{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 3,
  "global_step": 282,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.031914893617021274,
      "grad_norm": 212.81973266601562,
      "learning_rate": 3.7450394203144474e-05,
      "loss": 2.2611,
      "step": 3
    },
    {
      "epoch": 0.031914893617021274,
      "eval_loss": 1.6954327821731567,
      "eval_runtime": 90.1393,
      "eval_samples_per_second": 3.328,
      "eval_steps_per_second": 0.111,
      "step": 3
    },
    {
      "epoch": 0.06382978723404255,
      "grad_norm": 94.31295776367188,
      "learning_rate": 3.704770179235797e-05,
      "loss": 2.1851,
      "step": 6
    },
    {
      "epoch": 0.06382978723404255,
      "eval_loss": 1.83690345287323,
      "eval_runtime": 45.0365,
      "eval_samples_per_second": 6.661,
      "eval_steps_per_second": 0.222,
      "step": 6
    },
    {
      "epoch": 0.09574468085106383,
      "grad_norm": 88.24703216552734,
      "learning_rate": 3.6645009381571474e-05,
      "loss": 1.9095,
      "step": 9
    },
    {
      "epoch": 0.09574468085106383,
      "eval_loss": 1.8263834714889526,
      "eval_runtime": 44.4544,
      "eval_samples_per_second": 6.748,
      "eval_steps_per_second": 0.225,
      "step": 9
    },
    {
      "epoch": 0.1276595744680851,
      "grad_norm": 73.16114044189453,
      "learning_rate": 3.6242316970784977e-05,
      "loss": 1.6022,
      "step": 12
    },
    {
      "epoch": 0.1276595744680851,
      "eval_loss": 1.6566990613937378,
      "eval_runtime": 45.0842,
      "eval_samples_per_second": 6.654,
      "eval_steps_per_second": 0.222,
      "step": 12
    },
    {
      "epoch": 0.1595744680851064,
      "grad_norm": 74.73725128173828,
      "learning_rate": 3.583962455999847e-05,
      "loss": 1.5181,
      "step": 15
    },
    {
      "epoch": 0.1595744680851064,
      "eval_loss": 1.6280070543289185,
      "eval_runtime": 45.1381,
      "eval_samples_per_second": 6.646,
      "eval_steps_per_second": 0.222,
      "step": 15
    },
    {
      "epoch": 0.19148936170212766,
      "grad_norm": 79.7146987915039,
      "learning_rate": 3.5436932149211976e-05,
      "loss": 1.8478,
      "step": 18
    },
    {
      "epoch": 0.19148936170212766,
      "eval_loss": 1.5796161890029907,
      "eval_runtime": 45.9387,
      "eval_samples_per_second": 6.53,
      "eval_steps_per_second": 0.218,
      "step": 18
    },
    {
      "epoch": 0.22340425531914893,
      "grad_norm": 65.3478775024414,
      "learning_rate": 3.503423973842547e-05,
      "loss": 1.3884,
      "step": 21
    },
    {
      "epoch": 0.22340425531914893,
      "eval_loss": 1.5002487897872925,
      "eval_runtime": 45.7227,
      "eval_samples_per_second": 6.561,
      "eval_steps_per_second": 0.219,
      "step": 21
    },
    {
      "epoch": 0.2553191489361702,
      "grad_norm": 66.32596588134766,
      "learning_rate": 3.4631547327638975e-05,
      "loss": 1.4368,
      "step": 24
    },
    {
      "epoch": 0.2553191489361702,
      "eval_loss": 1.5191569328308105,
      "eval_runtime": 44.4497,
      "eval_samples_per_second": 6.749,
      "eval_steps_per_second": 0.225,
      "step": 24
    },
    {
      "epoch": 0.2872340425531915,
      "grad_norm": 67.5133056640625,
      "learning_rate": 3.422885491685248e-05,
      "loss": 1.6515,
      "step": 27
    },
    {
      "epoch": 0.2872340425531915,
      "eval_loss": 1.4516819715499878,
      "eval_runtime": 44.968,
      "eval_samples_per_second": 6.671,
      "eval_steps_per_second": 0.222,
      "step": 27
    },
    {
      "epoch": 0.3191489361702128,
      "grad_norm": 56.178218841552734,
      "learning_rate": 3.3826162506065975e-05,
      "loss": 1.5518,
      "step": 30
    },
    {
      "epoch": 0.3191489361702128,
      "eval_loss": 1.4015744924545288,
      "eval_runtime": 44.5216,
      "eval_samples_per_second": 6.738,
      "eval_steps_per_second": 0.225,
      "step": 30
    },
    {
      "epoch": 0.35106382978723405,
      "grad_norm": 62.902313232421875,
      "learning_rate": 3.342347009527948e-05,
      "loss": 1.339,
      "step": 33
    },
    {
      "epoch": 0.35106382978723405,
      "eval_loss": 1.3566683530807495,
      "eval_runtime": 44.9809,
      "eval_samples_per_second": 6.669,
      "eval_steps_per_second": 0.222,
      "step": 33
    },
    {
      "epoch": 0.3829787234042553,
      "grad_norm": 57.751197814941406,
      "learning_rate": 3.3020777684492974e-05,
      "loss": 1.6338,
      "step": 36
    },
    {
      "epoch": 0.3829787234042553,
