{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9993247805536799,
  "eval_steps": 500,
  "global_step": 370,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 21.415120223600994,
      "learning_rate": 4.1666666666666665e-05,
      "loss": 8.6562,
      "step": 1
    },
    {
      "epoch": 0.01,
      "grad_norm": 13.773433414832953,
      "learning_rate": 8.333333333333333e-05,
      "loss": 8.7148,
      "step": 2
    },
    {
      "epoch": 0.01,
      "grad_norm": 5.8226925278473605,
      "learning_rate": 0.000125,
      "loss": 6.1523,
      "step": 3
    },
    {
      "epoch": 0.01,
      "grad_norm": 6.913578305289376,
      "learning_rate": 0.00016666666666666666,
      "loss": 4.8105,
      "step": 4
    },
    {
      "epoch": 0.01,
      "grad_norm": 7.564457947295859,
      "learning_rate": 0.00020833333333333335,
      "loss": 4.7148,
      "step": 5
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.4572492282918157,
      "learning_rate": 0.00025,
      "loss": 4.3516,
      "step": 6
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.1297987720642504,
      "learning_rate": 0.0002916666666666667,
      "loss": 3.9824,
      "step": 7
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.1136160564583724,
      "learning_rate": 0.0003333333333333333,
      "loss": 3.7812,
      "step": 8
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.8904819206285229,
      "learning_rate": 0.000375,
      "loss": 3.7344,
      "step": 9
    },
    {
      "epoch": 0.03,
      "grad_norm": 2.11038283184139,
      "learning_rate": 0.0004166666666666667,
      "loss": 3.7041,
      "step": 10
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.802508097218576,
      "learning_rate": 0.0004583333333333333,
      "loss": 3.6377,
      "step": 11
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.4105103213318609,
      "learning_rate": 0.0005,
      "loss": 3.5898,
      "step": 12
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.28889375510308085,
      "learning_rate": 0.0004999903741094759,
      "loss": 3.5059,
      "step": 13
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.1903491115661804,
      "learning_rate": 0.0004999614971791658,
      "loss": 3.5312,
      "step": 14
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.18406971686632234,
      "learning_rate": 0.0004999133714327992,
      "loss": 3.5264,
      "step": 15
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.19187377072404146,
      "learning_rate": 0.0004998460005764011,
      "loss": 3.4863,
      "step": 16
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.18894966967720278,
      "learning_rate": 0.0004997593897980076,
      "loss": 3.5098,
      "step": 17
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.16990770049278048,
      "learning_rate": 0.0004996535457672657,
      "loss": 3.46,
      "step": 18
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.14341359539921433,
      "learning_rate": 0.0004995284766349196,
      "loss": 3.4102,
      "step": 19
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.13853022483240476,
      "learning_rate": 0.0004993841920321838,
      "loss": 3.3721,
      "step": 20
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.14882163330247777,
      "learning_rate": 0.0004992207030700005,
      "loss": 3.4062,
      "step": 21
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.12082174803609903,
      "learning_rate": 0.0004990380223381846,
      "loss": 3.3682,
      "step": 22
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.11072980253911602,
      "learning_rate": 0.0004988361639044537,
      "loss": 3.335,
      "step": 23
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.10689352906545974,
      "learning_rate": 0.0004986151433133455,
      "loss": 3.3447,
      "step": 24
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.10490544779612036,
      "learning_rate": 0.0004983749775850199,
      "loss": 3.3252,
      "step": 25
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.10182457932796082,
      "learning_rate": 0.0004981156852139491,
      "loss": 3.3037,
      "step": 26
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.1056648298216298,
      "learning_rate": 0.0004978372861674928,
      "loss": 3.3164,
      "step": 27
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.08205846396464361,
      "learning_rate": 0.0004975398018843612,
      "loss": 3.2939,
      "step": 28
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.08913381428371997,
      "learning_rate": 0.0004972232552729632,
      "loss": 3.2949,
      "step": 29
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.07539237478136081,
      "learning_rate": 0.000496887670709643,
      "loss": 3.2461,
      "step": 30
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.07628363158131596,
      "learning_rate": 0.0004965330740368029,
      "loss": 3.2578,
      "step": 31
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.07982185655162036,
      "learning_rate": 0.0004961594925609129,
      "loss": 3.2451,
      "step": 32
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.08695759890018029,
      "learning_rate": 0.000495766955050408,
      "loss": 3.2607,
      "step": 33
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.06301153004824052,
      "learning_rate": 0.0004953554917334731,
      "loss": 3.2061,
      "step": 34
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.06474729877299198,
      "learning_rate": 0.0004949251342957151,
      "loss": 3.25,
      "step": 35
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.07883442367683367,
      "learning_rate": 0.0004944759158777223,
      "loss": 3.2236,
      "step": 36
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.07015055529327788,
      "learning_rate": 0.0004940078710725135,
      "loss": 3.2227,
      "step": 37
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.07119572625376963,
      "learning_rate": 0.0004935210359228729,
      "loss": 3.2031,
      "step": 38
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.06367666239627473,
      "learning_rate": 0.0004930154479185756,
      "loss": 3.252,
      "step": 39
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.06300151867440586,
      "learning_rate": 0.0004924911459934997,
      "loss": 3.1777,
      "step": 40
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.07472850979326924,
      "learning_rate": 0.0004919481705226285,
      "loss": 3.2314,
      "step": 41
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.08114730974850456,
      "learning_rate": 0.0004913865633189419,
      "loss": 3.2119,
      "step": 42
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.05895132991107217,
      "learning_rate": 0.0004908063676301953,
      "loss": 3.2188,
      "step": 43
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.05010637153257128,
      "learning_rate": 0.0004902076281355902,
      "loss": 3.1689,
      "step": 44
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.06316741853726789,
      "learning_rate": 0.0004895903909423332,
      "loss": 3.1904,
      "step": 45
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.069290214948921,
      "learning_rate": 0.0004889547035820856,
      "loss": 3.1689,
      "step": 46
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.06229058085155359,
      "learning_rate": 0.0004883006150073028,
