{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.975786924939467,
  "eval_steps": 500,
  "global_step": 2060,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004842615012106538,
      "grad_norm": 3.9375,
      "learning_rate": 9.70873786407767e-07,
      "loss": 5.486,
      "step": 1
    },
    {
      "epoch": 0.024213075060532687,
      "grad_norm": 3.875,
      "learning_rate": 4.854368932038835e-06,
      "loss": 5.4688,
      "step": 5
    },
    {
      "epoch": 0.048426150121065374,
      "grad_norm": 3.9375,
      "learning_rate": 9.70873786407767e-06,
      "loss": 5.4581,
      "step": 10
    },
    {
      "epoch": 0.07263922518159806,
      "grad_norm": 3.796875,
      "learning_rate": 1.4563106796116505e-05,
      "loss": 5.3204,
      "step": 15
    },
    {
      "epoch": 0.09685230024213075,
      "grad_norm": 3.578125,
      "learning_rate": 1.941747572815534e-05,
      "loss": 5.0021,
      "step": 20
    },
    {
      "epoch": 0.12106537530266344,
      "grad_norm": 3.46875,
      "learning_rate": 2.4271844660194176e-05,
      "loss": 4.5571,
      "step": 25
    },
    {
      "epoch": 0.14527845036319612,
      "grad_norm": 3.265625,
      "learning_rate": 2.912621359223301e-05,
      "loss": 3.9862,
      "step": 30
    },
    {
      "epoch": 0.1694915254237288,
      "grad_norm": 2.84375,
      "learning_rate": 3.398058252427185e-05,
      "loss": 3.3741,
      "step": 35
    },
    {
      "epoch": 0.1937046004842615,
      "grad_norm": 2.34375,
      "learning_rate": 3.883495145631068e-05,
      "loss": 3.0035,
      "step": 40
    },
    {
      "epoch": 0.2179176755447942,
      "grad_norm": 2.8125,
      "learning_rate": 4.368932038834951e-05,
      "loss": 2.7678,
      "step": 45
    },
    {
      "epoch": 0.24213075060532688,
      "grad_norm": 3.046875,
      "learning_rate": 4.854368932038835e-05,
      "loss": 2.4982,
      "step": 50
    },
    {
      "epoch": 0.26634382566585957,
      "grad_norm": 3.28125,
      "learning_rate": 5.339805825242719e-05,
      "loss": 2.2149,
      "step": 55
    },
    {
      "epoch": 0.29055690072639223,
      "grad_norm": 8.625,
      "learning_rate": 5.825242718446602e-05,
      "loss": 1.9807,
      "step": 60
    },
    {
      "epoch": 0.31476997578692495,
      "grad_norm": 8.875,
      "learning_rate": 6.310679611650486e-05,
      "loss": 1.9918,
      "step": 65
    },
    {
      "epoch": 0.3389830508474576,
      "grad_norm": 2.234375,
      "learning_rate": 6.79611650485437e-05,
      "loss": 1.8154,
      "step": 70
    },
    {
      "epoch": 0.36319612590799033,
      "grad_norm": 2.671875,
      "learning_rate": 7.281553398058253e-05,
      "loss": 1.7162,
      "step": 75
    },
    {
      "epoch": 0.387409200968523,
      "grad_norm": 1.671875,
      "learning_rate": 7.766990291262136e-05,
      "loss": 1.6152,
      "step": 80
    },
    {
      "epoch": 0.4116222760290557,
      "grad_norm": 0.63671875,
      "learning_rate": 8.25242718446602e-05,
      "loss": 1.5063,
      "step": 85
    },
    {
      "epoch": 0.4358353510895884,
      "grad_norm": 0.69921875,
      "learning_rate": 8.737864077669902e-05,
      "loss": 1.4152,
      "step": 90
    },
    {
      "epoch": 0.4600484261501211,
      "grad_norm": 0.73046875,
      "learning_rate": 9.223300970873788e-05,
      "loss": 1.3459,
      "step": 95
    },
    {
      "epoch": 0.48426150121065376,
      "grad_norm": 0.41015625,
      "learning_rate": 9.70873786407767e-05,
      "loss": 1.3005,
      "step": 100
    },
    {
      "epoch": 0.5084745762711864,
      "grad_norm": 0.8125,
      "learning_rate": 0.00010194174757281553,
      "loss": 1.2498,
      "step": 105
    },
    {
      "epoch": 0.5326876513317191,
      "grad_norm": 0.5234375,
      "learning_rate": 0.00010679611650485437,
      "loss": 1.218,
      "step": 110
    },
    {
      "epoch": 0.5569007263922519,
      "grad_norm": 0.4453125,
      "learning_rate": 0.0001116504854368932,
      "loss": 1.1886,
      "step": 115
    },
    {
      "epoch": 0.5811138014527845,
      "grad_norm": 0.279296875,
      "learning_rate": 0.00011650485436893204,
      "loss": 1.1448,
      "step": 120
    },
    {
      "epoch": 0.6053268765133172,
      "grad_norm": 0.6328125,
      "learning_rate": 0.00012135922330097087,
      "loss": 1.1324,
      "step": 125
    },
    {
      "epoch": 0.6295399515738499,
      "grad_norm": 0.443359375,
      "learning_rate": 0.00012621359223300972,
      "loss": 1.1027,
      "step": 130
    },
    {
      "epoch": 0.6537530266343826,
      "grad_norm": 0.65234375,
      "learning_rate": 0.00013106796116504855,
      "loss": 1.081,
      "step": 135
    },
    {
      "epoch": 0.6779661016949152,
      "grad_norm": 0.412109375,
      "learning_rate": 0.0001359223300970874,
      "loss": 1.0642,
      "step": 140
    },
    {
      "epoch": 0.7021791767554479,
      "grad_norm": 0.5078125,
      "learning_rate": 0.00014077669902912623,
      "loss": 1.0491,
      "step": 145
    },
    {
      "epoch": 0.7263922518159807,
      "grad_norm": 0.423828125,
      "learning_rate": 0.00014563106796116506,
      "loss": 1.04,
      "step": 150
    },
    {
      "epoch": 0.7506053268765133,
      "grad_norm": 0.26953125,
      "learning_rate": 0.00015048543689320389,
      "loss": 1.027,
      "step": 155
    },
    {
      "epoch": 0.774818401937046,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0001553398058252427,
      "loss": 1.0164,
      "step": 160
    },
    {
      "epoch": 0.7990314769975787,
      "grad_norm": 0.33203125,
      "learning_rate": 0.00016019417475728157,
      "loss": 1.0065,
      "step": 165
    },
    {
      "epoch": 0.8232445520581114,
      "grad_norm": 0.30078125,
      "learning_rate": 0.0001650485436893204,
      "loss": 0.996,
      "step": 170
    },
    {
      "epoch": 0.847457627118644,
      "grad_norm": 0.28125,
      "learning_rate": 0.00016990291262135922,
      "loss": 0.9893,
      "step": 175
    },
    {
      "epoch": 0.8716707021791767,
      "grad_norm": 0.310546875,
      "learning_rate": 0.00017475728155339805,
      "loss": 0.9829,
      "step": 180
    },
    {
      "epoch": 0.8958837772397095,
      "grad_norm": 0.412109375,
      "learning_rate": 0.0001796116504854369,
      "loss": 0.9746,
      "step": 185
    },
    {
      "epoch": 0.9200968523002422,
      "grad_norm": 0.279296875,
      "learning_rate": 0.00018446601941747576,
      "loss": 0.9717,
      "step": 190
    },
    {
      "epoch": 0.9443099273607748,
      "grad_norm": 0.30859375,
      "learning_rate": 0.00018932038834951458,
      "loss": 0.9658,
      "step": 195
    },
    {
      "epoch": 0.9685230024213075,
      "grad_norm": 0.298828125,
      "learning_rate": 0.0001941747572815534,
      "loss": 0.9654,
      "step": 200
    },
    {
      "epoch": 0.9927360774818402,
      "grad_norm": 0.279296875,
      "learning_rate": 0.00019902912621359224,
      "loss": 0.9633,
      "step": 205
    },
    {
      "epoch": 0.9975786924939467,
      "eval_loss": 2.6959354877471924,
      "eval_runtime": 0.2681,
      "eval_samples_per_second": 37.305,
      "eval_steps_per_second": 3.73,
      "step": 206
    },
    {
      "epoch": 1.0169491525423728,
      "grad_norm": 0.7265625,
      "learning_rate": 0.0001999977029597241,
      "loss": 0.9554,
      "step": 210
    },
    {
      "epoch": 1.0411622276029056,
      "grad_norm": 0.306640625,
      "learning_rate": 0.00019998837141446378,
      "loss": 0.9453,
      "step": 215
    },
    {
      "epoch": 1.0653753026634383,
      "grad_norm": 0.267578125,
      "learning_rate": 0.00019997186246852679,
      "loss": 0.9391,
      "step": 220
    },
    {
      "epoch": 1.089588377723971,
      "grad_norm": 0.302734375,
      "learning_rate": 0.00019994817730696412,
      "loss": 0.9382,
      "step": 225
    },
    {
      "epoch": 1.1138014527845037,
      "grad_norm": 0.228515625,
      "learning_rate": 0.00019991731762995243,
      "loss": 0.9398,
      "step": 230
    },
    {
      "epoch": 1.1380145278450362,
      "grad_norm": 0.275390625,
      "learning_rate": 0.000199879285652672,
      "loss": 0.9253,
      "step": 235
    },
    {
      "epoch": 1.162227602905569,
      "grad_norm": 0.27734375,
      "learning_rate": 0.00019983408410514765,
      "loss": 0.9284,
      "step": 240
    },
    {
      "epoch": 1.1864406779661016,
      "grad_norm": 0.326171875,
      "learning_rate": 0.00019978171623205292,
      "loss": 0.9222,
      "step": 245
    },
    {
      "epoch": 1.2106537530266344,
      "grad_norm": 0.265625,
      "learning_rate": 0.00019972218579247697,
      "loss": 0.9177,
      "step": 250
    },
    {
      "epoch": 1.234866828087167,
      "grad_norm": 0.267578125,
      "learning_rate": 0.00019965549705965488,