      "eval_loss": 1.396578073501587,
      "eval_runtime": 45.3817,
      "eval_samples_per_second": 6.611,
      "eval_steps_per_second": 0.22,
      "step": 36
    },
    {
      "epoch": 0.4148936170212766,
      "grad_norm": 57.7042121887207,
      "learning_rate": 3.261808527370648e-05,
      "loss": 1.4372,
      "step": 39
    },
    {
      "epoch": 0.4148936170212766,
      "eval_loss": 1.3969274759292603,
      "eval_runtime": 63.0438,
      "eval_samples_per_second": 4.759,
      "eval_steps_per_second": 0.159,
      "step": 39
    },
    {
      "epoch": 0.44680851063829785,
      "grad_norm": 43.087100982666016,
      "learning_rate": 3.221539286291997e-05,
      "loss": 1.2751,
      "step": 42
    },
    {
      "epoch": 0.44680851063829785,
      "eval_loss": 1.3432629108428955,
      "eval_runtime": 43.0776,
      "eval_samples_per_second": 6.964,
      "eval_steps_per_second": 0.232,
      "step": 42
    },
    {
      "epoch": 0.4787234042553192,
      "grad_norm": 58.544620513916016,
      "learning_rate": 3.1812700452133476e-05,
      "loss": 1.484,
      "step": 45
    },
    {
      "epoch": 0.4787234042553192,
      "eval_loss": 1.389102578163147,
      "eval_runtime": 42.2187,
      "eval_samples_per_second": 7.106,
      "eval_steps_per_second": 0.237,
      "step": 45
    },
    {
      "epoch": 0.5106382978723404,
      "grad_norm": 40.23255920410156,
      "learning_rate": 3.141000804134697e-05,
      "loss": 1.4936,
      "step": 48
    },
    {
      "epoch": 0.5106382978723404,
      "eval_loss": 1.408040165901184,
      "eval_runtime": 42.8634,
      "eval_samples_per_second": 6.999,
      "eval_steps_per_second": 0.233,
      "step": 48
    },
    {
      "epoch": 0.5425531914893617,
      "grad_norm": 58.45737075805664,
      "learning_rate": 3.1007315630560475e-05,
      "loss": 1.6771,
      "step": 51
    },
    {
      "epoch": 0.5425531914893617,
      "eval_loss": 1.3876069784164429,
      "eval_runtime": 47.7351,
      "eval_samples_per_second": 6.285,
      "eval_steps_per_second": 0.209,
      "step": 51
    },
    {
      "epoch": 0.574468085106383,
      "grad_norm": 59.58686447143555,
      "learning_rate": 3.060462321977397e-05,
      "loss": 1.7328,
      "step": 54
    },
    {
      "epoch": 0.574468085106383,
      "eval_loss": 1.3280422687530518,
      "eval_runtime": 45.1191,
      "eval_samples_per_second": 6.649,
      "eval_steps_per_second": 0.222,
      "step": 54
    },
    {
      "epoch": 0.6063829787234043,
      "grad_norm": 49.796451568603516,
      "learning_rate": 3.0201930808987475e-05,
      "loss": 1.3566,
      "step": 57
    },
    {
      "epoch": 0.6063829787234043,
      "eval_loss": 1.3299312591552734,
      "eval_runtime": 45.4372,
      "eval_samples_per_second": 6.603,
      "eval_steps_per_second": 0.22,
      "step": 57
    },
    {
      "epoch": 0.6382978723404256,
      "grad_norm": 47.63386154174805,
      "learning_rate": 2.9799238398200978e-05,
      "loss": 1.1698,
      "step": 60
    },
    {
      "epoch": 0.6382978723404256,
      "eval_loss": 1.3619319200515747,
      "eval_runtime": 44.5898,
      "eval_samples_per_second": 6.728,
      "eval_steps_per_second": 0.224,
      "step": 60
    },
    {
      "epoch": 0.6702127659574468,
      "grad_norm": 51.1750602722168,
      "learning_rate": 2.9396545987414478e-05,
      "loss": 1.4553,
      "step": 63
    },
    {
      "epoch": 0.6702127659574468,
      "eval_loss": 1.2944597005844116,
      "eval_runtime": 44.0116,
      "eval_samples_per_second": 6.816,
      "eval_steps_per_second": 0.227,
      "step": 63
    },
    {
      "epoch": 0.7021276595744681,
      "grad_norm": 56.35239028930664,
      "learning_rate": 2.8993853576627977e-05,
      "loss": 1.4532,
      "step": 66
    },
    {
      "epoch": 0.7021276595744681,
      "eval_loss": 1.3396683931350708,
      "eval_runtime": 57.5395,
      "eval_samples_per_second": 5.214,
      "eval_steps_per_second": 0.174,
      "step": 66
    },
    {
      "epoch": 0.7340425531914894,
      "grad_norm": 55.66822052001953,
      "learning_rate": 2.859116116584148e-05,
      "loss": 1.5623,
      "step": 69
    },
    {
      "epoch": 0.7340425531914894,
      "eval_loss": 1.3019959926605225,
      "eval_runtime": 42.7505,
      "eval_samples_per_second": 7.017,
      "eval_steps_per_second": 0.234,
      "step": 69
    },
    {
      "epoch": 0.7659574468085106,
      "grad_norm": 83.2627944946289,
      "learning_rate": 2.8188468755054976e-05,
      "loss": 1.5575,
      "step": 72
    },
    {
