{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 25000,
  "global_step": 125000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 0.6675467491149902,
      "learning_rate": 0.0002988,
      "loss": 0.2058,
      "step": 500
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.40607237815856934,
      "learning_rate": 0.00029759999999999997,
      "loss": 0.2096,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9344772100448608,
      "learning_rate": 0.0002964,
      "loss": 0.237,
      "step": 1500
    },
    {
      "epoch": 0.02,
      "grad_norm": 2.254749059677124,
      "learning_rate": 0.00029519999999999997,
      "loss": 0.2275,
      "step": 2000
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.8561826348304749,
      "learning_rate": 0.000294,
      "loss": 0.2355,
      "step": 2500
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.0209927558898926,
      "learning_rate": 0.00029279999999999996,
      "loss": 0.2275,
      "step": 3000
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.9906947612762451,
      "learning_rate": 0.0002916,
      "loss": 0.2188,
      "step": 3500
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.9113153219223022,
      "learning_rate": 0.00029039999999999996,
      "loss": 0.2272,
      "step": 4000
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.4784849882125854,
      "learning_rate": 0.0002892,
      "loss": 0.2437,
      "step": 4500
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.9436748623847961,
      "learning_rate": 0.00028799999999999995,
      "loss": 0.2378,
      "step": 5000
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.9457006454467773,
      "learning_rate": 0.0002868,
      "loss": 0.2244,
      "step": 5500
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.6395999193191528,
      "learning_rate": 0.00028559999999999995,
      "loss": 0.2251,
      "step": 6000
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.4920419156551361,
      "learning_rate": 0.0002844,
      "loss": 0.2249,
      "step": 6500
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.67146235704422,
      "learning_rate": 0.00028319999999999994,
      "loss": 0.2223,
      "step": 7000
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.5474592447280884,
      "learning_rate": 0.00028199999999999997,
      "loss": 0.2231,
      "step": 7500
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.2913522720336914,
      "learning_rate": 0.0002808,
      "loss": 0.2344,
      "step": 8000
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.6871858835220337,
      "learning_rate": 0.00027959999999999997,
      "loss": 0.2325,
      "step": 8500
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.6918925046920776,
      "learning_rate": 0.0002784,
      "loss": 0.2329,
      "step": 9000
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.9310051798820496,
      "learning_rate": 0.0002772,
      "loss": 0.2309,
      "step": 9500
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.6937859058380127,
      "learning_rate": 0.000276,
      "loss": 0.2362,
      "step": 10000
    },
    {
      "epoch": 0.08,
      "grad_norm": 3.504669666290283,
      "learning_rate": 0.0002748,
      "loss": 0.2295,
      "step": 10500
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.7235252261161804,
      "learning_rate": 0.0002736,
      "loss": 0.2286,
      "step": 11000
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.0144588947296143,
      "learning_rate": 0.0002724,
      "loss": 0.2302,
      "step": 11500
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.7903403043746948,
      "learning_rate": 0.0002712,
      "loss": 0.2266,
      "step": 12000
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.6691497564315796,
      "learning_rate": 0.00027,
      "loss": 0.2356,
      "step": 12500
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.6409358978271484,
      "learning_rate": 0.0002688,
      "loss": 0.2292,
      "step": 13000
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.7465757131576538,
      "learning_rate": 0.0002676,
      "loss": 0.2347,
      "step": 13500
    },
    {
      "epoch": 0.11,
      "grad_norm": 1.411289930343628,
      "learning_rate": 0.00026639999999999997,
      "loss": 0.2275,
      "step": 14000
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.5167527198791504,
      "learning_rate": 0.0002652,
      "loss": 0.2364,
      "step": 14500
    },
    {
      "epoch": 0.12,
      "grad_norm": 1.128173589706421,
      "learning_rate": 0.00026399999999999997,
      "loss": 0.2197,
      "step": 15000
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.8312121033668518,
      "learning_rate": 0.0002628,
      "loss": 0.236,
      "step": 15500
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.7007321715354919,
      "learning_rate": 0.00026159999999999996,
      "loss": 0.2229,
      "step": 16000
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.0870940685272217,
