{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9398359161349132,
  "eval_steps": 1.0,
  "global_step": 129,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 9.694531367351816,
      "learning_rate": 0.0,
      "loss": 0.9609,
      "step": 1
    },
    {
      "epoch": 0.05,
      "grad_norm": 6.0698578450559015,
      "learning_rate": 5e-06,
      "loss": 0.8693,
      "step": 2
    },
    {
      "epoch": 0.07,
      "grad_norm": 6.5557897846625535,
      "learning_rate": 7.924812503605782e-06,
      "loss": 0.8571,
      "step": 3
    },
    {
      "epoch": 0.09,
      "grad_norm": 5.8249565635426235,
      "learning_rate": 1e-05,
      "loss": 0.8098,
      "step": 4
    },
    {
      "epoch": 0.11,
      "grad_norm": 7.157097194857861,
      "learning_rate": 1e-05,
      "loss": 0.7421,
      "step": 5
    },
    {
      "epoch": 0.14,
      "grad_norm": 6.7490063693246185,
      "learning_rate": 1e-05,
      "loss": 0.7184,
      "step": 6
    },
    {
      "epoch": 0.16,
      "grad_norm": 6.989582963023535,
      "learning_rate": 1e-05,
      "loss": 0.7101,
      "step": 7
    },
    {
      "epoch": 0.18,
      "grad_norm": 4.856325406983797,
      "learning_rate": 1e-05,
      "loss": 0.6815,
      "step": 8
    },
    {
      "epoch": 0.21,
      "grad_norm": 4.893753189326447,
      "learning_rate": 1e-05,
      "loss": 0.6975,
      "step": 9
    },
    {
      "epoch": 0.23,
      "grad_norm": 4.8564394287744745,
      "learning_rate": 1e-05,
      "loss": 0.65,
      "step": 10
    },
    {
      "epoch": 0.25,
      "grad_norm": 4.638777805285931,
      "learning_rate": 1e-05,
      "loss": 0.5817,
      "step": 11
    },
    {
      "epoch": 0.27,
      "grad_norm": 3.701804268792858,
      "learning_rate": 1e-05,
      "loss": 0.6005,
      "step": 12
    },
    {
      "epoch": 0.3,
      "grad_norm": 3.08244950250501,
      "learning_rate": 1e-05,
      "loss": 0.5529,
      "step": 13
    },
    {
      "epoch": 0.32,
      "grad_norm": 3.210541622720776,
      "learning_rate": 1e-05,
      "loss": 0.5654,
      "step": 14
    },
    {
      "epoch": 0.34,
      "grad_norm": 3.0384959005478867,
      "learning_rate": 1e-05,
      "loss": 0.5184,
      "step": 15
    },
    {
      "epoch": 0.36,
      "grad_norm": 2.6850393439015092,
      "learning_rate": 1e-05,
      "loss": 0.5234,
      "step": 16
    },
    {
      "epoch": 0.39,
      "grad_norm": 2.8596154120661748,
      "learning_rate": 1e-05,
      "loss": 0.5005,
      "step": 17
    },
    {
      "epoch": 0.41,
      "grad_norm": 2.726391191331952,
      "learning_rate": 1e-05,
      "loss": 0.4691,
      "step": 18
    },
    {
      "epoch": 0.43,
      "grad_norm": 2.934188314387199,
      "learning_rate": 1e-05,
      "loss": 0.5265,
      "step": 19
    },
    {
      "epoch": 0.46,
      "grad_norm": 2.496046815968877,
      "learning_rate": 1e-05,
      "loss": 0.46,
      "step": 20
    },
    {
      "epoch": 0.48,
      "grad_norm": 2.6527493698931703,
      "learning_rate": 1e-05,
      "loss": 0.4633,
      "step": 21
    },
    {
      "epoch": 0.5,
      "grad_norm": 2.3371216207823364,
