
      "grad_norm": 3.988151788711548,
      "learning_rate": 1.4824650608433099e-05,
      "loss": 0.8568,
      "step": 1489
    },
    {
      "epoch": 0.7617586912065439,
      "grad_norm": 4.378921985626221,
      "learning_rate": 1.4764553041224926e-05,
      "loss": 0.7963,
      "step": 1490
    },
    {
      "epoch": 0.7622699386503068,
      "grad_norm": 4.452109336853027,
      "learning_rate": 1.4704556428005478e-05,
      "loss": 0.835,
      "step": 1491
    },
    {
      "epoch": 0.7627811860940695,
      "grad_norm": 6.035761833190918,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.9188,
      "step": 1492
    },
    {
      "epoch": 0.7632924335378323,
      "grad_norm": 4.4452996253967285,
      "learning_rate": 1.4584866750834464e-05,
      "loss": 0.7396,
      "step": 1493
    },
    {
      "epoch": 0.7638036809815951,
      "grad_norm": 4.476807594299316,
      "learning_rate": 1.4525174029808858e-05,
      "loss": 0.7628,
      "step": 1494
    },
    {
      "epoch": 0.7643149284253579,
      "grad_norm": 4.541752338409424,
      "learning_rate": 1.4465582948622986e-05,
      "loss": 0.6893,
      "step": 1495
    },
    {
      "epoch": 0.7648261758691206,
      "grad_norm": 4.8742899894714355,
      "learning_rate": 1.4406093678012766e-05,
      "loss": 0.7884,
      "step": 1496
    },
    {
      "epoch": 0.7653374233128835,
      "grad_norm": 5.151451110839844,
      "learning_rate": 1.4346706388422493e-05,
      "loss": 0.683,
      "step": 1497
    },
    {
      "epoch": 0.7658486707566462,
      "grad_norm": 5.411539554595947,
      "learning_rate": 1.4287421250004157e-05,
      "loss": 0.79,
      "step": 1498
    },
    {
      "epoch": 0.766359918200409,
      "grad_norm": 5.503897666931152,
      "learning_rate": 1.4228238432617186e-05,
      "loss": 0.556,
      "step": 1499
    },
    {
      "epoch": 0.7668711656441718,
      "grad_norm": 6.515267372131348,
      "learning_rate": 1.4169158105827768e-05,
      "loss": 0.393,
      "step": 1500
    },
    {
      "epoch": 0.7673824130879345,
      "grad_norm": 2.1720798015594482,
      "learning_rate": 1.4110180438908505e-05,
      "loss": 0.7376,
      "step": 1501
    },
    {
      "epoch": 0.7678936605316974,
      "grad_norm": 2.3930561542510986,
      "learning_rate": 1.4051305600837799e-05,
      "loss": 1.0402,
      "step": 1502
    },
    {
      "epoch": 0.7684049079754601,
      "grad_norm": 2.3610873222351074,
      "learning_rate": 1.3992533760299498e-05,
      "loss": 0.9187,
      "step": 1503
    },
    {
      "epoch": 0.7689161554192229,
      "grad_norm": 2.4556972980499268,
      "learning_rate": 1.3933865085682312e-05,
      "loss": 0.9221,
      "step": 1504
    },
    {
      "epoch": 0.7694274028629857,
      "grad_norm": 2.7687621116638184,
      "learning_rate": 1.387529974507935e-05,
      "loss": 1.0278,
      "step": 1505
    },
    {
      "epoch": 0.7699386503067485,
      "grad_norm": 2.8164188861846924,
      "learning_rate": 1.3816837906287722e-05,
      "loss": 0.9995,
      "step": 1506
    },
    {
      "epoch": 0.7704498977505112,
      "grad_norm": 2.5187416076660156,
      "learning_rate": 1.3758479736807928e-05,
      "loss": 0.813,
      "step": 1507
    },
    {
      "epoch": 0.7709611451942741,
      "grad_norm": 2.9586594104766846,
      "learning_rate": 1.3700225403843469e-05,
      "loss": 0.9001,
      "step": 1508
    },
    {
      "epoch": 0.7714723926380368,
      "grad_norm": 3.797866106033325,
      "learning_rate": 1.3642075074300325e-05,
      "loss": 0.9324,
      "step": 1509
    },
    {
      "epoch": 0.7719836400817995,
      "grad_norm": 2.9937219619750977,
      "learning_rate": 1.3584028914786539e-05,
      "loss": 0.922,
      "step": 1510
    },
    {
      "epoch": 0.7724948875255624,
      "grad_norm": 2.9317002296447754,
      "learning_rate": 1.3526087091611623e-05,
      "loss": 0.9387,
      "step": 1511
    },
    {
      "epoch": 0.7730061349693251,
      "grad_norm": 3.1819257736206055,
      "learning_rate": 1.3468249770786223e-05,
      "loss": 0.8964,
      "step": 1512
    },
    {
      "epoch": 0.773517382413088,
      "grad_norm": 2.7915236949920654,
      "learning_rate": 1.341051711802151e-05,
      "loss": 0.9163,
      "step": 1513
    },
    {
      "epoch": 0.7740286298568507,
      "grad_norm": 2.946099281311035,
      "learning_rate": 1.3352889298728832e-05,
      "loss": 0.9546,
      "step": 1514
    },
    {
      "epoch": 0.7745398773006135,
      "grad_norm": 3.0270745754241943,
      "learning_rate": 1.3295366478019112e-05,
      "loss": 0.9431,
      "step": 1515
    },
    {
      "epoch": 0.7750511247443763,
      "grad_norm": 3.3205618858337402,
      "learning_rate": 1.3237948820702495e-05,
      "loss": 0.9058,
      "step": 1516
    },
    {
      "epoch": 0.7755623721881391,
      "grad_norm": 3.179041624069214,
      "learning_rate": 1.3180636491287773e-05,
      "loss": 0.8804,
      "step": 1517
    },
    {
      "epoch": 0.7760736196319018,
      "grad_norm": 3.682973623275757,
      "learning_rate": 1.3123429653981995e-05,
      "loss": 0.985,
      "step": 1518
    },
    {
      "epoch": 0.7765848670756647,
      "grad_norm": 3.1981019973754883,
      "learning_rate": 1.3066328472689932e-05,
      "loss": 0.8148,
      "step": 1519
    },
    {
      "epoch": 0.7770961145194274,
      "grad_norm": 3.3223016262054443,
      "learning_rate": 1.300933311101365e-05,
      "loss": 0.8839,
      "step": 1520
    },
    {
      "epoch": 0.7776073619631901,
      "grad_norm": 3.214383602142334,
      "learning_rate": 1.2952443732252057e-05,
      "loss": 0.9364,
      "step": 1521
    },
    {
      "epoch": 0.778118609406953,
      "grad_norm": 3.4616458415985107,
      "learning_rate": 1.2895660499400348e-05,
      "loss": 0.8707,
      "step": 1522
    },
    {
      "epoch": 0.7786298568507157,
      "grad_norm": 3.4007458686828613,
      "learning_rate": 1.283898357514966e-05,
      "loss": 0.9236,
      "step": 1523
    },
    {
      "epoch": 0.7791411042944786,
      "grad_norm": 3.370945930480957,
      "learning_rate": 1.2782413121886483e-05,
      "loss": 0.8499,
      "step": 1524
    },
    {
      "epoch": 0.7796523517382413,
      "grad_norm": 3.3444669246673584,
      "learning_rate": 1.2725949301692314e-05,
      "loss": 0.8872,
      "step": 1525
    },
    {
      "epoch": 0.7801635991820041,
      "grad_norm": 3.51122784614563,
      "learning_rate": 1.2669592276343084e-05,
      "loss": 0.911,
      "step": 1526
    },
    {
      "epoch": 0.7806748466257669,
      "grad_norm": 3.6887714862823486,
      "learning_rate": 1.2613342207308764e-05,
      "loss": 0.9304,
      "step": 1527
    },
    {
      "epoch": 0.7811860940695297,
      "grad_norm": 3.856081485748291,
      "learning_rate": 1.2557199255752867e-05,
      "loss": 0.9293,
      "step": 1528
    },
    {
      "epoch": 0.7816973415132924,
      "grad_norm": 3.9948816299438477,
      "learning_rate": 1.2501163582532038e-05,
      "loss": 0.837,
      "step": 1529
    },
    {
      "epoch": 0.7822085889570553,
      "grad_norm": 3.6386806964874268,
      "learning_rate": 1.24452353481955e-05,
      "loss": 0.8118,
      "step": 1530
    },
    {
      "epoch": 0.782719836400818,
      "grad_norm": 3.716519355773926,
      "learning_rate": 1.2389414712984715e-05,
      "loss": 0.7942,
      "step": 1531
    },
    {
      "epoch": 0.7832310838445807,
      "grad_norm": 3.621734142303467,
      "learning_rate": 1.2333701836832812e-05,
      "loss": 0.7823,
      "step": 1532
    },
    {
      "epoch": 0.7837423312883436,
      "grad_norm": 3.675961494445801,
      "learning_rate": 1.227809687936417e-05,
      "loss": 0.8813,
      "step": 1533
    },
    {
      "epoch": 0.7842535787321063,
      "grad_norm": 4.009909629821777,
      "learning_rate": 1.2222599999894018e-05,
      "loss": 0.9671,
      "step": 1534
    },
    {
      "epoch": 0.7847648261758691,
      "grad_norm": 3.6856648921966553,
      "learning_rate": 1.2167211357427878e-05,
      "loss": 0.7759,
      "step": 1535
    },
    {
      "epoch": 0.7852760736196319,
      "grad_norm": 4.252887725830078,
      "learning_rate": 1.2111931110661212e-05,
      "loss": 0.8245,
      "step": 1536
    },
    {
      "epoch": 0.7857873210633947,
      "grad_norm": 4.382503986358643,
      "learning_rate": 1.2056759417978835e-05,
      "loss": 1.0605,
      "step": 1537
    },
    {
      "epoch": 0.7862985685071575,
      "grad_norm": 4.000999927520752,
      "learning_rate": 1.2001696437454624e-05,
      "loss": 0.8371,
      "step": 1538
    },
    {
      "epoch": 0.7868098159509203,
      "grad_norm": 4.442836761474609,
      "learning_rate": 1.1946742326850912e-05,
      "loss": 0.8388,
      "step": 1539
    },
    {
      "epoch": 0.787321063394683,
      "grad_norm": 4.2866530418396,
      "learning_rate": 1.1891897243618182e-05,
      "loss": 0.7989,
      "step": 1540
    },
    {
      "epoch": 0.7878323108384458,
      "grad_norm": 4.245499610900879,
      "learning_rate": 1.183716134489446e-05,
      "loss": 0.9314,
      "step": 1541
    },
    {
      "epoch": 0.7883435582822086,
      "grad_norm": 4.35511589050293,
      "learning_rate": 1.1782534787505017e-05,
      "loss": 0.7689,
      "step": 1542
    },
    {
      "epoch": 0.7888548057259713,
      "grad_norm": 4.421148777008057,
      "learning_rate": 1.1728017727961794e-05,
      "loss": 0.7314,
      "step": 1543
    },
    {
      "epoch": 0.7893660531697342,
      "grad_norm": 4.419631481170654,
      "learning_rate": 1.1673610322463014e-05,
      "loss": 0.7021,
      "step": 1544
    },
    {
      "epoch": 0.7898773006134969,
      "grad_norm": 4.954163074493408,
      "learning_rate": 1.1619312726892762e-05,
      "loss": 0.7313,
      "step": 1545
    },
    {
      "epoch": 0.7903885480572597,
      "grad_norm": 4.796021461486816,
      "learning_rate": 1.1565125096820473e-05,
      "loss": 0.6311,
      "step": 1546
    },
    {
      "epoch": 0.7908997955010225,
      "grad_norm": 4.741822242736816,
      "learning_rate": 1.1511047587500523e-05,