      "loss": 3.1514,
      "step": 47
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.05042088665341411,
      "learning_rate": 0.000487628175587465,
      "loss": 3.1816,
      "step": 48
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.0545027205027528,
      "learning_rate": 0.00048693743710519807,
      "loss": 3.1875,
      "step": 49
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.06583086092960536,
      "learning_rate": 0.00048622845275228614,
      "loss": 3.1816,
      "step": 50
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.06171597888295892,
      "learning_rate": 0.00048550127712557515,
      "loss": 3.1621,
      "step": 51
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.05936904165187056,
      "learning_rate": 0.0004847559662227691,
      "loss": 3.1689,
      "step": 52
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.053617959255968184,
      "learning_rate": 0.0004839925774381172,
      "loss": 3.1826,
      "step": 53
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.0584940883576519,
      "learning_rate": 0.00048321116955799426,
      "loss": 3.1445,
      "step": 54
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.05534676948352036,
      "learning_rate": 0.0004824118027563742,
      "loss": 3.1367,
      "step": 55
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.05079053816193199,
      "learning_rate": 0.00048159453859019535,
      "loss": 3.1221,
      "step": 56
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.051247127284195455,
      "learning_rate": 0.000480759439994621,
      "loss": 3.1533,
      "step": 57
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.05840283966287256,
      "learning_rate": 0.00047990657127819234,
      "loss": 3.1592,
      "step": 58
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.06275081599011795,
      "learning_rate": 0.0004790359981178766,
      "loss": 3.1309,
      "step": 59
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.061480429010200704,
      "learning_rate": 0.0004781477875540091,
      "loss": 3.1484,
      "step": 60
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.06296237796551907,
      "learning_rate": 0.0004772420079851313,
      "loss": 3.1338,
      "step": 61
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.05939876669569374,
      "learning_rate": 0.00047631872916272265,
      "loss": 3.1758,
      "step": 62
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.05697500202970297,
      "learning_rate": 0.00047537802218583036,
      "loss": 3.1631,
      "step": 63
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.05235594093980006,
      "learning_rate": 0.00047441995949559323,
      "loss": 3.1592,
      "step": 64
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.05136699721805155,
      "learning_rate": 0.0004734446148696641,
      "loss": 3.1348,
      "step": 65
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.056572106168923254,
      "learning_rate": 0.00047245206341652734,
      "loss": 3.1475,
      "step": 66
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.0680276012617714,
      "learning_rate": 0.0004714423815697162,
      "loss": 3.1221,
      "step": 67
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.058203239195036796,
      "learning_rate": 0.0004704156470819261,
      "loss": 3.1387,
      "step": 68
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.05654758029632964,
      "learning_rate": 0.00046937193901902703,
      "loss": 3.1152,
      "step": 69
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.07079606102138672,
      "learning_rate": 0.0004683113377539756,
      "loss": 3.0811,
      "step": 70
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.06311116636541605,
      "learning_rate": 0.00046723392496062513,
      "loss": 3.1299,
      "step": 71
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.04930899192235485,
      "learning_rate": 0.0004661397836074362,
      "loss": 3.0898,
      "step": 72
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.07242568044263648,
      "learning_rate": 0.0004650289979510883,
      "loss": 3.1328,
      "step": 73
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.06815171525302943,
      "learning_rate": 0.00046390165352998996,
      "loss": 3.0713,
      "step": 74
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.06138422128904437,
      "learning_rate": 0.0004627578371576933,
      "loss": 3.0703,
      "step": 75
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.059644963475315685,
      "learning_rate": 0.0004615976369162077,
      "loss": 3.0898,
      "step": 76
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.06496043006168827,
      "learning_rate": 0.00046042114214921723,
      "loss": 3.0986,
      "step": 77
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.0634821846883854,
      "learning_rate": 0.00045922844345520046,
      "loss": 3.1025,
      "step": 78
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.06014574410746669,
      "learning_rate": 0.000458019632680454,
      "loss": 3.1055,
      "step": 79
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.06497852199883221,
      "learning_rate": 0.00045679480291201916,
      "loss": 3.0781,
      "step": 80
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.07113303653242763,
      "learning_rate": 0.00045555404847051406,
      "loss": 3.0742,
      "step": 81
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.06890895616862315,
      "learning_rate": 0.0004542974649028701,
      "loss": 3.0605,
      "step": 82
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.06609605331363849,
      "learning_rate": 0.00045302514897497414,
      "loss": 3.0859,
      "step": 83
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.06783983704880324,
      "learning_rate": 0.0004517371986642168,
      "loss": 3.0693,
      "step": 84
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.07647246769844311,
      "learning_rate": 0.0004504337131519477,
      "loss": 3.0918,
      "step": 85
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.07603324518310302,
      "learning_rate": 0.00044911479281583747,
      "loss": 3.0547,
      "step": 86
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.08035265561784409,
      "learning_rate": 0.0004477805392221483,
      "loss": 3.0547,
      "step": 87
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.07818094146866135,
      "learning_rate": 0.00044643105511791236,
      "loss": 3.0938,
      "step": 88
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.08705128277730868,
      "learning_rate": 0.0004450664444230197,
      "loss": 3.0811,
      "step": 89
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.08927687330837643,
      "learning_rate": 0.00044368681222221553,
      "loss": 3.0566,
      "step": 90
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.08533504666800651,
      "learning_rate": 0.00044229226475700815,
      "loss": 3.0889,
      "step": 91
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.08432813563363031,
      "learning_rate": 0.0004408829094174872,
      "loss": 3.0312,
      "step": 92
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.09460079385850353,
      "learning_rate": 0.0004394588547340546,
      "loss": 3.041,
      "step": 93
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.12021954624176756,