      "epoch": 0.7659574468085106,
      "eval_loss": 1.2901865243911743,
      "eval_runtime": 42.3414,
      "eval_samples_per_second": 7.085,
      "eval_steps_per_second": 0.236,
      "step": 72
    },
    {
      "epoch": 0.7978723404255319,
      "grad_norm": 66.81884002685547,
      "learning_rate": 2.778577634426848e-05,
      "loss": 1.4042,
      "step": 75
    },
    {
      "epoch": 0.7978723404255319,
      "eval_loss": 1.2550606727600098,
      "eval_runtime": 42.5955,
      "eval_samples_per_second": 7.043,
      "eval_steps_per_second": 0.235,
      "step": 75
    },
    {
      "epoch": 0.8297872340425532,
      "grad_norm": 47.388999938964844,
      "learning_rate": 2.738308393348198e-05,
      "loss": 1.3868,
      "step": 78
    },
    {
      "epoch": 0.8297872340425532,
      "eval_loss": 1.2457486391067505,
      "eval_runtime": 42.5244,
      "eval_samples_per_second": 7.055,
      "eval_steps_per_second": 0.235,
      "step": 78
    },
    {
      "epoch": 0.8617021276595744,
      "grad_norm": 51.87979507446289,
      "learning_rate": 2.698039152269548e-05,
      "loss": 1.3286,
      "step": 81
    },
    {
      "epoch": 0.8617021276595744,
      "eval_loss": 1.2376227378845215,
      "eval_runtime": 42.3825,
      "eval_samples_per_second": 7.078,
      "eval_steps_per_second": 0.236,
      "step": 81
    },
    {
      "epoch": 0.8936170212765957,
      "grad_norm": 61.56190490722656,
      "learning_rate": 2.6577699111908982e-05,
      "loss": 1.2435,
      "step": 84
    },
    {
      "epoch": 0.8936170212765957,
      "eval_loss": 1.1932413578033447,
      "eval_runtime": 43.0229,
      "eval_samples_per_second": 6.973,
      "eval_steps_per_second": 0.232,
      "step": 84
    },
    {
      "epoch": 0.925531914893617,
      "grad_norm": 56.688331604003906,
      "learning_rate": 2.617500670112248e-05,
      "loss": 1.1201,
      "step": 87
    },
    {
      "epoch": 0.925531914893617,
      "eval_loss": 1.1663594245910645,
      "eval_runtime": 42.6963,
      "eval_samples_per_second": 7.026,
      "eval_steps_per_second": 0.234,
      "step": 87
    },
    {
      "epoch": 0.9574468085106383,
      "grad_norm": 54.14950180053711,
      "learning_rate": 2.577231429033598e-05,
      "loss": 1.386,
      "step": 90
    },
    {
      "epoch": 0.9574468085106383,
      "eval_loss": 1.1541115045547485,
      "eval_runtime": 42.4377,
      "eval_samples_per_second": 7.069,
      "eval_steps_per_second": 0.236,
      "step": 90
    },
    {
      "epoch": 0.9893617021276596,
      "grad_norm": 63.0470085144043,
      "learning_rate": 2.536962187954948e-05,
      "loss": 1.0404,
      "step": 93
    },
    {
      "epoch": 0.9893617021276596,
      "eval_loss": 1.1485306024551392,
      "eval_runtime": 42.3281,
      "eval_samples_per_second": 7.087,
      "eval_steps_per_second": 0.236,
      "step": 93
    },
    {
      "epoch": 1.0212765957446808,
      "grad_norm": 29.81873321533203,
      "learning_rate": 2.496692946876298e-05,
      "loss": 0.7278,
      "step": 96
    },
    {
      "epoch": 1.0212765957446808,
      "eval_loss": 1.1454874277114868,
      "eval_runtime": 4.1684,
      "eval_samples_per_second": 71.97,
      "eval_steps_per_second": 2.399,
      "step": 96
    },
    {
      "epoch": 1.053191489361702,
      "grad_norm": 33.473297119140625,
      "learning_rate": 2.456423705797648e-05,
      "loss": 0.3543,
      "step": 99
    },
    {
      "epoch": 1.053191489361702,
      "eval_loss": 1.177135944366455,
      "eval_runtime": 42.0082,
      "eval_samples_per_second": 7.141,
      "eval_steps_per_second": 0.238,
      "step": 99
    },
    {
      "epoch": 1.0851063829787233,
      "grad_norm": 23.966768264770508,
      "learning_rate": 2.4161544647189983e-05,
      "loss": 0.2839,
      "step": 102
    },
    {
      "epoch": 1.0851063829787233,
      "eval_loss": 1.2594685554504395,
      "eval_runtime": 4.1814,
      "eval_samples_per_second": 71.746,
      "eval_steps_per_second": 2.392,
      "step": 102
    },
    {
      "epoch": 1.1170212765957448,
      "grad_norm": 23.973447799682617,
      "learning_rate": 2.375885223640348e-05,
      "loss": 0.2303,
      "step": 105
    },
    {
      "epoch": 1.1170212765957448,
      "eval_loss": 1.3750962018966675,
      "eval_runtime": 42.0396,
      "eval_samples_per_second": 7.136,
      "eval_steps_per_second": 0.238,
      "step": 105
    },
    {
      "epoch": 1.148936170212766,
      "grad_norm": 36.35573959350586,
      "learning_rate": 2.3356159825616983e-05,
      "loss": 0.3702,
