{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.997134670487107,
  "eval_steps": 50,
  "global_step": 1919,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14,
      "grad_norm": 0.1376492828130722,
      "learning_rate": 2.5e-05,
      "loss": 0.9919,
      "step": 25
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.14622816443443298,
      "learning_rate": 5e-05,
      "loss": 0.8765,
      "step": 50
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.1424524486064911,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.6284,
      "step": 75
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.09082940965890884,
      "learning_rate": 0.0001,
      "loss": 0.4119,
      "step": 100
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.11173515766859055,
      "learning_rate": 9.92603550295858e-05,
      "loss": 0.3504,
      "step": 125
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.09382566064596176,
      "learning_rate": 9.85207100591716e-05,
      "loss": 0.3142,
      "step": 150
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.11800573766231537,
      "learning_rate": 9.77810650887574e-05,
      "loss": 0.299,
      "step": 175
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.1385473757982254,
      "learning_rate": 9.70414201183432e-05,
      "loss": 0.2783,
      "step": 200
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.1731497198343277,
      "learning_rate": 9.6301775147929e-05,
      "loss": 0.2583,
      "step": 225
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.16235513985157013,
      "learning_rate": 9.55621301775148e-05,
      "loss": 0.2408,
      "step": 250
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.2389882653951645,
      "learning_rate": 9.48224852071006e-05,
      "loss": 0.2279,
      "step": 275
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.2514766454696655,
      "learning_rate": 9.408284023668639e-05,
      "loss": 0.2109,
      "step": 300
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.3435331881046295,
      "learning_rate": 9.33431952662722e-05,
      "loss": 0.2009,
      "step": 325
    },
    {
      "epoch": 2.01,
      "grad_norm": 0.2297317236661911,
      "learning_rate": 9.260355029585799e-05,
      "loss": 0.1904,
      "step": 350
    },
    {
      "epoch": 2.15,
      "grad_norm": 0.23601436614990234,
      "learning_rate": 9.186390532544379e-05,
      "loss": 0.1728,
      "step": 375
    },
    {
      "epoch": 2.29,
      "grad_norm": 0.26974865794181824,
      "learning_rate": 9.112426035502959e-05,
      "loss": 0.1658,
      "step": 400
    },
    {
      "epoch": 2.44,
      "grad_norm": 0.3188651502132416,
      "learning_rate": 9.038461538461538e-05,
      "loss": 0.1619,
      "step": 425
    },
    {
      "epoch": 2.58,
      "grad_norm": 0.40681496262550354,
      "learning_rate": 8.96449704142012e-05,
      "loss": 0.1496,
      "step": 450
    },
    {
      "epoch": 2.72,
      "grad_norm": 0.32123222947120667,
      "learning_rate": 8.8905325443787e-05,
      "loss": 0.1498,
      "step": 475
    },
    {
      "epoch": 2.87,
      "grad_norm": 0.2911372482776642,
      "learning_rate": 8.816568047337278e-05,
      "loss": 0.1378,
      "step": 500
    },
    {
      "epoch": 3.01,
      "grad_norm": 0.28380608558654785,
      "learning_rate": 8.742603550295858e-05,
      "loss": 0.1355,
      "step": 525
    },
    {
      "epoch": 3.15,
      "grad_norm": 0.3216865062713623,
      "learning_rate": 8.668639053254438e-05,
      "loss": 0.1302,
      "step": 550
    },
    {
      "epoch": 3.3,
      "grad_norm": 0.29950952529907227,
      "learning_rate": 8.594674556213019e-05,
      "loss": 0.1209,
      "step": 575
    },
    {
      "epoch": 3.44,
      "grad_norm": 0.2643551826477051,
      "learning_rate": 8.520710059171599e-05,
      "loss": 0.1222,
      "step": 600
    },
    {
      "epoch": 3.58,
      "grad_norm": 0.31945034861564636,
      "learning_rate": 8.446745562130178e-05,
      "loss": 0.1172,
      "step": 625
    },
    {
      "epoch": 3.72,
      "grad_norm": 0.28063860535621643,
      "learning_rate": 8.372781065088757e-05,