      "loss": 0.9149,
      "step": 255
    },
    {
      "epoch": 1.2590799031476998,
      "grad_norm": 0.2333984375,
      "learning_rate": 0.00019958165482066094,
      "loss": 0.9102,
      "step": 260
    },
    {
      "epoch": 1.2832929782082325,
      "grad_norm": 0.330078125,
      "learning_rate": 0.00019950066437606485,
      "loss": 0.9108,
      "step": 265
    },
    {
      "epoch": 1.307506053268765,
      "grad_norm": 0.265625,
      "learning_rate": 0.00019941253153955138,
      "loss": 0.9108,
      "step": 270
    },
    {
      "epoch": 1.331719128329298,
      "grad_norm": 0.232421875,
      "learning_rate": 0.00019931726263750302,
      "loss": 0.9057,
      "step": 275
    },
    {
      "epoch": 1.3559322033898304,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.0001992148645085458,
      "loss": 0.904,
      "step": 280
    },
    {
      "epoch": 1.3801452784503632,
      "grad_norm": 0.384765625,
      "learning_rate": 0.0001991053445030585,
      "loss": 0.9011,
      "step": 285
    },
    {
      "epoch": 1.4043583535108959,
      "grad_norm": 0.34765625,
      "learning_rate": 0.0001989887104826449,
      "loss": 0.9005,
      "step": 290
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.322265625,
      "learning_rate": 0.00019886497081956955,
      "loss": 0.8899,
      "step": 295
    },
    {
      "epoch": 1.4527845036319613,
      "grad_norm": 0.2734375,
      "learning_rate": 0.00019873413439615675,
      "loss": 0.8849,
      "step": 300
    },
    {
      "epoch": 1.4769975786924938,
      "grad_norm": 0.24609375,
      "learning_rate": 0.000198596210604153,
      "loss": 0.8902,
      "step": 305
    },
    {
      "epoch": 1.5012106537530268,
      "grad_norm": 0.28515625,
      "learning_rate": 0.00019845120934405274,
      "loss": 0.8939,
      "step": 310
    },
    {
      "epoch": 1.5254237288135593,
      "grad_norm": 0.251953125,
      "learning_rate": 0.00019829914102438779,
      "loss": 0.8943,
      "step": 315
    },
    {
      "epoch": 1.549636803874092,
      "grad_norm": 0.4375,
      "learning_rate": 0.00019814001656098001,
      "loss": 0.8892,
      "step": 320
    },
    {
      "epoch": 1.5738498789346247,
      "grad_norm": 0.35546875,
      "learning_rate": 0.0001979738473761581,
      "loss": 0.881,
      "step": 325
    },
    {
      "epoch": 1.5980629539951574,
      "grad_norm": 0.3046875,
      "learning_rate": 0.00019780064539793723,
      "loss": 0.8819,
      "step": 330
    },
    {
      "epoch": 1.6222760290556901,
      "grad_norm": 0.384765625,
      "learning_rate": 0.00019762042305916316,
      "loss": 0.8727,
      "step": 335
    },
    {
      "epoch": 1.6464891041162226,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0001974331932966196,
      "loss": 0.8815,
      "step": 340
    },
    {
      "epoch": 1.6707021791767556,
      "grad_norm": 0.328125,
      "learning_rate": 0.00019723896955009965,
      "loss": 0.8728,
      "step": 345
    },
    {
      "epoch": 1.694915254237288,
      "grad_norm": 0.58984375,
      "learning_rate": 0.00019703776576144105,
      "loss": 0.8793,
      "step": 350
    },
    {
      "epoch": 1.7191283292978208,
      "grad_norm": 0.296875,
      "learning_rate": 0.00019682959637352541,
      "loss": 0.8735,
      "step": 355
    },
    {
      "epoch": 1.7433414043583535,
      "grad_norm": 0.357421875,
      "learning_rate": 0.00019661447632924144,
      "loss": 0.8746,
      "step": 360
    },
    {
      "epoch": 1.7675544794188862,
      "grad_norm": 0.671875,
      "learning_rate": 0.00019639242107041227,
      "loss": 0.8721,
      "step": 365
    },
    {
      "epoch": 1.791767554479419,
      "grad_norm": 0.3984375,
      "learning_rate": 0.00019616344653668711,
      "loss": 0.8697,
      "step": 370
    },
    {
      "epoch": 1.8159806295399514,
      "grad_norm": 0.349609375,
      "learning_rate": 0.00019592756916439705,
      "loss": 0.8705,
      "step": 375
    },
    {
      "epoch": 1.8401937046004844,
      "grad_norm": 0.4140625,
      "learning_rate": 0.00019568480588537512,
      "loss": 0.8676,
      "step": 380
    },
    {
      "epoch": 1.8644067796610169,
      "grad_norm": 0.283203125,
      "learning_rate": 0.00019543517412574097,
      "loss": 0.8736,
      "step": 385
    },
    {
      "epoch": 1.8886198547215496,
      "grad_norm": 0.453125,
      "learning_rate": 0.00019517869180464992,
      "loss": 0.8663,
      "step": 390
    },
    {
      "epoch": 1.9128329297820823,
      "grad_norm": 0.2890625,
      "learning_rate": 0.00019491537733300676,
      "loss": 0.8699,
      "step": 395
    },
    {
      "epoch": 1.937046004842615,
      "grad_norm": 0.28125,
      "learning_rate": 0.00019464524961214415,
      "loss": 0.8657,
      "step": 400
    },
    {
      "epoch": 1.9612590799031477,
      "grad_norm": 0.28515625,
      "learning_rate": 0.00019436832803246574,
      "loss": 0.8621,
      "step": 405
    },
    {
      "epoch": 1.9854721549636802,
      "grad_norm": 0.23046875,
      "learning_rate": 0.0001940846324720544,
      "loss": 0.865,
      "step": 410
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.6510560512542725,
      "eval_runtime": 0.237,
      "eval_samples_per_second": 42.194,
      "eval_steps_per_second": 4.219,
      "step": 413
    },
    {
      "epoch": 2.009685230024213,
      "grad_norm": 0.27734375,
      "learning_rate": 0.00019379418329524534,
      "loss": 0.857,
      "step": 415
    },
    {
      "epoch": 2.0338983050847457,
      "grad_norm": 0.279296875,
      "learning_rate": 0.00019349700135116416,
      "loss": 0.8445,
      "step": 420
    },
    {
      "epoch": 2.0581113801452786,
      "grad_norm": 0.37109375,
      "learning_rate": 0.00019319310797223036,
      "loss": 0.8468,
      "step": 425
    },
    {
      "epoch": 2.082324455205811,
      "grad_norm": 0.306640625,
      "learning_rate": 0.00019288252497262612,
      "loss": 0.8413,
      "step": 430
    },
    {
      "epoch": 2.106537530266344,
      "grad_norm": 0.259765625,
      "learning_rate": 0.00019256527464673025,
      "loss": 0.8442,
      "step": 435
    },
    {
      "epoch": 2.1307506053268765,
      "grad_norm": 0.259765625,
      "learning_rate": 0.00019224137976751795,
      "loss": 0.8374,
      "step": 440
    },
    {
      "epoch": 2.154963680387409,
      "grad_norm": 0.3125,
      "learning_rate": 0.0001919108635849261,
      "loss": 0.8399,
      "step": 445
    },
    {
      "epoch": 2.179176755447942,
      "grad_norm": 0.306640625,
      "learning_rate": 0.00019157374982418435,
      "loss": 0.837,
      "step": 450
    },
    {
      "epoch": 2.2033898305084745,
      "grad_norm": 0.41796875,
      "learning_rate": 0.00019123006268411205,
      "loss": 0.8363,
      "step": 455
    },
    {
      "epoch": 2.2276029055690074,
      "grad_norm": 0.375,
      "learning_rate": 0.00019087982683538114,
      "loss": 0.8331,
      "step": 460
    },
    {
      "epoch": 2.25181598062954,
      "grad_norm": 0.365234375,
      "learning_rate": 0.00019052306741874532,
      "loss": 0.8318,
      "step": 465
    },
    {
      "epoch": 2.2760290556900724,
      "grad_norm": 0.259765625,
      "learning_rate": 0.00019015981004323536,
      "loss": 0.84,
      "step": 470
    },
    {
      "epoch": 2.3002421307506054,
      "grad_norm": 0.26953125,
      "learning_rate": 0.00018979008078432073,
      "loss": 0.8364,
      "step": 475
    },
    {
      "epoch": 2.324455205811138,
      "grad_norm": 0.326171875,
      "learning_rate": 0.00018941390618203806,
      "loss": 0.8352,
      "step": 480
    },
    {
      "epoch": 2.348668280871671,
      "grad_norm": 0.296875,
      "learning_rate": 0.00018903131323908578,
      "loss": 0.8437,
      "step": 485
    },
    {
      "epoch": 2.3728813559322033,
      "grad_norm": 0.322265625,
      "learning_rate": 0.000188642329418886,
      "loss": 0.8322,
      "step": 490
    },
    {
      "epoch": 2.3970944309927362,
      "grad_norm": 0.34765625,
      "learning_rate": 0.00018824698264361306,
      "loss": 0.8349,
      "step": 495
    },
    {
      "epoch": 2.4213075060532687,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0001878453012921891,
      "loss": 0.8324,
      "step": 500
    },
    {
      "epoch": 2.4455205811138017,
      "grad_norm": 0.375,
      "learning_rate": 0.0001874373141982471,
      "loss": 0.8378,
      "step": 505
    },
    {
      "epoch": 2.469733656174334,
      "grad_norm": 0.28125,
      "learning_rate": 0.00018702305064806108,
      "loss": 0.8362,
      "step": 510
    },
    {
      "epoch": 2.4939467312348667,
      "grad_norm": 0.248046875,
      "learning_rate": 0.00018660254037844388,
      "loss": 0.8287,
      "step": 515
    },
    {
      "epoch": 2.5181598062953996,