      "step": 108
    },
    {
      "epoch": 1.148936170212766,
      "eval_loss": 1.4052996635437012,
      "eval_runtime": 4.1732,
      "eval_samples_per_second": 71.887,
      "eval_steps_per_second": 2.396,
      "step": 108
    },
    {
      "epoch": 1.1808510638297873,
      "grad_norm": 37.33869171142578,
      "learning_rate": 2.2953467414830486e-05,
      "loss": 0.37,
      "step": 111
    },
    {
      "epoch": 1.1808510638297873,
      "eval_loss": 1.4478391408920288,
      "eval_runtime": 42.2951,
      "eval_samples_per_second": 7.093,
      "eval_steps_per_second": 0.236,
      "step": 111
    },
    {
      "epoch": 1.2127659574468086,
      "grad_norm": 32.96900177001953,
      "learning_rate": 2.2550775004043982e-05,
      "loss": 0.2412,
      "step": 114
    },
    {
      "epoch": 1.2127659574468086,
      "eval_loss": 1.4774221181869507,
      "eval_runtime": 5.0258,
      "eval_samples_per_second": 59.693,
      "eval_steps_per_second": 1.99,
      "step": 114
    },
    {
      "epoch": 1.2446808510638299,
      "grad_norm": 45.11040496826172,
      "learning_rate": 2.2148082593257485e-05,
      "loss": 0.3028,
      "step": 117
    },
    {
      "epoch": 1.2446808510638299,
      "eval_loss": 1.4591237306594849,
      "eval_runtime": 41.5496,
      "eval_samples_per_second": 7.22,
      "eval_steps_per_second": 0.241,
      "step": 117
    },
    {
      "epoch": 1.2765957446808511,
      "grad_norm": 31.62494659423828,
      "learning_rate": 2.1745390182470985e-05,
      "loss": 0.322,
      "step": 120
    },
    {
      "epoch": 1.2765957446808511,
      "eval_loss": 1.483196496963501,
      "eval_runtime": 4.901,
      "eval_samples_per_second": 61.212,
      "eval_steps_per_second": 2.04,
      "step": 120
    },
    {
      "epoch": 1.3085106382978724,
      "grad_norm": 53.78076934814453,
      "learning_rate": 2.1342697771684484e-05,
      "loss": 0.4919,
      "step": 123
    },
    {
      "epoch": 1.3085106382978724,
      "eval_loss": 1.4939560890197754,
      "eval_runtime": 41.6747,
      "eval_samples_per_second": 7.199,
      "eval_steps_per_second": 0.24,
      "step": 123
    },
    {
      "epoch": 1.3404255319148937,
      "grad_norm": 54.71748352050781,
      "learning_rate": 2.0940005360897984e-05,
      "loss": 0.3868,
      "step": 126
    },
    {
      "epoch": 1.3404255319148937,
      "eval_loss": 1.4461126327514648,
      "eval_runtime": 4.3166,
      "eval_samples_per_second": 69.498,
      "eval_steps_per_second": 2.317,
      "step": 126
    },
    {
      "epoch": 1.372340425531915,
      "grad_norm": 39.04596710205078,
      "learning_rate": 2.0537312950111484e-05,
      "loss": 0.3795,
      "step": 129
    },
    {
      "epoch": 1.372340425531915,
      "eval_loss": 1.421697735786438,
      "eval_runtime": 41.8322,
      "eval_samples_per_second": 7.172,
      "eval_steps_per_second": 0.239,
      "step": 129
    },
    {
      "epoch": 1.4042553191489362,
      "grad_norm": 43.464900970458984,
      "learning_rate": 2.0134620539324983e-05,
      "loss": 0.2185,
      "step": 132
    },
    {
      "epoch": 1.4042553191489362,
      "eval_loss": 1.4404044151306152,
      "eval_runtime": 4.1662,
      "eval_samples_per_second": 72.008,
      "eval_steps_per_second": 2.4,
      "step": 132
    },
    {
      "epoch": 1.4361702127659575,
      "grad_norm": 39.858341217041016,
      "learning_rate": 1.9731928128538486e-05,
      "loss": 0.361,
      "step": 135
    },
    {
      "epoch": 1.4361702127659575,
      "eval_loss": 1.445768117904663,
      "eval_runtime": 41.5719,
      "eval_samples_per_second": 7.216,
      "eval_steps_per_second": 0.241,
      "step": 135
    },
    {
      "epoch": 1.4680851063829787,
      "grad_norm": 35.612701416015625,
      "learning_rate": 1.9329235717751983e-05,
      "loss": 0.3308,
      "step": 138
    },
    {
      "epoch": 1.4680851063829787,
      "eval_loss": 1.447102665901184,
      "eval_runtime": 4.2095,
      "eval_samples_per_second": 71.267,
      "eval_steps_per_second": 2.376,
      "step": 138
    },
    {
      "epoch": 1.5,
      "grad_norm": 40.08439636230469,
      "learning_rate": 1.8926543306965486e-05,
      "loss": 0.29,
      "step": 141
    },
    {
      "epoch": 1.5,
      "eval_loss": 1.4261672496795654,
      "eval_runtime": 41.3155,
      "eval_samples_per_second": 7.261,
      "eval_steps_per_second": 0.242,
      "step": 141
    },
    {
      "epoch": 1.5319148936170213,
      "grad_norm": 34.78931427001953,
      "learning_rate": 1.8523850896178985e-05,