      "learning_rate": 0.0002604,
      "loss": 0.2201,
      "step": 16500
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.8884884715080261,
      "learning_rate": 0.00025919999999999996,
      "loss": 0.2178,
      "step": 17000
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.5293123126029968,
      "learning_rate": 0.000258,
      "loss": 0.2248,
      "step": 17500
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.8335347771644592,
      "learning_rate": 0.00025679999999999995,
      "loss": 0.2236,
      "step": 18000
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.7105595469474792,
      "learning_rate": 0.0002556,
      "loss": 0.2294,
      "step": 18500
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.0978660583496094,
      "learning_rate": 0.00025439999999999995,
      "loss": 0.2236,
      "step": 19000
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.7025924921035767,
      "learning_rate": 0.0002532,
      "loss": 0.219,
      "step": 19500
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.9559372067451477,
      "learning_rate": 0.00025199999999999995,
      "loss": 0.2147,
      "step": 20000
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.550178587436676,
      "learning_rate": 0.00025079999999999997,
      "loss": 0.2141,
      "step": 20500
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.0202014446258545,
      "learning_rate": 0.00024959999999999994,
      "loss": 0.2166,
      "step": 21000
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.5799145102500916,
      "learning_rate": 0.00024839999999999997,
      "loss": 0.2219,
      "step": 21500
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.9509211778640747,
      "learning_rate": 0.0002472,
      "loss": 0.2144,
      "step": 22000
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6535309553146362,
      "learning_rate": 0.00024599999999999996,
      "loss": 0.2189,
      "step": 22500
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.4835234880447388,
      "learning_rate": 0.0002448,
      "loss": 0.2105,
      "step": 23000
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.50291907787323,
      "learning_rate": 0.00024359999999999999,
      "loss": 0.2076,
      "step": 23500
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.6879478693008423,
      "learning_rate": 0.00024239999999999998,
      "loss": 0.2231,
      "step": 24000
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.3261083364486694,
      "learning_rate": 0.00024119999999999998,
      "loss": 0.2255,
      "step": 24500
    },
    {
      "epoch": 0.2,
      "grad_norm": 1.1070926189422607,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.229,
      "step": 25000
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.14503039419651031,
      "eval_runtime": 197.5245,
      "eval_samples_per_second": 101.253,
      "eval_steps_per_second": 25.313,
      "step": 25000
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.4342971742153168,
      "learning_rate": 0.0002388,
      "loss": 0.2178,
      "step": 25500
    },
    {
      "epoch": 0.21,
      "grad_norm": 4.339428901672363,
      "learning_rate": 0.0002376,
      "loss": 0.2107,
      "step": 26000
    },
    {
      "epoch": 0.21,
      "grad_norm": 3.38554310798645,
      "learning_rate": 0.0002364,
      "loss": 0.2227,
      "step": 26500
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.249841332435608,
      "learning_rate": 0.0002352,
      "loss": 0.2119,
      "step": 27000
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.1185526847839355,
      "learning_rate": 0.000234,
      "loss": 0.2085,
      "step": 27500
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.008499264717102,
      "learning_rate": 0.0002328,
      "loss": 0.214,
      "step": 28000
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.0491580963134766,
      "learning_rate": 0.0002316,
      "loss": 0.2086,
      "step": 28500
    },
    {
      "epoch": 0.23,
      "grad_norm": 1.0968835353851318,
      "learning_rate": 0.0002304,
      "loss": 0.2094,
      "step": 29000
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.4313267767429352,
      "learning_rate": 0.0002292,
      "loss": 0.2048,
      "step": 29500
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.5634821653366089,
      "learning_rate": 0.00022799999999999999,
      "loss": 0.211,
      "step": 30000
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.8901171684265137,
      "learning_rate": 0.00022679999999999998,
      "loss": 0.2123,
      "step": 30500
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.5377653241157532,
      "learning_rate": 0.00022559999999999998,
      "loss": 0.2097,
      "step": 31000
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.8237270712852478,
      "learning_rate": 0.00022439999999999998,
      "loss": 0.2066,
      "step": 31500
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5759382247924805,