      "loss": 0.5983,
      "step": 1547
    },
    {
      "epoch": 0.7914110429447853,
      "grad_norm": 6.696097373962402,
      "learning_rate": 1.1457080353871769e-05,
      "loss": 0.9485,
      "step": 1548
    },
    {
      "epoch": 0.7919222903885481,
      "grad_norm": 5.633409023284912,
      "learning_rate": 1.1403223550557146e-05,
      "loss": 0.6431,
      "step": 1549
    },
    {
      "epoch": 0.7924335378323109,
      "grad_norm": 7.090452194213867,
      "learning_rate": 1.134947733186315e-05,
      "loss": 0.4912,
      "step": 1550
    },
    {
      "epoch": 0.7929447852760736,
      "grad_norm": 1.965378761291504,
      "learning_rate": 1.1295841851779488e-05,
      "loss": 0.9309,
      "step": 1551
    },
    {
      "epoch": 0.7934560327198364,
      "grad_norm": 2.6269702911376953,
      "learning_rate": 1.1242317263978525e-05,
      "loss": 1.0566,
      "step": 1552
    },
    {
      "epoch": 0.7939672801635992,
      "grad_norm": 2.5522866249084473,
      "learning_rate": 1.118890372181497e-05,
      "loss": 0.9996,
      "step": 1553
    },
    {
      "epoch": 0.7944785276073619,
      "grad_norm": 2.6898603439331055,
      "learning_rate": 1.1135601378325316e-05,
      "loss": 0.8959,
      "step": 1554
    },
    {
      "epoch": 0.7949897750511248,
      "grad_norm": 2.5996921062469482,
      "learning_rate": 1.1082410386227527e-05,
      "loss": 0.9355,
      "step": 1555
    },
    {
      "epoch": 0.7955010224948875,
      "grad_norm": 2.5710082054138184,
      "learning_rate": 1.102933089792042e-05,
      "loss": 0.9206,
      "step": 1556
    },
    {
      "epoch": 0.7960122699386503,
      "grad_norm": 2.8860228061676025,
      "learning_rate": 1.0976363065483464e-05,
      "loss": 0.9861,
      "step": 1557
    },
    {
      "epoch": 0.7965235173824131,
      "grad_norm": 3.0384457111358643,
      "learning_rate": 1.092350704067614e-05,
      "loss": 1.0165,
      "step": 1558
    },
    {
      "epoch": 0.7970347648261759,
      "grad_norm": 2.8058857917785645,
      "learning_rate": 1.0870762974937598e-05,
      "loss": 0.8881,
      "step": 1559
    },
    {
      "epoch": 0.7975460122699386,
      "grad_norm": 2.571446418762207,
      "learning_rate": 1.0818131019386252e-05,
      "loss": 0.8571,
      "step": 1560
    },
    {
      "epoch": 0.7980572597137015,
      "grad_norm": 3.1314709186553955,
      "learning_rate": 1.0765611324819247e-05,
      "loss": 0.8383,
      "step": 1561
    },
    {
      "epoch": 0.7985685071574642,
      "grad_norm": 2.915090799331665,
      "learning_rate": 1.0713204041712145e-05,
      "loss": 0.8493,
      "step": 1562
    },
    {
      "epoch": 0.799079754601227,
      "grad_norm": 3.2132315635681152,
      "learning_rate": 1.066090932021837e-05,
      "loss": 0.9181,
      "step": 1563
    },
    {
      "epoch": 0.7995910020449898,
      "grad_norm": 3.073258399963379,
      "learning_rate": 1.060872731016892e-05,
      "loss": 0.8863,
      "step": 1564
    },
    {
      "epoch": 0.8001022494887525,
      "grad_norm": 3.104003667831421,
      "learning_rate": 1.0556658161071792e-05,
      "loss": 0.9108,
      "step": 1565
    },
    {
      "epoch": 0.8006134969325154,
      "grad_norm": 3.0709762573242188,
      "learning_rate": 1.0504702022111661e-05,
      "loss": 0.8055,
      "step": 1566
    },
    {
      "epoch": 0.8011247443762781,
      "grad_norm": 3.2626142501831055,
      "learning_rate": 1.0452859042149382e-05,
      "loss": 0.8732,
      "step": 1567
    },
    {
      "epoch": 0.8016359918200409,
      "grad_norm": 3.4502737522125244,
      "learning_rate": 1.040112936972164e-05,
      "loss": 0.9525,
      "step": 1568
    },
    {
      "epoch": 0.8021472392638037,
      "grad_norm": 3.2807888984680176,
      "learning_rate": 1.0349513153040436e-05,
      "loss": 0.937,
      "step": 1569
    },
    {
      "epoch": 0.8026584867075665,
      "grad_norm": 3.3237321376800537,
      "learning_rate": 1.0298010539992748e-05,
      "loss": 0.8712,
      "step": 1570
    },
    {
      "epoch": 0.8031697341513292,
      "grad_norm": 3.9305286407470703,
      "learning_rate": 1.0246621678140023e-05,
      "loss": 0.9547,
      "step": 1571
    },
    {
      "epoch": 0.803680981595092,
      "grad_norm": 3.3259196281433105,
      "learning_rate": 1.0195346714717813e-05,
      "loss": 0.8361,
      "step": 1572
    },
    {
      "epoch": 0.8041922290388548,
      "grad_norm": 3.3074426651000977,
      "learning_rate": 1.0144185796635359e-05,
      "loss": 0.7841,
      "step": 1573
    },
    {
      "epoch": 0.8047034764826176,
      "grad_norm": 3.421103000640869,
      "learning_rate": 1.00931390704751e-05,
      "loss": 0.8898,
      "step": 1574
    },
    {
      "epoch": 0.8052147239263804,
      "grad_norm": 3.6833465099334717,
      "learning_rate": 1.0042206682492372e-05,
      "loss": 0.9386,
      "step": 1575
    },
    {
      "epoch": 0.8057259713701431,
      "grad_norm": 3.3095312118530273,
      "learning_rate": 9.991388778614824e-06,
      "loss": 0.8398,
      "step": 1576
    },
    {
      "epoch": 0.806237218813906,
      "grad_norm": 3.4535927772521973,
      "learning_rate": 9.940685504442183e-06,
      "loss": 0.86,
      "step": 1577
    },
    {
      "epoch": 0.8067484662576687,
      "grad_norm": 3.7982585430145264,
      "learning_rate": 9.89009700524568e-06,
      "loss": 0.9577,
      "step": 1578
    },
    {
      "epoch": 0.8072597137014315,
      "grad_norm": 3.5163521766662598,
      "learning_rate": 9.83962342596776e-06,
      "loss": 0.7962,
      "step": 1579
    },
    {
      "epoch": 0.8077709611451943,
      "grad_norm": 3.4356038570404053,
      "learning_rate": 9.789264911221546e-06,
      "loss": 0.8193,
      "step": 1580
    },
    {
      "epoch": 0.808282208588957,
      "grad_norm": 3.920682907104492,
      "learning_rate": 9.739021605290549e-06,
      "loss": 0.8903,
      "step": 1581
    },
    {
      "epoch": 0.8087934560327198,
      "grad_norm": 3.7626466751098633,
      "learning_rate": 9.688893652128151e-06,
      "loss": 0.8662,
      "step": 1582
    },
    {
      "epoch": 0.8093047034764826,
      "grad_norm": 4.124518394470215,
      "learning_rate": 9.638881195357224e-06,
      "loss": 0.8829,
      "step": 1583
    },
    {
      "epoch": 0.8098159509202454,
      "grad_norm": 3.74548602104187,
      "learning_rate": 9.588984378269783e-06,
      "loss": 0.7028,
      "step": 1584
    },
    {
      "epoch": 0.8103271983640081,
      "grad_norm": 3.769134998321533,
      "learning_rate": 9.539203343826469e-06,
      "loss": 0.9024,
      "step": 1585
    },
    {
      "epoch": 0.810838445807771,
      "grad_norm": 3.955554723739624,
      "learning_rate": 9.489538234656214e-06,
      "loss": 0.8195,
      "step": 1586
    },
    {
      "epoch": 0.8113496932515337,
      "grad_norm": 3.704868793487549,
      "learning_rate": 9.439989193055788e-06,
      "loss": 0.7934,
      "step": 1587
    },
    {
      "epoch": 0.8118609406952966,
      "grad_norm": 3.6619386672973633,
      "learning_rate": 9.39055636098945e-06,
      "loss": 0.868,
      "step": 1588
    },
    {
      "epoch": 0.8123721881390593,
      "grad_norm": 4.098423480987549,
      "learning_rate": 9.341239880088465e-06,
      "loss": 0.8754,
      "step": 1589
    },
    {
      "epoch": 0.8128834355828221,
      "grad_norm": 4.314637184143066,
      "learning_rate": 9.292039891650784e-06,
      "loss": 0.8408,
      "step": 1590
    },
    {
      "epoch": 0.8133946830265849,
      "grad_norm": 4.415641784667969,
      "learning_rate": 9.24295653664053e-06,
      "loss": 0.8727,
      "step": 1591
    },
    {
      "epoch": 0.8139059304703476,
      "grad_norm": 4.379350185394287,
      "learning_rate": 9.193989955687715e-06,
      "loss": 0.8209,
      "step": 1592
    },
    {
      "epoch": 0.8144171779141104,
      "grad_norm": 4.419501304626465,
      "learning_rate": 9.145140289087755e-06,
      "loss": 0.9432,
      "step": 1593
    },
    {
      "epoch": 0.8149284253578732,
      "grad_norm": 4.199825763702393,
      "learning_rate": 9.096407676801077e-06,
      "loss": 0.7578,
      "step": 1594
    },
    {
      "epoch": 0.815439672801636,
      "grad_norm": 5.105489253997803,
      "learning_rate": 9.047792258452742e-06,
      "loss": 0.8932,
      "step": 1595
    },
    {
      "epoch": 0.8159509202453987,
      "grad_norm": 5.17318058013916,
      "learning_rate": 8.999294173332058e-06,
      "loss": 0.9014,
      "step": 1596
    },
    {
      "epoch": 0.8164621676891616,
      "grad_norm": 4.764847755432129,
      "learning_rate": 8.950913560392132e-06,
      "loss": 0.7204,
      "step": 1597
    },
    {
      "epoch": 0.8169734151329243,
      "grad_norm": 5.115185737609863,
      "learning_rate": 8.902650558249499e-06,
      "loss": 0.7984,
      "step": 1598
    },
    {
      "epoch": 0.8174846625766872,
      "grad_norm": 5.964483261108398,
      "learning_rate": 8.854505305183752e-06,
      "loss": 0.709,
      "step": 1599
    },
    {
      "epoch": 0.8179959100204499,
      "grad_norm": 5.853104591369629,
      "learning_rate": 8.80647793913708e-06,
      "loss": 0.4204,
      "step": 1600
    },
    {
      "epoch": 0.8185071574642127,
      "grad_norm": 1.909232258796692,
      "learning_rate": 8.758568597713946e-06,
      "loss": 0.9726,
      "step": 1601
    },
    {
      "epoch": 0.8190184049079755,
      "grad_norm": 2.361161231994629,
      "learning_rate": 8.710777418180615e-06,
      "loss": 0.939,
      "step": 1602
    },
    {
      "epoch": 0.8195296523517382,
      "grad_norm": 2.425341844558716,
      "learning_rate": 8.663104537464866e-06,
      "loss": 0.8918,
      "step": 1603
    },
    {
      "epoch": 0.820040899795501,
      "grad_norm": 2.4736742973327637,
      "learning_rate": 8.615550092155478e-06,
      "loss": 0.9242,
      "step": 1604
    },
    {
      "epoch": 0.8205521472392638,
      "grad_norm": 2.6613001823425293,
      "learning_rate": 8.568114218501922e-06,
      "loss": 0.9088,
      "step": 1605
    },
    {
      "epoch": 0.8210633946830266,