      "loss": 0.2372,
      "step": 144
    },
    {
      "epoch": 1.5319148936170213,
      "eval_loss": 1.4144420623779297,
      "eval_runtime": 4.1971,
      "eval_samples_per_second": 71.478,
      "eval_steps_per_second": 2.383,
      "step": 144
    },
    {
      "epoch": 1.5638297872340425,
      "grad_norm": 45.00236892700195,
      "learning_rate": 1.8121158485392488e-05,
      "loss": 0.2666,
      "step": 147
    },
    {
      "epoch": 1.5638297872340425,
      "eval_loss": 1.4330610036849976,
      "eval_runtime": 41.5952,
      "eval_samples_per_second": 7.212,
      "eval_steps_per_second": 0.24,
      "step": 147
    },
    {
      "epoch": 1.5957446808510638,
      "grad_norm": 53.813011169433594,
      "learning_rate": 1.7718466074605988e-05,
      "loss": 0.357,
      "step": 150
    },
    {
      "epoch": 1.5957446808510638,
      "eval_loss": 1.49522864818573,
      "eval_runtime": 4.8438,
      "eval_samples_per_second": 61.935,
      "eval_steps_per_second": 2.065,
      "step": 150
    },
    {
      "epoch": 1.627659574468085,
      "grad_norm": 50.32698440551758,
      "learning_rate": 1.7315773663819488e-05,
      "loss": 0.3668,
      "step": 153
    },
    {
      "epoch": 1.627659574468085,
      "eval_loss": 1.5075035095214844,
      "eval_runtime": 42.1649,
      "eval_samples_per_second": 7.115,
      "eval_steps_per_second": 0.237,
      "step": 153
    },
    {
      "epoch": 1.6595744680851063,
      "grad_norm": 29.047195434570312,
      "learning_rate": 1.6913081253032987e-05,
      "loss": 0.257,
      "step": 156
    },
    {
      "epoch": 1.6595744680851063,
      "eval_loss": 1.499000072479248,
      "eval_runtime": 5.1976,
      "eval_samples_per_second": 57.719,
      "eval_steps_per_second": 1.924,
      "step": 156
    },
    {
      "epoch": 1.6914893617021276,
      "grad_norm": 44.64079666137695,
      "learning_rate": 1.6510388842246487e-05,
      "loss": 0.2812,
      "step": 159
    },
    {
      "epoch": 1.6914893617021276,
      "eval_loss": 1.5078258514404297,
      "eval_runtime": 41.7195,
      "eval_samples_per_second": 7.191,
      "eval_steps_per_second": 0.24,
      "step": 159
    },
    {
      "epoch": 1.7234042553191489,
      "grad_norm": 45.4018669128418,
      "learning_rate": 1.6107696431459987e-05,
      "loss": 0.3697,
      "step": 162
    },
    {
      "epoch": 1.7234042553191489,
      "eval_loss": 1.4968957901000977,
      "eval_runtime": 4.6701,
      "eval_samples_per_second": 64.239,
      "eval_steps_per_second": 2.141,
      "step": 162
    },
    {
      "epoch": 1.7553191489361701,
      "grad_norm": 23.335609436035156,
      "learning_rate": 1.5705004020673486e-05,
      "loss": 0.2935,
      "step": 165
    },
    {
      "epoch": 1.7553191489361701,
      "eval_loss": 1.4559073448181152,
      "eval_runtime": 41.376,
      "eval_samples_per_second": 7.251,
      "eval_steps_per_second": 0.242,
      "step": 165
    },
    {
      "epoch": 1.7872340425531914,
      "grad_norm": 38.07520294189453,
      "learning_rate": 1.5302311609886986e-05,
      "loss": 0.2622,
      "step": 168
    },
    {
      "epoch": 1.7872340425531914,
      "eval_loss": 1.4034123420715332,
      "eval_runtime": 4.1798,
      "eval_samples_per_second": 71.773,
      "eval_steps_per_second": 2.392,
      "step": 168
    },
    {
      "epoch": 1.8191489361702127,
      "grad_norm": 79.16899871826172,
      "learning_rate": 1.4899619199100489e-05,
      "loss": 0.4383,
      "step": 171
    },
    {
      "epoch": 1.8191489361702127,
      "eval_loss": 1.379632830619812,
      "eval_runtime": 41.4885,
      "eval_samples_per_second": 7.231,
      "eval_steps_per_second": 0.241,
      "step": 171
    },
    {
      "epoch": 1.851063829787234,
      "grad_norm": 56.25248718261719,
      "learning_rate": 1.4496926788313989e-05,
      "loss": 0.2953,
      "step": 174
    },
    {
      "epoch": 1.851063829787234,
      "eval_loss": 1.3718376159667969,
      "eval_runtime": 4.2567,
      "eval_samples_per_second": 70.478,
      "eval_steps_per_second": 2.349,
      "step": 174
    },
    {
      "epoch": 1.8829787234042552,
      "grad_norm": 45.190975189208984,
      "learning_rate": 1.4094234377527488e-05,
      "loss": 0.3427,
      "step": 177
    },
    {
      "epoch": 1.8829787234042552,
      "eval_loss": 1.3516465425491333,
      "eval_runtime": 41.6735,
      "eval_samples_per_second": 7.199,
      "eval_steps_per_second": 0.24,
      "step": 177
    },
    {
      "epoch": 1.9148936170212765,
      "grad_norm": 28.82175064086914,