      "learning_rate": 0.00022319999999999998,
      "loss": 0.2098,
      "step": 32000
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.7323036193847656,
      "learning_rate": 0.00022199999999999998,
      "loss": 0.2087,
      "step": 32500
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.8658179640769958,
      "learning_rate": 0.00022079999999999997,
      "loss": 0.2038,
      "step": 33000
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.8712558150291443,
      "learning_rate": 0.00021959999999999997,
      "loss": 0.2202,
      "step": 33500
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.005462884902954,
      "learning_rate": 0.00021839999999999997,
      "loss": 0.2052,
      "step": 34000
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.4281785786151886,
      "learning_rate": 0.00021719999999999997,
      "loss": 0.2049,
      "step": 34500
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5366469621658325,
      "learning_rate": 0.00021599999999999996,
      "loss": 0.2067,
      "step": 35000
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.9057306051254272,
      "learning_rate": 0.00021479999999999996,
      "loss": 0.2036,
      "step": 35500
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.9041026830673218,
      "learning_rate": 0.00021359999999999996,
      "loss": 0.2092,
      "step": 36000
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.8272420167922974,
      "learning_rate": 0.00021239999999999996,
      "loss": 0.2043,
      "step": 36500
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.302566409111023,
      "learning_rate": 0.00021119999999999996,
      "loss": 0.2247,
      "step": 37000
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.45851215720176697,
      "learning_rate": 0.00020999999999999998,
      "loss": 0.2119,
      "step": 37500
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.9625339508056641,
      "learning_rate": 0.00020879999999999998,
      "loss": 0.2036,
      "step": 38000
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.465646505355835,
      "learning_rate": 0.00020759999999999998,
      "loss": 0.1981,
      "step": 38500
    },
    {
      "epoch": 0.31,
      "grad_norm": 3.8826301097869873,
      "learning_rate": 0.00020639999999999998,
      "loss": 0.2008,
      "step": 39000
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.9201263785362244,
      "learning_rate": 0.0002052,
      "loss": 0.1959,
      "step": 39500
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.8154922723770142,
      "learning_rate": 0.000204,
      "loss": 0.2032,
      "step": 40000
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.0132577419281006,
      "learning_rate": 0.0002028,
      "loss": 0.2012,
      "step": 40500
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.34237733483314514,
      "learning_rate": 0.0002016,
      "loss": 0.2068,
      "step": 41000
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.5986165404319763,
      "learning_rate": 0.0002004,
      "loss": 0.204,
      "step": 41500
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.2128850221633911,
      "learning_rate": 0.0001992,
      "loss": 0.1969,
      "step": 42000
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.2548407316207886,
      "learning_rate": 0.000198,
      "loss": 0.2055,
      "step": 42500
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.9521819949150085,
      "learning_rate": 0.00019679999999999999,
      "loss": 0.2018,
      "step": 43000
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.6435508728027344,
      "learning_rate": 0.00019559999999999998,
      "loss": 0.2013,
      "step": 43500
    },
    {
      "epoch": 0.35,
      "grad_norm": 2.0959296226501465,
      "learning_rate": 0.00019439999999999998,
      "loss": 0.2032,
      "step": 44000
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.5862271189689636,
      "learning_rate": 0.00019319999999999998,
      "loss": 0.1961,
      "step": 44500
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.31843170523643494,
      "learning_rate": 0.00019199999999999998,
      "loss": 0.196,
      "step": 45000
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.3819515705108643,
      "learning_rate": 0.00019079999999999998,
      "loss": 0.1966,
      "step": 45500
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.03080415725708,
      "learning_rate": 0.00018959999999999997,
      "loss": 0.1985,
      "step": 46000
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.47353965044021606,
      "learning_rate": 0.00018839999999999997,
      "loss": 0.1928,
      "step": 46500
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.774486780166626,
      "learning_rate": 0.0001872,
      "loss": 0.1985,
      "step": 47000
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.371840000152588,
      "learning_rate": 0.000186,
      "loss": 0.2053,
      "step": 47500