      "learning_rate": 0.00043802021036906604,
      "loss": 3.0293,
      "step": 94
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.09741352747719122,
      "learning_rate": 0.00043656708710838687,
      "loss": 3.0381,
      "step": 95
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.1005763508223488,
      "learning_rate": 0.00043509959685286047,
      "loss": 3.043,
      "step": 96
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.11242499386359363,
      "learning_rate": 0.00043361785260969126,
      "loss": 3.0674,
      "step": 97
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.10975775405702991,
      "learning_rate": 0.0004321219684837422,
      "loss": 3.0312,
      "step": 98
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.11737592618628766,
      "learning_rate": 0.00043061205966874784,
      "loss": 3.0547,
      "step": 99
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.13051148309222815,
      "learning_rate": 0.00042908824243844383,
      "loss": 3.0225,
      "step": 100
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.1293140100108899,
      "learning_rate": 0.00042755063413761304,
      "loss": 3.0352,
      "step": 101
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.11252955969439887,
      "learning_rate": 0.0004259993531730486,
      "loss": 3.0303,
      "step": 102
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.1328900419617491,
      "learning_rate": 0.00042443451900443666,
      "loss": 3.0146,
      "step": 103
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.12693126792922263,
      "learning_rate": 0.0004228562521351562,
      "loss": 3.0244,
      "step": 104
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.1251814736102281,
      "learning_rate": 0.0004212646741030001,
      "loss": 3.0264,
      "step": 105
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.13416576734949723,
      "learning_rate": 0.0004196599074708156,
      "loss": 3.0166,
      "step": 106
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.12552446417030455,
      "learning_rate": 0.0004180420758170659,
      "loss": 3.0332,
      "step": 107
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.13647763499691762,
      "learning_rate": 0.00041641130372631417,
      "loss": 3.0049,
      "step": 108
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.12497753797378944,
      "learning_rate": 0.0004147677167796294,
      "loss": 2.9766,
      "step": 109
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.16181335617568174,
      "learning_rate": 0.00041311144154491567,
      "loss": 3.0078,
      "step": 110
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.1231265891953706,
      "learning_rate": 0.0004114426055671655,
      "loss": 3.0371,
      "step": 111
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.1500040260456989,
      "learning_rate": 0.0004097613373586385,
      "loss": 3.0137,
      "step": 112
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.14493053467762143,
      "learning_rate": 0.0004080677663889644,
      "loss": 3.0059,
      "step": 113
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.13498422720011877,
      "learning_rate": 0.00040636202307517313,
      "loss": 2.9834,
      "step": 114
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.13758614368752517,
      "learning_rate": 0.0004046442387716519,
      "loss": 2.9883,
      "step": 115
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.1547910759873873,
      "learning_rate": 0.0004029145457600298,
      "loss": 3.0146,
      "step": 116
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.1114547142061138,
      "learning_rate": 0.0004011730772389916,
      "loss": 2.9502,
      "step": 117
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.158851085984572,
      "learning_rate": 0.0003994199673140199,
      "loss": 2.9707,
      "step": 118
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.15035124478352418,
      "learning_rate": 0.0003976553509870683,
      "loss": 2.959,
      "step": 119
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.18562089234260126,
      "learning_rate": 0.00039587936414616556,
      "loss": 2.9893,
      "step": 120
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.1500050006066447,
      "learning_rate": 0.00039409214355495093,
      "loss": 2.9912,
      "step": 121
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.15173547245397362,
      "learning_rate": 0.00039229382684214233,
      "loss": 2.9971,
      "step": 122
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.17021016573674075,
      "learning_rate": 0.0003904845524909382,
      "loss": 2.9766,
      "step": 123
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.16160325963256653,
      "learning_rate": 0.0003886644598283533,
      "loss": 2.9463,
      "step": 124
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.16620594275883835,
      "learning_rate": 0.00038683368901448934,
      "loss": 2.9951,
      "step": 125
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.18996567972484682,
      "learning_rate": 0.00038499238103174166,
      "loss": 2.9492,
      "step": 126
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.17307569767495753,
      "learning_rate": 0.00038314067767394286,
      "loss": 2.9805,
      "step": 127
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.19685088775869003,
      "learning_rate": 0.0003812787215354433,
      "loss": 2.9424,
      "step": 128
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.16189642626507808,
      "learning_rate": 0.00037940665600013057,
      "loss": 2.9512,
      "step": 129
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.17612947304733076,
      "learning_rate": 0.00037752462523038793,
      "loss": 2.9639,
      "step": 130
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.18179176164130736,
      "learning_rate": 0.0003756327741559925,
      "loss": 2.9688,
      "step": 131
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.16591267900078252,
      "learning_rate": 0.00037373124846295505,
      "loss": 2.9482,
      "step": 132
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.18798955074723858,
      "learning_rate": 0.00037182019458230066,
      "loss": 2.9395,
      "step": 133
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.17543591147515358,
      "learning_rate": 0.000369899759678793,
      "loss": 2.9404,
      "step": 134
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.18490619032970695,
      "learning_rate": 0.0003679700916396009,
      "loss": 2.9736,
      "step": 135
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.1804894673337707,
      "learning_rate": 0.00036603133906291095,
      "loss": 2.9561,
      "step": 136
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.18154491542277257,
      "learning_rate": 0.0003640836512464837,
      "loss": 2.9609,
      "step": 137
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.15353929430307406,
      "learning_rate": 0.00036212717817615635,
      "loss": 2.9355,
      "step": 138
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.21621902507983504,
      "learning_rate": 0.000360162070514294,
      "loss": 2.957,
      "step": 139
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.1829941353882955,
      "learning_rate": 0.0003581884795881861,
      "loss": 2.9521,