      "learning_rate": 1e-05,
      "loss": 0.4512,
      "step": 22
    },
    {
      "epoch": 0.52,
      "grad_norm": 2.5029789662415576,
      "learning_rate": 1e-05,
      "loss": 0.4538,
      "step": 23
    },
    {
      "epoch": 0.55,
      "grad_norm": 2.3654775885807435,
      "learning_rate": 1e-05,
      "loss": 0.4503,
      "step": 24
    },
    {
      "epoch": 0.57,
      "grad_norm": 2.520897440850751,
      "learning_rate": 1e-05,
      "loss": 0.4393,
      "step": 25
    },
    {
      "epoch": 0.59,
      "grad_norm": 2.2973148314047998,
      "learning_rate": 1e-05,
      "loss": 0.4246,
      "step": 26
    },
    {
      "epoch": 0.62,
      "grad_norm": 2.064310434148654,
      "learning_rate": 1e-05,
      "loss": 0.4466,
      "step": 27
    },
    {
      "epoch": 0.64,
      "grad_norm": 2.1518579274793614,
      "learning_rate": 1e-05,
      "loss": 0.4157,
      "step": 28
    },
    {
      "epoch": 0.66,
      "grad_norm": 2.101663348817292,
      "learning_rate": 1e-05,
      "loss": 0.4142,
      "step": 29
    },
    {
      "epoch": 0.68,
      "grad_norm": 2.0792549392682624,
      "learning_rate": 1e-05,
      "loss": 0.3814,
      "step": 30
    },
    {
      "epoch": 0.71,
      "grad_norm": 2.4317303715104868,
      "learning_rate": 1e-05,
      "loss": 0.4042,
      "step": 31
    },
    {
      "epoch": 0.73,
      "grad_norm": 2.6323607639867443,
      "learning_rate": 1e-05,
      "loss": 0.3745,
      "step": 32
    },
    {
      "epoch": 0.75,
      "grad_norm": 2.2663973989673987,
      "learning_rate": 1e-05,
      "loss": 0.3781,
      "step": 33
    },
    {
      "epoch": 0.77,
      "grad_norm": 2.2288967526596193,
      "learning_rate": 1e-05,
      "loss": 0.3894,
      "step": 34
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.2598649631650223,
      "learning_rate": 1e-05,
      "loss": 0.3734,
      "step": 35
    },
    {
      "epoch": 0.82,
      "grad_norm": 2.046710491098762,
      "learning_rate": 1e-05,
      "loss": 0.3436,
      "step": 36
    },
    {
      "epoch": 0.84,
      "grad_norm": 2.104660502205773,
      "learning_rate": 1e-05,
      "loss": 0.3636,
      "step": 37
    },
    {
      "epoch": 0.87,
      "grad_norm": 2.2194274529750335,
      "learning_rate": 1e-05,
      "loss": 0.3687,
      "step": 38
    },
    {
      "epoch": 0.89,
      "grad_norm": 2.380578087514372,
      "learning_rate": 1e-05,
      "loss": 0.3386,
      "step": 39
    },
    {
      "epoch": 0.91,
      "grad_norm": 2.1539059012744675,
      "learning_rate": 1e-05,
      "loss": 0.3663,
      "step": 40
    },
    {
      "epoch": 0.93,
      "grad_norm": 2.2975008924550884,
      "learning_rate": 1e-05,
      "loss": 0.3529,
      "step": 41
    },
    {
      "epoch": 0.96,
      "grad_norm": 2.0215009917813864,
      "learning_rate": 1e-05,
      "loss": 0.3255,
      "step": 42
    },
    {
      "epoch": 0.98,
      "grad_norm": 2.260279893370542,
      "learning_rate": 1e-05,
      "loss": 0.3496,
      "step": 43
    },
    {
      "epoch": 1.0,
      "grad_norm": 2.1998031778618596,