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.9633282423019409,
      "learning_rate": 0.0001848,
      "loss": 0.1967,
      "step": 48000
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.7216082215309143,
      "learning_rate": 0.0001836,
      "loss": 0.1935,
      "step": 48500
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.8321841955184937,
      "learning_rate": 0.0001824,
      "loss": 0.1909,
      "step": 49000
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.924881637096405,
      "learning_rate": 0.00018119999999999999,
      "loss": 0.1914,
      "step": 49500
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.1322112083435059,
      "learning_rate": 0.00017999999999999998,
      "loss": 0.1823,
      "step": 50000
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.1376880556344986,
      "eval_runtime": 197.1453,
      "eval_samples_per_second": 101.448,
      "eval_steps_per_second": 25.362,
      "step": 50000
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5036086440086365,
      "learning_rate": 0.00017879999999999998,
      "loss": 0.19,
      "step": 50500
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.8745198249816895,
      "learning_rate": 0.00017759999999999998,
      "loss": 0.1952,
      "step": 51000
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.9014063477516174,
      "learning_rate": 0.00017639999999999998,
      "loss": 0.1825,
      "step": 51500
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.5580583214759827,
      "learning_rate": 0.00017519999999999998,
      "loss": 0.1958,
      "step": 52000
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.6560271382331848,
      "learning_rate": 0.00017399999999999997,
      "loss": 0.1986,
      "step": 52500
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.9042519330978394,
      "learning_rate": 0.00017279999999999997,
      "loss": 0.2037,
      "step": 53000
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.7758679986000061,
      "learning_rate": 0.00017159999999999997,
      "loss": 0.1891,
      "step": 53500
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.9086706042289734,
      "learning_rate": 0.00017039999999999997,
      "loss": 0.1864,
      "step": 54000
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.0545058250427246,
      "learning_rate": 0.00016919999999999997,
      "loss": 0.199,
      "step": 54500
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.20504042506217957,
      "learning_rate": 0.000168,
      "loss": 0.1902,
      "step": 55000
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.7427380084991455,
      "learning_rate": 0.0001668,
      "loss": 0.192,
      "step": 55500
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.42541056871414185,
      "learning_rate": 0.0001656,
      "loss": 0.1939,
      "step": 56000
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.1101672649383545,
      "learning_rate": 0.0001644,
      "loss": 0.195,
      "step": 56500
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.7569801807403564,
      "learning_rate": 0.0001632,
      "loss": 0.1838,
      "step": 57000
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.2808530032634735,
      "learning_rate": 0.000162,
      "loss": 0.1948,
      "step": 57500
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.3836987018585205,
      "learning_rate": 0.0001608,
      "loss": 0.1908,
      "step": 58000
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.8910863995552063,
      "learning_rate": 0.0001596,
      "loss": 0.1916,
      "step": 58500
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.40731602907180786,
      "learning_rate": 0.0001584,
      "loss": 0.178,
      "step": 59000
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.8736116886138916,
      "learning_rate": 0.0001572,
      "loss": 0.1932,
      "step": 59500
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.457201600074768,
      "learning_rate": 0.000156,
      "loss": 0.1815,
      "step": 60000
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.6906099319458008,
      "learning_rate": 0.0001548,
      "loss": 0.1923,
      "step": 60500
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.0163846015930176,
      "learning_rate": 0.0001536,
      "loss": 0.1843,
      "step": 61000
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.36051931977272034,
      "learning_rate": 0.0001524,
      "loss": 0.1722,
      "step": 61500
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.3554840087890625,
      "learning_rate": 0.0001512,
      "loss": 0.1868,
      "step": 62000
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.4623011350631714,
      "learning_rate": 0.00015,
      "loss": 0.1833,
      "step": 62500
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.31331172585487366,
      "learning_rate": 0.00014879999999999998,
      "loss": 0.1826,
      "step": 63000
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.1543946266174316,