      "grad_norm": 0.2431640625,
      "learning_rate": 0.00018617581357461253,
      "loss": 0.8367,
      "step": 520
    },
    {
      "epoch": 2.542372881355932,
      "grad_norm": 0.318359375,
      "learning_rate": 0.00018574290086802156,
      "loss": 0.8317,
      "step": 525
    },
    {
      "epoch": 2.566585956416465,
      "grad_norm": 0.2734375,
      "learning_rate": 0.00018530383333416418,
      "loss": 0.8361,
      "step": 530
    },
    {
      "epoch": 2.5907990314769975,
      "grad_norm": 0.279296875,
      "learning_rate": 0.00018485864249034146,
      "loss": 0.8303,
      "step": 535
    },
    {
      "epoch": 2.61501210653753,
      "grad_norm": 0.2890625,
      "learning_rate": 0.00018440736029340022,
      "loss": 0.8296,
      "step": 540
    },
    {
      "epoch": 2.639225181598063,
      "grad_norm": 0.30078125,
      "learning_rate": 0.00018395001913743892,
      "loss": 0.8348,
      "step": 545
    },
    {
      "epoch": 2.663438256658596,
      "grad_norm": 0.275390625,
      "learning_rate": 0.00018348665185148228,
      "loss": 0.8307,
      "step": 550
    },
    {
      "epoch": 2.6876513317191284,
      "grad_norm": 0.34375,
      "learning_rate": 0.0001830172916971249,
      "loss": 0.8316,
      "step": 555
    },
    {
      "epoch": 2.711864406779661,
      "grad_norm": 0.486328125,
      "learning_rate": 0.00018254197236614354,
      "loss": 0.8299,
      "step": 560
    },
    {
      "epoch": 2.736077481840194,
      "grad_norm": 0.3125,
      "learning_rate": 0.0001820607279780787,
      "loss": 0.8278,
      "step": 565
    },
    {
      "epoch": 2.7602905569007263,
      "grad_norm": 0.3359375,
      "learning_rate": 0.00018157359307778545,
      "loss": 0.8317,
      "step": 570
    },
    {
      "epoch": 2.7845036319612593,
      "grad_norm": 0.2734375,
      "learning_rate": 0.00018108060263295362,
      "loss": 0.8246,
      "step": 575
    },
    {
      "epoch": 2.8087167070217918,
      "grad_norm": 0.373046875,
      "learning_rate": 0.00018058179203159788,
      "loss": 0.8284,
      "step": 580
    },
    {
      "epoch": 2.8329297820823243,
      "grad_norm": 0.30859375,
      "learning_rate": 0.00018007719707951745,
      "loss": 0.8273,
      "step": 585
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.279296875,
      "learning_rate": 0.00017956685399772576,
      "loss": 0.8227,
      "step": 590
    },
    {
      "epoch": 2.8813559322033897,
      "grad_norm": 0.341796875,
      "learning_rate": 0.00017905079941985064,
      "loss": 0.8237,
      "step": 595
    },
    {
      "epoch": 2.9055690072639226,
      "grad_norm": 0.259765625,
      "learning_rate": 0.0001785290703895045,
      "loss": 0.8243,
      "step": 600
    },
    {
      "epoch": 2.929782082324455,
      "grad_norm": 0.25,
      "learning_rate": 0.0001780017043576253,
      "loss": 0.82,
      "step": 605
    },
    {
      "epoch": 2.9539951573849876,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0001774687391797882,
      "loss": 0.8223,
      "step": 610
    },
    {
      "epoch": 2.9782082324455206,
      "grad_norm": 0.26953125,
      "learning_rate": 0.00017693021311348837,
      "loss": 0.8266,
      "step": 615
    },
    {
      "epoch": 2.9975786924939465,
      "eval_loss": 2.6475303173065186,
      "eval_runtime": 0.3372,
      "eval_samples_per_second": 29.654,
      "eval_steps_per_second": 2.965,
      "step": 619
    },
    {
      "epoch": 3.002421307506053,
      "grad_norm": 0.30078125,
      "learning_rate": 0.0001763861648153945,
      "loss": 0.8135,
      "step": 620
    },
    {
      "epoch": 3.026634382566586,
      "grad_norm": 0.244140625,
      "learning_rate": 0.00017583663333857417,
      "loss": 0.8029,
      "step": 625
    },
    {
      "epoch": 3.0508474576271185,
      "grad_norm": 0.2294921875,
      "learning_rate": 0.00017528165812969037,
      "loss": 0.799,
      "step": 630
    },
    {
      "epoch": 3.0750605326876514,
      "grad_norm": 0.31640625,
      "learning_rate": 0.00017472127902617009,
      "loss": 0.8072,
      "step": 635
    },
    {
      "epoch": 3.099273607748184,
      "grad_norm": 0.3671875,
      "learning_rate": 0.00017415553625334447,
      "loss": 0.8003,
      "step": 640
    },
    {
      "epoch": 3.123486682808717,
      "grad_norm": 0.291015625,
      "learning_rate": 0.00017358447042156155,
      "loss": 0.8047,
      "step": 645
    },
    {
      "epoch": 3.1476997578692494,
      "grad_norm": 0.35546875,
      "learning_rate": 0.00017300812252327104,
      "loss": 0.8009,
      "step": 650
    },
    {
      "epoch": 3.171912832929782,
      "grad_norm": 0.26953125,
      "learning_rate": 0.00017242653393008183,
      "loss": 0.8049,
      "step": 655
    },
    {
      "epoch": 3.196125907990315,
      "grad_norm": 0.287109375,
      "learning_rate": 0.00017183974638979222,
      "loss": 0.8073,
      "step": 660
    },
    {
      "epoch": 3.2203389830508473,
      "grad_norm": 0.357421875,
      "learning_rate": 0.0001712478020233932,
      "loss": 0.7974,
      "step": 665
    },
    {
      "epoch": 3.2445520581113803,
      "grad_norm": 0.265625,
      "learning_rate": 0.0001706507433220448,
      "loss": 0.803,
      "step": 670
    },
    {
      "epoch": 3.2687651331719128,
      "grad_norm": 0.353515625,
      "learning_rate": 0.00017004861314402615,
      "loss": 0.7959,
      "step": 675
    },
    {
      "epoch": 3.2929782082324457,
      "grad_norm": 0.337890625,
      "learning_rate": 0.0001694414547116588,
      "loss": 0.8046,
      "step": 680
    },
    {
      "epoch": 3.317191283292978,
      "grad_norm": 0.34375,
      "learning_rate": 0.0001688293116082044,
      "loss": 0.805,
      "step": 685
    },
    {
      "epoch": 3.341404358353511,
      "grad_norm": 0.35546875,
      "learning_rate": 0.00016821222777473583,
      "loss": 0.8041,
      "step": 690
    },
    {
      "epoch": 3.3656174334140436,
      "grad_norm": 0.3125,
      "learning_rate": 0.00016759024750698335,
      "loss": 0.7956,
      "step": 695
    },
    {
      "epoch": 3.389830508474576,
      "grad_norm": 0.26953125,
      "learning_rate": 0.00016696341545215475,
      "loss": 0.8002,
      "step": 700
    },
    {
      "epoch": 3.414043583535109,
      "grad_norm": 0.27734375,
      "learning_rate": 0.00016633177660573047,
      "loss": 0.7987,
      "step": 705
    },
    {
      "epoch": 3.4382566585956416,
      "grad_norm": 0.263671875,
      "learning_rate": 0.00016569537630823383,
      "loss": 0.8008,
      "step": 710
    },
    {
      "epoch": 3.4624697336561745,
      "grad_norm": 0.28515625,
      "learning_rate": 0.0001650542602419763,
      "loss": 0.7969,
      "step": 715
    },
    {
      "epoch": 3.486682808716707,
      "grad_norm": 0.30078125,
      "learning_rate": 0.00016440847442777834,
      "loss": 0.7994,
      "step": 720
    },
    {
      "epoch": 3.5108958837772395,
      "grad_norm": 0.318359375,
      "learning_rate": 0.00016375806522166586,
      "loss": 0.7991,
      "step": 725
    },
    {
      "epoch": 3.5351089588377724,
      "grad_norm": 0.283203125,
      "learning_rate": 0.0001631030793115428,
      "loss": 0.8044,
      "step": 730
    },
    {
      "epoch": 3.559322033898305,
      "grad_norm": 0.53515625,
      "learning_rate": 0.00016244356371383962,
      "loss": 0.796,
      "step": 735
    },
    {
      "epoch": 3.583535108958838,
      "grad_norm": 0.279296875,
      "learning_rate": 0.00016177956577013847,
      "loss": 0.7946,
      "step": 740
    },
    {
      "epoch": 3.6077481840193704,
      "grad_norm": 0.25390625,
      "learning_rate": 0.00016111113314377484,
      "loss": 0.8053,
      "step": 745
    },
    {
      "epoch": 3.6319612590799033,
      "grad_norm": 0.251953125,
      "learning_rate": 0.00016043831381641616,
      "loss": 0.8015,
      "step": 750
    },
    {
      "epoch": 3.656174334140436,
      "grad_norm": 0.283203125,
      "learning_rate": 0.00015976115608461758,
      "loss": 0.7998,
      "step": 755
    },
    {
      "epoch": 3.6803874092009687,
      "grad_norm": 0.24609375,
      "learning_rate": 0.00015907970855635512,
      "loss": 0.8,
      "step": 760
    },
    {
      "epoch": 3.7046004842615012,
      "grad_norm": 0.298828125,
      "learning_rate": 0.00015839402014753654,
      "loss": 0.7956,
      "step": 765
    },
    {
      "epoch": 3.7288135593220337,
      "grad_norm": 0.296875,
      "learning_rate": 0.00015770414007848995,
      "loss": 0.7938,
      "step": 770
    },
    {
      "epoch": 3.7530266343825667,
      "grad_norm": 0.25390625,
      "learning_rate": 0.00015701011787043077,
      "loss": 0.7987,
      "step": 775
    },
    {
      "epoch": 3.777239709443099,
      "grad_norm": 0.29296875,