      "step": 140
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.18716180371474261,
      "learning_rate": 0.0003562065573783945,
      "loss": 2.9434,
      "step": 141
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.2275322416310555,
      "learning_rate": 0.0003542164565070486,
      "loss": 2.9365,
      "step": 142
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.19125954171713724,
      "learning_rate": 0.00035221833022609366,
      "loss": 2.9121,
      "step": 143
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.23978782914992927,
      "learning_rate": 0.0003502123324054882,
      "loss": 2.9473,
      "step": 144
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.18956138126965327,
      "learning_rate": 0.0003481986175213556,
      "loss": 2.9531,
      "step": 145
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.20333907762683046,
      "learning_rate": 0.0003461773406440879,
      "loss": 2.9229,
      "step": 146
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.21650596272683872,
      "learning_rate": 0.0003441486574264048,
      "loss": 2.9639,
      "step": 147
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.20068147784646773,
      "learning_rate": 0.00034211272409136667,
      "loss": 2.9395,
      "step": 148
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.20053669008991853,
      "learning_rate": 0.00034006969742034465,
      "loss": 2.9541,
      "step": 149
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.19123221320191458,
      "learning_rate": 0.00033801973474094745,
      "loss": 2.9668,
      "step": 150
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.20055890299268714,
      "learning_rate": 0.00033596299391490565,
      "loss": 2.918,
      "step": 151
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.21961462852023858,
      "learning_rate": 0.0003338996333259155,
      "loss": 2.8887,
      "step": 152
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.19923388725160776,
      "learning_rate": 0.000331829811867442,
      "loss": 2.915,
      "step": 153
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.2202467020445118,
      "learning_rate": 0.0003297536889304835,
      "loss": 2.9072,
      "step": 154
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.1896310412062489,
      "learning_rate": 0.0003276714243912966,
      "loss": 2.9219,
      "step": 155
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.24063275517469532,
      "learning_rate": 0.0003255831785990854,
      "loss": 2.9268,
      "step": 156
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.2093352162026971,
      "learning_rate": 0.000323489112363653,
      "loss": 2.9102,
      "step": 157
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.19084865959587785,
      "learning_rate": 0.000321389386943018,
      "loss": 2.916,
      "step": 158
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.24890451630400445,
      "learning_rate": 0.0003192841640309966,
      "loss": 2.9492,
      "step": 159
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.18002995253459828,
      "learning_rate": 0.0003171736057447511,
      "loss": 2.9111,
      "step": 160
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.22447733600257028,
      "learning_rate": 0.00031505787461230553,
      "loss": 2.8936,
      "step": 161
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.2467901168622779,
      "learning_rate": 0.00031293713356003,
      "loss": 2.9316,
      "step": 162
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.20550840515489216,
      "learning_rate": 0.00031081154590009404,
      "loss": 2.9453,
      "step": 163
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.21166602849404573,
      "learning_rate": 0.0003086812753178907,
      "loss": 2.916,
      "step": 164
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.3602799328889876,
      "learning_rate": 0.0003065464858594311,
      "loss": 2.9102,
      "step": 165
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.18298535380931488,
      "learning_rate": 0.00030440734191871234,
      "loss": 2.9043,
      "step": 166
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.2572700716773119,
      "learning_rate": 0.00030226400822505736,
      "loss": 2.9033,
      "step": 167
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.25835419996657133,
      "learning_rate": 0.00030011664983043025,
      "loss": 2.9062,
      "step": 168
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.17683277759979313,
      "learning_rate": 0.0002979654320967255,
      "loss": 2.916,
      "step": 169
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.2894378726624145,
      "learning_rate": 0.0002958105206830343,
      "loss": 2.9053,
      "step": 170
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.23685763825987682,
      "learning_rate": 0.0002936520815328876,
      "loss": 2.8857,
      "step": 171
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.2008573100452373,
      "learning_rate": 0.000291490280861477,
      "loss": 2.9297,
      "step": 172
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.26444755166927053,
      "learning_rate": 0.00028932528514285534,
      "loss": 2.9316,
      "step": 173
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.24379542789804157,
      "learning_rate": 0.00028715726109711665,
      "loss": 2.9131,
      "step": 174
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.20673216784596224,
      "learning_rate": 0.00028498637567755817,
      "loss": 2.9082,
      "step": 175
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.21944969615374402,
      "learning_rate": 0.0002828127960578227,
      "loss": 2.9199,
      "step": 176
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.20478834301866145,
      "learning_rate": 0.00028063668961902623,
      "loss": 2.9033,
      "step": 177
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.21481876173786985,
      "learning_rate": 0.00027845822393686724,
      "loss": 2.8809,
      "step": 178
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.22811931240082148,
      "learning_rate": 0.00027627756676872344,
      "loss": 2.8916,
      "step": 179
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.25707311204201183,
      "learning_rate": 0.00027409488604073184,
      "loss": 2.9258,
      "step": 180
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.22655991900141323,
      "learning_rate": 0.0002719103498348586,
      "loss": 2.8936,
      "step": 181
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.2462832484307767,
      "learning_rate": 0.00026972412637595445,
      "loss": 2.8789,
      "step": 182
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.16995673551677068,
      "learning_rate": 0.00026753638401880094,
      "loss": 2.876,
      "step": 183
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.2536950503045075,
      "learning_rate": 0.0002653472912351454,
      "loss": 2.9189,
      "step": 184
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.20069386810955497,
      "learning_rate": 0.0002631570166007276,
      "loss": 2.8838,
      "step": 185
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.17449546614470834,
      "learning_rate": 0.00026096572878229855,
      "loss": 2.8936,
      "step": 186
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.20654114988441505,