      "learning_rate": 0.00014759999999999998,
      "loss": 0.181,
      "step": 63500
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.758339524269104,
      "learning_rate": 0.00014639999999999998,
      "loss": 0.1813,
      "step": 64000
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.7100640535354614,
      "learning_rate": 0.00014519999999999998,
      "loss": 0.1822,
      "step": 64500
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.7758503556251526,
      "learning_rate": 0.00014399999999999998,
      "loss": 0.1809,
      "step": 65000
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.6668974161148071,
      "learning_rate": 0.00014279999999999997,
      "loss": 0.1888,
      "step": 65500
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.6467080116271973,
      "learning_rate": 0.00014159999999999997,
      "loss": 0.1809,
      "step": 66000
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.47351643443107605,
      "learning_rate": 0.0001404,
      "loss": 0.1822,
      "step": 66500
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.6044811606407166,
      "learning_rate": 0.0001392,
      "loss": 0.1822,
      "step": 67000
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.485547423362732,
      "learning_rate": 0.000138,
      "loss": 0.1789,
      "step": 67500
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.9173816442489624,
      "learning_rate": 0.0001368,
      "loss": 0.184,
      "step": 68000
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.2925781011581421,
      "learning_rate": 0.0001356,
      "loss": 0.183,
      "step": 68500
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.29401320219039917,
      "learning_rate": 0.0001344,
      "loss": 0.1799,
      "step": 69000
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.5869783163070679,
      "learning_rate": 0.00013319999999999999,
      "loss": 0.1768,
      "step": 69500
    },
    {
      "epoch": 0.56,
      "grad_norm": 1.3995858430862427,
      "learning_rate": 0.00013199999999999998,
      "loss": 0.1848,
      "step": 70000
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.703525185585022,
      "learning_rate": 0.00013079999999999998,
      "loss": 0.1734,
      "step": 70500
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.8634682893753052,
      "learning_rate": 0.00012959999999999998,
      "loss": 0.1789,
      "step": 71000
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.5816720724105835,
      "learning_rate": 0.00012839999999999998,
      "loss": 0.1772,
      "step": 71500
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.1784312725067139,
      "learning_rate": 0.00012719999999999997,
      "loss": 0.1759,
      "step": 72000
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.4554961025714874,
      "learning_rate": 0.00012599999999999997,
      "loss": 0.184,
      "step": 72500
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.8070880174636841,
      "learning_rate": 0.00012479999999999997,
      "loss": 0.1813,
      "step": 73000
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.8661233186721802,
      "learning_rate": 0.0001236,
      "loss": 0.1733,
      "step": 73500
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.788370668888092,
      "learning_rate": 0.0001224,
      "loss": 0.1788,
      "step": 74000
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.25436270236969,
      "learning_rate": 0.00012119999999999999,
      "loss": 0.1808,
      "step": 74500
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.562129020690918,
      "learning_rate": 0.00011999999999999999,
      "loss": 0.1757,
      "step": 75000
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.1267654001712799,
      "eval_runtime": 198.6188,
      "eval_samples_per_second": 100.695,
      "eval_steps_per_second": 25.174,
      "step": 75000
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.4803348481655121,
      "learning_rate": 0.0001188,
      "loss": 0.1827,
      "step": 75500
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.7015312910079956,
      "learning_rate": 0.0001176,
      "loss": 0.1731,
      "step": 76000
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.7549095153808594,
      "learning_rate": 0.0001164,
      "loss": 0.1815,
      "step": 76500
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.0105984210968018,
      "learning_rate": 0.0001152,
      "loss": 0.1695,
      "step": 77000
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.8554935455322266,
      "learning_rate": 0.00011399999999999999,
      "loss": 0.1767,
      "step": 77500
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.7114706039428711,
      "learning_rate": 0.00011279999999999999,
      "loss": 0.1667,
      "step": 78000
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.6958938241004944,
      "learning_rate": 0.00011159999999999999,
      "loss": 0.1765,
      "step": 78500