      "grad_norm": 2.5696263313293457,
      "learning_rate": 8.520797052413931e-06,
      "loss": 0.8776,
      "step": 1606
    },
    {
      "epoch": 0.8215746421267893,
      "grad_norm": 2.944065570831299,
      "learning_rate": 8.473598729461163e-06,
      "loss": 0.9622,
      "step": 1607
    },
    {
      "epoch": 0.8220858895705522,
      "grad_norm": 2.8112123012542725,
      "learning_rate": 8.426519384872733e-06,
      "loss": 0.905,
      "step": 1608
    },
    {
      "epoch": 0.8225971370143149,
      "grad_norm": 2.9660046100616455,
      "learning_rate": 8.379559153536909e-06,
      "loss": 0.9772,
      "step": 1609
    },
    {
      "epoch": 0.8231083844580777,
      "grad_norm": 2.898458242416382,
      "learning_rate": 8.332718170000647e-06,
      "loss": 0.9601,
      "step": 1610
    },
    {
      "epoch": 0.8236196319018405,
      "grad_norm": 2.934843063354492,
      "learning_rate": 8.285996568469245e-06,
      "loss": 0.9621,
      "step": 1611
    },
    {
      "epoch": 0.8241308793456033,
      "grad_norm": 2.774526834487915,
      "learning_rate": 8.239394482805996e-06,
      "loss": 0.8366,
      "step": 1612
    },
    {
      "epoch": 0.8246421267893661,
      "grad_norm": 3.195984363555908,
      "learning_rate": 8.192912046531732e-06,
      "loss": 0.8993,
      "step": 1613
    },
    {
      "epoch": 0.8251533742331288,
      "grad_norm": 2.9150772094726562,
      "learning_rate": 8.14654939282447e-06,
      "loss": 0.818,
      "step": 1614
    },
    {
      "epoch": 0.8256646216768916,
      "grad_norm": 3.1671900749206543,
      "learning_rate": 8.10030665451904e-06,
      "loss": 0.9031,
      "step": 1615
    },
    {
      "epoch": 0.8261758691206544,
      "grad_norm": 3.0718796253204346,
      "learning_rate": 8.054183964106738e-06,
      "loss": 0.902,
      "step": 1616
    },
    {
      "epoch": 0.8266871165644172,
      "grad_norm": 3.13236403465271,
      "learning_rate": 8.008181453734832e-06,
      "loss": 0.8966,
      "step": 1617
    },
    {
      "epoch": 0.8271983640081799,
      "grad_norm": 3.3213281631469727,
      "learning_rate": 7.96229925520634e-06,
      "loss": 0.9051,
      "step": 1618
    },
    {
      "epoch": 0.8277096114519428,
      "grad_norm": 3.212383508682251,
      "learning_rate": 7.916537499979509e-06,
      "loss": 0.8877,
      "step": 1619
    },
    {
      "epoch": 0.8282208588957055,
      "grad_norm": 3.477694272994995,
      "learning_rate": 7.870896319167548e-06,
      "loss": 0.8837,
      "step": 1620
    },
    {
      "epoch": 0.8287321063394683,
      "grad_norm": 3.3036673069000244,
      "learning_rate": 7.825375843538163e-06,
      "loss": 0.8893,
      "step": 1621
    },
    {
      "epoch": 0.8292433537832311,
      "grad_norm": 3.0736191272735596,
      "learning_rate": 7.77997620351324e-06,
      "loss": 0.7746,
      "step": 1622
    },
    {
      "epoch": 0.8297546012269938,
      "grad_norm": 3.514597177505493,
      "learning_rate": 7.734697529168483e-06,
      "loss": 0.9255,
      "step": 1623
    },
    {
      "epoch": 0.8302658486707567,
      "grad_norm": 3.152017831802368,
      "learning_rate": 7.689539950232977e-06,
      "loss": 0.8449,
      "step": 1624
    },
    {
      "epoch": 0.8307770961145194,
      "grad_norm": 3.652573823928833,
      "learning_rate": 7.644503596088865e-06,
      "loss": 0.884,
      "step": 1625
    },
    {
      "epoch": 0.8312883435582822,
      "grad_norm": 3.5253610610961914,
      "learning_rate": 7.599588595770957e-06,
      "loss": 0.8573,
      "step": 1626
    },
    {
      "epoch": 0.831799591002045,
      "grad_norm": 3.6805419921875,
      "learning_rate": 7.554795077966409e-06,
      "loss": 0.7815,
      "step": 1627
    },
    {
      "epoch": 0.8323108384458078,
      "grad_norm": 3.5299296379089355,
      "learning_rate": 7.510123171014255e-06,
      "loss": 0.9062,
      "step": 1628
    },
    {
      "epoch": 0.8328220858895705,
      "grad_norm": 3.8087053298950195,
      "learning_rate": 7.4655730029051575e-06,
      "loss": 0.9056,
      "step": 1629
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 3.496819496154785,
      "learning_rate": 7.42114470128093e-06,
      "loss": 0.7842,
      "step": 1630
    },
    {
      "epoch": 0.8338445807770961,
      "grad_norm": 3.998002767562866,
      "learning_rate": 7.376838393434265e-06,
      "loss": 0.8565,
      "step": 1631
    },
    {
      "epoch": 0.8343558282208589,
      "grad_norm": 3.822242498397827,
      "learning_rate": 7.332654206308298e-06,
      "loss": 0.8017,
      "step": 1632
    },
    {
      "epoch": 0.8348670756646217,
      "grad_norm": 3.5486857891082764,
      "learning_rate": 7.288592266496286e-06,
      "loss": 0.8265,
      "step": 1633
    },
    {
      "epoch": 0.8353783231083844,
      "grad_norm": 3.6276445388793945,
      "learning_rate": 7.2446527002412225e-06,
      "loss": 0.851,
      "step": 1634
    },
    {
      "epoch": 0.8358895705521472,
      "grad_norm": 3.7170212268829346,
      "learning_rate": 7.20083563343551e-06,
      "loss": 0.7564,
      "step": 1635
    },
    {
      "epoch": 0.83640081799591,
      "grad_norm": 3.684058666229248,
      "learning_rate": 7.157141191620548e-06,
      "loss": 0.8008,
      "step": 1636
    },
    {
      "epoch": 0.8369120654396728,
      "grad_norm": 3.752002000808716,
      "learning_rate": 7.1135694999864e-06,
      "loss": 0.7732,
      "step": 1637
    },
    {
      "epoch": 0.8374233128834356,
      "grad_norm": 4.312587738037109,
      "learning_rate": 7.070120683371462e-06,
      "loss": 0.808,
      "step": 1638
    },
    {
      "epoch": 0.8379345603271984,
      "grad_norm": 4.659461498260498,
      "learning_rate": 7.026794866262048e-06,
      "loss": 0.8293,
      "step": 1639
    },
    {
      "epoch": 0.8384458077709611,
      "grad_norm": 4.59619665145874,
      "learning_rate": 6.983592172792086e-06,
      "loss": 0.851,
      "step": 1640
    },
    {
      "epoch": 0.838957055214724,
      "grad_norm": 4.543966293334961,
      "learning_rate": 6.940512726742715e-06,
      "loss": 0.8155,
      "step": 1641
    },
    {
      "epoch": 0.8394683026584867,
      "grad_norm": 4.165318489074707,
      "learning_rate": 6.897556651542003e-06,
      "loss": 0.7734,
      "step": 1642
    },
    {
      "epoch": 0.8399795501022495,
      "grad_norm": 4.547415733337402,
      "learning_rate": 6.854724070264451e-06,
      "loss": 0.9181,
      "step": 1643
    },
    {
      "epoch": 0.8404907975460123,
      "grad_norm": 4.19881010055542,
      "learning_rate": 6.812015105630842e-06,
      "loss": 0.7814,
      "step": 1644
    },
    {
      "epoch": 0.841002044989775,
      "grad_norm": 4.672346115112305,
      "learning_rate": 6.769429880007705e-06,
      "loss": 0.6833,
      "step": 1645
    },
    {
      "epoch": 0.8415132924335378,
      "grad_norm": 4.165514945983887,
      "learning_rate": 6.7269685154070895e-06,
      "loss": 0.5298,
      "step": 1646
    },
    {
      "epoch": 0.8420245398773006,
      "grad_norm": 5.330099105834961,
      "learning_rate": 6.6846311334861415e-06,
      "loss": 0.6836,
      "step": 1647
    },
    {
      "epoch": 0.8425357873210634,
      "grad_norm": 4.8200860023498535,
      "learning_rate": 6.642417855546768e-06,
      "loss": 0.7025,
      "step": 1648
    },
    {
      "epoch": 0.8430470347648262,
      "grad_norm": 5.211686611175537,
      "learning_rate": 6.600328802535355e-06,
      "loss": 0.5772,
      "step": 1649
    },
    {
      "epoch": 0.843558282208589,
      "grad_norm": 5.803190231323242,
      "learning_rate": 6.558364095042302e-06,
      "loss": 0.632,
      "step": 1650
    },
    {
      "epoch": 0.8440695296523517,
      "grad_norm": 2.0162458419799805,
      "learning_rate": 6.516523853301804e-06,
      "loss": 0.7469,
      "step": 1651
    },
    {
      "epoch": 0.8445807770961146,
      "grad_norm": 2.1648364067077637,
      "learning_rate": 6.474808197191401e-06,
      "loss": 0.9474,
      "step": 1652
    },
    {
      "epoch": 0.8450920245398773,
      "grad_norm": 2.424511194229126,
      "learning_rate": 6.433217246231704e-06,
      "loss": 0.9164,
      "step": 1653
    },
    {
      "epoch": 0.84560327198364,
      "grad_norm": 2.5511527061462402,
      "learning_rate": 6.391751119586003e-06,
      "loss": 0.9075,
      "step": 1654
    },
    {
      "epoch": 0.8461145194274029,
      "grad_norm": 2.9107165336608887,
      "learning_rate": 6.350409936059998e-06,
      "loss": 0.9481,
      "step": 1655
    },
    {
      "epoch": 0.8466257668711656,
      "grad_norm": 2.7874770164489746,
      "learning_rate": 6.3091938141013495e-06,
      "loss": 0.9119,
      "step": 1656
    },
    {
      "epoch": 0.8471370143149284,
      "grad_norm": 2.7134737968444824,
      "learning_rate": 6.268102871799459e-06,
      "loss": 0.8897,
      "step": 1657
    },
    {
      "epoch": 0.8476482617586912,
      "grad_norm": 2.6800270080566406,
      "learning_rate": 6.227137226885027e-06,
      "loss": 0.8625,
      "step": 1658
    },
    {
      "epoch": 0.848159509202454,
      "grad_norm": 3.039987564086914,
      "learning_rate": 6.186296996729796e-06,
      "loss": 0.9901,
      "step": 1659
    },
    {
      "epoch": 0.8486707566462167,
      "grad_norm": 2.883843183517456,
      "learning_rate": 6.145582298346153e-06,
      "loss": 0.8557,
      "step": 1660
    },
    {
      "epoch": 0.8491820040899796,
      "grad_norm": 3.040339469909668,
      "learning_rate": 6.104993248386831e-06,
      "loss": 0.9266,
      "step": 1661
    },
    {
      "epoch": 0.8496932515337423,
      "grad_norm": 3.1167523860931396,
      "learning_rate": 6.064529963144583e-06,
      "loss": 0.922,
      "step": 1662
    },
    {
      "epoch": 0.8502044989775052,
      "grad_norm": 3.086543560028076,
      "learning_rate": 6.024192558551784e-06,
      "loss": 0.8847,
      "step": 1663
    },
    {
      "epoch": 0.8507157464212679,
      "grad_norm": 3.4161489009857178,
      "learning_rate": 5.98398115018019e-06,
      "loss": 0.8947,
      "step": 1664
    },
    {