      "learning_rate": 0.00015631200334190675,
      "loss": 0.7965,
      "step": 780
    },
    {
      "epoch": 3.801452784503632,
      "grad_norm": 0.302734375,
      "learning_rate": 0.00015560984660522223,
      "loss": 0.7904,
      "step": 785
    },
    {
      "epoch": 3.8256658595641646,
      "grad_norm": 0.423828125,
      "learning_rate": 0.00015490369806284057,
      "loss": 0.7949,
      "step": 790
    },
    {
      "epoch": 3.849878934624697,
      "grad_norm": 0.32421875,
      "learning_rate": 0.00015419360840376655,
      "loss": 0.7921,
      "step": 795
    },
    {
      "epoch": 3.87409200968523,
      "grad_norm": 0.375,
      "learning_rate": 0.00015347962859990744,
      "loss": 0.7957,
      "step": 800
    },
    {
      "epoch": 3.898305084745763,
      "grad_norm": 0.298828125,
      "learning_rate": 0.00015276180990241427,
      "loss": 0.7894,
      "step": 805
    },
    {
      "epoch": 3.9225181598062955,
      "grad_norm": 0.36328125,
      "learning_rate": 0.00015204020383800292,
      "loss": 0.7923,
      "step": 810
    },
    {
      "epoch": 3.946731234866828,
      "grad_norm": 0.384765625,
      "learning_rate": 0.00015131486220525543,
      "loss": 0.7907,
      "step": 815
    },
    {
      "epoch": 3.970944309927361,
      "grad_norm": 0.283203125,
      "learning_rate": 0.00015058583707090164,
      "loss": 0.7921,
      "step": 820
    },
    {
      "epoch": 3.9951573849878934,
      "grad_norm": 0.291015625,
      "learning_rate": 0.00014985318076608188,
      "loss": 0.7953,
      "step": 825
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.6602656841278076,
      "eval_runtime": 0.2371,
      "eval_samples_per_second": 42.178,
      "eval_steps_per_second": 4.218,
      "step": 826
    },
    {
      "epoch": 4.019370460048426,
      "grad_norm": 0.29296875,
      "learning_rate": 0.00014911694588259038,
      "loss": 0.7705,
      "step": 830
    },
    {
      "epoch": 4.043583535108959,
      "grad_norm": 0.291015625,
      "learning_rate": 0.00014837718526910018,
      "loss": 0.7729,
      "step": 835
    },
    {
      "epoch": 4.067796610169491,
      "grad_norm": 0.279296875,
      "learning_rate": 0.0001476339520273696,
      "loss": 0.7692,
      "step": 840
    },
    {
      "epoch": 4.092009685230024,
      "grad_norm": 0.294921875,
      "learning_rate": 0.00014688729950843035,
      "loss": 0.7757,
      "step": 845
    },
    {
      "epoch": 4.116222760290557,
      "grad_norm": 0.32421875,
      "learning_rate": 0.0001461372813087578,
      "loss": 0.7793,
      "step": 850
    },
    {
      "epoch": 4.14043583535109,
      "grad_norm": 0.27734375,
      "learning_rate": 0.00014538395126642394,
      "loss": 0.7784,
      "step": 855
    },
    {
      "epoch": 4.164648910411622,
      "grad_norm": 0.3671875,
      "learning_rate": 0.0001446273634572326,
      "loss": 0.7735,
      "step": 860
    },
    {
      "epoch": 4.188861985472155,
      "grad_norm": 0.296875,
      "learning_rate": 0.0001438675721908377,
      "loss": 0.7789,
      "step": 865
    },
    {
      "epoch": 4.213075060532688,
      "grad_norm": 0.26171875,
      "learning_rate": 0.00014310463200684502,
      "loss": 0.7703,
      "step": 870
    },
    {
      "epoch": 4.237288135593221,
      "grad_norm": 0.30859375,
      "learning_rate": 0.00014233859767089687,
      "loss": 0.775,
      "step": 875
    },
    {
      "epoch": 4.261501210653753,
      "grad_norm": 0.30859375,
      "learning_rate": 0.00014156952417074115,
      "loss": 0.7792,
      "step": 880
    },
    {
      "epoch": 4.285714285714286,
      "grad_norm": 0.314453125,
      "learning_rate": 0.00014079746671228418,
      "loss": 0.78,
      "step": 885
    },
    {
      "epoch": 4.309927360774818,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0001400224807156278,
      "loss": 0.774,
      "step": 890
    },
    {
      "epoch": 4.3341404358353515,
      "grad_norm": 0.26171875,
      "learning_rate": 0.00013924462181109115,
      "loss": 0.7728,
      "step": 895
    },
    {
      "epoch": 4.358353510895884,
      "grad_norm": 0.306640625,
      "learning_rate": 0.0001384639458352175,
      "loss": 0.7791,
      "step": 900
    },
    {
      "epoch": 4.3825665859564165,
      "grad_norm": 0.279296875,
      "learning_rate": 0.0001376805088267662,
      "loss": 0.773,
      "step": 905
    },
    {
      "epoch": 4.406779661016949,
      "grad_norm": 0.291015625,
      "learning_rate": 0.00013689436702268992,
      "loss": 0.7719,
      "step": 910
    },
    {
      "epoch": 4.4309927360774815,
      "grad_norm": 0.2578125,
      "learning_rate": 0.00013610557685409796,
      "loss": 0.7744,
      "step": 915
    },
    {
      "epoch": 4.455205811138015,
      "grad_norm": 0.3125,
      "learning_rate": 0.00013531419494220548,
      "loss": 0.7737,
      "step": 920
    },
    {
      "epoch": 4.479418886198547,
      "grad_norm": 0.28125,
      "learning_rate": 0.00013452027809426903,
      "loss": 0.7787,
      "step": 925
    },
    {
      "epoch": 4.50363196125908,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0001337238832995089,
      "loss": 0.7687,
      "step": 930
    },
    {
      "epoch": 4.527845036319612,
      "grad_norm": 0.259765625,
      "learning_rate": 0.00013292506772501819,
      "loss": 0.7721,
      "step": 935
    },
    {
      "epoch": 4.552058111380145,
      "grad_norm": 0.30078125,
      "learning_rate": 0.00013212388871165927,
      "loss": 0.7715,
      "step": 940
    },
    {
      "epoch": 4.576271186440678,
      "grad_norm": 0.28125,
      "learning_rate": 0.00013132040376994772,
      "loss": 0.7792,
      "step": 945
    },
    {
      "epoch": 4.600484261501211,
      "grad_norm": 0.251953125,
      "learning_rate": 0.00013051467057592414,
      "loss": 0.7711,
      "step": 950
    },
    {
      "epoch": 4.624697336561743,
      "grad_norm": 0.28515625,
      "learning_rate": 0.00012970674696701387,
      "loss": 0.7781,
      "step": 955
    },
    {
      "epoch": 4.648910411622276,
      "grad_norm": 0.330078125,
      "learning_rate": 0.00012889669093787547,
      "loss": 0.7788,
      "step": 960
    },
    {
      "epoch": 4.673123486682809,
      "grad_norm": 0.296875,
      "learning_rate": 0.0001280845606362376,
      "loss": 0.7768,
      "step": 965
    },
    {
      "epoch": 4.697336561743342,
      "grad_norm": 0.31640625,
      "learning_rate": 0.0001272704143587251,
      "loss": 0.7722,
      "step": 970
    },
    {
      "epoch": 4.721549636803874,
      "grad_norm": 0.275390625,
      "learning_rate": 0.00012645431054667436,
      "loss": 0.771,
      "step": 975
    },
    {
      "epoch": 4.745762711864407,
      "grad_norm": 0.259765625,
      "learning_rate": 0.00012563630778193805,
      "loss": 0.775,
      "step": 980
    },
    {
      "epoch": 4.76997578692494,
      "grad_norm": 0.302734375,
      "learning_rate": 0.00012481646478268028,
      "loss": 0.7726,
      "step": 985
    },
    {
      "epoch": 4.7941888619854724,
      "grad_norm": 0.36328125,
      "learning_rate": 0.00012399484039916147,
      "loss": 0.7687,
      "step": 990
    },
    {
      "epoch": 4.818401937046005,
      "grad_norm": 0.279296875,
      "learning_rate": 0.00012317149360951407,
      "loss": 0.7716,
      "step": 995
    },
    {
      "epoch": 4.842615012106537,
      "grad_norm": 0.275390625,
      "learning_rate": 0.0001223464835155087,
      "loss": 0.7717,
      "step": 1000
    },
    {
      "epoch": 4.86682808716707,
      "grad_norm": 0.291015625,
      "learning_rate": 0.00012151986933831213,
      "loss": 0.773,
      "step": 1005
    },
    {
      "epoch": 4.891041162227603,
      "grad_norm": 0.2490234375,
      "learning_rate": 0.00012069171041423583,
      "loss": 0.7726,
      "step": 1010
    },
    {
      "epoch": 4.915254237288136,
      "grad_norm": 0.25,
      "learning_rate": 0.00011986206619047693,
      "loss": 0.7721,
      "step": 1015
    },
    {
      "epoch": 4.939467312348668,
      "grad_norm": 0.25390625,
      "learning_rate": 0.00011903099622085077,
      "loss": 0.7773,
      "step": 1020
    },
    {
      "epoch": 4.963680387409201,
      "grad_norm": 0.296875,
      "learning_rate": 0.00011819856016151615,
      "loss": 0.776,
      "step": 1025
    },
    {
      "epoch": 4.987893462469733,
      "grad_norm": 0.259765625,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.7708,
      "step": 1030
    },
    {
      "epoch": 4.9975786924939465,
      "eval_loss": 2.6720097064971924,
      "eval_runtime": 0.2452,
      "eval_samples_per_second": 40.776,
      "eval_steps_per_second": 4.078,
      "step": 1032
    },
    {
      "epoch": 5.012106537530267,
      "grad_norm": 0.328125,
      "learning_rate": 0.00011652982888437316,