      "learning_rate": 1e-05,
      "loss": 0.3295,
      "step": 44
    },
    {
      "epoch": 1.03,
      "grad_norm": 1.9950078809811038,
      "learning_rate": 1e-05,
      "loss": 0.2919,
      "step": 45
    },
    {
      "epoch": 1.05,
      "grad_norm": 1.8802987071836594,
      "learning_rate": 1e-05,
      "loss": 0.2774,
      "step": 46
    },
    {
      "epoch": 1.07,
      "grad_norm": 2.330358996793001,
      "learning_rate": 1e-05,
      "loss": 0.2953,
      "step": 47
    },
    {
      "epoch": 1.09,
      "grad_norm": 2.187255642034782,
      "learning_rate": 1e-05,
      "loss": 0.2637,
      "step": 48
    },
    {
      "epoch": 1.12,
      "grad_norm": 2.246935471483204,
      "learning_rate": 1e-05,
      "loss": 0.2806,
      "step": 49
    },
    {
      "epoch": 1.14,
      "grad_norm": 2.129600005729623,
      "learning_rate": 1e-05,
      "loss": 0.2566,
      "step": 50
    },
    {
      "epoch": 1.16,
      "grad_norm": 2.1613544928929347,
      "learning_rate": 1e-05,
      "loss": 0.2703,
      "step": 51
    },
    {
      "epoch": 1.19,
      "grad_norm": 2.097651271845284,
      "learning_rate": 1e-05,
      "loss": 0.2578,
      "step": 52
    },
    {
      "epoch": 1.21,
      "grad_norm": 2.1698187501885404,
      "learning_rate": 1e-05,
      "loss": 0.266,
      "step": 53
    },
    {
      "epoch": 1.23,
      "grad_norm": 2.0534602028652733,
      "learning_rate": 1e-05,
      "loss": 0.2617,
      "step": 54
    },
    {
      "epoch": 1.25,
      "grad_norm": 2.280247094380378,
      "learning_rate": 1e-05,
      "loss": 0.2725,
      "step": 55
    },
    {
      "epoch": 1.28,
      "grad_norm": 2.282981855579156,
      "learning_rate": 1e-05,
      "loss": 0.2696,
      "step": 56
    },
    {
      "epoch": 1.3,
      "grad_norm": 2.1723199295603246,
      "learning_rate": 1e-05,
      "loss": 0.2801,
      "step": 57
    },
    {
      "epoch": 1.32,
      "grad_norm": 2.1013790638047833,
      "learning_rate": 1e-05,
      "loss": 0.2462,
      "step": 58
    },
    {
      "epoch": 1.34,
      "grad_norm": 2.1486192844939187,
      "learning_rate": 1e-05,
      "loss": 0.2634,
      "step": 59
    },
    {
      "epoch": 1.37,
      "grad_norm": 2.2011921615871874,
      "learning_rate": 1e-05,
      "loss": 0.2644,
      "step": 60
    },
    {
      "epoch": 1.39,
      "grad_norm": 2.0994036223733907,
      "learning_rate": 1e-05,
      "loss": 0.2366,
      "step": 61
    },
    {
      "epoch": 1.41,
      "grad_norm": 1.8682173502881247,
      "learning_rate": 1e-05,
      "loss": 0.2369,
      "step": 62
    },
    {
      "epoch": 1.44,
      "grad_norm": 1.9469094762833548,
      "learning_rate": 1e-05,
      "loss": 0.243,
      "step": 63
    },
    {
      "epoch": 1.46,
      "grad_norm": 2.198810958517713,
      "learning_rate": 1e-05,
      "loss": 0.2538,
      "step": 64
    },
    {
      "epoch": 1.48,
      "grad_norm": 1.9614171222986219,
      "learning_rate": 1e-05,
      "loss": 0.2209,
      "step": 65
    },
    {
      "epoch": 1.5,
      "grad_norm": 2.1909792689278924,
      "learning_rate": 1e-05,