      "loss": 0.1162,
      "step": 650
    },
    {
      "epoch": 3.87,
      "grad_norm": 0.30673155188560486,
      "learning_rate": 8.298816568047337e-05,
      "loss": 0.1127,
      "step": 675
    },
    {
      "epoch": 4.01,
      "grad_norm": 0.26293325424194336,
      "learning_rate": 8.224852071005918e-05,
      "loss": 0.1078,
      "step": 700
    },
    {
      "epoch": 4.15,
      "grad_norm": 0.3006896674633026,
      "learning_rate": 8.150887573964498e-05,
      "loss": 0.1039,
      "step": 725
    },
    {
      "epoch": 4.3,
      "grad_norm": 0.22531060874462128,
      "learning_rate": 8.076923076923078e-05,
      "loss": 0.1026,
      "step": 750
    },
    {
      "epoch": 4.44,
      "grad_norm": 0.27850422263145447,
      "learning_rate": 8.002958579881658e-05,
      "loss": 0.1038,
      "step": 775
    },
    {
      "epoch": 4.58,
      "grad_norm": 0.24622277915477753,
      "learning_rate": 7.928994082840237e-05,
      "loss": 0.1004,
      "step": 800
    },
    {
      "epoch": 4.73,
      "grad_norm": 0.2631712555885315,
      "learning_rate": 7.855029585798817e-05,
      "loss": 0.0944,
      "step": 825
    },
    {
      "epoch": 4.87,
      "grad_norm": 0.2231552004814148,
      "learning_rate": 7.781065088757397e-05,
      "loss": 0.0937,
      "step": 850
    },
    {
      "epoch": 5.01,
      "grad_norm": 0.2672629952430725,
      "learning_rate": 7.707100591715977e-05,
      "loss": 0.0908,
      "step": 875
    },
    {
      "epoch": 5.16,
      "grad_norm": 0.32010477781295776,
      "learning_rate": 7.633136094674557e-05,
      "loss": 0.0949,
      "step": 900
    },
    {
      "epoch": 5.3,
      "grad_norm": 0.20605669915676117,
      "learning_rate": 7.559171597633137e-05,
      "loss": 0.0883,
      "step": 925
    },
    {
      "epoch": 5.44,
      "grad_norm": 0.2745099663734436,
      "learning_rate": 7.485207100591716e-05,
      "loss": 0.0881,
      "step": 950
    },
    {
      "epoch": 5.59,
      "grad_norm": 0.22129909694194794,
      "learning_rate": 7.411242603550296e-05,
      "loss": 0.0861,
      "step": 975
    },
    {
      "epoch": 5.73,
      "grad_norm": 0.19070565700531006,
      "learning_rate": 7.337278106508876e-05,
      "loss": 0.0874,
      "step": 1000
    },
    {
      "epoch": 5.87,
      "grad_norm": 0.27825799584388733,
      "learning_rate": 7.263313609467456e-05,
      "loss": 0.0846,
      "step": 1025
    },
    {
      "epoch": 6.02,
      "grad_norm": 0.28191158175468445,
      "learning_rate": 7.189349112426036e-05,
      "loss": 0.0859,
      "step": 1050
    },
    {
      "epoch": 6.16,
      "grad_norm": 0.2130414992570877,
      "learning_rate": 7.115384615384616e-05,
      "loss": 0.0843,
      "step": 1075
    },
    {
      "epoch": 6.3,
      "grad_norm": 0.2572125494480133,
      "learning_rate": 7.041420118343195e-05,
      "loss": 0.0829,
      "step": 1100
    },
    {
      "epoch": 6.45,
      "grad_norm": 0.2799431383609772,
      "learning_rate": 6.967455621301775e-05,
      "loss": 0.0794,
      "step": 1125
    },
    {
      "epoch": 6.59,
      "grad_norm": 0.22224412858486176,
      "learning_rate": 6.893491124260355e-05,
      "loss": 0.0807,
      "step": 1150
    },
    {
      "epoch": 6.73,
      "grad_norm": 0.19804546236991882,
      "learning_rate": 6.819526627218935e-05,
      "loss": 0.077,
      "step": 1175
    },
    {
      "epoch": 6.88,
      "grad_norm": 0.2524961233139038,
      "learning_rate": 6.745562130177515e-05,
      "loss": 0.0802,
      "step": 1200
    },
    {
      "epoch": 7.02,
      "grad_norm": 0.18262654542922974,
      "learning_rate": 6.671597633136095e-05,
      "loss": 0.0757,
      "step": 1225
    },
    {
      "epoch": 7.16,
      "grad_norm": 0.1993337720632553,
      "learning_rate": 6.597633136094676e-05,
      "loss": 0.0766,
      "step": 1250
    },
    {
      "epoch": 7.31,
      "grad_norm": 0.19785763323307037,
      "learning_rate": 6.523668639053254e-05,
      "loss": 0.0751,
      "step": 1275
    },
    {
      "epoch": 7.45,
      "grad_norm": 0.19745734333992004,