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.3407191336154938,
      "learning_rate": 0.00011039999999999999,
      "loss": 0.1834,
      "step": 79000
    },
    {
      "epoch": 0.64,
      "grad_norm": 1.1113251447677612,
      "learning_rate": 0.00010919999999999998,
      "loss": 0.173,
      "step": 79500
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.9301660060882568,
      "learning_rate": 0.00010799999999999998,
      "loss": 0.1731,
      "step": 80000
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.45441198348999023,
      "learning_rate": 0.00010679999999999998,
      "loss": 0.1747,
      "step": 80500
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.413922518491745,
      "learning_rate": 0.00010559999999999998,
      "loss": 0.1783,
      "step": 81000
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.6205990314483643,
      "learning_rate": 0.00010439999999999999,
      "loss": 0.174,
      "step": 81500
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.4416259527206421,
      "learning_rate": 0.00010319999999999999,
      "loss": 0.1773,
      "step": 82000
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.6363853216171265,
      "learning_rate": 0.000102,
      "loss": 0.1781,
      "step": 82500
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.5858229398727417,
      "learning_rate": 0.0001008,
      "loss": 0.1675,
      "step": 83000
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.49044686555862427,
      "learning_rate": 9.96e-05,
      "loss": 0.1672,
      "step": 83500
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.8010206818580627,
      "learning_rate": 9.839999999999999e-05,
      "loss": 0.1658,
      "step": 84000
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.1232587099075317,
      "learning_rate": 9.719999999999999e-05,
      "loss": 0.171,
      "step": 84500
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.5122153759002686,
      "learning_rate": 9.599999999999999e-05,
      "loss": 0.1768,
      "step": 85000
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.524728775024414,
      "learning_rate": 9.479999999999999e-05,
      "loss": 0.1741,
      "step": 85500
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.6309654116630554,
      "learning_rate": 9.36e-05,
      "loss": 0.168,
      "step": 86000
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.7848625183105469,
      "learning_rate": 9.24e-05,
      "loss": 0.1727,
      "step": 86500
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.2840665578842163,
      "learning_rate": 9.12e-05,
      "loss": 0.1681,
      "step": 87000
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.561613917350769,
      "learning_rate": 8.999999999999999e-05,
      "loss": 0.174,
      "step": 87500
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.3210696578025818,
      "learning_rate": 8.879999999999999e-05,
      "loss": 0.1665,
      "step": 88000
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.6899012327194214,
      "learning_rate": 8.759999999999999e-05,
      "loss": 0.1713,
      "step": 88500
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.5571711659431458,
      "learning_rate": 8.639999999999999e-05,
      "loss": 0.1742,
      "step": 89000
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.3232167065143585,
      "learning_rate": 8.519999999999998e-05,
      "loss": 0.1642,
      "step": 89500
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5947475433349609,
      "learning_rate": 8.4e-05,
      "loss": 0.1607,
      "step": 90000
    },
    {
      "epoch": 0.72,
      "grad_norm": 1.0395877361297607,
      "learning_rate": 8.28e-05,
      "loss": 0.1668,
      "step": 90500
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.8020333051681519,
      "learning_rate": 8.16e-05,
      "loss": 0.1705,
      "step": 91000
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.420878142118454,
      "learning_rate": 8.04e-05,
      "loss": 0.1495,
      "step": 91500
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.2120175063610077,
      "learning_rate": 7.92e-05,
      "loss": 0.1684,
      "step": 92000
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.321204662322998,
      "learning_rate": 7.8e-05,
      "loss": 0.1677,
      "step": 92500
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.5741947889328003,
      "learning_rate": 7.68e-05,
      "loss": 0.1643,
      "step": 93000
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.2746628224849701,
      "learning_rate": 7.56e-05,
      "loss": 0.169,
      "step": 93500
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.5869113206863403,
      "learning_rate": 7.439999999999999e-05,
      "loss": 0.158,
      "step": 94000
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.5085694789886475,
      "learning_rate": 7.319999999999999e-05,
      "loss": 0.1766,
      "step": 94500
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.8071849942207336,
