{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 8.0,
  "global_step": 76416,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "learning_rate": 2.9803706030150755e-05,
      "loss": 1.2294,
      "step": 500
    },
    {
      "epoch": 0.1,
      "learning_rate": 2.960741206030151e-05,
      "loss": 0.9625,
      "step": 1000
    },
    {
      "epoch": 0.16,
      "learning_rate": 2.9411118090452263e-05,
      "loss": 0.9145,
      "step": 1500
    },
    {
      "epoch": 0.21,
      "learning_rate": 2.9214824120603013e-05,
      "loss": 0.8595,
      "step": 2000
    },
    {
      "epoch": 0.26,
      "learning_rate": 2.901853015075377e-05,
      "loss": 0.8622,
      "step": 2500
    },
    {
      "epoch": 0.31,
      "learning_rate": 2.8822236180904525e-05,
      "loss": 0.8504,
      "step": 3000
    },
    {
      "epoch": 0.37,
      "learning_rate": 2.8625942211055275e-05,
      "loss": 0.8308,
      "step": 3500
    },
    {
      "epoch": 0.42,
      "learning_rate": 2.842964824120603e-05,
      "loss": 0.8292,
      "step": 4000
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.8233354271356787e-05,
      "loss": 0.8085,
      "step": 4500
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.8037060301507537e-05,
      "loss": 0.7914,
      "step": 5000
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.784076633165829e-05,
      "loss": 0.8014,
      "step": 5500
    },
    {
      "epoch": 0.63,
      "learning_rate": 2.7644472361809045e-05,
      "loss": 0.7876,
      "step": 6000
    },
    {
      "epoch": 0.68,
      "learning_rate": 2.74481783919598e-05,
      "loss": 0.7936,
      "step": 6500
    },
    {
      "epoch": 0.73,
      "learning_rate": 2.7251884422110553e-05,
      "loss": 0.7975,
      "step": 7000
    },
    {
      "epoch": 0.79,
      "learning_rate": 2.7055590452261307e-05,
      "loss": 0.7748,
      "step": 7500
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.685929648241206e-05,
      "loss": 0.7804,
      "step": 8000
    },
    {
      "epoch": 0.89,
      "learning_rate": 2.6663002512562815e-05,
      "loss": 0.773,
      "step": 8500
    },
    {
      "epoch": 0.94,
      "learning_rate": 2.646670854271357e-05,
      "loss": 0.7652,
      "step": 9000
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.6270414572864323e-05,
      "loss": 0.7727,
      "step": 9500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.8350390195846558,
      "eval_runtime": 8.4552,
      "eval_samples_per_second": 1312.323,
      "eval_steps_per_second": 41.04,
      "step": 9552
    },
    {
      "epoch": 1.05,
      "learning_rate": 2.6074120603015074e-05,
      "loss": 0.7397,
      "step": 10000
    },
    {
      "epoch": 1.1,
      "learning_rate": 2.587782663316583e-05,
      "loss": 0.7335,
      "step": 10500
    },
    {
      "epoch": 1.15,
      "learning_rate": 2.5681532663316585e-05,
      "loss": 0.7095,
      "step": 11000
    },
    {
      "epoch": 1.2,
      "learning_rate": 2.5485238693467336e-05,
      "loss": 0.7407,
      "step": 11500
    },
    {
      "epoch": 1.26,
      "learning_rate": 2.528894472361809e-05,
      "loss": 0.7235,
      "step": 12000
    },
    {
      "epoch": 1.31,
      "learning_rate": 2.5092650753768844e-05,
      "loss": 0.709,
      "step": 12500
    },
    {
      "epoch": 1.36,
      "learning_rate": 2.4896356783919598e-05,