      "loss": 0.2289,
      "step": 66
    },
    {
      "epoch": 1.53,
      "grad_norm": 2.219018585854138,
      "learning_rate": 1e-05,
      "loss": 0.2438,
      "step": 67
    },
    {
      "epoch": 1.55,
      "grad_norm": 2.095328922178155,
      "learning_rate": 1e-05,
      "loss": 0.2384,
      "step": 68
    },
    {
      "epoch": 1.57,
      "grad_norm": 2.1047747885459596,
      "learning_rate": 1e-05,
      "loss": 0.2511,
      "step": 69
    },
    {
      "epoch": 1.6,
      "grad_norm": 2.217218508465867,
      "learning_rate": 1e-05,
      "loss": 0.2326,
      "step": 70
    },
    {
      "epoch": 1.62,
      "grad_norm": 2.2940614424982364,
      "learning_rate": 1e-05,
      "loss": 0.2095,
      "step": 71
    },
    {
      "epoch": 1.64,
      "grad_norm": 2.0023231563553012,
      "learning_rate": 1e-05,
      "loss": 0.2178,
      "step": 72
    },
    {
      "epoch": 1.66,
      "grad_norm": 1.8803020902826912,
      "learning_rate": 1e-05,
      "loss": 0.2299,
      "step": 73
    },
    {
      "epoch": 1.69,
      "grad_norm": 1.9874036980658476,
      "learning_rate": 1e-05,
      "loss": 0.2085,
      "step": 74
    },
    {
      "epoch": 1.71,
      "grad_norm": 1.9993267804137187,
      "learning_rate": 1e-05,
      "loss": 0.2189,
      "step": 75
    },
    {
      "epoch": 1.73,
      "grad_norm": 2.075032021080106,
      "learning_rate": 1e-05,
      "loss": 0.2122,
      "step": 76
    },
    {
      "epoch": 1.75,
      "grad_norm": 1.9908316147368204,
      "learning_rate": 1e-05,
      "loss": 0.2169,
      "step": 77
    },
    {
      "epoch": 1.78,
      "grad_norm": 1.9737029328293805,
      "learning_rate": 1e-05,
      "loss": 0.1976,
      "step": 78
    },
    {
      "epoch": 1.8,
      "grad_norm": 2.069169880586868,
      "learning_rate": 1e-05,
      "loss": 0.2193,
      "step": 79
    },
    {
      "epoch": 1.82,
      "grad_norm": 1.9730042900890021,
      "learning_rate": 1e-05,
      "loss": 0.1959,
      "step": 80
    },
    {
      "epoch": 1.85,
      "grad_norm": 1.9487990391437768,
      "learning_rate": 1e-05,
      "loss": 0.2099,
      "step": 81
    },
    {
      "epoch": 1.87,
      "grad_norm": 2.2136709023857923,
      "learning_rate": 1e-05,
      "loss": 0.2064,
      "step": 82
    },
    {
      "epoch": 1.89,
      "grad_norm": 1.867453149915136,
      "learning_rate": 1e-05,
      "loss": 0.2148,
      "step": 83
    },
    {
      "epoch": 1.91,
      "grad_norm": 2.2341488312462374,
      "learning_rate": 1e-05,
      "loss": 0.2152,
      "step": 84
    },
    {
      "epoch": 1.94,
      "grad_norm": 1.917448307480571,
      "learning_rate": 1e-05,
      "loss": 0.2087,
      "step": 85
    },
    {
      "epoch": 1.96,
      "grad_norm": 2.1010824249510938,
      "learning_rate": 1e-05,
      "loss": 0.1989,
      "step": 86
    },
    {
      "epoch": 1.98,
      "grad_norm": 1.9948658919746771,
      "learning_rate": 1e-05,
      "loss": 0.184,
      "step": 87
    },
    {
      "epoch": 2.01,
      "grad_norm": 1.7700018111142861,
      "learning_rate": 1e-05,
      "loss": 0.1755,
      "step": 88