      "learning_rate": 0.00025877359652463124,
      "loss": 2.8965,
      "step": 187
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.19593843356526874,
      "learning_rate": 0.0002565807886375267,
      "loss": 2.8984,
      "step": 188
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.20292351001271516,
      "learning_rate": 0.00025438747398281433,
      "loss": 2.9004,
      "step": 189
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.2043498399445655,
      "learning_rate": 0.00025219382146134806,
      "loss": 2.8516,
      "step": 190
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.1661636250320572,
      "learning_rate": 0.00025,
      "loss": 2.8545,
      "step": 191
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.19022198317016406,
      "learning_rate": 0.00024780617853865195,
      "loss": 2.9092,
      "step": 192
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.22844383852438707,
      "learning_rate": 0.0002456125260171857,
      "loss": 2.9111,
      "step": 193
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.17676802172340947,
      "learning_rate": 0.00024341921136247338,
      "loss": 2.8857,
      "step": 194
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.17406205780039047,
      "learning_rate": 0.00024122640347536885,
      "loss": 2.8896,
      "step": 195
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.21596180880748264,
      "learning_rate": 0.00023903427121770148,
      "loss": 2.8643,
      "step": 196
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.2302727016974318,
      "learning_rate": 0.0002368429833992724,
      "loss": 2.8936,
      "step": 197
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.193768899062196,
      "learning_rate": 0.00023465270876485474,
      "loss": 2.8496,
      "step": 198
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.1784013570789268,
      "learning_rate": 0.0002324636159811991,
      "loss": 2.8379,
      "step": 199
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.20783430292995747,
      "learning_rate": 0.00023027587362404558,
      "loss": 2.8799,
      "step": 200
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.20944704580582277,
      "learning_rate": 0.00022808965016514143,
      "loss": 2.8965,
      "step": 201
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.21496187154467147,
      "learning_rate": 0.00022590511395926811,
      "loss": 2.8672,
      "step": 202
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.21633737213344018,
      "learning_rate": 0.00022372243323127663,
      "loss": 2.875,
      "step": 203
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.19865551825102143,
      "learning_rate": 0.00022154177606313277,
      "loss": 2.8672,
      "step": 204
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.18963996378916614,
      "learning_rate": 0.00021936331038097386,
      "loss": 2.9023,
      "step": 205
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.20425004554589565,
      "learning_rate": 0.00021718720394217725,
      "loss": 2.8574,
      "step": 206
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.2555973427913448,
      "learning_rate": 0.00021501362432244193,
      "loss": 2.8701,
      "step": 207
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.1953962081104273,
      "learning_rate": 0.00021284273890288336,
      "loss": 2.8779,
      "step": 208
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.2248900981224502,
      "learning_rate": 0.00021067471485714472,
      "loss": 2.8398,
      "step": 209
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.26993244613823164,
      "learning_rate": 0.00020850971913852307,
      "loss": 2.8613,
      "step": 210
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.2022265920119534,
      "learning_rate": 0.00020634791846711243,
      "loss": 2.8574,
      "step": 211
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.24664106252564721,
      "learning_rate": 0.00020418947931696575,
      "loss": 2.8564,
      "step": 212
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.1967992968536126,
      "learning_rate": 0.00020203456790327453,
      "loss": 2.8555,
      "step": 213
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.22710937800094108,
      "learning_rate": 0.0001998833501695698,
      "loss": 2.8594,
      "step": 214
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.2609592105615878,
      "learning_rate": 0.00019773599177494265,
      "loss": 2.8682,
      "step": 215
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.20377916758575193,
      "learning_rate": 0.00019559265808128778,
      "loss": 2.8496,
      "step": 216
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.2298131342082367,
      "learning_rate": 0.00019345351414056894,
      "loss": 2.8398,
      "step": 217
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.234771024880693,
      "learning_rate": 0.00019131872468210938,
      "loss": 2.8467,
      "step": 218
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.23844110318408976,
      "learning_rate": 0.00018918845409990592,
      "loss": 2.917,
      "step": 219
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.19036660800238525,
      "learning_rate": 0.00018706286643996997,
      "loss": 2.8564,
      "step": 220
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.20622165918252797,
      "learning_rate": 0.0001849421253876945,
      "loss": 2.8848,
      "step": 221
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.1883236349273076,
      "learning_rate": 0.000182826394255249,
      "loss": 2.8643,
      "step": 222
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.22937231306412564,
      "learning_rate": 0.00018071583596900343,
      "loss": 2.8623,
      "step": 223
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.21538020857939882,
      "learning_rate": 0.00017861061305698205,
      "loss": 2.8438,
      "step": 224
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.1918870218278195,
      "learning_rate": 0.0001765108876363471,
      "loss": 2.8467,
      "step": 225
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.21091693539511583,
      "learning_rate": 0.00017441682140091464,
      "loss": 2.8281,
      "step": 226
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.30476893084177187,
      "learning_rate": 0.0001723285756087034,
      "loss": 2.873,
      "step": 227
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.17816221080916358,
      "learning_rate": 0.0001702463110695166,
      "loss": 2.8525,
      "step": 228
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.195493340516346,
      "learning_rate": 0.00016817018813255796,
      "loss": 2.8359,
      "step": 229
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.23565534271641125,
      "learning_rate": 0.00016610036667408462,
      "loss": 2.8535,
      "step": 230
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.2033674101509759,
      "learning_rate": 0.00016403700608509438,
      "loss": 2.832,
      "step": 231
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.1643046553266504,
      "learning_rate": 0.00016198026525905256,
      "loss": 2.8525,
      "step": 232
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.18680871871908325,
      "learning_rate": 0.00015993030257965536,
      "loss": 2.8496,
      "step": 233