      "epoch": 0.8512269938650306,
      "grad_norm": 3.196218252182007,
      "learning_rate": 5.943895853240533e-06,
      "loss": 0.8973,
      "step": 1665
    },
    {
      "epoch": 0.8517382413087935,
      "grad_norm": 3.152571201324463,
      "learning_rate": 5.903936782582253e-06,
      "loss": 0.8851,
      "step": 1666
    },
    {
      "epoch": 0.8522494887525562,
      "grad_norm": 3.339434862136841,
      "learning_rate": 5.8641040526930925e-06,
      "loss": 0.9871,
      "step": 1667
    },
    {
      "epoch": 0.852760736196319,
      "grad_norm": 3.3059394359588623,
      "learning_rate": 5.824397777698859e-06,
      "loss": 0.8634,
      "step": 1668
    },
    {
      "epoch": 0.8532719836400818,
      "grad_norm": 3.992922306060791,
      "learning_rate": 5.784818071362996e-06,
      "loss": 0.9209,
      "step": 1669
    },
    {
      "epoch": 0.8537832310838446,
      "grad_norm": 3.1529664993286133,
      "learning_rate": 5.74536504708637e-06,
      "loss": 0.889,
      "step": 1670
    },
    {
      "epoch": 0.8542944785276073,
      "grad_norm": 3.0963938236236572,
      "learning_rate": 5.706038817906845e-06,
      "loss": 0.8722,
      "step": 1671
    },
    {
      "epoch": 0.8548057259713702,
      "grad_norm": 3.323939085006714,
      "learning_rate": 5.666839496499022e-06,
      "loss": 0.8879,
      "step": 1672
    },
    {
      "epoch": 0.8553169734151329,
      "grad_norm": 3.5231525897979736,
      "learning_rate": 5.6277671951738716e-06,
      "loss": 0.7867,
      "step": 1673
    },
    {
      "epoch": 0.8558282208588958,
      "grad_norm": 3.041142225265503,
      "learning_rate": 5.588822025878476e-06,
      "loss": 0.7922,
      "step": 1674
    },
    {
      "epoch": 0.8563394683026585,
      "grad_norm": 3.526838779449463,
      "learning_rate": 5.550004100195639e-06,
      "loss": 0.9025,
      "step": 1675
    },
    {
      "epoch": 0.8568507157464212,
      "grad_norm": 3.6863765716552734,
      "learning_rate": 5.5113135293435815e-06,
      "loss": 0.9744,
      "step": 1676
    },
    {
      "epoch": 0.8573619631901841,
      "grad_norm": 3.2295031547546387,
      "learning_rate": 5.4727504241756874e-06,
      "loss": 0.8475,
      "step": 1677
    },
    {
      "epoch": 0.8578732106339468,
      "grad_norm": 3.341581344604492,
      "learning_rate": 5.434314895180082e-06,
      "loss": 0.8515,
      "step": 1678
    },
    {
      "epoch": 0.8583844580777096,
      "grad_norm": 3.328876495361328,
      "learning_rate": 5.396007052479407e-06,
      "loss": 0.8321,
      "step": 1679
    },
    {
      "epoch": 0.8588957055214724,
      "grad_norm": 3.497668743133545,
      "learning_rate": 5.357827005830435e-06,
      "loss": 0.8929,
      "step": 1680
    },
    {
      "epoch": 0.8594069529652352,
      "grad_norm": 3.690748691558838,
      "learning_rate": 5.319774864623834e-06,
      "loss": 0.8603,
      "step": 1681
    },
    {
      "epoch": 0.8599182004089979,
      "grad_norm": 3.709012508392334,
      "learning_rate": 5.281850737883731e-06,
      "loss": 0.9677,
      "step": 1682
    },
    {
      "epoch": 0.8604294478527608,
      "grad_norm": 3.8703484535217285,
      "learning_rate": 5.2440547342675614e-06,
      "loss": 0.8876,
      "step": 1683
    },
    {
      "epoch": 0.8609406952965235,
      "grad_norm": 3.9959239959716797,
      "learning_rate": 5.206386962065602e-06,
      "loss": 0.791,
      "step": 1684
    },
    {
      "epoch": 0.8614519427402862,
      "grad_norm": 3.8929603099823,
      "learning_rate": 5.168847529200782e-06,
      "loss": 0.8654,
      "step": 1685
    },
    {
      "epoch": 0.8619631901840491,
      "grad_norm": 4.002877235412598,
      "learning_rate": 5.1314365432282904e-06,
      "loss": 0.8284,
      "step": 1686
    },
    {
      "epoch": 0.8624744376278118,
      "grad_norm": 4.066626071929932,
      "learning_rate": 5.094154111335292e-06,
      "loss": 0.8603,
      "step": 1687
    },
    {
      "epoch": 0.8629856850715747,
      "grad_norm": 4.232320785522461,
      "learning_rate": 5.057000340340678e-06,
      "loss": 0.9426,
      "step": 1688
    },
    {
      "epoch": 0.8634969325153374,
      "grad_norm": 3.965895175933838,
      "learning_rate": 5.019975336694649e-06,
      "loss": 0.7798,
      "step": 1689
    },
    {
      "epoch": 0.8640081799591002,
      "grad_norm": 4.249181270599365,
      "learning_rate": 4.983079206478513e-06,
      "loss": 0.8211,
      "step": 1690
    },
    {
      "epoch": 0.864519427402863,
      "grad_norm": 4.230957508087158,
      "learning_rate": 4.946312055404328e-06,
      "loss": 0.9142,
      "step": 1691
    },
    {
      "epoch": 0.8650306748466258,
      "grad_norm": 4.479368209838867,
      "learning_rate": 4.909673988814601e-06,
      "loss": 0.9803,
      "step": 1692
    },
    {
      "epoch": 0.8655419222903885,
      "grad_norm": 4.021700859069824,
      "learning_rate": 4.873165111681993e-06,
      "loss": 0.7705,
      "step": 1693
    },
    {
      "epoch": 0.8660531697341514,
      "grad_norm": 4.131351470947266,
      "learning_rate": 4.836785528609051e-06,
      "loss": 0.7206,
      "step": 1694
    },
    {
      "epoch": 0.8665644171779141,
      "grad_norm": 4.708834171295166,
      "learning_rate": 4.800535343827833e-06,
      "loss": 0.8698,
      "step": 1695
    },
    {
      "epoch": 0.8670756646216768,
      "grad_norm": 4.927844047546387,
      "learning_rate": 4.764414661199707e-06,
      "loss": 0.7615,
      "step": 1696
    },
    {
      "epoch": 0.8675869120654397,
      "grad_norm": 5.510052680969238,
      "learning_rate": 4.728423584214947e-06,
      "loss": 0.9676,
      "step": 1697
    },
    {
      "epoch": 0.8680981595092024,
      "grad_norm": 5.557268142700195,
      "learning_rate": 4.692562215992541e-06,
      "loss": 0.6631,
      "step": 1698
    },
    {
      "epoch": 0.8686094069529653,
      "grad_norm": 6.219789028167725,
      "learning_rate": 4.656830659279804e-06,
      "loss": 0.8028,
      "step": 1699
    },
    {
      "epoch": 0.869120654396728,
      "grad_norm": 5.629024028778076,
      "learning_rate": 4.621229016452156e-06,
      "loss": 0.4033,
      "step": 1700
    },
    {
      "epoch": 0.8696319018404908,
      "grad_norm": 1.9563968181610107,
      "learning_rate": 4.585757389512768e-06,
      "loss": 0.8627,
      "step": 1701
    },
    {
      "epoch": 0.8701431492842536,
      "grad_norm": 2.113006591796875,
      "learning_rate": 4.550415880092313e-06,
      "loss": 0.8872,
      "step": 1702
    },
    {
      "epoch": 0.8706543967280164,
      "grad_norm": 2.489034652709961,
      "learning_rate": 4.515204589448674e-06,
      "loss": 0.9124,
      "step": 1703
    },
    {
      "epoch": 0.8711656441717791,
      "grad_norm": 2.6170196533203125,
      "learning_rate": 4.48012361846662e-06,
      "loss": 1.0461,
      "step": 1704
    },
    {
      "epoch": 0.871676891615542,
      "grad_norm": 2.774226188659668,
      "learning_rate": 4.445173067657554e-06,
      "loss": 1.0186,
      "step": 1705
    },
    {
      "epoch": 0.8721881390593047,
      "grad_norm": 2.7719779014587402,
      "learning_rate": 4.410353037159193e-06,
      "loss": 0.9851,
      "step": 1706
    },
    {
      "epoch": 0.8726993865030674,
      "grad_norm": 2.8371808528900146,
      "learning_rate": 4.3756636267353214e-06,
      "loss": 1.0079,
      "step": 1707
    },
    {
      "epoch": 0.8732106339468303,
      "grad_norm": 3.129444122314453,
      "learning_rate": 4.341104935775442e-06,
      "loss": 0.8973,
      "step": 1708
    },
    {
      "epoch": 0.873721881390593,
      "grad_norm": 2.850924491882324,
      "learning_rate": 4.306677063294573e-06,
      "loss": 0.918,
      "step": 1709
    },
    {
      "epoch": 0.8742331288343558,
      "grad_norm": 2.839162826538086,
      "learning_rate": 4.272380107932888e-06,
      "loss": 0.9134,
      "step": 1710
    },
    {
      "epoch": 0.8747443762781186,
      "grad_norm": 3.149712562561035,
      "learning_rate": 4.238214167955484e-06,
      "loss": 0.9679,
      "step": 1711
    },
    {
      "epoch": 0.8752556237218814,
      "grad_norm": 2.945416212081909,
      "learning_rate": 4.2041793412520734e-06,
      "loss": 0.8602,
      "step": 1712
    },
    {
      "epoch": 0.8757668711656442,
      "grad_norm": 3.0397250652313232,
      "learning_rate": 4.17027572533672e-06,
      "loss": 0.9156,
      "step": 1713
    },
    {
      "epoch": 0.876278118609407,
      "grad_norm": 3.256863832473755,
      "learning_rate": 4.136503417347554e-06,
      "loss": 0.7461,
      "step": 1714
    },
    {
      "epoch": 0.8767893660531697,
      "grad_norm": 3.1958820819854736,
      "learning_rate": 4.102862514046474e-06,
      "loss": 0.8761,
      "step": 1715
    },
    {
      "epoch": 0.8773006134969326,
      "grad_norm": 3.4146556854248047,
      "learning_rate": 4.069353111818913e-06,
      "loss": 0.9106,
      "step": 1716
    },
    {
      "epoch": 0.8778118609406953,
      "grad_norm": 3.057286024093628,
      "learning_rate": 4.035975306673517e-06,
      "loss": 0.8755,
      "step": 1717
    },
    {
      "epoch": 0.878323108384458,
      "grad_norm": 3.0085763931274414,
      "learning_rate": 4.0027291942419055e-06,
      "loss": 0.7805,
      "step": 1718
    },
    {
      "epoch": 0.8788343558282209,
      "grad_norm": 3.2898752689361572,
      "learning_rate": 3.969614869778354e-06,
      "loss": 0.8877,
      "step": 1719
    },
    {
      "epoch": 0.8793456032719836,
      "grad_norm": 3.1025335788726807,
      "learning_rate": 3.936632428159609e-06,
      "loss": 0.808,
      "step": 1720
    },
    {
      "epoch": 0.8798568507157464,
      "grad_norm": 3.2930991649627686,
      "learning_rate": 3.903781963884467e-06,
      "loss": 0.8736,
      "step": 1721
    },
    {
      "epoch": 0.8803680981595092,
      "grad_norm": 3.3883445262908936,
      "learning_rate": 3.871063571073668e-06,
      "loss": 0.9969,
      "step": 1722
    },
    {
      "epoch": 0.880879345603272,
      "grad_norm": 3.6220452785491943,
      "learning_rate": 3.838477343469516e-06,