      "loss": 0.7161,
      "step": 13000
    },
    {
      "epoch": 1.41,
      "learning_rate": 2.4700062814070352e-05,
      "loss": 0.7278,
      "step": 13500
    },
    {
      "epoch": 1.47,
      "learning_rate": 2.4503768844221106e-05,
      "loss": 0.74,
      "step": 14000
    },
    {
      "epoch": 1.52,
      "learning_rate": 2.430747487437186e-05,
      "loss": 0.7323,
      "step": 14500
    },
    {
      "epoch": 1.57,
      "learning_rate": 2.4111180904522614e-05,
      "loss": 0.7298,
      "step": 15000
    },
    {
      "epoch": 1.62,
      "learning_rate": 2.3914886934673368e-05,
      "loss": 0.7394,
      "step": 15500
    },
    {
      "epoch": 1.68,
      "learning_rate": 2.3718592964824122e-05,
      "loss": 0.7393,
      "step": 16000
    },
    {
      "epoch": 1.73,
      "learning_rate": 2.3522298994974873e-05,
      "loss": 0.715,
      "step": 16500
    },
    {
      "epoch": 1.78,
      "learning_rate": 2.332600502512563e-05,
      "loss": 0.7139,
      "step": 17000
    },
    {
      "epoch": 1.83,
      "learning_rate": 2.3129711055276384e-05,
      "loss": 0.7215,
      "step": 17500
    },
    {
      "epoch": 1.88,
      "learning_rate": 2.2933417085427135e-05,
      "loss": 0.7122,
      "step": 18000
    },
    {
      "epoch": 1.94,
      "learning_rate": 2.273712311557789e-05,
      "loss": 0.7119,
      "step": 18500
    },
    {
      "epoch": 1.99,
      "learning_rate": 2.2540829145728646e-05,
      "loss": 0.7113,
      "step": 19000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8123675584793091,
      "eval_runtime": 8.2517,
      "eval_samples_per_second": 1344.695,
      "eval_steps_per_second": 42.052,
      "step": 19104
    },
    {
      "epoch": 2.04,
      "learning_rate": 2.2344535175879397e-05,
      "loss": 0.6898,
      "step": 19500
    },
    {
      "epoch": 2.09,
      "learning_rate": 2.214824120603015e-05,
      "loss": 0.6819,
      "step": 20000
    },
    {
      "epoch": 2.15,
      "learning_rate": 2.1951947236180905e-05,
      "loss": 0.6797,
      "step": 20500
    },
    {
      "epoch": 2.2,
      "learning_rate": 2.175565326633166e-05,
      "loss": 0.6808,
      "step": 21000
    },
    {
      "epoch": 2.25,
      "learning_rate": 2.1559359296482413e-05,
      "loss": 0.6949,
      "step": 21500
    },
    {
      "epoch": 2.3,
      "learning_rate": 2.1363065326633167e-05,
      "loss": 0.689,
      "step": 22000
    },
    {
      "epoch": 2.36,
      "learning_rate": 2.116677135678392e-05,
      "loss": 0.6993,
      "step": 22500
    },
    {
      "epoch": 2.41,
      "learning_rate": 2.0970477386934675e-05,
      "loss": 0.6795,
      "step": 23000
    },
    {
      "epoch": 2.46,
      "learning_rate": 2.077418341708543e-05,
      "loss": 0.6972,
      "step": 23500
    },
    {
      "epoch": 2.51,
      "learning_rate": 2.0577889447236183e-05,
      "loss": 0.6831,
      "step": 24000
    },
    {
      "epoch": 2.56,
      "learning_rate": 2.0381595477386933e-05,
      "loss": 0.678,
      "step": 24500
    },
    {
      "epoch": 2.62,
      "learning_rate": 2.0185301507537687e-05,
      "loss": 0.6871,
      "step": 25000
    },
    {
      "epoch": 2.67,
      "learning_rate": 1.9989007537688445e-05,
      "loss": 0.6793,
      "step": 25500
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.9792713567839195e-05,
      "loss": 0.6808,