      "learning_rate": 7.199999999999999e-05,
      "loss": 0.1623,
      "step": 95000
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.6275350451469421,
      "learning_rate": 7.079999999999999e-05,
      "loss": 0.1624,
      "step": 95500
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.35898515582084656,
      "learning_rate": 6.96e-05,
      "loss": 0.1659,
      "step": 96000
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.5679383873939514,
      "learning_rate": 6.84e-05,
      "loss": 0.1621,
      "step": 96500
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.2387058734893799,
      "learning_rate": 6.72e-05,
      "loss": 0.1603,
      "step": 97000
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.4777892529964447,
      "learning_rate": 6.599999999999999e-05,
      "loss": 0.1584,
      "step": 97500
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.789397120475769,
      "learning_rate": 6.479999999999999e-05,
      "loss": 0.1621,
      "step": 98000
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.8871356248855591,
      "learning_rate": 6.359999999999999e-05,
      "loss": 0.1559,
      "step": 98500
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.0920441150665283,
      "learning_rate": 6.239999999999999e-05,
      "loss": 0.169,
      "step": 99000
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.3875184059143066,
      "learning_rate": 6.12e-05,
      "loss": 0.1623,
      "step": 99500
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.1163883209228516,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 0.1663,
      "step": 100000
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.1176806390285492,
      "eval_runtime": 202.7443,
      "eval_samples_per_second": 98.646,
      "eval_steps_per_second": 24.662,
      "step": 100000
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.8637445569038391,
      "learning_rate": 5.88e-05,
      "loss": 0.1663,
      "step": 100500
    },
    {
      "epoch": 0.81,
      "grad_norm": 3.794600486755371,
      "learning_rate": 5.76e-05,
      "loss": 0.1638,
      "step": 101000
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.7344540953636169,
      "learning_rate": 5.6399999999999995e-05,
      "loss": 0.1589,
      "step": 101500
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.591451108455658,
      "learning_rate": 5.519999999999999e-05,
      "loss": 0.1613,
      "step": 102000
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.9407438635826111,
      "learning_rate": 5.399999999999999e-05,
      "loss": 0.1588,
      "step": 102500
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.5834544897079468,
      "learning_rate": 5.279999999999999e-05,
      "loss": 0.1708,
      "step": 103000
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.6752551198005676,
      "learning_rate": 5.1599999999999994e-05,
      "loss": 0.1548,
      "step": 103500
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.6357858180999756,
      "learning_rate": 5.04e-05,
      "loss": 0.1641,
      "step": 104000
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.6341969966888428,
      "learning_rate": 4.9199999999999997e-05,
      "loss": 0.162,
      "step": 104500
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3042418956756592,
      "learning_rate": 4.7999999999999994e-05,
      "loss": 0.1519,
      "step": 105000
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.0411969423294067,
      "learning_rate": 4.68e-05,
      "loss": 0.1552,
      "step": 105500
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.1914784461259842,
      "learning_rate": 4.56e-05,
      "loss": 0.1614,
      "step": 106000
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.5716798305511475,
      "learning_rate": 4.4399999999999995e-05,
      "loss": 0.1639,
      "step": 106500
    },
    {
      "epoch": 0.86,
      "grad_norm": 1.2206032276153564,
      "learning_rate": 4.319999999999999e-05,
      "loss": 0.1504,
      "step": 107000
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.9124048352241516,
      "learning_rate": 4.2e-05,
      "loss": 0.1575,
      "step": 107500
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.4684576392173767,
      "learning_rate": 4.08e-05,
      "loss": 0.1678,
      "step": 108000
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4757069945335388,
      "learning_rate": 3.96e-05,
      "loss": 0.1609,
      "step": 108500
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.4235193431377411,
      "learning_rate": 3.84e-05,
      "loss": 0.1633,
      "step": 109000
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.6009053587913513,
      "learning_rate": 3.7199999999999996e-05,
      "loss": 0.1528,
      "step": 109500
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.9195963144302368,
      "learning_rate": 3.5999999999999994e-05,
      "loss": 0.1616,
      "step": 110000