      "loss": 0.88,
      "step": 1723
    },
    {
      "epoch": 0.8813905930470347,
      "grad_norm": 3.937267303466797,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.9832,
      "step": 1724
    },
    {
      "epoch": 0.8819018404907976,
      "grad_norm": 3.5126397609710693,
      "learning_rate": 3.77370175695681e-06,
      "loss": 0.8513,
      "step": 1725
    },
    {
      "epoch": 0.8824130879345603,
      "grad_norm": 3.5317177772521973,
      "learning_rate": 3.74151258363844e-06,
      "loss": 0.8387,
      "step": 1726
    },
    {
      "epoch": 0.8829243353783232,
      "grad_norm": 3.232076406478882,
      "learning_rate": 3.7094559467066083e-06,
      "loss": 0.8725,
      "step": 1727
    },
    {
      "epoch": 0.8834355828220859,
      "grad_norm": 3.6095595359802246,
      "learning_rate": 3.6775319380076e-06,
      "loss": 0.9068,
      "step": 1728
    },
    {
      "epoch": 0.8839468302658486,
      "grad_norm": 3.2999720573425293,
      "learning_rate": 3.645740649007734e-06,
      "loss": 0.806,
      "step": 1729
    },
    {
      "epoch": 0.8844580777096115,
      "grad_norm": 3.733455181121826,
      "learning_rate": 3.614082170793021e-06,
      "loss": 0.8415,
      "step": 1730
    },
    {
      "epoch": 0.8849693251533742,
      "grad_norm": 3.5478620529174805,
      "learning_rate": 3.5825565940690087e-06,
      "loss": 0.8471,
      "step": 1731
    },
    {
      "epoch": 0.885480572597137,
      "grad_norm": 3.889519214630127,
      "learning_rate": 3.551164009160429e-06,
      "loss": 0.8128,
      "step": 1732
    },
    {
      "epoch": 0.8859918200408998,
      "grad_norm": 3.7842485904693604,
      "learning_rate": 3.5199045060110013e-06,
      "loss": 0.9556,
      "step": 1733
    },
    {
      "epoch": 0.8865030674846626,
      "grad_norm": 3.5816397666931152,
      "learning_rate": 3.488778174183116e-06,
      "loss": 0.8108,
      "step": 1734
    },
    {
      "epoch": 0.8870143149284253,
      "grad_norm": 3.812117338180542,
      "learning_rate": 3.4577851028576523e-06,
      "loss": 0.7997,
      "step": 1735
    },
    {
      "epoch": 0.8875255623721882,
      "grad_norm": 3.7888643741607666,
      "learning_rate": 3.4269253808336455e-06,
      "loss": 0.8451,
      "step": 1736
    },
    {
      "epoch": 0.8880368098159509,
      "grad_norm": 3.8936595916748047,
      "learning_rate": 3.3961990965280745e-06,
      "loss": 0.8154,
      "step": 1737
    },
    {
      "epoch": 0.8885480572597138,
      "grad_norm": 4.134406566619873,
      "learning_rate": 3.36560633797563e-06,
      "loss": 0.8866,
      "step": 1738
    },
    {
      "epoch": 0.8890593047034765,
      "grad_norm": 4.125121116638184,
      "learning_rate": 3.335147192828403e-06,
      "loss": 0.7727,
      "step": 1739
    },
    {
      "epoch": 0.8895705521472392,
      "grad_norm": 4.411681175231934,
      "learning_rate": 3.3048217483556744e-06,
      "loss": 0.8563,
      "step": 1740
    },
    {
      "epoch": 0.8900817995910021,
      "grad_norm": 3.7549989223480225,
      "learning_rate": 3.2746300914436534e-06,
      "loss": 0.7357,
      "step": 1741
    },
    {
      "epoch": 0.8905930470347648,
      "grad_norm": 4.538980960845947,
      "learning_rate": 3.2445723085952504e-06,
      "loss": 0.818,
      "step": 1742
    },
    {
      "epoch": 0.8911042944785276,
      "grad_norm": 4.389502048492432,
      "learning_rate": 3.214648485929783e-06,
      "loss": 0.8799,
      "step": 1743
    },
    {
      "epoch": 0.8916155419222904,
      "grad_norm": 4.548320293426514,
      "learning_rate": 3.184858709182775e-06,
      "loss": 0.6832,
      "step": 1744
    },
    {
      "epoch": 0.8921267893660532,
      "grad_norm": 4.614025115966797,
      "learning_rate": 3.1552030637056806e-06,
      "loss": 0.8305,
      "step": 1745
    },
    {
      "epoch": 0.8926380368098159,
      "grad_norm": 4.867095470428467,
      "learning_rate": 3.1256816344656602e-06,
      "loss": 0.8145,
      "step": 1746
    },
    {
      "epoch": 0.8931492842535788,
      "grad_norm": 4.788604259490967,
      "learning_rate": 3.096294506045311e-06,
      "loss": 0.8077,
      "step": 1747
    },
    {
      "epoch": 0.8936605316973415,
      "grad_norm": 5.478014945983887,
      "learning_rate": 3.067041762642475e-06,
      "loss": 0.771,
      "step": 1748
    },
    {
      "epoch": 0.8941717791411042,
      "grad_norm": 6.572366237640381,
      "learning_rate": 3.037923488069927e-06,
      "loss": 0.7604,
      "step": 1749
    },
    {
      "epoch": 0.8946830265848671,
      "grad_norm": 5.753463268280029,
      "learning_rate": 3.0089397657551865e-06,
      "loss": 0.4072,
      "step": 1750
    },
    {
      "epoch": 0.8951942740286298,
      "grad_norm": 1.9616332054138184,
      "learning_rate": 2.9800906787402716e-06,
      "loss": 0.9257,
      "step": 1751
    },
    {
      "epoch": 0.8957055214723927,
      "grad_norm": 2.392256259918213,
      "learning_rate": 2.9513763096814305e-06,
      "loss": 1.0013,
      "step": 1752
    },
    {
      "epoch": 0.8962167689161554,
      "grad_norm": 2.3039324283599854,
      "learning_rate": 2.9227967408489653e-06,
      "loss": 0.9961,
      "step": 1753
    },
    {
      "epoch": 0.8967280163599182,
      "grad_norm": 2.534642457962036,
      "learning_rate": 2.89435205412692e-06,
      "loss": 0.9289,
      "step": 1754
    },
    {
      "epoch": 0.897239263803681,
      "grad_norm": 2.941516160964966,
      "learning_rate": 2.8660423310129135e-06,
      "loss": 0.957,
      "step": 1755
    },
    {
      "epoch": 0.8977505112474438,
      "grad_norm": 2.806290626525879,
      "learning_rate": 2.8378676526178482e-06,
      "loss": 1.0622,
      "step": 1756
    },
    {
      "epoch": 0.8982617586912065,
      "grad_norm": 2.6641461849212646,
      "learning_rate": 2.8098280996657456e-06,
      "loss": 0.9309,
      "step": 1757
    },
    {
      "epoch": 0.8987730061349694,
      "grad_norm": 2.6802642345428467,
      "learning_rate": 2.781923752493437e-06,
      "loss": 0.8945,
      "step": 1758
    },
    {
      "epoch": 0.8992842535787321,
      "grad_norm": 2.851447820663452,
      "learning_rate": 2.754154691050387e-06,
      "loss": 0.8722,
      "step": 1759
    },
    {
      "epoch": 0.8997955010224948,
      "grad_norm": 2.958563804626465,
      "learning_rate": 2.7265209948984514e-06,
      "loss": 0.8529,
      "step": 1760
    },
    {
      "epoch": 0.9003067484662577,
      "grad_norm": 2.867089033126831,
      "learning_rate": 2.6990227432116544e-06,
      "loss": 0.8957,
      "step": 1761
    },
    {
      "epoch": 0.9008179959100204,
      "grad_norm": 3.3198704719543457,
      "learning_rate": 2.671660014775934e-06,
      "loss": 0.8905,
      "step": 1762
    },
    {
      "epoch": 0.9013292433537833,
      "grad_norm": 3.0108888149261475,
      "learning_rate": 2.6444328879889622e-06,
      "loss": 0.8434,
      "step": 1763
    },
    {
      "epoch": 0.901840490797546,
      "grad_norm": 2.9155540466308594,
      "learning_rate": 2.6173414408598827e-06,
      "loss": 0.8414,
      "step": 1764
    },
    {
      "epoch": 0.9023517382413088,
      "grad_norm": 3.178889036178589,
      "learning_rate": 2.5903857510090835e-06,
      "loss": 0.9461,
      "step": 1765
    },
    {
      "epoch": 0.9028629856850716,
      "grad_norm": 3.119640588760376,
      "learning_rate": 2.56356589566803e-06,
      "loss": 0.8996,
      "step": 1766
    },
    {
      "epoch": 0.9033742331288344,
      "grad_norm": 3.1831552982330322,
      "learning_rate": 2.53688195167896e-06,
      "loss": 0.9139,
      "step": 1767
    },
    {
      "epoch": 0.9038854805725971,
      "grad_norm": 3.3082258701324463,
      "learning_rate": 2.5103339954947626e-06,
      "loss": 0.8465,
      "step": 1768
    },
    {
      "epoch": 0.90439672801636,
      "grad_norm": 3.1402430534362793,
      "learning_rate": 2.483922103178632e-06,
      "loss": 0.9071,
      "step": 1769
    },
    {
      "epoch": 0.9049079754601227,
      "grad_norm": 3.3881165981292725,
      "learning_rate": 2.4576463504039913e-06,
      "loss": 0.9479,
      "step": 1770
    },
    {
      "epoch": 0.9054192229038854,
      "grad_norm": 3.3849008083343506,
      "learning_rate": 2.4315068124541597e-06,
      "loss": 0.8833,
      "step": 1771
    },
    {
      "epoch": 0.9059304703476483,
      "grad_norm": 3.725773334503174,
      "learning_rate": 2.4055035642222224e-06,
      "loss": 0.8946,
      "step": 1772
    },
    {
      "epoch": 0.906441717791411,
      "grad_norm": 4.276130676269531,
      "learning_rate": 2.3796366802107394e-06,
      "loss": 0.9649,
      "step": 1773
    },
    {
      "epoch": 0.9069529652351738,
      "grad_norm": 3.543367862701416,
      "learning_rate": 2.3539062345316e-06,
      "loss": 0.8461,
      "step": 1774
    },
    {
      "epoch": 0.9074642126789366,
      "grad_norm": 3.076871156692505,
      "learning_rate": 2.3283123009057607e-06,
      "loss": 0.8243,
      "step": 1775
    },
    {
      "epoch": 0.9079754601226994,
      "grad_norm": 3.6242263317108154,
      "learning_rate": 2.3028549526630583e-06,
      "loss": 0.8345,
      "step": 1776
    },
    {
      "epoch": 0.9084867075664622,
      "grad_norm": 3.5291264057159424,
      "learning_rate": 2.277534262742015e-06,
      "loss": 0.892,
      "step": 1777
    },
    {
      "epoch": 0.908997955010225,
      "grad_norm": 3.4097137451171875,
      "learning_rate": 2.2523503036895764e-06,
      "loss": 0.8539,
      "step": 1778
    },
    {
      "epoch": 0.9095092024539877,
      "grad_norm": 3.7175967693328857,
      "learning_rate": 2.227303147660964e-06,
      "loss": 0.9083,
      "step": 1779
    },
    {
      "epoch": 0.9100204498977505,
      "grad_norm": 3.9345943927764893,
      "learning_rate": 2.202392866419423e-06,
      "loss": 0.9569,
      "step": 1780
    },
    {
      "epoch": 0.9105316973415133,
      "grad_norm": 3.6897552013397217,
      "learning_rate": 2.1776195313360505e-06,
      "loss": 0.8444,
      "step": 1781
    },
    {
      "epoch": 0.911042944785276,
      "grad_norm": 3.9733729362487793,