      "learning_rate": 1.369154196674099e-05,
      "loss": 0.2204,
      "step": 180
    },
    {
      "epoch": 1.9148936170212765,
      "eval_loss": 1.3585424423217773,
      "eval_runtime": 4.1577,
      "eval_samples_per_second": 72.155,
      "eval_steps_per_second": 2.405,
      "step": 180
    },
    {
      "epoch": 1.9468085106382977,
      "grad_norm": 46.982139587402344,
      "learning_rate": 1.3288849555954491e-05,
      "loss": 0.2,
      "step": 183
    },
    {
      "epoch": 1.9468085106382977,
      "eval_loss": 1.3777296543121338,
      "eval_runtime": 41.8859,
      "eval_samples_per_second": 7.162,
      "eval_steps_per_second": 0.239,
      "step": 183
    },
    {
      "epoch": 1.978723404255319,
      "grad_norm": 22.6469669342041,
      "learning_rate": 1.288615714516799e-05,
      "loss": 0.2074,
      "step": 186
    },
    {
      "epoch": 1.978723404255319,
      "eval_loss": 1.3690061569213867,
      "eval_runtime": 4.6763,
      "eval_samples_per_second": 64.154,
      "eval_steps_per_second": 2.138,
      "step": 186
    },
    {
      "epoch": 2.0106382978723403,
      "grad_norm": 12.022841453552246,
      "learning_rate": 1.248346473438149e-05,
      "loss": 0.0972,
      "step": 189
    },
    {
      "epoch": 2.0106382978723403,
      "eval_loss": 1.354335069656372,
      "eval_runtime": 55.4961,
      "eval_samples_per_second": 5.406,
      "eval_steps_per_second": 0.18,
      "step": 189
    },
    {
      "epoch": 2.0425531914893615,
      "grad_norm": 32.32695007324219,
      "learning_rate": 1.2080772323594992e-05,
      "loss": 0.1505,
      "step": 192
    },
    {
      "epoch": 2.0425531914893615,
      "eval_loss": 1.3698673248291016,
      "eval_runtime": 34.6113,
      "eval_samples_per_second": 8.668,
      "eval_steps_per_second": 0.289,
      "step": 192
    },
    {
      "epoch": 2.074468085106383,
      "grad_norm": 20.132957458496094,
      "learning_rate": 1.1678079912808491e-05,
      "loss": 0.0651,
      "step": 195
    },
    {
      "epoch": 2.074468085106383,
      "eval_loss": 1.4110251665115356,
      "eval_runtime": 12.5667,
      "eval_samples_per_second": 23.873,
      "eval_steps_per_second": 0.796,
      "step": 195
    },
    {
      "epoch": 2.106382978723404,
      "grad_norm": 16.97626495361328,
      "learning_rate": 1.1275387502021991e-05,
      "loss": 0.045,
      "step": 198
    },
    {
      "epoch": 2.106382978723404,
      "eval_loss": 1.4494677782058716,
      "eval_runtime": 34.2066,
      "eval_samples_per_second": 8.77,
      "eval_steps_per_second": 0.292,
      "step": 198
    },
    {
      "epoch": 2.1382978723404253,
      "grad_norm": 27.48331642150879,
      "learning_rate": 1.0872695091235492e-05,
      "loss": 0.1105,
      "step": 201
    },
    {
      "epoch": 2.1382978723404253,
      "eval_loss": 1.4678298234939575,
      "eval_runtime": 12.6413,
      "eval_samples_per_second": 23.732,
      "eval_steps_per_second": 0.791,
      "step": 201
    },
    {
      "epoch": 2.1702127659574466,
      "grad_norm": 9.6218900680542,
      "learning_rate": 1.0470002680448992e-05,
      "loss": 0.0703,
      "step": 204
    },
    {
      "epoch": 2.1702127659574466,
      "eval_loss": 1.4693076610565186,
      "eval_runtime": 33.7057,
      "eval_samples_per_second": 8.901,
      "eval_steps_per_second": 0.297,
      "step": 204
    },
    {
      "epoch": 2.202127659574468,
      "grad_norm": 12.747748374938965,
      "learning_rate": 1.0067310269662492e-05,
      "loss": 0.0793,
      "step": 207
    },
    {
      "epoch": 2.202127659574468,
      "eval_loss": 1.4696707725524902,
      "eval_runtime": 11.7313,
      "eval_samples_per_second": 25.573,
      "eval_steps_per_second": 0.852,
      "step": 207
    },
    {
      "epoch": 2.2340425531914896,
      "grad_norm": 18.998815536499023,
      "learning_rate": 9.664617858875991e-06,
      "loss": 0.0709,
      "step": 210
    },
    {
      "epoch": 2.2340425531914896,
      "eval_loss": 1.4826494455337524,
      "eval_runtime": 33.8588,
      "eval_samples_per_second": 8.86,
      "eval_steps_per_second": 0.295,
      "step": 210
    },
    {
      "epoch": 2.2659574468085104,
      "grad_norm": 26.24040412902832,
      "learning_rate": 9.261925448089493e-06,
      "loss": 0.133,
      "step": 213
    },
    {
      "epoch": 2.2659574468085104,
      "eval_loss": 1.484723687171936,
      "eval_runtime": 11.9592,
      "eval_samples_per_second": 25.085,
      "eval_steps_per_second": 0.836,
      "step": 213
    },
    {