      "step": 26000
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.959641959798995e-05,
      "loss": 0.6894,
      "step": 26500
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.9400125628140703e-05,
      "loss": 0.6876,
      "step": 27000
    },
    {
      "epoch": 2.88,
      "learning_rate": 1.9203831658291457e-05,
      "loss": 0.6919,
      "step": 27500
    },
    {
      "epoch": 2.93,
      "learning_rate": 1.900753768844221e-05,
      "loss": 0.6856,
      "step": 28000
    },
    {
      "epoch": 2.98,
      "learning_rate": 1.8811243718592965e-05,
      "loss": 0.6708,
      "step": 28500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.7998350262641907,
      "eval_runtime": 8.2332,
      "eval_samples_per_second": 1347.713,
      "eval_steps_per_second": 42.146,
      "step": 28656
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.861494974874372e-05,
      "loss": 0.6758,
      "step": 29000
    },
    {
      "epoch": 3.09,
      "learning_rate": 1.8418655778894473e-05,
      "loss": 0.6638,
      "step": 29500
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.8222361809045227e-05,
      "loss": 0.654,
      "step": 30000
    },
    {
      "epoch": 3.19,
      "learning_rate": 1.802606783919598e-05,
      "loss": 0.6792,
      "step": 30500
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.7829773869346732e-05,
      "loss": 0.6999,
      "step": 31000
    },
    {
      "epoch": 3.3,
      "learning_rate": 1.763347989949749e-05,
      "loss": 0.6834,
      "step": 31500
    },
    {
      "epoch": 3.35,
      "learning_rate": 1.7437185929648243e-05,
      "loss": 0.647,
      "step": 32000
    },
    {
      "epoch": 3.4,
      "learning_rate": 1.7240891959798994e-05,
      "loss": 0.662,
      "step": 32500
    },
    {
      "epoch": 3.45,
      "learning_rate": 1.7044597989949748e-05,
      "loss": 0.6674,
      "step": 33000
    },
    {
      "epoch": 3.51,
      "learning_rate": 1.6848304020100505e-05,
      "loss": 0.6409,
      "step": 33500
    },
    {
      "epoch": 3.56,
      "learning_rate": 1.6652010050251256e-05,
      "loss": 0.6766,
      "step": 34000
    },
    {
      "epoch": 3.61,
      "learning_rate": 1.645571608040201e-05,
      "loss": 0.6658,
      "step": 34500
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.6259422110552764e-05,
      "loss": 0.6765,
      "step": 35000
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.6063128140703518e-05,
      "loss": 0.6478,
      "step": 35500
    },
    {
      "epoch": 3.77,
      "learning_rate": 1.5866834170854272e-05,
      "loss": 0.6604,
      "step": 36000
    },
    {
      "epoch": 3.82,
      "learning_rate": 1.5670540201005026e-05,
      "loss": 0.6398,
      "step": 36500
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.547424623115578e-05,
      "loss": 0.6554,
      "step": 37000
    },
    {
      "epoch": 3.93,
      "learning_rate": 1.527795226130653e-05,
      "loss": 0.6442,
      "step": 37500
    },
    {
      "epoch": 3.98,
      "learning_rate": 1.5081658291457286e-05,
      "loss": 0.6552,
      "step": 38000
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.793069064617157,
      "eval_runtime": 8.0925,
      "eval_samples_per_second": 1371.141,
      "eval_steps_per_second": 42.879,
      "step": 38208
    },
    {
      "epoch": 4.03,
      "learning_rate": 1.488536432160804e-05,