      "learning_rate": 2.152983213389559e-06,
      "loss": 0.9426,
      "step": 1782
    },
    {
      "epoch": 0.9115541922290389,
      "grad_norm": 4.125808238983154,
      "learning_rate": 2.1284839831661075e-06,
      "loss": 0.8886,
      "step": 1783
    },
    {
      "epoch": 0.9120654396728016,
      "grad_norm": 3.4890663623809814,
      "learning_rate": 2.1041219108590692e-06,
      "loss": 0.8138,
      "step": 1784
    },
    {
      "epoch": 0.9125766871165644,
      "grad_norm": 3.9719114303588867,
      "learning_rate": 2.0798970662688545e-06,
      "loss": 0.8747,
      "step": 1785
    },
    {
      "epoch": 0.9130879345603272,
      "grad_norm": 4.040510654449463,
      "learning_rate": 2.055809518802676e-06,
      "loss": 1.0,
      "step": 1786
    },
    {
      "epoch": 0.91359918200409,
      "grad_norm": 4.081876754760742,
      "learning_rate": 2.031859337474407e-06,
      "loss": 0.7761,
      "step": 1787
    },
    {
      "epoch": 0.9141104294478528,
      "grad_norm": 4.227837562561035,
      "learning_rate": 2.0080465909043113e-06,
      "loss": 0.8716,
      "step": 1788
    },
    {
      "epoch": 0.9146216768916156,
      "grad_norm": 4.269420623779297,
      "learning_rate": 1.984371347318914e-06,
      "loss": 0.8066,
      "step": 1789
    },
    {
      "epoch": 0.9151329243353783,
      "grad_norm": 4.166466236114502,
      "learning_rate": 1.9608336745507716e-06,
      "loss": 0.802,
      "step": 1790
    },
    {
      "epoch": 0.9156441717791411,
      "grad_norm": 4.287985801696777,
      "learning_rate": 1.937433640038261e-06,
      "loss": 0.7958,
      "step": 1791
    },
    {
      "epoch": 0.9161554192229039,
      "grad_norm": 4.09846830368042,
      "learning_rate": 1.914171310825441e-06,
      "loss": 0.7164,
      "step": 1792
    },
    {
      "epoch": 0.9166666666666666,
      "grad_norm": 4.546944618225098,
      "learning_rate": 1.8910467535617983e-06,
      "loss": 0.7325,
      "step": 1793
    },
    {
      "epoch": 0.9171779141104295,
      "grad_norm": 4.4474897384643555,
      "learning_rate": 1.8680600345021171e-06,
      "loss": 0.7291,
      "step": 1794
    },
    {
      "epoch": 0.9176891615541922,
      "grad_norm": 4.722184658050537,
      "learning_rate": 1.845211219506221e-06,
      "loss": 0.6715,
      "step": 1795
    },
    {
      "epoch": 0.918200408997955,
      "grad_norm": 5.06058931350708,
      "learning_rate": 1.8225003740388547e-06,
      "loss": 0.9135,
      "step": 1796
    },
    {
      "epoch": 0.9187116564417178,
      "grad_norm": 5.043420791625977,
      "learning_rate": 1.79992756316944e-06,
      "loss": 0.7258,
      "step": 1797
    },
    {
      "epoch": 0.9192229038854806,
      "grad_norm": 5.176082134246826,
      "learning_rate": 1.7774928515719157e-06,
      "loss": 0.7435,
      "step": 1798
    },
    {
      "epoch": 0.9197341513292433,
      "grad_norm": 6.064294815063477,
      "learning_rate": 1.7551963035245588e-06,
      "loss": 0.7142,
      "step": 1799
    },
    {
      "epoch": 0.9202453987730062,
      "grad_norm": 6.77680778503418,
      "learning_rate": 1.733037982909791e-06,
      "loss": 0.4713,
      "step": 1800
    },
    {
      "epoch": 0.9207566462167689,
      "grad_norm": 1.9696067571640015,
      "learning_rate": 1.7110179532139781e-06,
      "loss": 0.8423,
      "step": 1801
    },
    {
      "epoch": 0.9212678936605317,
      "grad_norm": 2.4090499877929688,
      "learning_rate": 1.6891362775272812e-06,
      "loss": 0.9596,
      "step": 1802
    },
    {
      "epoch": 0.9217791411042945,
      "grad_norm": 2.441483974456787,
      "learning_rate": 1.6673930185434561e-06,
      "loss": 0.8638,
      "step": 1803
    },
    {
      "epoch": 0.9222903885480572,
      "grad_norm": 2.631037712097168,
      "learning_rate": 1.6457882385596646e-06,
      "loss": 1.0107,
      "step": 1804
    },
    {
      "epoch": 0.9228016359918201,
      "grad_norm": 2.69331431388855,
      "learning_rate": 1.6243219994763304e-06,
      "loss": 0.9641,
      "step": 1805
    },
    {
      "epoch": 0.9233128834355828,
      "grad_norm": 2.4592180252075195,
      "learning_rate": 1.6029943627969223e-06,
      "loss": 0.8723,
      "step": 1806
    },
    {
      "epoch": 0.9238241308793456,
      "grad_norm": 2.729130744934082,
      "learning_rate": 1.5818053896278162e-06,
      "loss": 0.9765,
      "step": 1807
    },
    {
      "epoch": 0.9243353783231084,
      "grad_norm": 2.6787896156311035,
      "learning_rate": 1.5607551406780717e-06,
      "loss": 0.8601,
      "step": 1808
    },
    {
      "epoch": 0.9248466257668712,
      "grad_norm": 2.9896671772003174,
      "learning_rate": 1.5398436762593061e-06,
      "loss": 0.8525,
      "step": 1809
    },
    {
      "epoch": 0.9253578732106339,
      "grad_norm": 3.0496749877929688,
      "learning_rate": 1.519071056285487e-06,
      "loss": 0.9679,
      "step": 1810
    },
    {
      "epoch": 0.9258691206543967,
      "grad_norm": 2.976200819015503,
      "learning_rate": 1.4984373402728014e-06,
      "loss": 0.8789,
      "step": 1811
    },
    {
      "epoch": 0.9263803680981595,
      "grad_norm": 3.0850539207458496,
      "learning_rate": 1.4779425873394259e-06,
      "loss": 0.85,
      "step": 1812
    },
    {
      "epoch": 0.9268916155419223,
      "grad_norm": 3.117995023727417,
      "learning_rate": 1.4575868562054228e-06,
      "loss": 0.8234,
      "step": 1813
    },
    {
      "epoch": 0.9274028629856851,
      "grad_norm": 2.9029812812805176,
      "learning_rate": 1.4373702051925065e-06,
      "loss": 0.9211,
      "step": 1814
    },
    {
      "epoch": 0.9279141104294478,
      "grad_norm": 3.243557929992676,
      "learning_rate": 1.4172926922239315e-06,
      "loss": 0.8936,
      "step": 1815
    },
    {
      "epoch": 0.9284253578732107,
      "grad_norm": 2.932993173599243,
      "learning_rate": 1.3973543748243e-06,
      "loss": 0.8516,
      "step": 1816
    },
    {
      "epoch": 0.9289366053169734,
      "grad_norm": 3.073636054992676,
      "learning_rate": 1.377555310119405e-06,
      "loss": 0.8922,
      "step": 1817
    },
    {
      "epoch": 0.9294478527607362,
      "grad_norm": 3.3185694217681885,
      "learning_rate": 1.3578955548360473e-06,
      "loss": 0.8148,
      "step": 1818
    },
    {
      "epoch": 0.929959100204499,
      "grad_norm": 3.1685848236083984,
      "learning_rate": 1.3383751653019029e-06,
      "loss": 0.9387,
      "step": 1819
    },
    {
      "epoch": 0.9304703476482618,
      "grad_norm": 3.10244083404541,
      "learning_rate": 1.31899419744535e-06,
      "loss": 0.7719,
      "step": 1820
    },
    {
      "epoch": 0.9309815950920245,
      "grad_norm": 3.1995668411254883,
      "learning_rate": 1.2997527067952875e-06,
      "loss": 0.9624,
      "step": 1821
    },
    {
      "epoch": 0.9314928425357873,
      "grad_norm": 3.6962995529174805,
      "learning_rate": 1.2806507484810215e-06,
      "loss": 0.8737,
      "step": 1822
    },
    {
      "epoch": 0.9320040899795501,
      "grad_norm": 3.9050567150115967,
      "learning_rate": 1.2616883772320508e-06,
      "loss": 1.0082,
      "step": 1823
    },
    {
      "epoch": 0.9325153374233128,
      "grad_norm": 3.417041540145874,
      "learning_rate": 1.2428656473779721e-06,
      "loss": 0.8719,
      "step": 1824
    },
    {
      "epoch": 0.9330265848670757,
      "grad_norm": 3.355666160583496,
      "learning_rate": 1.2241826128482625e-06,
      "loss": 0.9281,
      "step": 1825
    },
    {
      "epoch": 0.9335378323108384,
      "grad_norm": 3.6375527381896973,
      "learning_rate": 1.20563932717217e-06,
      "loss": 0.8613,
      "step": 1826
    },
    {
      "epoch": 0.9340490797546013,
      "grad_norm": 3.6801555156707764,
      "learning_rate": 1.1872358434785346e-06,
      "loss": 0.7958,
      "step": 1827
    },
    {
      "epoch": 0.934560327198364,
      "grad_norm": 3.775987148284912,
      "learning_rate": 1.1689722144956671e-06,
      "loss": 0.8864,
      "step": 1828
    },
    {
      "epoch": 0.9350715746421268,
      "grad_norm": 3.3887085914611816,
      "learning_rate": 1.1508484925511542e-06,
      "loss": 0.8428,
      "step": 1829
    },
    {
      "epoch": 0.9355828220858896,
      "grad_norm": 3.701101064682007,
      "learning_rate": 1.132864729571731e-06,
      "loss": 0.8598,
      "step": 1830
    },
    {
      "epoch": 0.9360940695296524,
      "grad_norm": 3.671170711517334,
      "learning_rate": 1.1150209770831588e-06,
      "loss": 0.8273,
      "step": 1831
    },
    {
      "epoch": 0.9366053169734151,
      "grad_norm": 3.679232120513916,
      "learning_rate": 1.0973172862100145e-06,
      "loss": 0.899,
      "step": 1832
    },
    {
      "epoch": 0.9371165644171779,
      "grad_norm": 3.9870572090148926,
      "learning_rate": 1.0797537076756127e-06,
      "loss": 0.9478,
      "step": 1833
    },
    {
      "epoch": 0.9376278118609407,
      "grad_norm": 3.767942190170288,
      "learning_rate": 1.0623302918018108e-06,
      "loss": 0.8134,
      "step": 1834
    },
    {
      "epoch": 0.9381390593047034,
      "grad_norm": 4.18915319442749,
      "learning_rate": 1.0450470885088937e-06,
      "loss": 0.9433,
      "step": 1835
    },
    {
      "epoch": 0.9386503067484663,
      "grad_norm": 4.106919765472412,
      "learning_rate": 1.0279041473154116e-06,
      "loss": 0.8786,
      "step": 1836
    },
    {
      "epoch": 0.939161554192229,
      "grad_norm": 4.100769519805908,
      "learning_rate": 1.010901517338042e-06,
      "loss": 0.8459,
      "step": 1837
    },
    {
      "epoch": 0.9396728016359919,
      "grad_norm": 3.8171629905700684,
      "learning_rate": 9.94039247291456e-07,
      "loss": 0.7617,
      "step": 1838
    },
    {
      "epoch": 0.9401840490797546,
      "grad_norm": 4.406624794006348,
      "learning_rate": 9.773173854881913e-07,
      "loss": 0.7923,
      "step": 1839
    },
    {
      "epoch": 0.9406952965235174,
      "grad_norm": 4.20837926864624,
      "learning_rate": 9.607359798384785e-07,
      "loss": 0.9773,