      "loss": 0.7584,
      "step": 1035
    },
    {
      "epoch": 5.036319612590799,
      "grad_norm": 0.265625,
      "learning_rate": 0.00011569365345202414,
      "loss": 0.7465,
      "step": 1040
    },
    {
      "epoch": 5.060532687651332,
      "grad_norm": 0.251953125,
      "learning_rate": 0.00011485635149228681,
      "loss": 0.7568,
      "step": 1045
    },
    {
      "epoch": 5.084745762711864,
      "grad_norm": 0.318359375,
      "learning_rate": 0.00011401798310866693,
      "loss": 0.7503,
      "step": 1050
    },
    {
      "epoch": 5.108958837772397,
      "grad_norm": 0.271484375,
      "learning_rate": 0.00011317860848122066,
      "loss": 0.7532,
      "step": 1055
    },
    {
      "epoch": 5.13317191283293,
      "grad_norm": 0.271484375,
      "learning_rate": 0.00011233828786223464,
      "loss": 0.7531,
      "step": 1060
    },
    {
      "epoch": 5.157384987893463,
      "grad_norm": 0.310546875,
      "learning_rate": 0.00011149708157190101,
      "loss": 0.7628,
      "step": 1065
    },
    {
      "epoch": 5.181598062953995,
      "grad_norm": 0.2431640625,
      "learning_rate": 0.00011065504999398762,
      "loss": 0.7586,
      "step": 1070
    },
    {
      "epoch": 5.2058111380145276,
      "grad_norm": 0.26953125,
      "learning_rate": 0.00010981225357150327,
      "loss": 0.7605,
      "step": 1075
    },
    {
      "epoch": 5.230024213075061,
      "grad_norm": 0.267578125,
      "learning_rate": 0.00010896875280235907,
      "loss": 0.7627,
      "step": 1080
    },
    {
      "epoch": 5.254237288135593,
      "grad_norm": 0.27734375,
      "learning_rate": 0.00010812460823502597,
      "loss": 0.7562,
      "step": 1085
    },
    {
      "epoch": 5.278450363196126,
      "grad_norm": 0.275390625,
      "learning_rate": 0.00010727988046418809,
      "loss": 0.7571,
      "step": 1090
    },
    {
      "epoch": 5.302663438256658,
      "grad_norm": 0.255859375,
      "learning_rate": 0.00010643463012639337,
      "loss": 0.7571,
      "step": 1095
    },
    {
      "epoch": 5.326876513317191,
      "grad_norm": 0.251953125,
      "learning_rate": 0.00010558891789570082,
      "loss": 0.7583,
      "step": 1100
    },
    {
      "epoch": 5.351089588377724,
      "grad_norm": 0.26953125,
      "learning_rate": 0.00010474280447932519,
      "loss": 0.7622,
      "step": 1105
    },
    {
      "epoch": 5.375302663438257,
      "grad_norm": 0.3203125,
      "learning_rate": 0.00010389635061327925,
      "loss": 0.7554,
      "step": 1110
    },
    {
      "epoch": 5.399515738498789,
      "grad_norm": 0.279296875,
      "learning_rate": 0.00010304961705801415,
      "loss": 0.7598,
      "step": 1115
    },
    {
      "epoch": 5.423728813559322,
      "grad_norm": 0.265625,
      "learning_rate": 0.00010220266459405772,
      "loss": 0.7546,
      "step": 1120
    },
    {
      "epoch": 5.447941888619855,
      "grad_norm": 0.283203125,
      "learning_rate": 0.00010135555401765153,
      "loss": 0.756,
      "step": 1125
    },
    {
      "epoch": 5.472154963680388,
      "grad_norm": 0.296875,
      "learning_rate": 0.00010050834613638695,
      "loss": 0.7612,
      "step": 1130
    },
    {
      "epoch": 5.49636803874092,
      "grad_norm": 0.275390625,
      "learning_rate": 9.966110176484004e-05,
      "loss": 0.7531,
      "step": 1135
    },
    {
      "epoch": 5.520581113801453,
      "grad_norm": 0.2734375,
      "learning_rate": 9.881388172020624e-05,
      "loss": 0.7479,
      "step": 1140
    },
    {
      "epoch": 5.544794188861985,
      "grad_norm": 0.287109375,
      "learning_rate": 9.796674681793471e-05,
      "loss": 0.76,
      "step": 1145
    },
    {
      "epoch": 5.5690072639225185,
      "grad_norm": 0.255859375,
      "learning_rate": 9.711975786736296e-05,
      "loss": 0.7556,
      "step": 1150
    },
    {
      "epoch": 5.593220338983051,
      "grad_norm": 0.2578125,
      "learning_rate": 9.627297566735167e-05,
      "loss": 0.7554,
      "step": 1155
    },
    {
      "epoch": 5.6174334140435835,
      "grad_norm": 0.25390625,
      "learning_rate": 9.542646100192056e-05,
      "loss": 0.7617,
      "step": 1160
    },
    {
      "epoch": 5.641646489104116,
      "grad_norm": 0.2734375,
      "learning_rate": 9.458027463588502e-05,
      "loss": 0.7589,
      "step": 1165
    },
    {
      "epoch": 5.6658595641646485,
      "grad_norm": 0.26171875,
      "learning_rate": 9.373447731049436e-05,
      "loss": 0.7505,
      "step": 1170
    },
    {
      "epoch": 5.690072639225182,
      "grad_norm": 0.30078125,
      "learning_rate": 9.288912973907166e-05,
      "loss": 0.7578,
      "step": 1175
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 0.341796875,
      "learning_rate": 9.20442926026555e-05,
      "loss": 0.7538,
      "step": 1180
    },
    {
      "epoch": 5.738498789346247,
      "grad_norm": 0.28125,
      "learning_rate": 9.120002654564445e-05,
      "loss": 0.7563,
      "step": 1185
    },
    {
      "epoch": 5.762711864406779,
      "grad_norm": 0.2734375,
      "learning_rate": 9.035639217144335e-05,
      "loss": 0.7517,
      "step": 1190
    },
    {
      "epoch": 5.786924939467312,
      "grad_norm": 0.275390625,
      "learning_rate": 8.951345003811366e-05,
      "loss": 0.7513,
      "step": 1195
    },
    {
      "epoch": 5.811138014527845,
      "grad_norm": 0.271484375,
      "learning_rate": 8.867126065402596e-05,
      "loss": 0.7614,
      "step": 1200
    },
    {
      "epoch": 5.835351089588378,
      "grad_norm": 0.255859375,
      "learning_rate": 8.782988447351685e-05,
      "loss": 0.7584,
      "step": 1205
    },
    {
      "epoch": 5.85956416464891,
      "grad_norm": 0.291015625,
      "learning_rate": 8.698938189254912e-05,
      "loss": 0.7596,
      "step": 1210
    },
    {
      "epoch": 5.883777239709443,
      "grad_norm": 0.265625,
      "learning_rate": 8.614981324437667e-05,
      "loss": 0.7523,
      "step": 1215
    },
    {
      "epoch": 5.907990314769976,
      "grad_norm": 0.283203125,
      "learning_rate": 8.53112387952134e-05,
      "loss": 0.7553,
      "step": 1220
    },
    {
      "epoch": 5.932203389830509,
      "grad_norm": 0.302734375,
      "learning_rate": 8.447371873990736e-05,
      "loss": 0.7519,
      "step": 1225
    },
    {
      "epoch": 5.956416464891041,
      "grad_norm": 0.310546875,
      "learning_rate": 8.363731319761964e-05,
      "loss": 0.7574,
      "step": 1230
    },
    {
      "epoch": 5.980629539951574,
      "grad_norm": 0.265625,
      "learning_rate": 8.280208220750896e-05,
      "loss": 0.75,
      "step": 1235
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.6897997856140137,
      "eval_runtime": 0.2374,
      "eval_samples_per_second": 42.117,
      "eval_steps_per_second": 4.212,
      "step": 1239
    },
    {
      "epoch": 6.004842615012106,
      "grad_norm": 0.27734375,
      "learning_rate": 8.196808572442203e-05,
      "loss": 0.7479,
      "step": 1240
    },
    {
      "epoch": 6.0290556900726395,
      "grad_norm": 0.279296875,
      "learning_rate": 8.113538361458963e-05,
      "loss": 0.7448,
      "step": 1245
    },
    {
      "epoch": 6.053268765133172,
      "grad_norm": 0.294921875,
      "learning_rate": 8.030403565132942e-05,
      "loss": 0.7442,
      "step": 1250
    },
    {
      "epoch": 6.0774818401937045,
      "grad_norm": 0.255859375,
      "learning_rate": 7.94741015107552e-05,
      "loss": 0.7455,
      "step": 1255
    },
    {
      "epoch": 6.101694915254237,
      "grad_norm": 0.263671875,
      "learning_rate": 7.864564076749335e-05,
      "loss": 0.7419,
      "step": 1260
    },
    {
      "epoch": 6.12590799031477,
      "grad_norm": 0.26953125,
      "learning_rate": 7.781871289040618e-05,
      "loss": 0.7415,
      "step": 1265
    },
    {
      "epoch": 6.150121065375303,
      "grad_norm": 0.255859375,
      "learning_rate": 7.699337723832337e-05,
      "loss": 0.7472,
      "step": 1270
    },
    {
      "epoch": 6.174334140435835,
      "grad_norm": 0.26953125,
      "learning_rate": 7.616969305578084e-05,
      "loss": 0.7426,
      "step": 1275
    },
    {
      "epoch": 6.198547215496368,
      "grad_norm": 0.255859375,
      "learning_rate": 7.53477194687683e-05,
      "loss": 0.7405,
      "step": 1280
    },
    {
      "epoch": 6.2227602905569,
      "grad_norm": 0.259765625,
      "learning_rate": 7.452751548048471e-05,
      "loss": 0.7412,
      "step": 1285
    },
    {
      "epoch": 6.246973365617434,
      "grad_norm": 0.26953125,
      "learning_rate": 7.370913996710316e-05,
      "loss": 0.7352,
      "step": 1290
    },
    {
      "epoch": 6.271186440677966,
      "grad_norm": 0.322265625,
      "learning_rate": 7.289265167354449e-05,
      "loss": 0.7438,
      "step": 1295
    },