      "loss": 0.6526,
      "step": 38500
    },
    {
      "epoch": 4.08,
      "learning_rate": 1.4689070351758794e-05,
      "loss": 0.656,
      "step": 39000
    },
    {
      "epoch": 4.14,
      "learning_rate": 1.4492776381909548e-05,
      "loss": 0.6499,
      "step": 39500
    },
    {
      "epoch": 4.19,
      "learning_rate": 1.4296482412060302e-05,
      "loss": 0.6319,
      "step": 40000
    },
    {
      "epoch": 4.24,
      "learning_rate": 1.4100188442211055e-05,
      "loss": 0.6511,
      "step": 40500
    },
    {
      "epoch": 4.29,
      "learning_rate": 1.390389447236181e-05,
      "loss": 0.6461,
      "step": 41000
    },
    {
      "epoch": 4.34,
      "learning_rate": 1.3707600502512563e-05,
      "loss": 0.6472,
      "step": 41500
    },
    {
      "epoch": 4.4,
      "learning_rate": 1.3511306532663317e-05,
      "loss": 0.6334,
      "step": 42000
    },
    {
      "epoch": 4.45,
      "learning_rate": 1.331501256281407e-05,
      "loss": 0.6302,
      "step": 42500
    },
    {
      "epoch": 4.5,
      "learning_rate": 1.3118718592964825e-05,
      "loss": 0.6584,
      "step": 43000
    },
    {
      "epoch": 4.55,
      "learning_rate": 1.2922424623115579e-05,
      "loss": 0.635,
      "step": 43500
    },
    {
      "epoch": 4.61,
      "learning_rate": 1.2726130653266331e-05,
      "loss": 0.6305,
      "step": 44000
    },
    {
      "epoch": 4.66,
      "learning_rate": 1.2529836683417085e-05,
      "loss": 0.6448,
      "step": 44500
    },
    {
      "epoch": 4.71,
      "learning_rate": 1.233354271356784e-05,
      "loss": 0.6546,
      "step": 45000
    },
    {
      "epoch": 4.76,
      "learning_rate": 1.2137248743718593e-05,
      "loss": 0.6426,
      "step": 45500
    },
    {
      "epoch": 4.82,
      "learning_rate": 1.1940954773869347e-05,
      "loss": 0.6405,
      "step": 46000
    },
    {
      "epoch": 4.87,
      "learning_rate": 1.1744660804020101e-05,
      "loss": 0.6207,
      "step": 46500
    },
    {
      "epoch": 4.92,
      "learning_rate": 1.1548366834170855e-05,
      "loss": 0.6319,
      "step": 47000
    },
    {
      "epoch": 4.97,
      "learning_rate": 1.1352072864321609e-05,
      "loss": 0.6419,
      "step": 47500
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.7893060445785522,
      "eval_runtime": 8.4012,
      "eval_samples_per_second": 1320.763,
      "eval_steps_per_second": 41.304,
      "step": 47760
    },
    {
      "epoch": 5.03,
      "learning_rate": 1.1155778894472361e-05,
      "loss": 0.6382,
      "step": 48000
    },
    {
      "epoch": 5.08,
      "learning_rate": 1.0959484924623115e-05,
      "loss": 0.6264,
      "step": 48500
    },
    {
      "epoch": 5.13,
      "learning_rate": 1.076319095477387e-05,
      "loss": 0.627,
      "step": 49000
    },
    {
      "epoch": 5.18,
      "learning_rate": 1.0566896984924623e-05,
      "loss": 0.6284,
      "step": 49500
    },
    {
      "epoch": 5.23,
      "learning_rate": 1.0370603015075377e-05,
      "loss": 0.6256,
      "step": 50000
    },
    {
      "epoch": 5.29,
      "learning_rate": 1.0174309045226131e-05,
      "loss": 0.6423,
      "step": 50500
    },
    {
      "epoch": 5.34,
      "learning_rate": 9.978015075376884e-06,
      "loss": 0.634,
      "step": 51000
    },
    {
      "epoch": 5.39,
      "learning_rate": 9.78172110552764e-06,