      "step": 1840
    },
    {
      "epoch": 0.9412065439672802,
      "grad_norm": 4.484842777252197,
      "learning_rate": 9.442950778501325e-07,
      "loss": 0.8711,
      "step": 1841
    },
    {
      "epoch": 0.941717791411043,
      "grad_norm": 4.719925403594971,
      "learning_rate": 9.279947266284061e-07,
      "loss": 0.8392,
      "step": 1842
    },
    {
      "epoch": 0.9422290388548057,
      "grad_norm": 4.476531982421875,
      "learning_rate": 9.118349728758468e-07,
      "loss": 0.767,
      "step": 1843
    },
    {
      "epoch": 0.9427402862985685,
      "grad_norm": 4.48267126083374,
      "learning_rate": 8.958158628922019e-07,
      "loss": 0.6777,
      "step": 1844
    },
    {
      "epoch": 0.9432515337423313,
      "grad_norm": 4.858860015869141,
      "learning_rate": 8.799374425742246e-07,
      "loss": 0.7469,
      "step": 1845
    },
    {
      "epoch": 0.943762781186094,
      "grad_norm": 4.577967166900635,
      "learning_rate": 8.641997574155846e-07,
      "loss": 0.676,
      "step": 1846
    },
    {
      "epoch": 0.9442740286298569,
      "grad_norm": 5.150705337524414,
      "learning_rate": 8.486028525067358e-07,
      "loss": 0.7106,
      "step": 1847
    },
    {
      "epoch": 0.9447852760736196,
      "grad_norm": 4.760867595672607,
      "learning_rate": 8.331467725347708e-07,
      "loss": 0.7528,
      "step": 1848
    },
    {
      "epoch": 0.9452965235173824,
      "grad_norm": 6.085768222808838,
      "learning_rate": 8.178315617832999e-07,
      "loss": 0.784,
      "step": 1849
    },
    {
      "epoch": 0.9458077709611452,
      "grad_norm": 6.886128902435303,
      "learning_rate": 8.026572641323393e-07,
      "loss": 0.5051,
      "step": 1850
    },
    {
      "epoch": 0.946319018404908,
      "grad_norm": 2.127772808074951,
      "learning_rate": 7.876239230581506e-07,
      "loss": 0.9763,
      "step": 1851
    },
    {
      "epoch": 0.9468302658486708,
      "grad_norm": 2.3633391857147217,
      "learning_rate": 7.727315816331515e-07,
      "loss": 1.012,
      "step": 1852
    },
    {
      "epoch": 0.9473415132924335,
      "grad_norm": 2.347792625427246,
      "learning_rate": 7.579802825257775e-07,
      "loss": 0.9064,
      "step": 1853
    },
    {
      "epoch": 0.9478527607361963,
      "grad_norm": 2.4126203060150146,
      "learning_rate": 7.43370068000343e-07,
      "loss": 0.8778,
      "step": 1854
    },
    {
      "epoch": 0.9483640081799591,
      "grad_norm": 2.6466588973999023,
      "learning_rate": 7.289009799169688e-07,
      "loss": 0.9782,
      "step": 1855
    },
    {
      "epoch": 0.9488752556237219,
      "grad_norm": 2.6971752643585205,
      "learning_rate": 7.145730597314049e-07,
      "loss": 0.979,
      "step": 1856
    },
    {
      "epoch": 0.9493865030674846,
      "grad_norm": 2.6623589992523193,
      "learning_rate": 7.003863484949413e-07,
      "loss": 0.9151,
      "step": 1857
    },
    {
      "epoch": 0.9498977505112475,
      "grad_norm": 2.865541696548462,
      "learning_rate": 6.86340886854292e-07,
      "loss": 0.8949,
      "step": 1858
    },
    {
      "epoch": 0.9504089979550102,
      "grad_norm": 3.0056710243225098,
      "learning_rate": 6.724367150514777e-07,
      "loss": 0.8936,
      "step": 1859
    },
    {
      "epoch": 0.950920245398773,
      "grad_norm": 2.805217742919922,
      "learning_rate": 6.58673872923693e-07,
      "loss": 0.831,
      "step": 1860
    },
    {
      "epoch": 0.9514314928425358,
      "grad_norm": 3.1351966857910156,
      "learning_rate": 6.450523999032177e-07,
      "loss": 1.0224,
      "step": 1861
    },
    {
      "epoch": 0.9519427402862985,
      "grad_norm": 2.93005633354187,
      "learning_rate": 6.315723350172775e-07,
      "loss": 0.8546,
      "step": 1862
    },
    {
      "epoch": 0.9524539877300614,
      "grad_norm": 3.181835651397705,
      "learning_rate": 6.182337168879671e-07,
      "loss": 0.903,
      "step": 1863
    },
    {
      "epoch": 0.9529652351738241,
      "grad_norm": 3.401155948638916,
      "learning_rate": 6.050365837320992e-07,
      "loss": 0.9181,
      "step": 1864
    },
    {
      "epoch": 0.9534764826175869,
      "grad_norm": 3.0578250885009766,
      "learning_rate": 5.919809733611171e-07,
      "loss": 0.8978,
      "step": 1865
    },
    {
      "epoch": 0.9539877300613497,
      "grad_norm": 3.142280101776123,
      "learning_rate": 5.790669231809875e-07,
      "loss": 0.9351,
      "step": 1866
    },
    {
      "epoch": 0.9544989775051125,
      "grad_norm": 3.1850273609161377,
      "learning_rate": 5.66294470192097e-07,
      "loss": 0.9621,
      "step": 1867
    },
    {
      "epoch": 0.9550102249488752,
      "grad_norm": 3.274036169052124,
      "learning_rate": 5.536636509891225e-07,
      "loss": 0.8428,
      "step": 1868
    },
    {
      "epoch": 0.9555214723926381,
      "grad_norm": 3.323293924331665,
      "learning_rate": 5.411745017609493e-07,
      "loss": 0.9623,
      "step": 1869
    },
    {
      "epoch": 0.9560327198364008,
      "grad_norm": 3.296380043029785,
      "learning_rate": 5.288270582905708e-07,
      "loss": 0.86,
      "step": 1870
    },
    {
      "epoch": 0.9565439672801636,
      "grad_norm": 3.7464077472686768,
      "learning_rate": 5.166213559549549e-07,
      "loss": 0.8665,
      "step": 1871
    },
    {
      "epoch": 0.9570552147239264,
      "grad_norm": 3.5171990394592285,
      "learning_rate": 5.045574297249833e-07,
      "loss": 0.9063,
      "step": 1872
    },
    {
      "epoch": 0.9575664621676891,
      "grad_norm": 3.684037685394287,
      "learning_rate": 4.926353141653184e-07,
      "loss": 0.9079,
      "step": 1873
    },
    {
      "epoch": 0.9580777096114519,
      "grad_norm": 3.424814462661743,
      "learning_rate": 4.80855043434325e-07,
      "loss": 0.8508,
      "step": 1874
    },
    {
      "epoch": 0.9585889570552147,
      "grad_norm": 3.3258156776428223,
      "learning_rate": 4.692166512839491e-07,
      "loss": 0.8615,
      "step": 1875
    },
    {
      "epoch": 0.9591002044989775,
      "grad_norm": 3.7672901153564453,
      "learning_rate": 4.577201710596612e-07,
      "loss": 0.9087,
      "step": 1876
    },
    {
      "epoch": 0.9596114519427403,
      "grad_norm": 3.638936996459961,
      "learning_rate": 4.4636563570031873e-07,
      "loss": 0.921,
      "step": 1877
    },
    {
      "epoch": 0.9601226993865031,
      "grad_norm": 3.7481777667999268,
      "learning_rate": 4.3515307773809855e-07,
      "loss": 0.8998,
      "step": 1878
    },
    {
      "epoch": 0.9606339468302658,
      "grad_norm": 4.0111212730407715,
      "learning_rate": 4.240825292983808e-07,
      "loss": 0.8379,
      "step": 1879
    },
    {
      "epoch": 0.9611451942740287,
      "grad_norm": 3.8063735961914062,
      "learning_rate": 4.131540220996877e-07,
      "loss": 1.0216,
      "step": 1880
    },
    {
      "epoch": 0.9616564417177914,
      "grad_norm": 3.818110942840576,
      "learning_rate": 4.023675874535671e-07,
      "loss": 0.9189,
      "step": 1881
    },
    {
      "epoch": 0.9621676891615542,
      "grad_norm": 3.307727098464966,
      "learning_rate": 3.917232562645035e-07,
      "loss": 0.7723,
      "step": 1882
    },
    {
      "epoch": 0.962678936605317,
      "grad_norm": 3.7070207595825195,
      "learning_rate": 3.812210590298515e-07,
      "loss": 0.8208,
      "step": 1883
    },
    {
      "epoch": 0.9631901840490797,
      "grad_norm": 3.999943971633911,
      "learning_rate": 3.7086102583972494e-07,
      "loss": 0.9011,
      "step": 1884
    },
    {
      "epoch": 0.9637014314928425,
      "grad_norm": 3.9105939865112305,
      "learning_rate": 3.6064318637693e-07,
      "loss": 0.9319,
      "step": 1885
    },
    {
      "epoch": 0.9642126789366053,
      "grad_norm": 3.4953696727752686,
      "learning_rate": 3.505675699168487e-07,
      "loss": 0.836,
      "step": 1886
    },
    {
      "epoch": 0.9647239263803681,
      "grad_norm": 4.094799995422363,
      "learning_rate": 3.406342053274003e-07,
      "loss": 0.7909,
      "step": 1887
    },
    {
      "epoch": 0.9652351738241309,
      "grad_norm": 3.863297700881958,
      "learning_rate": 3.3084312106892446e-07,
      "loss": 0.8657,
      "step": 1888
    },
    {
      "epoch": 0.9657464212678937,
      "grad_norm": 4.07185173034668,
      "learning_rate": 3.211943451941035e-07,
      "loss": 0.8233,
      "step": 1889
    },
    {
      "epoch": 0.9662576687116564,
      "grad_norm": 4.35256814956665,
      "learning_rate": 3.1168790534789605e-07,
      "loss": 0.9713,
      "step": 1890
    },
    {
      "epoch": 0.9667689161554193,
      "grad_norm": 3.9577109813690186,
      "learning_rate": 3.023238287674479e-07,
      "loss": 0.8875,
      "step": 1891
    },
    {
      "epoch": 0.967280163599182,
      "grad_norm": 4.198614597320557,
      "learning_rate": 2.9310214228202013e-07,
      "loss": 0.7918,
      "step": 1892
    },
    {
      "epoch": 0.9677914110429447,
      "grad_norm": 4.020992279052734,
      "learning_rate": 2.840228723129001e-07,
      "loss": 0.7969,
      "step": 1893
    },
    {
      "epoch": 0.9683026584867076,
      "grad_norm": 4.857772350311279,
      "learning_rate": 2.750860448733461e-07,
      "loss": 0.7989,
      "step": 1894
    },
    {
      "epoch": 0.9688139059304703,
      "grad_norm": 4.296475887298584,
      "learning_rate": 2.662916855684816e-07,
      "loss": 0.7896,
      "step": 1895
    },
    {
      "epoch": 0.9693251533742331,
      "grad_norm": 4.845669269561768,
      "learning_rate": 2.5763981959526786e-07,
      "loss": 0.9473,
      "step": 1896
    },
    {
      "epoch": 0.9698364008179959,
      "grad_norm": 5.276214599609375,
      "learning_rate": 2.4913047174237035e-07,
      "loss": 0.8389,
      "step": 1897
    },
    {
      "epoch": 0.9703476482617587,
      "grad_norm": 5.026844501495361,
      "learning_rate": 2.407636663901591e-07,
      "loss": 0.6934,
      "step": 1898
    },
    {
      "epoch": 0.9708588957055214,
      "grad_norm": 5.138829231262207,
      "learning_rate": 2.3253942751056968e-07,