      "epoch": 2.297872340425532,
      "grad_norm": 38.74705123901367,
      "learning_rate": 8.859233037302994e-06,
      "loss": 0.1055,
      "step": 216
    },
    {
      "epoch": 2.297872340425532,
      "eval_loss": 1.4781628847122192,
      "eval_runtime": 34.0608,
      "eval_samples_per_second": 8.808,
      "eval_steps_per_second": 0.294,
      "step": 216
    },
    {
      "epoch": 2.329787234042553,
      "grad_norm": 51.491939544677734,
      "learning_rate": 8.456540626516494e-06,
      "loss": 0.106,
      "step": 219
    },
    {
      "epoch": 2.329787234042553,
      "eval_loss": 1.4520379304885864,
      "eval_runtime": 12.1072,
      "eval_samples_per_second": 24.779,
      "eval_steps_per_second": 0.826,
      "step": 219
    },
    {
      "epoch": 2.3617021276595747,
      "grad_norm": 15.932838439941406,
      "learning_rate": 8.053848215729993e-06,
      "loss": 0.0553,
      "step": 222
    },
    {
      "epoch": 2.3617021276595747,
      "eval_loss": 1.4323525428771973,
      "eval_runtime": 34.1547,
      "eval_samples_per_second": 8.784,
      "eval_steps_per_second": 0.293,
      "step": 222
    },
    {
      "epoch": 2.393617021276596,
      "grad_norm": 14.132286071777344,
      "learning_rate": 7.651155804943493e-06,
      "loss": 0.0598,
      "step": 225
    },
    {
      "epoch": 2.393617021276596,
      "eval_loss": 1.4186018705368042,
      "eval_runtime": 12.5329,
      "eval_samples_per_second": 23.937,
      "eval_steps_per_second": 0.798,
      "step": 225
    },
    {
      "epoch": 2.425531914893617,
      "grad_norm": 43.08507537841797,
      "learning_rate": 7.248463394156994e-06,
      "loss": 0.0903,
      "step": 228
    },
    {
      "epoch": 2.425531914893617,
      "eval_loss": 1.4079523086547852,
      "eval_runtime": 33.5371,
      "eval_samples_per_second": 8.945,
      "eval_steps_per_second": 0.298,
      "step": 228
    },
    {
      "epoch": 2.4574468085106385,
      "grad_norm": 18.148130416870117,
      "learning_rate": 6.845770983370495e-06,
      "loss": 0.0379,
      "step": 231
    },
    {
      "epoch": 2.4574468085106385,
      "eval_loss": 1.402553677558899,
      "eval_runtime": 12.5951,
      "eval_samples_per_second": 23.819,
      "eval_steps_per_second": 0.794,
      "step": 231
    },
    {
      "epoch": 2.4893617021276597,
      "grad_norm": 18.774492263793945,
      "learning_rate": 6.443078572583995e-06,
      "loss": 0.056,
      "step": 234
    },
    {
      "epoch": 2.4893617021276597,
      "eval_loss": 1.4105722904205322,
      "eval_runtime": 34.0705,
      "eval_samples_per_second": 8.805,
      "eval_steps_per_second": 0.294,
      "step": 234
    },
    {
      "epoch": 2.521276595744681,
      "grad_norm": 21.43934440612793,
      "learning_rate": 6.040386161797496e-06,
      "loss": 0.0753,
      "step": 237
    },
    {
      "epoch": 2.521276595744681,
      "eval_loss": 1.429024338722229,
      "eval_runtime": 11.7342,
      "eval_samples_per_second": 25.566,
      "eval_steps_per_second": 0.852,
      "step": 237
    },
    {
      "epoch": 2.5531914893617023,
      "grad_norm": 21.858001708984375,
      "learning_rate": 5.6376937510109955e-06,
      "loss": 0.0677,
      "step": 240
    },
    {
      "epoch": 2.5531914893617023,
      "eval_loss": 1.449249267578125,
      "eval_runtime": 34.3754,
      "eval_samples_per_second": 8.727,
      "eval_steps_per_second": 0.291,
      "step": 240
    },
    {
      "epoch": 2.5851063829787235,
      "grad_norm": 18.75592803955078,
      "learning_rate": 5.235001340224496e-06,
      "loss": 0.0636,
      "step": 243
    },
    {
      "epoch": 2.5851063829787235,
      "eval_loss": 1.4570808410644531,
      "eval_runtime": 11.6793,
      "eval_samples_per_second": 25.687,
      "eval_steps_per_second": 0.856,
      "step": 243
    },
    {
      "epoch": 2.617021276595745,
      "grad_norm": 23.944276809692383,
      "learning_rate": 4.832308929437996e-06,
      "loss": 0.0748,
      "step": 246
    },
    {
      "epoch": 2.617021276595745,
      "eval_loss": 1.4605278968811035,
      "eval_runtime": 33.9241,
      "eval_samples_per_second": 8.843,
      "eval_steps_per_second": 0.295,
      "step": 246
    },
    {
      "epoch": 2.648936170212766,
      "grad_norm": 9.529204368591309,
      "learning_rate": 4.429616518651497e-06,
      "loss": 0.0271,
      "step": 249
    },
    {
      "epoch": 2.648936170212766,
      "eval_loss": 1.4531209468841553,
      "eval_runtime": 11.8506,
      "eval_samples_per_second": 25.315,
      "eval_steps_per_second": 0.844,