    },
    {
      "epoch": 2.03,
      "grad_norm": 1.7706928580089987,
      "learning_rate": 1e-05,
      "loss": 0.1524,
      "step": 89
    },
    {
      "epoch": 2.05,
      "grad_norm": 1.924950952533756,
      "learning_rate": 1e-05,
      "loss": 0.1639,
      "step": 90
    },
    {
      "epoch": 2.07,
      "grad_norm": 1.8520437499999318,
      "learning_rate": 1e-05,
      "loss": 0.1628,
      "step": 91
    },
    {
      "epoch": 2.1,
      "grad_norm": 1.9282660088754877,
      "learning_rate": 1e-05,
      "loss": 0.1738,
      "step": 92
    },
    {
      "epoch": 2.12,
      "grad_norm": 2.152060500211321,
      "learning_rate": 1e-05,
      "loss": 0.1533,
      "step": 93
    },
    {
      "epoch": 2.14,
      "grad_norm": 2.1072066102911204,
      "learning_rate": 1e-05,
      "loss": 0.1699,
      "step": 94
    },
    {
      "epoch": 2.16,
      "grad_norm": 1.9056526630981285,
      "learning_rate": 1e-05,
      "loss": 0.1542,
      "step": 95
    },
    {
      "epoch": 2.19,
      "grad_norm": 2.0541561878206815,
      "learning_rate": 1e-05,
      "loss": 0.1654,
      "step": 96
    },
    {
      "epoch": 2.21,
      "grad_norm": 1.9010874038699952,
      "learning_rate": 1e-05,
      "loss": 0.1284,
      "step": 97
    },
    {
      "epoch": 2.23,
      "grad_norm": 1.8180452474072457,
      "learning_rate": 1e-05,
      "loss": 0.1294,
      "step": 98
    },
    {
      "epoch": 2.26,
      "grad_norm": 1.9510507921351066,
      "learning_rate": 1e-05,
      "loss": 0.1555,
      "step": 99
    },
    {
      "epoch": 2.28,
      "grad_norm": 2.16385305667214,
      "learning_rate": 1e-05,
      "loss": 0.1568,
      "step": 100
    },
    {
      "epoch": 2.3,
      "grad_norm": 1.778578736374391,
      "learning_rate": 1e-05,
      "loss": 0.1493,
      "step": 101
    },
    {
      "epoch": 2.32,
      "grad_norm": 1.7020886640236346,
      "learning_rate": 1e-05,
      "loss": 0.1365,
      "step": 102
    },
    {
      "epoch": 2.35,
      "grad_norm": 1.9239197373726553,
      "learning_rate": 1e-05,
      "loss": 0.1529,
      "step": 103
    },
    {
      "epoch": 2.37,
      "grad_norm": 1.8300125885434804,
      "learning_rate": 1e-05,
      "loss": 0.1507,
      "step": 104
    },
    {
      "epoch": 2.39,
      "grad_norm": 1.7740533757030397,
      "learning_rate": 1e-05,
      "loss": 0.1404,
      "step": 105
    },
    {
      "epoch": 2.42,
      "grad_norm": 1.921607088125746,
      "learning_rate": 1e-05,
      "loss": 0.1403,
      "step": 106
    },
    {
      "epoch": 2.44,
      "grad_norm": 1.7244219031686432,
      "learning_rate": 1e-05,
      "loss": 0.1493,
      "step": 107
    },
    {
      "epoch": 2.46,
      "grad_norm": 1.840637334327782,
      "learning_rate": 1e-05,
      "loss": 0.1555,
      "step": 108
    },
    {
      "epoch": 2.48,
      "grad_norm": 1.7138834314865232,
      "learning_rate": 1e-05,
      "loss": 0.1507,
      "step": 109
    },
    {
      "epoch": 2.51,
      "grad_norm": 1.8869772327906467,
      "learning_rate": 1e-05,
      "loss": 0.1519,
      "step": 110
    },
    {