    {
      "epoch": 6.295399515738499,
      "grad_norm": 0.28515625,
      "learning_rate": 7.20781092092604e-05,
      "loss": 0.7374,
      "step": 1300
    },
    {
      "epoch": 6.319612590799031,
      "grad_norm": 0.25,
      "learning_rate": 7.126557104402655e-05,
      "loss": 0.7315,
      "step": 1305
    },
    {
      "epoch": 6.343825665859564,
      "grad_norm": 0.259765625,
      "learning_rate": 7.045509550374509e-05,
      "loss": 0.7378,
      "step": 1310
    },
    {
      "epoch": 6.368038740920097,
      "grad_norm": 0.28515625,
      "learning_rate": 6.964674076625827e-05,
      "loss": 0.7408,
      "step": 1315
    },
    {
      "epoch": 6.39225181598063,
      "grad_norm": 0.318359375,
      "learning_rate": 6.884056485717198e-05,
      "loss": 0.7394,
      "step": 1320
    },
    {
      "epoch": 6.416464891041162,
      "grad_norm": 0.2578125,
      "learning_rate": 6.80366256456908e-05,
      "loss": 0.7362,
      "step": 1325
    },
    {
      "epoch": 6.440677966101695,
      "grad_norm": 0.251953125,
      "learning_rate": 6.723498084046378e-05,
      "loss": 0.7436,
      "step": 1330
    },
    {
      "epoch": 6.464891041162228,
      "grad_norm": 0.2734375,
      "learning_rate": 6.64356879854422e-05,
      "loss": 0.7382,
      "step": 1335
    },
    {
      "epoch": 6.4891041162227605,
      "grad_norm": 0.32421875,
      "learning_rate": 6.563880445574873e-05,
      "loss": 0.7433,
      "step": 1340
    },
    {
      "epoch": 6.513317191283293,
      "grad_norm": 0.310546875,
      "learning_rate": 6.484438745355897e-05,
      "loss": 0.7443,
      "step": 1345
    },
    {
      "epoch": 6.5375302663438255,
      "grad_norm": 0.275390625,
      "learning_rate": 6.405249400399554e-05,
      "loss": 0.7381,
      "step": 1350
    },
    {
      "epoch": 6.561743341404358,
      "grad_norm": 0.322265625,
      "learning_rate": 6.326318095103432e-05,
      "loss": 0.7446,
      "step": 1355
    },
    {
      "epoch": 6.585956416464891,
      "grad_norm": 0.29296875,
      "learning_rate": 6.247650495342435e-05,
      "loss": 0.7485,
      "step": 1360
    },
    {
      "epoch": 6.610169491525424,
      "grad_norm": 0.3125,
      "learning_rate": 6.169252248062057e-05,
      "loss": 0.7444,
      "step": 1365
    },
    {
      "epoch": 6.634382566585956,
      "grad_norm": 0.259765625,
      "learning_rate": 6.091128980873047e-05,
      "loss": 0.7421,
      "step": 1370
    },
    {
      "epoch": 6.658595641646489,
      "grad_norm": 0.2890625,
      "learning_rate": 6.013286301647423e-05,
      "loss": 0.7514,
      "step": 1375
    },
    {
      "epoch": 6.682808716707022,
      "grad_norm": 0.275390625,
      "learning_rate": 5.9357297981159497e-05,
      "loss": 0.7429,
      "step": 1380
    },
    {
      "epoch": 6.707021791767555,
      "grad_norm": 0.259765625,
      "learning_rate": 5.8584650374670135e-05,
      "loss": 0.744,
      "step": 1385
    },
    {
      "epoch": 6.731234866828087,
      "grad_norm": 0.259765625,
      "learning_rate": 5.7814975659470294e-05,
      "loss": 0.7445,
      "step": 1390
    },
    {
      "epoch": 6.75544794188862,
      "grad_norm": 0.267578125,
      "learning_rate": 5.704832908462282e-05,
      "loss": 0.7528,
      "step": 1395
    },
    {
      "epoch": 6.779661016949152,
      "grad_norm": 0.265625,
      "learning_rate": 5.628476568182349e-05,
      "loss": 0.7422,
      "step": 1400
    },
    {
      "epoch": 6.803874092009686,
      "grad_norm": 0.25390625,
      "learning_rate": 5.552434026145097e-05,
      "loss": 0.7377,
      "step": 1405
    },
    {
      "epoch": 6.828087167070218,
      "grad_norm": 0.26953125,
      "learning_rate": 5.476710740863189e-05,
      "loss": 0.7412,
      "step": 1410
    },
    {
      "epoch": 6.852300242130751,
      "grad_norm": 0.259765625,
      "learning_rate": 5.4013121479323026e-05,
      "loss": 0.7445,
      "step": 1415
    },
    {
      "epoch": 6.876513317191283,
      "grad_norm": 0.259765625,
      "learning_rate": 5.326243659640928e-05,
      "loss": 0.7414,
      "step": 1420
    },
    {
      "epoch": 6.900726392251816,
      "grad_norm": 0.279296875,
      "learning_rate": 5.251510664581869e-05,
      "loss": 0.7455,
      "step": 1425
    },
    {
      "epoch": 6.924939467312349,
      "grad_norm": 0.26953125,
      "learning_rate": 5.177118527265438e-05,
      "loss": 0.7436,
      "step": 1430
    },
    {
      "epoch": 6.9491525423728815,
      "grad_norm": 0.248046875,
      "learning_rate": 5.1030725877343744e-05,
      "loss": 0.7402,
      "step": 1435
    },
    {
      "epoch": 6.973365617433414,
      "grad_norm": 0.255859375,
      "learning_rate": 5.0293781611805146e-05,
      "loss": 0.7432,
      "step": 1440
    },
    {
      "epoch": 6.9975786924939465,
      "grad_norm": 0.248046875,
      "learning_rate": 4.956040537563291e-05,
      "loss": 0.7446,
      "step": 1445
    },
    {
      "epoch": 6.9975786924939465,
      "eval_loss": 2.702648639678955,
      "eval_runtime": 0.2447,
      "eval_samples_per_second": 40.862,
      "eval_steps_per_second": 4.086,
      "step": 1445
    },
    {
      "epoch": 7.021791767554479,
      "grad_norm": 0.279296875,
      "learning_rate": 4.8830649812299555e-05,
      "loss": 0.7359,
      "step": 1450
    },
    {
      "epoch": 7.046004842615012,
      "grad_norm": 0.32421875,
      "learning_rate": 4.8104567305377336e-05,
      "loss": 0.7284,
      "step": 1455
    },
    {
      "epoch": 7.070217917675545,
      "grad_norm": 0.28515625,
      "learning_rate": 4.738220997477785e-05,
      "loss": 0.7293,
      "step": 1460
    },
    {
      "epoch": 7.094430992736077,
      "grad_norm": 0.267578125,
      "learning_rate": 4.6663629673010636e-05,
      "loss": 0.7348,
      "step": 1465
    },
    {
      "epoch": 7.11864406779661,
      "grad_norm": 0.267578125,
      "learning_rate": 4.594887798146147e-05,
      "loss": 0.7407,
      "step": 1470
    },
    {
      "epoch": 7.142857142857143,
      "grad_norm": 0.29296875,
      "learning_rate": 4.523800620668921e-05,
      "loss": 0.7301,
      "step": 1475
    },
    {
      "epoch": 7.167070217917676,
      "grad_norm": 0.248046875,
      "learning_rate": 4.4531065376743316e-05,
      "loss": 0.7354,
      "step": 1480
    },
    {
      "epoch": 7.191283292978208,
      "grad_norm": 0.251953125,
      "learning_rate": 4.382810623750073e-05,
      "loss": 0.7351,
      "step": 1485
    },
    {
      "epoch": 7.215496368038741,
      "grad_norm": 0.271484375,
      "learning_rate": 4.312917924902328e-05,
      "loss": 0.7316,
      "step": 1490
    },
    {
      "epoch": 7.239709443099273,
      "grad_norm": 0.267578125,
      "learning_rate": 4.243433458193541e-05,
      "loss": 0.7325,
      "step": 1495
    },
    {
      "epoch": 7.263922518159807,
      "grad_norm": 0.26171875,
      "learning_rate": 4.174362211382315e-05,
      "loss": 0.7304,
      "step": 1500
    },
    {
      "epoch": 7.288135593220339,
      "grad_norm": 0.255859375,
      "learning_rate": 4.1057091425653306e-05,
      "loss": 0.7348,
      "step": 1505
    },
    {
      "epoch": 7.312348668280872,
      "grad_norm": 0.2578125,
      "learning_rate": 4.037479179821485e-05,
      "loss": 0.7358,
      "step": 1510
    },
    {
      "epoch": 7.336561743341404,
      "grad_norm": 0.255859375,
      "learning_rate": 3.969677220858123e-05,
      "loss": 0.7333,
      "step": 1515
    },
    {
      "epoch": 7.3607748184019375,
      "grad_norm": 0.263671875,
      "learning_rate": 3.902308132659457e-05,
      "loss": 0.7344,
      "step": 1520
    },
    {
      "epoch": 7.38498789346247,
      "grad_norm": 0.255859375,
      "learning_rate": 3.8353767511372406e-05,
      "loss": 0.7329,
      "step": 1525
    },
    {
      "epoch": 7.4092009685230025,
      "grad_norm": 0.259765625,
      "learning_rate": 3.768887880783589e-05,
      "loss": 0.7388,
      "step": 1530
    },
    {
      "epoch": 7.433414043583535,
      "grad_norm": 0.25,
      "learning_rate": 3.702846294326137e-05,
      "loss": 0.7419,
      "step": 1535
    },
    {
      "epoch": 7.4576271186440675,
      "grad_norm": 0.26171875,
      "learning_rate": 3.637256732385423e-05,
      "loss": 0.7328,
      "step": 1540
    },
    {
      "epoch": 7.481840193704601,
      "grad_norm": 0.26171875,
      "learning_rate": 3.5721239031346066e-05,
      "loss": 0.7352,
      "step": 1545
    },
    {
      "epoch": 7.506053268765133,
      "grad_norm": 0.255859375,
      "learning_rate": 3.507452481961495e-05,
      "loss": 0.733,
      "step": 1550
    },
    {
      "epoch": 7.530266343825666,
      "grad_norm": 0.26171875,
      "learning_rate": 3.443247111132942e-05,
      "loss": 0.733,
      "step": 1555
    },
    {