      "learning_rate": 6.449704142011834e-05,
      "loss": 0.0768,
      "step": 1300
    },
    {
      "epoch": 7.59,
      "grad_norm": 0.17973637580871582,
      "learning_rate": 6.375739644970414e-05,
      "loss": 0.0738,
      "step": 1325
    },
    {
      "epoch": 7.74,
      "grad_norm": 0.1808304488658905,
      "learning_rate": 6.301775147928994e-05,
      "loss": 0.0757,
      "step": 1350
    },
    {
      "epoch": 7.88,
      "grad_norm": 0.19356884062290192,
      "learning_rate": 6.227810650887575e-05,
      "loss": 0.0738,
      "step": 1375
    },
    {
      "epoch": 8.02,
      "grad_norm": 0.19427159428596497,
      "learning_rate": 6.153846153846155e-05,
      "loss": 0.0748,
      "step": 1400
    },
    {
      "epoch": 8.17,
      "grad_norm": 0.16624993085861206,
      "learning_rate": 6.079881656804735e-05,
      "loss": 0.0737,
      "step": 1425
    },
    {
      "epoch": 8.31,
      "grad_norm": 0.2472502738237381,
      "learning_rate": 6.005917159763313e-05,
      "loss": 0.0726,
      "step": 1450
    },
    {
      "epoch": 8.45,
      "grad_norm": 0.20945917069911957,
      "learning_rate": 5.931952662721894e-05,
      "loss": 0.0703,
      "step": 1475
    },
    {
      "epoch": 8.6,
      "grad_norm": 0.21044062077999115,
      "learning_rate": 5.8579881656804736e-05,
      "loss": 0.0705,
      "step": 1500
    },
    {
      "epoch": 8.74,
      "grad_norm": 0.1783844232559204,
      "learning_rate": 5.7840236686390534e-05,
      "loss": 0.0711,
      "step": 1525
    },
    {
      "epoch": 8.88,
      "grad_norm": 0.1689521223306656,
      "learning_rate": 5.710059171597634e-05,
      "loss": 0.0738,
      "step": 1550
    },
    {
      "epoch": 9.03,
      "grad_norm": 0.1530502885580063,
      "learning_rate": 5.636094674556214e-05,
      "loss": 0.0684,
      "step": 1575
    },
    {
      "epoch": 9.17,
      "grad_norm": 0.1769784688949585,
      "learning_rate": 5.562130177514793e-05,
      "loss": 0.0699,
      "step": 1600
    },
    {
      "epoch": 9.31,
      "grad_norm": 0.19619758427143097,
      "learning_rate": 5.488165680473373e-05,
      "loss": 0.0681,
      "step": 1625
    },
    {
      "epoch": 9.46,
      "grad_norm": 0.19810470938682556,
      "learning_rate": 5.4142011834319526e-05,
      "loss": 0.071,
      "step": 1650
    },
    {
      "epoch": 9.6,
      "grad_norm": 0.183139830827713,
      "learning_rate": 5.340236686390533e-05,
      "loss": 0.0675,
      "step": 1675
    },
    {
      "epoch": 9.74,
      "grad_norm": 0.22580993175506592,
      "learning_rate": 5.266272189349113e-05,
      "loss": 0.068,
      "step": 1700
    },
    {
      "epoch": 9.89,
      "grad_norm": 0.1878005862236023,
      "learning_rate": 5.192307692307693e-05,
      "loss": 0.0682,
      "step": 1725
    },
    {
      "epoch": 10.03,
      "grad_norm": 0.17333678901195526,
      "learning_rate": 5.118343195266272e-05,
      "loss": 0.0683,
      "step": 1750
    },
    {
      "epoch": 10.17,
      "grad_norm": 0.16736529767513275,
      "learning_rate": 5.044378698224852e-05,
      "loss": 0.0639,
      "step": 1775
    },
    {
      "epoch": 10.32,
      "grad_norm": 0.15685293078422546,
      "learning_rate": 4.970414201183432e-05,
      "loss": 0.0682,
      "step": 1800
    },
    {
      "epoch": 10.46,
      "grad_norm": 0.16026218235492706,
      "learning_rate": 4.896449704142012e-05,
      "loss": 0.0657,
      "step": 1825
    },
    {
      "epoch": 10.6,
      "grad_norm": 0.17992612719535828,
      "learning_rate": 4.822485207100592e-05,
      "loss": 0.0648,
      "step": 1850
    },
    {
      "epoch": 10.74,
      "grad_norm": 0.17420756816864014,
      "learning_rate": 4.748520710059172e-05,
      "loss": 0.0664,
      "step": 1875
    },
    {
      "epoch": 10.89,
      "grad_norm": 0.14792782068252563,
      "learning_rate": 4.674556213017752e-05,
      "loss": 0.0668,
      "step": 1900
    }
  ],
  "logging_steps": 25,
  "max_steps": 3480,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 100,
  "total_flos": 1.371816806339666e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}