      "loss": 0.5662,
      "step": 1899
    },
    {
      "epoch": 0.9713701431492843,
      "grad_norm": 8.76285457611084,
      "learning_rate": 2.2445777866709205e-07,
      "loss": 0.8929,
      "step": 1900
    },
    {
      "epoch": 0.971881390593047,
      "grad_norm": 2.1116087436676025,
      "learning_rate": 2.1651874301465979e-07,
      "loss": 0.8929,
      "step": 1901
    },
    {
      "epoch": 0.9723926380368099,
      "grad_norm": 2.1844208240509033,
      "learning_rate": 2.087223432996166e-07,
      "loss": 0.8937,
      "step": 1902
    },
    {
      "epoch": 0.9729038854805726,
      "grad_norm": 2.286012887954712,
      "learning_rate": 2.0106860185962194e-07,
      "loss": 0.8114,
      "step": 1903
    },
    {
      "epoch": 0.9734151329243353,
      "grad_norm": 2.505859613418579,
      "learning_rate": 1.935575406236123e-07,
      "loss": 0.9787,
      "step": 1904
    },
    {
      "epoch": 0.9739263803680982,
      "grad_norm": 2.708432674407959,
      "learning_rate": 1.861891811117178e-07,
      "loss": 0.9853,
      "step": 1905
    },
    {
      "epoch": 0.9744376278118609,
      "grad_norm": 2.945042610168457,
      "learning_rate": 1.7896354443521778e-07,
      "loss": 0.9986,
      "step": 1906
    },
    {
      "epoch": 0.9749488752556237,
      "grad_norm": 2.5554847717285156,
      "learning_rate": 1.7188065129647435e-07,
      "loss": 0.9626,
      "step": 1907
    },
    {
      "epoch": 0.9754601226993865,
      "grad_norm": 2.838167190551758,
      "learning_rate": 1.6494052198886555e-07,
      "loss": 0.9229,
      "step": 1908
    },
    {
      "epoch": 0.9759713701431493,
      "grad_norm": 2.6479218006134033,
      "learning_rate": 1.5814317639673005e-07,
      "loss": 0.8479,
      "step": 1909
    },
    {
      "epoch": 0.976482617586912,
      "grad_norm": 3.0106728076934814,
      "learning_rate": 1.5148863399532254e-07,
      "loss": 0.8836,
      "step": 1910
    },
    {
      "epoch": 0.9769938650306749,
      "grad_norm": 3.02433705329895,
      "learning_rate": 1.4497691385074175e-07,
      "loss": 0.9304,
      "step": 1911
    },
    {
      "epoch": 0.9775051124744376,
      "grad_norm": 3.085106134414673,
      "learning_rate": 1.3860803461989146e-07,
      "loss": 0.8676,
      "step": 1912
    },
    {
      "epoch": 0.9780163599182005,
      "grad_norm": 2.835423469543457,
      "learning_rate": 1.3238201455040844e-07,
      "loss": 0.8435,
      "step": 1913
    },
    {
      "epoch": 0.9785276073619632,
      "grad_norm": 3.2276833057403564,
      "learning_rate": 1.2629887148061792e-07,
      "loss": 0.9043,
      "step": 1914
    },
    {
      "epoch": 0.9790388548057259,
      "grad_norm": 3.260972261428833,
      "learning_rate": 1.203586228395004e-07,
      "loss": 0.9502,
      "step": 1915
    },
    {
      "epoch": 0.9795501022494888,
      "grad_norm": 3.3064229488372803,
      "learning_rate": 1.1456128564660273e-07,
      "loss": 0.9968,
      "step": 1916
    },
    {
      "epoch": 0.9800613496932515,
      "grad_norm": 3.274178981781006,
      "learning_rate": 1.0890687651203823e-07,
      "loss": 0.8302,
      "step": 1917
    },
    {
      "epoch": 0.9805725971370143,
      "grad_norm": 3.076536178588867,
      "learning_rate": 1.0339541163639776e-07,
      "loss": 0.9421,
      "step": 1918
    },
    {
      "epoch": 0.9810838445807771,
      "grad_norm": 3.247903823852539,
      "learning_rate": 9.802690681071647e-08,
      "loss": 0.9819,
      "step": 1919
    },
    {
      "epoch": 0.9815950920245399,
      "grad_norm": 3.3538260459899902,
      "learning_rate": 9.280137741643491e-08,
      "loss": 0.8744,
      "step": 1920
    },
    {
      "epoch": 0.9821063394683026,
      "grad_norm": 3.515782356262207,
      "learning_rate": 8.771883842536021e-08,
      "loss": 0.9124,
      "step": 1921
    },
    {
      "epoch": 0.9826175869120655,
      "grad_norm": 3.6226806640625,
      "learning_rate": 8.277930439959946e-08,
      "loss": 0.9011,
      "step": 1922
    },
    {
      "epoch": 0.9831288343558282,
      "grad_norm": 3.3394203186035156,
      "learning_rate": 7.798278949154303e-08,
      "loss": 0.8316,
      "step": 1923
    },
    {
      "epoch": 0.983640081799591,
      "grad_norm": 3.246371030807495,
      "learning_rate": 7.332930744380906e-08,
      "loss": 0.8556,
      "step": 1924
    },
    {
      "epoch": 0.9841513292433538,
      "grad_norm": 3.402927875518799,
      "learning_rate": 6.881887158920464e-08,
      "loss": 0.7978,
      "step": 1925
    },
    {
      "epoch": 0.9846625766871165,
      "grad_norm": 3.8112809658050537,
      "learning_rate": 6.445149485070357e-08,
      "loss": 0.9133,
      "step": 1926
    },
    {
      "epoch": 0.9851738241308794,
      "grad_norm": 3.5460119247436523,
      "learning_rate": 6.022718974137975e-08,
      "loss": 0.8158,
      "step": 1927
    },
    {
      "epoch": 0.9856850715746421,
      "grad_norm": 3.341395854949951,
      "learning_rate": 5.614596836440722e-08,
      "loss": 0.8246,
      "step": 1928
    },
    {
      "epoch": 0.9861963190184049,
      "grad_norm": 3.6873090267181396,
      "learning_rate": 5.2207842412999034e-08,
      "loss": 0.8714,
      "step": 1929
    },
    {
      "epoch": 0.9867075664621677,
      "grad_norm": 3.4815688133239746,
      "learning_rate": 4.841282317037399e-08,
      "loss": 0.8948,
      "step": 1930
    },
    {
      "epoch": 0.9872188139059305,
      "grad_norm": 3.5316038131713867,
      "learning_rate": 4.476092150975109e-08,
      "loss": 0.8622,
      "step": 1931
    },
    {
      "epoch": 0.9877300613496932,
      "grad_norm": 3.5975794792175293,
      "learning_rate": 4.1252147894277336e-08,
      "loss": 0.881,
      "step": 1932
    },
    {
      "epoch": 0.9882413087934561,
      "grad_norm": 3.441171646118164,
      "learning_rate": 3.7886512377033334e-08,
      "loss": 0.8396,
      "step": 1933
    },
    {
      "epoch": 0.9887525562372188,
      "grad_norm": 3.8511383533477783,
      "learning_rate": 3.4664024600988835e-08,
      "loss": 0.9208,
      "step": 1934
    },
    {
      "epoch": 0.9892638036809815,
      "grad_norm": 3.8687822818756104,
      "learning_rate": 3.158469379898055e-08,
      "loss": 0.9135,
      "step": 1935
    },
    {
      "epoch": 0.9897750511247444,
      "grad_norm": 3.593276023864746,
      "learning_rate": 2.8648528793673302e-08,
      "loss": 0.8474,
      "step": 1936
    },
    {
      "epoch": 0.9902862985685071,
      "grad_norm": 4.0986127853393555,
      "learning_rate": 2.5855537997548917e-08,
      "loss": 0.8883,
      "step": 1937
    },
    {
      "epoch": 0.99079754601227,
      "grad_norm": 4.03285551071167,
      "learning_rate": 2.3205729412884016e-08,
      "loss": 0.7779,
      "step": 1938
    },
    {
      "epoch": 0.9913087934560327,
      "grad_norm": 4.346153736114502,
      "learning_rate": 2.0699110631711148e-08,
      "loss": 0.8757,
      "step": 1939
    },
    {
      "epoch": 0.9918200408997955,
      "grad_norm": 4.283609390258789,
      "learning_rate": 1.8335688835802167e-08,
      "loss": 0.8173,
      "step": 1940
    },
    {
      "epoch": 0.9923312883435583,
      "grad_norm": 4.301876068115234,
      "learning_rate": 1.6115470796662647e-08,
      "loss": 0.9134,
      "step": 1941
    },
    {
      "epoch": 0.9928425357873211,
      "grad_norm": 4.885223865509033,
      "learning_rate": 1.4038462875504143e-08,
      "loss": 0.8289,
      "step": 1942
    },
    {
      "epoch": 0.9933537832310838,
      "grad_norm": 4.63042688369751,
      "learning_rate": 1.2104671023199787e-08,
      "loss": 0.8625,
      "step": 1943
    },
    {
      "epoch": 0.9938650306748467,
      "grad_norm": 4.702084064483643,
      "learning_rate": 1.0314100780317581e-08,
      "loss": 0.9342,
      "step": 1944
    },
    {
      "epoch": 0.9943762781186094,
      "grad_norm": 4.4585771560668945,
      "learning_rate": 8.666757277064897e-09,
      "loss": 0.6828,
      "step": 1945
    },
    {
      "epoch": 0.9948875255623721,
      "grad_norm": 4.869369029998779,
      "learning_rate": 7.162645233282916e-09,
      "loss": 0.8505,
      "step": 1946
    },
    {
      "epoch": 0.995398773006135,
      "grad_norm": 4.623004913330078,
      "learning_rate": 5.8017689584521915e-09,
      "loss": 0.6772,
      "step": 1947
    },
    {
      "epoch": 0.9959100204498977,
      "grad_norm": 5.718740940093994,
      "learning_rate": 4.584132351642678e-09,
      "loss": 0.8251,
      "step": 1948
    },
    {
      "epoch": 0.9964212678936605,
      "grad_norm": 5.196649551391602,
      "learning_rate": 3.509738901547044e-09,
      "loss": 0.6039,
      "step": 1949
    },
    {
      "epoch": 0.9969325153374233,
      "grad_norm": 6.253082752227783,
      "learning_rate": 2.5785916864307092e-09,
      "loss": 0.5829,
      "step": 1950
    },
    {
      "epoch": 0.9974437627811861,
      "grad_norm": 2.393533229827881,
      "learning_rate": 1.7906933741484999e-09,
      "loss": 0.869,
      "step": 1951
    },
    {
      "epoch": 0.9979550102249489,
      "grad_norm": 2.994673728942871,
      "learning_rate": 1.1460462221279944e-09,
      "loss": 0.9103,
      "step": 1952
    },
    {
      "epoch": 0.9984662576687117,
      "grad_norm": 3.5846893787384033,
      "learning_rate": 6.446520773695231e-10,
      "loss": 0.8366,
      "step": 1953
    },
    {
      "epoch": 0.9989775051124744,
      "grad_norm": 3.5298497676849365,
      "learning_rate": 2.8651237642396414e-10,
      "loss": 0.7694,
      "step": 1954
    },
    {
      "epoch": 0.9994887525562373,
      "grad_norm": 3.686530828475952,
      "learning_rate": 7.162814541494811e-11,
      "loss": 0.67,
      "step": 1955
    },
    {
      "epoch": 1.0,
      "grad_norm": 5.693774223327637,
      "learning_rate": 0.0,
      "loss": 0.7087,
      "step": 1956
    }
  ],
  "logging_steps": 1,
  "max_steps": 1956,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 239,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.478583213010452e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}