      "loss": 0.6393,
      "step": 51500
    },
    {
      "epoch": 5.44,
      "learning_rate": 9.585427135678392e-06,
      "loss": 0.6259,
      "step": 52000
    },
    {
      "epoch": 5.5,
      "learning_rate": 9.389133165829146e-06,
      "loss": 0.628,
      "step": 52500
    },
    {
      "epoch": 5.55,
      "learning_rate": 9.1928391959799e-06,
      "loss": 0.6337,
      "step": 53000
    },
    {
      "epoch": 5.6,
      "learning_rate": 8.996545226130654e-06,
      "loss": 0.6128,
      "step": 53500
    },
    {
      "epoch": 5.65,
      "learning_rate": 8.800251256281408e-06,
      "loss": 0.6276,
      "step": 54000
    },
    {
      "epoch": 5.71,
      "learning_rate": 8.60395728643216e-06,
      "loss": 0.629,
      "step": 54500
    },
    {
      "epoch": 5.76,
      "learning_rate": 8.407663316582914e-06,
      "loss": 0.6261,
      "step": 55000
    },
    {
      "epoch": 5.81,
      "learning_rate": 8.21136934673367e-06,
      "loss": 0.6137,
      "step": 55500
    },
    {
      "epoch": 5.86,
      "learning_rate": 8.015075376884422e-06,
      "loss": 0.6353,
      "step": 56000
    },
    {
      "epoch": 5.91,
      "learning_rate": 7.818781407035176e-06,
      "loss": 0.6239,
      "step": 56500
    },
    {
      "epoch": 5.97,
      "learning_rate": 7.622487437185929e-06,
      "loss": 0.618,
      "step": 57000
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.7882150411605835,
      "eval_runtime": 8.1734,
      "eval_samples_per_second": 1357.582,
      "eval_steps_per_second": 42.455,
      "step": 57312
    },
    {
      "epoch": 6.02,
      "learning_rate": 7.426193467336683e-06,
      "loss": 0.6427,
      "step": 57500
    },
    {
      "epoch": 6.07,
      "learning_rate": 7.229899497487438e-06,
      "loss": 0.6283,
      "step": 58000
    },
    {
      "epoch": 6.12,
      "learning_rate": 7.033605527638191e-06,
      "loss": 0.6277,
      "step": 58500
    },
    {
      "epoch": 6.18,
      "learning_rate": 6.837311557788945e-06,
      "loss": 0.6131,
      "step": 59000
    },
    {
      "epoch": 6.23,
      "learning_rate": 6.641017587939698e-06,
      "loss": 0.6086,
      "step": 59500
    },
    {
      "epoch": 6.28,
      "learning_rate": 6.444723618090452e-06,
      "loss": 0.6154,
      "step": 60000
    },
    {
      "epoch": 6.33,
      "learning_rate": 6.248429648241206e-06,
      "loss": 0.6184,
      "step": 60500
    },
    {
      "epoch": 6.39,
      "learning_rate": 6.0521356783919595e-06,
      "loss": 0.6139,
      "step": 61000
    },
    {
      "epoch": 6.44,
      "learning_rate": 5.8558417085427135e-06,
      "loss": 0.6148,
      "step": 61500
    },
    {
      "epoch": 6.49,
      "learning_rate": 5.6595477386934675e-06,
      "loss": 0.6256,
      "step": 62000
    },
    {
      "epoch": 6.54,
      "learning_rate": 5.4632537688442215e-06,
      "loss": 0.6135,
      "step": 62500
    },
    {
      "epoch": 6.6,
      "learning_rate": 5.266959798994975e-06,
      "loss": 0.6274,
      "step": 63000
    },
    {
      "epoch": 6.65,
      "learning_rate": 5.070665829145729e-06,
      "loss": 0.623,
      "step": 63500
    },
    {
      "epoch": 6.7,
      "learning_rate": 4.874371859296483e-06,
      "loss": 0.6254,
      "step": 64000
    },
    {
      "epoch": 6.75,
      "learning_rate": 4.678077889447237e-06,