      "epoch": 7.554479418886198,
      "grad_norm": 0.2470703125,
      "learning_rate": 3.379512399461603e-05,
      "loss": 0.7331,
      "step": 1560
    },
    {
      "epoch": 7.578692493946731,
      "grad_norm": 0.2578125,
      "learning_rate": 3.316252921975116e-05,
      "loss": 0.7309,
      "step": 1565
    },
    {
      "epoch": 7.602905569007264,
      "grad_norm": 0.2431640625,
      "learning_rate": 3.2534732195876924e-05,
      "loss": 0.7362,
      "step": 1570
    },
    {
      "epoch": 7.627118644067797,
      "grad_norm": 0.25390625,
      "learning_rate": 3.191177798774155e-05,
      "loss": 0.7334,
      "step": 1575
    },
    {
      "epoch": 7.651331719128329,
      "grad_norm": 0.2578125,
      "learning_rate": 3.129371131246459e-05,
      "loss": 0.7329,
      "step": 1580
    },
    {
      "epoch": 7.675544794188862,
      "grad_norm": 0.26171875,
      "learning_rate": 3.0680576536326856e-05,
      "loss": 0.7381,
      "step": 1585
    },
    {
      "epoch": 7.699757869249394,
      "grad_norm": 0.25390625,
      "learning_rate": 3.007241767158604e-05,
      "loss": 0.7282,
      "step": 1590
    },
    {
      "epoch": 7.723970944309928,
      "grad_norm": 0.26953125,
      "learning_rate": 2.9469278373316987e-05,
      "loss": 0.737,
      "step": 1595
    },
    {
      "epoch": 7.74818401937046,
      "grad_norm": 0.25390625,
      "learning_rate": 2.887120193627838e-05,
      "loss": 0.734,
      "step": 1600
    },
    {
      "epoch": 7.772397094430993,
      "grad_norm": 0.255859375,
      "learning_rate": 2.827823129180477e-05,
      "loss": 0.7312,
      "step": 1605
    },
    {
      "epoch": 7.796610169491525,
      "grad_norm": 0.25390625,
      "learning_rate": 2.769040900472488e-05,
      "loss": 0.7346,
      "step": 1610
    },
    {
      "epoch": 7.8208232445520585,
      "grad_norm": 0.25,
      "learning_rate": 2.7107777270306267e-05,
      "loss": 0.7406,
      "step": 1615
    },
    {
      "epoch": 7.845036319612591,
      "grad_norm": 0.25390625,
      "learning_rate": 2.653037791122631e-05,
      "loss": 0.73,
      "step": 1620
    },
    {
      "epoch": 7.8692493946731235,
      "grad_norm": 0.26171875,
      "learning_rate": 2.595825237457026e-05,
      "loss": 0.7352,
      "step": 1625
    },
    {
      "epoch": 7.893462469733656,
      "grad_norm": 0.267578125,
      "learning_rate": 2.5391441728855925e-05,
      "loss": 0.7297,
      "step": 1630
    },
    {
      "epoch": 7.917675544794189,
      "grad_norm": 0.244140625,
      "learning_rate": 2.482998666108577e-05,
      "loss": 0.7303,
      "step": 1635
    },
    {
      "epoch": 7.941888619854722,
      "grad_norm": 0.26953125,
      "learning_rate": 2.427392747382623e-05,
      "loss": 0.7369,
      "step": 1640
    },
    {
      "epoch": 7.966101694915254,
      "grad_norm": 0.263671875,
      "learning_rate": 2.372330408231479e-05,
      "loss": 0.7285,
      "step": 1645
    },
    {
      "epoch": 7.990314769975787,
      "grad_norm": 0.25390625,
      "learning_rate": 2.3178156011594588e-05,
      "loss": 0.7301,
      "step": 1650
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.7081146240234375,
      "eval_runtime": 0.2373,
      "eval_samples_per_second": 42.146,
      "eval_steps_per_second": 4.215,
      "step": 1652
    },
    {
      "epoch": 8.01452784503632,
      "grad_norm": 0.2578125,
      "learning_rate": 2.2638522393677563e-05,
      "loss": 0.7319,
      "step": 1655
    },
    {
      "epoch": 8.038740920096853,
      "grad_norm": 0.255859375,
      "learning_rate": 2.2104441964735033e-05,
      "loss": 0.7268,
      "step": 1660
    },
    {
      "epoch": 8.062953995157384,
      "grad_norm": 0.25390625,
      "learning_rate": 2.1575953062317477e-05,
      "loss": 0.7282,
      "step": 1665
    },
    {
      "epoch": 8.087167070217918,
      "grad_norm": 0.2578125,
      "learning_rate": 2.1053093622602404e-05,
      "loss": 0.7299,
      "step": 1670
    },
    {
      "epoch": 8.111380145278451,
      "grad_norm": 0.2470703125,
      "learning_rate": 2.0535901177671158e-05,
      "loss": 0.7253,
      "step": 1675
    },
    {
      "epoch": 8.135593220338983,
      "grad_norm": 0.251953125,
      "learning_rate": 2.0024412852815032e-05,
      "loss": 0.7229,
      "step": 1680
    },
    {
      "epoch": 8.159806295399516,
      "grad_norm": 0.2490234375,
      "learning_rate": 1.951866536386997e-05,
      "loss": 0.7292,
      "step": 1685
    },
    {
      "epoch": 8.184019370460048,
      "grad_norm": 0.25,
      "learning_rate": 1.90186950145813e-05,
      "loss": 0.726,
      "step": 1690
    },
    {
      "epoch": 8.208232445520581,
      "grad_norm": 0.25390625,
      "learning_rate": 1.8524537693997644e-05,
      "loss": 0.7343,
      "step": 1695
    },
    {
      "epoch": 8.232445520581114,
      "grad_norm": 0.251953125,
      "learning_rate": 1.8036228873894746e-05,
      "loss": 0.7289,
      "step": 1700
    },
    {
      "epoch": 8.256658595641646,
      "grad_norm": 0.25,
      "learning_rate": 1.75538036062292e-05,
      "loss": 0.7272,
      "step": 1705
    },
    {
      "epoch": 8.28087167070218,
      "grad_norm": 0.248046875,
      "learning_rate": 1.707729652062241e-05,
      "loss": 0.7366,
      "step": 1710
    },
    {
      "epoch": 8.305084745762711,
      "grad_norm": 0.25,
      "learning_rate": 1.6606741821874595e-05,
      "loss": 0.7345,
      "step": 1715
    },
    {
      "epoch": 8.329297820823244,
      "grad_norm": 0.24609375,
      "learning_rate": 1.6142173287509866e-05,
      "loss": 0.7367,
      "step": 1720
    },
    {
      "epoch": 8.353510895883778,
      "grad_norm": 0.2451171875,
      "learning_rate": 1.5683624265351125e-05,
      "loss": 0.7297,
      "step": 1725
    },
    {
      "epoch": 8.37772397094431,
      "grad_norm": 0.25,
      "learning_rate": 1.5231127671126677e-05,
      "loss": 0.7269,
      "step": 1730
    },
    {
      "epoch": 8.401937046004843,
      "grad_norm": 0.2490234375,
      "learning_rate": 1.4784715986107267e-05,
      "loss": 0.7332,
      "step": 1735
    },
    {
      "epoch": 8.426150121065376,
      "grad_norm": 0.25,
      "learning_rate": 1.4344421254774431e-05,
      "loss": 0.724,
      "step": 1740
    },
    {
      "epoch": 8.450363196125908,
      "grad_norm": 0.25,
      "learning_rate": 1.3910275082520573e-05,
      "loss": 0.7241,
      "step": 1745
    },
    {
      "epoch": 8.474576271186441,
      "grad_norm": 0.251953125,
      "learning_rate": 1.3482308633379837e-05,
      "loss": 0.7256,
      "step": 1750
    },
    {
      "epoch": 8.498789346246973,
      "grad_norm": 0.251953125,
      "learning_rate": 1.3060552627791444e-05,
      "loss": 0.7281,
      "step": 1755
    },
    {
      "epoch": 8.523002421307506,
      "grad_norm": 0.25390625,
      "learning_rate": 1.2645037340394284e-05,
      "loss": 0.7291,
      "step": 1760
    },
    {
      "epoch": 8.54721549636804,
      "grad_norm": 0.25,
      "learning_rate": 1.2235792597853857e-05,
      "loss": 0.7321,
      "step": 1765
    },
    {
      "epoch": 8.571428571428571,
      "grad_norm": 0.2578125,
      "learning_rate": 1.183284777672109e-05,
      "loss": 0.7317,
      "step": 1770
    },
    {
      "epoch": 8.595641646489105,
      "grad_norm": 0.25390625,
      "learning_rate": 1.1436231801323882e-05,
      "loss": 0.7289,
      "step": 1775
    },
    {
      "epoch": 8.619854721549636,
      "grad_norm": 0.2431640625,
      "learning_rate": 1.1045973141690513e-05,
      "loss": 0.73,
      "step": 1780
    },
    {
      "epoch": 8.64406779661017,
      "grad_norm": 0.251953125,
      "learning_rate": 1.0662099811506332e-05,
      "loss": 0.7302,
      "step": 1785
    },
    {
      "epoch": 8.668280871670703,
      "grad_norm": 0.255859375,
      "learning_rate": 1.02846393661026e-05,
      "loss": 0.7322,
      "step": 1790
    },
    {
      "epoch": 8.692493946731235,
      "grad_norm": 0.2470703125,
      "learning_rate": 9.913618900478594e-06,
      "loss": 0.7335,
      "step": 1795
    },
    {
      "epoch": 8.716707021791768,
      "grad_norm": 0.2470703125,
      "learning_rate": 9.549065047356832e-06,
      "loss": 0.7315,
      "step": 1800
    },
    {
      "epoch": 8.7409200968523,
      "grad_norm": 0.25390625,
      "learning_rate": 9.191003975270985e-06,
      "loss": 0.7294,
      "step": 1805
    },
    {
      "epoch": 8.765133171912833,
      "grad_norm": 0.2490234375,
      "learning_rate": 8.839461386687719e-06,
      "loss": 0.7313,
      "step": 1810
    },
    {
      "epoch": 8.789346246973366,
      "grad_norm": 0.248046875,
      "learning_rate": 8.494462516161606e-06,
      "loss": 0.725,
      "step": 1815
    },
    {
      "epoch": 8.813559322033898,
      "grad_norm": 0.25,