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.21879472472240535,
      "learning_rate": 0.00015788727590863345,
      "loss": 2.8457,
      "step": 234
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.189459681090247,
      "learning_rate": 0.00015585134257359523,
      "loss": 2.8252,
      "step": 235
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.20698851721297937,
      "learning_rate": 0.00015382265935591212,
      "loss": 2.8545,
      "step": 236
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.21230475441950047,
      "learning_rate": 0.00015180138247864445,
      "loss": 2.8682,
      "step": 237
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.2170131840363288,
      "learning_rate": 0.00014978766759451187,
      "loss": 2.835,
      "step": 238
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.1747044380388941,
      "learning_rate": 0.00014778166977390643,
      "loss": 2.8193,
      "step": 239
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.2195091552174841,
      "learning_rate": 0.00014578354349295138,
      "loss": 2.8428,
      "step": 240
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.17839712623841394,
      "learning_rate": 0.0001437934426216056,
      "loss": 2.8496,
      "step": 241
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.2218268901944312,
      "learning_rate": 0.0001418115204118139,
      "loss": 2.8379,
      "step": 242
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.22291238199967986,
      "learning_rate": 0.00013983792948570606,
      "loss": 2.8418,
      "step": 243
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.2314705382253311,
      "learning_rate": 0.00013787282182384363,
      "loss": 2.835,
      "step": 244
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.2062566807258234,
      "learning_rate": 0.0001359163487535164,
      "loss": 2.8174,
      "step": 245
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.20906998538613936,
      "learning_rate": 0.00013396866093708898,
      "loss": 2.8184,
      "step": 246
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.2687448764467888,
      "learning_rate": 0.0001320299083603992,
      "loss": 2.835,
      "step": 247
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.1990426441612401,
      "learning_rate": 0.00013010024032120718,
      "loss": 2.877,
      "step": 248
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.20269787106945644,
      "learning_rate": 0.00012817980541769935,
      "loss": 2.8564,
      "step": 249
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.19805581453093807,
      "learning_rate": 0.000126268751537045,
      "loss": 2.8447,
      "step": 250
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.204837890537741,
      "learning_rate": 0.00012436722584400751,
      "loss": 2.8418,
      "step": 251
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.169050938877304,
      "learning_rate": 0.00012247537476961214,
      "loss": 2.8311,
      "step": 252
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.21928573521752834,
      "learning_rate": 0.00012059334399986949,
      "loss": 2.8066,
      "step": 253
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.18055284433976476,
      "learning_rate": 0.00011872127846455672,
      "loss": 2.8125,
      "step": 254
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.20659092646221383,
      "learning_rate": 0.00011685932232605718,
      "loss": 2.8066,
      "step": 255
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.20299552954457484,
      "learning_rate": 0.00011500761896825842,
      "loss": 2.8398,
      "step": 256
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.21714725029231918,
      "learning_rate": 0.00011316631098551067,
      "loss": 2.832,
      "step": 257
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.21963153956688958,
      "learning_rate": 0.00011133554017164671,
      "loss": 2.8193,
      "step": 258
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.19726411127227286,
      "learning_rate": 0.00010951544750906181,
      "loss": 2.8213,
      "step": 259
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.21949146356088167,
      "learning_rate": 0.00010770617315785766,
      "loss": 2.8418,
      "step": 260
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.2126404293875657,
      "learning_rate": 0.00010590785644504919,
      "loss": 2.833,
      "step": 261
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.2071663035323768,
      "learning_rate": 0.00010412063585383452,
      "loss": 2.8174,
      "step": 262
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.222005661427221,
      "learning_rate": 0.00010234464901293173,
      "loss": 2.8076,
      "step": 263
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.2233059412032165,
      "learning_rate": 0.00010058003268598015,
      "loss": 2.8389,
      "step": 264
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.20234119110348878,
      "learning_rate": 9.882692276100841e-05,
      "loss": 2.8242,
      "step": 265
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.22629912766899482,
      "learning_rate": 9.70854542399702e-05,
      "loss": 2.8232,
      "step": 266
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.19123353391124664,
      "learning_rate": 9.535576122834819e-05,
      "loss": 2.8418,
      "step": 267
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.20149824977959827,
      "learning_rate": 9.363797692482695e-05,
      "loss": 2.8115,
      "step": 268
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.17961635360231348,
      "learning_rate": 9.193223361103556e-05,
      "loss": 2.8291,
      "step": 269
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.2121856691428259,
      "learning_rate": 9.023866264136154e-05,
      "loss": 2.8457,
      "step": 270
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.18851239383885068,
      "learning_rate": 8.85573944328345e-05,
      "loss": 2.7979,
      "step": 271
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.17521924172534067,
      "learning_rate": 8.688855845508442e-05,
      "loss": 2.8086,
      "step": 272
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.18640071329178245,
      "learning_rate": 8.523228322037063e-05,
      "loss": 2.8369,
      "step": 273
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.16959390971032334,
      "learning_rate": 8.358869627368584e-05,
      "loss": 2.8398,
      "step": 274
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.21111788917507904,
      "learning_rate": 8.195792418293416e-05,
      "loss": 2.7852,
      "step": 275
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.17179812719144402,
      "learning_rate": 8.03400925291845e-05,
      "loss": 2.8115,
      "step": 276
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.20491937385204168,
      "learning_rate": 7.873532589699988e-05,
      "loss": 2.791,
      "step": 277
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.16357238878820704,
      "learning_rate": 7.71437478648438e-05,
      "loss": 2.8291,
      "step": 278
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.1835990750271033,
      "learning_rate": 7.556548099556338e-05,
      "loss": 2.7998,
      "step": 279
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.1907507274896681,