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.5454314947128296,
      "learning_rate": 3.48e-05,
      "loss": 0.1611,
      "step": 110500
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.5214249491691589,
      "learning_rate": 3.36e-05,
      "loss": 0.1586,
      "step": 111000
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.0026371479034424,
      "learning_rate": 3.2399999999999995e-05,
      "loss": 0.1641,
      "step": 111500
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.3591657876968384,
      "learning_rate": 3.119999999999999e-05,
      "loss": 0.1599,
      "step": 112000
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.6785110831260681,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 0.1585,
      "step": 112500
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.6680885553359985,
      "learning_rate": 2.88e-05,
      "loss": 0.1534,
      "step": 113000
    },
    {
      "epoch": 0.91,
      "grad_norm": 1.4302424192428589,
      "learning_rate": 2.7599999999999997e-05,
      "loss": 0.1603,
      "step": 113500
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.9608228206634521,
      "learning_rate": 2.6399999999999995e-05,
      "loss": 0.1509,
      "step": 114000
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.16929888725280762,
      "learning_rate": 2.52e-05,
      "loss": 0.1576,
      "step": 114500
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.19129659235477448,
      "learning_rate": 2.3999999999999997e-05,
      "loss": 0.1556,
      "step": 115000
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.7054457068443298,
      "learning_rate": 2.28e-05,
      "loss": 0.1615,
      "step": 115500
    },
    {
      "epoch": 0.93,
      "grad_norm": 1.2099674940109253,
      "learning_rate": 2.1599999999999996e-05,
      "loss": 0.152,
      "step": 116000
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.6213043332099915,
      "learning_rate": 2.04e-05,
      "loss": 0.159,
      "step": 116500
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5257019996643066,
      "learning_rate": 1.92e-05,
      "loss": 0.1643,
      "step": 117000
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.7175723314285278,
      "learning_rate": 1.7999999999999997e-05,
      "loss": 0.1561,
      "step": 117500
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.37755417823791504,
      "learning_rate": 1.68e-05,
      "loss": 0.1594,
      "step": 118000
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.6101952195167542,
      "learning_rate": 1.5599999999999996e-05,
      "loss": 0.1574,
      "step": 118500
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.0712097883224487,
      "learning_rate": 1.44e-05,
      "loss": 0.1468,
      "step": 119000
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5005315542221069,
      "learning_rate": 1.3199999999999997e-05,
      "loss": 0.1586,
      "step": 119500
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.7147447466850281,
      "learning_rate": 1.1999999999999999e-05,
      "loss": 0.1553,
      "step": 120000
    },
    {
      "epoch": 0.96,
      "grad_norm": 1.2048006057739258,
      "learning_rate": 1.0799999999999998e-05,
      "loss": 0.1603,
      "step": 120500
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.6054685711860657,
      "learning_rate": 9.6e-06,
      "loss": 0.161,
      "step": 121000
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.4145253300666809,
      "learning_rate": 8.4e-06,
      "loss": 0.1549,
      "step": 121500
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.2331937849521637,
      "learning_rate": 7.2e-06,
      "loss": 0.1584,
      "step": 122000
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.44909703731536865,
      "learning_rate": 5.999999999999999e-06,
      "loss": 0.161,
      "step": 122500
    },
    {
      "epoch": 0.98,
      "grad_norm": 6.39445686340332,
      "learning_rate": 4.8e-06,
      "loss": 0.159,
      "step": 123000
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.5847737193107605,
      "learning_rate": 3.6e-06,
      "loss": 0.1469,
      "step": 123500
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.45996806025505066,
      "learning_rate": 2.4e-06,
      "loss": 0.1606,
      "step": 124000
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.5662907361984253,
      "learning_rate": 1.2e-06,
      "loss": 0.1511,
      "step": 124500
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.9263651967048645,
      "learning_rate": 0.0,
      "loss": 0.1543,
      "step": 125000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.11344572901725769,
      "eval_runtime": 200.0914,
      "eval_samples_per_second": 99.954,
      "eval_steps_per_second": 24.989,
      "step": 125000
    }
  ],
  "logging_steps": 500,
  "max_steps": 125000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 62500,
  "total_flos": 6.527143521828864e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}