      "learning_rate": 8.156032128523694e-06,
      "loss": 0.7298,
      "step": 1820
    },
    {
      "epoch": 8.837772397094431,
      "grad_norm": 0.2490234375,
      "learning_rate": 7.8241945171039e-06,
      "loss": 0.7303,
      "step": 1825
    },
    {
      "epoch": 8.861985472154963,
      "grad_norm": 0.25,
      "learning_rate": 7.498973501987128e-06,
      "loss": 0.7286,
      "step": 1830
    },
    {
      "epoch": 8.886198547215496,
      "grad_norm": 0.2490234375,
      "learning_rate": 7.180392428303395e-06,
      "loss": 0.7309,
      "step": 1835
    },
    {
      "epoch": 8.91041162227603,
      "grad_norm": 0.25,
      "learning_rate": 6.868474164552185e-06,
      "loss": 0.7292,
      "step": 1840
    },
    {
      "epoch": 8.934624697336561,
      "grad_norm": 0.2470703125,
      "learning_rate": 6.563241100960682e-06,
      "loss": 0.7356,
      "step": 1845
    },
    {
      "epoch": 8.958837772397095,
      "grad_norm": 0.25,
      "learning_rate": 6.264715147876743e-06,
      "loss": 0.728,
      "step": 1850
    },
    {
      "epoch": 8.983050847457626,
      "grad_norm": 0.248046875,
      "learning_rate": 5.9729177341960265e-06,
      "loss": 0.7268,
      "step": 1855
    },
    {
      "epoch": 8.997578692493947,
      "eval_loss": 2.7127978801727295,
      "eval_runtime": 0.2436,
      "eval_samples_per_second": 41.055,
      "eval_steps_per_second": 4.105,
      "step": 1858
    },
    {
      "epoch": 9.00726392251816,
      "grad_norm": 0.2578125,
      "learning_rate": 5.687869805823765e-06,
      "loss": 0.7189,
      "step": 1860
    },
    {
      "epoch": 9.031476997578693,
      "grad_norm": 0.255859375,
      "learning_rate": 5.409591824171311e-06,
      "loss": 0.73,
      "step": 1865
    },
    {
      "epoch": 9.055690072639225,
      "grad_norm": 0.26171875,
      "learning_rate": 5.13810376468723e-06,
      "loss": 0.7255,
      "step": 1870
    },
    {
      "epoch": 9.079903147699758,
      "grad_norm": 0.24609375,
      "learning_rate": 4.8734251154235355e-06,
      "loss": 0.7256,
      "step": 1875
    },
    {
      "epoch": 9.104116222760291,
      "grad_norm": 0.25,
      "learning_rate": 4.61557487563673e-06,
      "loss": 0.7276,
      "step": 1880
    },
    {
      "epoch": 9.128329297820823,
      "grad_norm": 0.251953125,
      "learning_rate": 4.364571554424035e-06,
      "loss": 0.7269,
      "step": 1885
    },
    {
      "epoch": 9.152542372881356,
      "grad_norm": 0.2490234375,
      "learning_rate": 4.1204331693946995e-06,
      "loss": 0.7322,
      "step": 1890
    },
    {
      "epoch": 9.176755447941888,
      "grad_norm": 0.2578125,
      "learning_rate": 3.883177245376735e-06,
      "loss": 0.7291,
      "step": 1895
    },
    {
      "epoch": 9.200968523002421,
      "grad_norm": 0.251953125,
      "learning_rate": 3.6528208131588526e-06,
      "loss": 0.7239,
      "step": 1900
    },
    {
      "epoch": 9.225181598062955,
      "grad_norm": 0.2421875,
      "learning_rate": 3.4293804082680235e-06,
      "loss": 0.7319,
      "step": 1905
    },
    {
      "epoch": 9.249394673123486,
      "grad_norm": 0.248046875,
      "learning_rate": 3.212872069782513e-06,
      "loss": 0.7287,
      "step": 1910
    },
    {
      "epoch": 9.27360774818402,
      "grad_norm": 0.251953125,
      "learning_rate": 3.003311339180526e-06,
      "loss": 0.7316,
      "step": 1915
    },
    {
      "epoch": 9.297820823244551,
      "grad_norm": 0.2490234375,
      "learning_rate": 2.800713259224619e-06,
      "loss": 0.732,
      "step": 1920
    },
    {
      "epoch": 9.322033898305085,
      "grad_norm": 0.2470703125,
      "learning_rate": 2.6050923728818787e-06,
      "loss": 0.7283,
      "step": 1925
    },
    {
      "epoch": 9.346246973365618,
      "grad_norm": 0.251953125,
      "learning_rate": 2.416462722280055e-06,
      "loss": 0.7281,
      "step": 1930
    },
    {
      "epoch": 9.37046004842615,
      "grad_norm": 0.255859375,
      "learning_rate": 2.234837847699478e-06,
      "loss": 0.7308,
      "step": 1935
    },
    {
      "epoch": 9.394673123486683,
      "grad_norm": 0.25390625,
      "learning_rate": 2.060230786601225e-06,
      "loss": 0.7277,
      "step": 1940
    },
    {
      "epoch": 9.418886198547215,
      "grad_norm": 0.255859375,
      "learning_rate": 1.8926540726911469e-06,
      "loss": 0.7252,
      "step": 1945
    },
    {
      "epoch": 9.443099273607748,
      "grad_norm": 0.255859375,
      "learning_rate": 1.7321197350202767e-06,
      "loss": 0.7328,
      "step": 1950
    },
    {
      "epoch": 9.467312348668282,
      "grad_norm": 0.25390625,
      "learning_rate": 1.578639297121265e-06,
      "loss": 0.7235,
      "step": 1955
    },
    {
      "epoch": 9.491525423728813,
      "grad_norm": 0.25,
      "learning_rate": 1.4322237761812408e-06,
      "loss": 0.73,
      "step": 1960
    },
    {
      "epoch": 9.515738498789347,
      "grad_norm": 0.24609375,
      "learning_rate": 1.292883682250956e-06,
      "loss": 0.7251,
      "step": 1965
    },
    {
      "epoch": 9.539951573849878,
      "grad_norm": 0.25390625,
      "learning_rate": 1.160629017490389e-06,
      "loss": 0.7299,
      "step": 1970
    },
    {
      "epoch": 9.564164648910412,
      "grad_norm": 0.244140625,
      "learning_rate": 1.0354692754506845e-06,
      "loss": 0.7286,
      "step": 1975
    },
    {
      "epoch": 9.588377723970945,
      "grad_norm": 0.2490234375,
      "learning_rate": 9.174134403927781e-07,
      "loss": 0.7318,
      "step": 1980
    },
    {
      "epoch": 9.612590799031477,
      "grad_norm": 0.25390625,
      "learning_rate": 8.064699866424109e-07,
      "loss": 0.7284,
      "step": 1985
    },
    {
      "epoch": 9.63680387409201,
      "grad_norm": 0.2490234375,
      "learning_rate": 7.026468779818385e-07,
      "loss": 0.7244,
      "step": 1990
    },
    {
      "epoch": 9.661016949152543,
      "grad_norm": 0.25390625,
      "learning_rate": 6.059515670782334e-07,
      "loss": 0.737,
      "step": 1995
    },
    {
      "epoch": 9.685230024213075,
      "grad_norm": 0.255859375,
      "learning_rate": 5.163909949486234e-07,
      "loss": 0.7309,
      "step": 2000
    },
    {
      "epoch": 9.709443099273608,
      "grad_norm": 0.24609375,
      "learning_rate": 4.3397159046171254e-07,
      "loss": 0.7244,
      "step": 2005
    },
    {
      "epoch": 9.73365617433414,
      "grad_norm": 0.24609375,
      "learning_rate": 3.5869926987636135e-07,
      "loss": 0.725,
      "step": 2010
    },
    {
      "epoch": 9.757869249394673,
      "grad_norm": 0.251953125,
      "learning_rate": 2.9057943641693785e-07,
      "loss": 0.7296,
      "step": 2015
    },
    {
      "epoch": 9.782082324455207,
      "grad_norm": 0.24609375,
      "learning_rate": 2.2961697988544972e-07,
      "loss": 0.7248,
      "step": 2020
    },
    {
      "epoch": 9.806295399515738,
      "grad_norm": 0.251953125,
      "learning_rate": 1.7581627631053643e-07,
      "loss": 0.725,
      "step": 2025
    },
    {
      "epoch": 9.830508474576272,
      "grad_norm": 0.25,
      "learning_rate": 1.2918118763335373e-07,
      "loss": 0.7326,
      "step": 2030
    },
    {
      "epoch": 9.854721549636803,
      "grad_norm": 0.24609375,
      "learning_rate": 8.971506143032882e-08,
      "loss": 0.7304,
      "step": 2035
    },
    {
      "epoch": 9.878934624697337,
      "grad_norm": 0.25390625,
      "learning_rate": 5.7420730672919086e-08,
      "loss": 0.7255,
      "step": 2040
    },
    {
      "epoch": 9.90314769975787,
      "grad_norm": 0.255859375,
      "learning_rate": 3.230051352420826e-08,
      "loss": 0.7271,
      "step": 2045
    },
    {
      "epoch": 9.927360774818402,
      "grad_norm": 0.25390625,
      "learning_rate": 1.435621317251723e-08,
      "loss": 0.7292,
      "step": 2050
    },
    {
      "epoch": 9.951573849878935,
      "grad_norm": 0.24609375,
      "learning_rate": 3.5891177019520273e-09,
      "loss": 0.7296,
      "step": 2055
    },
    {
      "epoch": 9.975786924939467,
      "grad_norm": 0.25390625,
      "learning_rate": 0.0,
      "loss": 0.7318,
      "step": 2060
    },
    {
      "epoch": 9.975786924939467,
      "eval_loss": 2.712730884552002,
      "eval_runtime": 0.2369,
      "eval_samples_per_second": 42.216,
      "eval_steps_per_second": 4.222,
      "step": 2060
    },
    {
      "epoch": 9.975786924939467,
      "step": 2060,
      "total_flos": 3.242523538363515e+18,
      "train_loss": 0.8949183881861492,
      "train_runtime": 11275.6837,
      "train_samples_per_second": 23.387,
      "train_steps_per_second": 0.183
    }
  ],
  "logging_steps": 5,
  "max_steps": 2060,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.242523538363515e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}