      "step": 249
    },
    {
      "epoch": 2.6808510638297873,
      "grad_norm": 3.7583017349243164,
      "learning_rate": 4.026924107864997e-06,
      "loss": 0.0542,
      "step": 252
    },
    {
      "epoch": 2.6808510638297873,
      "eval_loss": 1.4430041313171387,
      "eval_runtime": 33.7906,
      "eval_samples_per_second": 8.878,
      "eval_steps_per_second": 0.296,
      "step": 252
    },
    {
      "epoch": 2.7127659574468086,
      "grad_norm": 12.694750785827637,
      "learning_rate": 3.624231697078497e-06,
      "loss": 0.0332,
      "step": 255
    },
    {
      "epoch": 2.7127659574468086,
      "eval_loss": 1.4361652135849,
      "eval_runtime": 11.6664,
      "eval_samples_per_second": 25.715,
      "eval_steps_per_second": 0.857,
      "step": 255
    },
    {
      "epoch": 2.74468085106383,
      "grad_norm": 36.67692184448242,
      "learning_rate": 3.2215392862919977e-06,
      "loss": 0.1035,
      "step": 258
    },
    {
      "epoch": 2.74468085106383,
      "eval_loss": 1.430528998374939,
      "eval_runtime": 33.9464,
      "eval_samples_per_second": 8.837,
      "eval_steps_per_second": 0.295,
      "step": 258
    },
    {
      "epoch": 2.776595744680851,
      "grad_norm": 30.15072250366211,
      "learning_rate": 2.8188468755054977e-06,
      "loss": 0.0643,
      "step": 261
    },
    {
      "epoch": 2.776595744680851,
      "eval_loss": 1.4278947114944458,
      "eval_runtime": 12.1733,
      "eval_samples_per_second": 24.644,
      "eval_steps_per_second": 0.821,
      "step": 261
    },
    {
      "epoch": 2.8085106382978724,
      "grad_norm": 5.219892978668213,
      "learning_rate": 2.416154464718998e-06,
      "loss": 0.0549,
      "step": 264
    },
    {
      "epoch": 2.8085106382978724,
      "eval_loss": 1.4259532690048218,
      "eval_runtime": 33.872,
      "eval_samples_per_second": 8.857,
      "eval_steps_per_second": 0.295,
      "step": 264
    },
    {
      "epoch": 2.8404255319148937,
      "grad_norm": 20.615497589111328,
      "learning_rate": 2.0134620539324983e-06,
      "loss": 0.0495,
      "step": 267
    },
    {
      "epoch": 2.8404255319148937,
      "eval_loss": 1.4268009662628174,
      "eval_runtime": 12.5283,
      "eval_samples_per_second": 23.946,
      "eval_steps_per_second": 0.798,
      "step": 267
    },
    {
      "epoch": 2.872340425531915,
      "grad_norm": 0.853824257850647,
      "learning_rate": 1.6107696431459988e-06,
      "loss": 0.0152,
      "step": 270
    },
    {
      "epoch": 2.872340425531915,
      "eval_loss": 1.427072525024414,
      "eval_runtime": 33.8909,
      "eval_samples_per_second": 8.852,
      "eval_steps_per_second": 0.295,
      "step": 270
    },
    {
      "epoch": 2.904255319148936,
      "grad_norm": 6.235328197479248,
      "learning_rate": 1.208077232359499e-06,
      "loss": 0.0806,
      "step": 273
    },
    {
      "epoch": 2.904255319148936,
      "eval_loss": 1.4273793697357178,
      "eval_runtime": 12.4911,
      "eval_samples_per_second": 24.017,
      "eval_steps_per_second": 0.801,
      "step": 273
    },
    {
      "epoch": 2.9361702127659575,
      "grad_norm": 17.21672248840332,
      "learning_rate": 8.053848215729994e-07,
      "loss": 0.0564,
      "step": 276
    },
    {
      "epoch": 2.9361702127659575,
      "eval_loss": 1.4273569583892822,
      "eval_runtime": 33.6977,
      "eval_samples_per_second": 8.903,
      "eval_steps_per_second": 0.297,
      "step": 276
    },
    {
      "epoch": 2.9680851063829787,
      "grad_norm": 23.0670166015625,
      "learning_rate": 4.026924107864997e-07,
      "loss": 0.0726,
      "step": 279
    },
    {
      "epoch": 2.9680851063829787,
      "eval_loss": 1.4270917177200317,
      "eval_runtime": 11.7754,
      "eval_samples_per_second": 25.477,
      "eval_steps_per_second": 0.849,
      "step": 279
    },
    {
      "epoch": 3.0,
      "grad_norm": 20.24632453918457,
      "learning_rate": 0.0,
      "loss": 0.0332,
      "step": 282
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.427086591720581,
      "eval_runtime": 34.0405,
      "eval_samples_per_second": 8.813,
      "eval_steps_per_second": 0.294,
      "step": 282
    }
  ],
  "logging_steps": 3,
  "max_steps": 282,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 523328480700102.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "_wandb": {},
    "assignments": {},
    "decay": 0.1,
    "learning_rate": 3.785308661393097e-05,
    "metric": "eval/loss",
    "per_device_train_batch_size": 32
  }
}