      "learning_rate": 7.40006468269514e-05,
      "loss": 2.8213,
      "step": 280
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.18564884527181547,
      "learning_rate": 7.244936586238704e-05,
      "loss": 2.8301,
      "step": 281
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.17700032167669388,
      "learning_rate": 7.091175756155618e-05,
      "loss": 2.8115,
      "step": 282
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.16847559163220463,
      "learning_rate": 6.938794033125225e-05,
      "loss": 2.833,
      "step": 283
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.18723560047102772,
      "learning_rate": 6.787803151625791e-05,
      "loss": 2.8281,
      "step": 284
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.18633675001733044,
      "learning_rate": 6.63821473903087e-05,
      "loss": 2.8184,
      "step": 285
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.1822056204472567,
      "learning_rate": 6.490040314713952e-05,
      "loss": 2.834,
      "step": 286
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.19871618333174051,
      "learning_rate": 6.343291289161316e-05,
      "loss": 2.8174,
      "step": 287
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.20546173536543255,
      "learning_rate": 6.197978963093401e-05,
      "loss": 2.8242,
      "step": 288
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.19374612820702158,
      "learning_rate": 6.054114526594548e-05,
      "loss": 2.835,
      "step": 289
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.18924589658433175,
      "learning_rate": 5.911709058251283e-05,
      "loss": 2.7988,
      "step": 290
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.22644302618589385,
      "learning_rate": 5.770773524299192e-05,
      "loss": 2.8525,
      "step": 291
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.18156812551515011,
      "learning_rate": 5.631318777778452e-05,
      "loss": 2.8047,
      "step": 292
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.19435635715822475,
      "learning_rate": 5.493355557698029e-05,
      "loss": 2.8203,
      "step": 293
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.20414237163125332,
      "learning_rate": 5.3568944882087604e-05,
      "loss": 2.8369,
      "step": 294
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.18567184675458911,
      "learning_rate": 5.2219460777851695e-05,
      "loss": 2.8398,
      "step": 295
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.1783969629345075,
      "learning_rate": 5.088520718416254e-05,
      "loss": 2.8057,
      "step": 296
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.18247845423738968,
      "learning_rate": 4.9566286848052363e-05,
      "loss": 2.835,
      "step": 297
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.18785911029833752,
      "learning_rate": 4.826280133578323e-05,
      "loss": 2.8242,
      "step": 298
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.18953403976774194,
      "learning_rate": 4.697485102502588e-05,
      "loss": 2.8252,
      "step": 299
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.18190067624900982,
      "learning_rate": 4.5702535097129906e-05,
      "loss": 2.835,
      "step": 300
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.20251912717942616,
      "learning_rate": 4.444595152948605e-05,
      "loss": 2.8242,
      "step": 301
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.18960132345131622,
      "learning_rate": 4.32051970879809e-05,
      "loss": 2.8418,
      "step": 302
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.20472286168933038,
      "learning_rate": 4.198036731954605e-05,
      "loss": 2.79,
      "step": 303
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.1885255817179192,
      "learning_rate": 4.077155654479955e-05,
      "loss": 2.8076,
      "step": 304
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.17197377072124265,
      "learning_rate": 3.957885785078277e-05,
      "loss": 2.8125,
      "step": 305
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.17243813839090957,
      "learning_rate": 3.840236308379233e-05,
      "loss": 2.7998,
      "step": 306
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.187479341506351,
      "learning_rate": 3.7242162842306724e-05,
      "loss": 2.8076,
      "step": 307
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.19619960697754343,
      "learning_rate": 3.609834647001001e-05,
      "loss": 2.8125,
      "step": 308
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.1969308587109652,
      "learning_rate": 3.497100204891177e-05,
      "loss": 2.7891,
      "step": 309
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.1856744385771311,
      "learning_rate": 3.386021639256373e-05,
      "loss": 2.8076,
      "step": 310
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.19472326340578114,
      "learning_rate": 3.2766075039374926e-05,
      "loss": 2.8135,
      "step": 311
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.16977810395348839,
      "learning_rate": 3.168866224602443e-05,
      "loss": 2.8008,
      "step": 312
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.16737273719653672,
      "learning_rate": 3.0628060980972956e-05,
      "loss": 2.8096,
      "step": 313
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.18706543053983507,
      "learning_rate": 2.958435291807396e-05,
      "loss": 2.8223,
      "step": 314
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.20138299122195422,
      "learning_rate": 2.8557618430283845e-05,
      "loss": 2.8145,
      "step": 315
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.1825577738379651,
      "learning_rate": 2.7547936583472682e-05,
      "loss": 2.8135,
      "step": 316
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.1681442717695743,
      "learning_rate": 2.655538513033598e-05,
      "loss": 2.8174,
      "step": 317
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.1886334015299787,
      "learning_rate": 2.5580040504406736e-05,
      "loss": 2.8105,
      "step": 318
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.18472284304096284,
      "learning_rate": 2.4621977814169668e-05,
      "loss": 2.8076,
      "step": 319
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.16272769219517882,
      "learning_rate": 2.3681270837277365e-05,
      "loss": 2.8174,
      "step": 320
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.19968898028239582,
      "learning_rate": 2.2757992014868806e-05,
      "loss": 2.8242,
      "step": 321
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.17043048449106335,
      "learning_rate": 2.1852212445990894e-05,
      "loss": 2.8135,
      "step": 322
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.1644894369897433,
      "learning_rate": 2.0964001882123466e-05,
      "loss": 2.7998,
      "step": 323
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.16881143272811544,
      "learning_rate": 2.0093428721807694e-05,
      "loss": 2.8252,
      "step": 324
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.16555621080410832,
      "learning_rate": 1.924056000537905e-05,
      "loss": 2.7939,
      "step": 325
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.16784278084821697,
      "learning_rate": 1.8405461409804676e-05,
      "loss": 2.8242,
      "step": 326