      "epoch": 2.53,
      "grad_norm": 1.722031710475424,
      "learning_rate": 1e-05,
      "loss": 0.1388,
      "step": 111
    },
    {
      "epoch": 2.55,
      "grad_norm": 1.8294129933303556,
      "learning_rate": 1e-05,
      "loss": 0.1429,
      "step": 112
    },
    {
      "epoch": 2.58,
      "grad_norm": 1.913994003850658,
      "learning_rate": 1e-05,
      "loss": 0.1391,
      "step": 113
    },
    {
      "epoch": 2.6,
      "grad_norm": 1.9673824351683147,
      "learning_rate": 1e-05,
      "loss": 0.1374,
      "step": 114
    },
    {
      "epoch": 2.62,
      "grad_norm": 1.9673399496447597,
      "learning_rate": 1e-05,
      "loss": 0.1458,
      "step": 115
    },
    {
      "epoch": 2.64,
      "grad_norm": 1.8027392959029291,
      "learning_rate": 1e-05,
      "loss": 0.1351,
      "step": 116
    },
    {
      "epoch": 2.67,
      "grad_norm": 1.667565922975832,
      "learning_rate": 1e-05,
      "loss": 0.1455,
      "step": 117
    },
    {
      "epoch": 2.69,
      "grad_norm": 1.8116940749067432,
      "learning_rate": 1e-05,
      "loss": 0.1502,
      "step": 118
    },
    {
      "epoch": 2.71,
      "grad_norm": 1.9922161139058774,
      "learning_rate": 1e-05,
      "loss": 0.1422,
      "step": 119
    },
    {
      "epoch": 2.73,
      "grad_norm": 2.0124481055349044,
      "learning_rate": 1e-05,
      "loss": 0.1373,
      "step": 120
    },
    {
      "epoch": 2.76,
      "grad_norm": 1.800014793589776,
      "learning_rate": 1e-05,
      "loss": 0.1342,
      "step": 121
    },
    {
      "epoch": 2.78,
      "grad_norm": 1.7532349977825892,
      "learning_rate": 1e-05,
      "loss": 0.1262,
      "step": 122
    },
    {
      "epoch": 2.8,
      "grad_norm": 1.7857219830778754,
      "learning_rate": 1e-05,
      "loss": 0.1375,
      "step": 123
    },
    {
      "epoch": 2.83,
      "grad_norm": 1.7910722257498661,
      "learning_rate": 1e-05,
      "loss": 0.148,
      "step": 124
    },
    {
      "epoch": 2.85,
      "grad_norm": 1.6958921797341877,
      "learning_rate": 1e-05,
      "loss": 0.1287,
      "step": 125
    },
    {
      "epoch": 2.87,
      "grad_norm": 1.9268954167687131,
      "learning_rate": 1e-05,
      "loss": 0.1334,
      "step": 126
    },
    {
      "epoch": 2.89,
      "grad_norm": 1.7925408289852067,
      "learning_rate": 1e-05,
      "loss": 0.1445,
      "step": 127
    },
    {
      "epoch": 2.92,
      "grad_norm": 1.9380903694792881,
      "learning_rate": 1e-05,
      "loss": 0.1478,
      "step": 128
    },
    {
      "epoch": 2.94,
      "grad_norm": 1.8550832947077076,
      "learning_rate": 1e-05,
      "loss": 0.1185,
      "step": 129
    },
    {
      "epoch": 2.94,
      "step": 129,
      "total_flos": 56540883877888.0,
      "train_loss": 0.30087860679441647,
      "train_runtime": 2488.5787,
      "train_samples_per_second": 6.613,
      "train_steps_per_second": 0.052
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 129,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1.0,
  "total_flos": 56540883877888.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}