      "loss": 0.6233,
      "step": 64500
    },
    {
      "epoch": 6.8,
      "learning_rate": 4.48178391959799e-06,
      "loss": 0.6124,
      "step": 65000
    },
    {
      "epoch": 6.86,
      "learning_rate": 4.285489949748744e-06,
      "loss": 0.6108,
      "step": 65500
    },
    {
      "epoch": 6.91,
      "learning_rate": 4.089195979899497e-06,
      "loss": 0.6182,
      "step": 66000
    },
    {
      "epoch": 6.96,
      "learning_rate": 3.892902010050252e-06,
      "loss": 0.5991,
      "step": 66500
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.7896326184272766,
      "eval_runtime": 8.35,
      "eval_samples_per_second": 1328.855,
      "eval_steps_per_second": 41.557,
      "step": 66864
    },
    {
      "epoch": 7.01,
      "learning_rate": 3.696608040201005e-06,
      "loss": 0.6139,
      "step": 67000
    },
    {
      "epoch": 7.07,
      "learning_rate": 3.5003140703517586e-06,
      "loss": 0.6092,
      "step": 67500
    },
    {
      "epoch": 7.12,
      "learning_rate": 3.3040201005025126e-06,
      "loss": 0.6021,
      "step": 68000
    },
    {
      "epoch": 7.17,
      "learning_rate": 3.107726130653266e-06,
      "loss": 0.6138,
      "step": 68500
    },
    {
      "epoch": 7.22,
      "learning_rate": 2.91143216080402e-06,
      "loss": 0.6279,
      "step": 69000
    },
    {
      "epoch": 7.28,
      "learning_rate": 2.7151381909547737e-06,
      "loss": 0.6086,
      "step": 69500
    },
    {
      "epoch": 7.33,
      "learning_rate": 2.5188442211055277e-06,
      "loss": 0.6108,
      "step": 70000
    },
    {
      "epoch": 7.38,
      "learning_rate": 2.3225502512562813e-06,
      "loss": 0.5933,
      "step": 70500
    },
    {
      "epoch": 7.43,
      "learning_rate": 2.1262562814070353e-06,
      "loss": 0.6089,
      "step": 71000
    },
    {
      "epoch": 7.49,
      "learning_rate": 1.929962311557789e-06,
      "loss": 0.6086,
      "step": 71500
    },
    {
      "epoch": 7.54,
      "learning_rate": 1.7336683417085427e-06,
      "loss": 0.622,
      "step": 72000
    },
    {
      "epoch": 7.59,
      "learning_rate": 1.5373743718592965e-06,
      "loss": 0.6041,
      "step": 72500
    },
    {
      "epoch": 7.64,
      "learning_rate": 1.3410804020100503e-06,
      "loss": 0.6146,
      "step": 73000
    },
    {
      "epoch": 7.69,
      "learning_rate": 1.144786432160804e-06,
      "loss": 0.5999,
      "step": 73500
    },
    {
      "epoch": 7.75,
      "learning_rate": 9.484924623115579e-07,
      "loss": 0.6094,
      "step": 74000
    },
    {
      "epoch": 7.8,
      "learning_rate": 7.521984924623115e-07,
      "loss": 0.6137,
      "step": 74500
    },
    {
      "epoch": 7.85,
      "learning_rate": 5.559045226130653e-07,
      "loss": 0.6065,
      "step": 75000
    },
    {
      "epoch": 7.9,
      "learning_rate": 3.5961055276381907e-07,
      "loss": 0.6149,
      "step": 75500
    },
    {
      "epoch": 7.96,
      "learning_rate": 1.6331658291457286e-07,
      "loss": 0.6159,
      "step": 76000
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.7893399000167847,
      "eval_runtime": 8.9446,
      "eval_samples_per_second": 1240.522,
      "eval_steps_per_second": 38.794,
      "step": 76416
    }
  ],
  "max_steps": 76416,
  "num_train_epochs": 8,
  "total_flos": 2.326706308251648e+16,
  "trial_name": null,
  "trial_params": null
}