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.1641280362316012,
      "learning_rate": 1.7588197243625816e-05,
      "loss": 2.8223,
      "step": 327
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.17345327239016983,
      "learning_rate": 1.6788830442005742e-05,
      "loss": 2.835,
      "step": 328
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.1678603553402573,
      "learning_rate": 1.600742256188287e-05,
      "loss": 2.8623,
      "step": 329
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.17995413273432045,
      "learning_rate": 1.5244033777230909e-05,
      "loss": 2.7979,
      "step": 330
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.17259350839299145,
      "learning_rate": 1.4498722874424859e-05,
      "loss": 2.8105,
      "step": 331
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.1741787680290243,
      "learning_rate": 1.3771547247713956e-05,
      "loss": 2.7979,
      "step": 332
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.1654260393599542,
      "learning_rate": 1.3062562894801955e-05,
      "loss": 2.7998,
      "step": 333
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.15975325006455904,
      "learning_rate": 1.237182441253501e-05,
      "loss": 2.8076,
      "step": 334
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.1831835291092947,
      "learning_rate": 1.169938499269721e-05,
      "loss": 2.833,
      "step": 335
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.17054992749801087,
      "learning_rate": 1.1045296417914408e-05,
      "loss": 2.8125,
      "step": 336
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.1914560652459072,
      "learning_rate": 1.040960905766683e-05,
      "loss": 2.8174,
      "step": 337
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.16125789541336907,
      "learning_rate": 9.792371864409844e-06,
      "loss": 2.8145,
      "step": 338
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.18343996045358885,
      "learning_rate": 9.19363236980475e-06,
      "loss": 2.8105,
      "step": 339
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.19474642229296382,
      "learning_rate": 8.613436681058128e-06,
      "loss": 2.7979,
      "step": 340
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.1541708717690255,
      "learning_rate": 8.051829477371458e-06,
      "loss": 2.8076,
      "step": 341
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.15274058388659703,
      "learning_rate": 7.508854006500393e-06,
      "loss": 2.8057,
      "step": 342
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.1778349122094768,
      "learning_rate": 6.984552081424456e-06,
      "loss": 2.8252,
      "step": 343
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.18535400361130824,
      "learning_rate": 6.4789640771270864e-06,
      "loss": 2.8506,
      "step": 344
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.17517598706340834,
      "learning_rate": 5.992128927486573e-06,
      "loss": 2.8047,
      "step": 345
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.1594767650558485,
      "learning_rate": 5.524084122277756e-06,
      "loss": 2.7871,
      "step": 346
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.17674478758195836,
      "learning_rate": 5.074865704284975e-06,
      "loss": 2.8242,
      "step": 347
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.1795205762704063,
      "learning_rate": 4.644508266526848e-06,
      "loss": 2.8037,
      "step": 348
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.16428283866890742,
      "learning_rate": 4.23304494959198e-06,
      "loss": 2.7852,
      "step": 349
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.1637667775135852,
      "learning_rate": 3.840507439087093e-06,
      "loss": 2.8252,
      "step": 350
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.1904128210839155,
      "learning_rate": 3.4669259631970873e-06,
      "loss": 2.8047,
      "step": 351
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.17893501780149162,
      "learning_rate": 3.1123292903569854e-06,
      "loss": 2.8115,
      "step": 352
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.17860019081470324,
      "learning_rate": 2.7767447270368453e-06,
      "loss": 2.8184,
      "step": 353
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.16537005007226285,
      "learning_rate": 2.4601981156388566e-06,
      "loss": 2.7988,
      "step": 354
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.1482593636447568,
      "learning_rate": 2.162713832507157e-06,
      "loss": 2.7979,
      "step": 355
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.17394817869790966,
      "learning_rate": 1.8843147860509445e-06,
      "loss": 2.7988,
      "step": 356
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.1728142685313597,
      "learning_rate": 1.625022414980165e-06,
      "loss": 2.792,
      "step": 357
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.17603370515257444,
      "learning_rate": 1.3848566866545842e-06,
      "loss": 2.7998,
      "step": 358
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.16591885986335803,
      "learning_rate": 1.1638360955463222e-06,
      "loss": 2.8135,
      "step": 359
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.17699991948505095,
      "learning_rate": 9.619776618154386e-07,
      "loss": 2.8076,
      "step": 360
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.1690937483022535,
      "learning_rate": 7.792969299994512e-07,
      "loss": 2.8438,
      "step": 361
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.18098457876096985,
      "learning_rate": 6.158079678161566e-07,
      "loss": 2.8408,
      "step": 362
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.18213680457305967,
      "learning_rate": 4.7152336508038407e-07,
      "loss": 2.8311,
      "step": 363
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.16488850320672738,
      "learning_rate": 3.464542327344111e-07,
      "loss": 2.7949,
      "step": 364
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.1606825195560946,
      "learning_rate": 2.406102019924805e-07,
      "loss": 2.8154,
      "step": 365
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.1743249393559749,
      "learning_rate": 1.5399942359897723e-07,
      "loss": 2.8174,
      "step": 366
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.1890660779935612,
      "learning_rate": 8.662856720090261e-08,
      "loss": 2.8359,
      "step": 367
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.17906338951545656,
      "learning_rate": 3.8502820834174226e-08,
      "loss": 2.7842,
      "step": 368
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.189945263964402,
      "learning_rate": 9.625890524084246e-09,
      "loss": 2.8037,
      "step": 369
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.16386427942822598,
      "learning_rate": 0.0,
      "loss": 2.8037,
      "step": 370
    },
    {
      "epoch": 1.0,
      "step": 370,
      "total_flos": 2.715544613577294e+18,
      "train_loss": 3.01240234375,
      "train_runtime": 27753.3023,
      "train_samples_per_second": 27.316,
      "train_steps_per_second": 0.013
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 370,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "total_flos": 2.715544613577294e+18,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}