{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.999818016378526,
  "eval_steps": 500,
  "global_step": 2747,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 1.9036518335342407,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 1.975,
      "step": 25
    },
    {
      "epoch": 0.02,
      "grad_norm": 1.182765007019043,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 1.7533,
      "step": 50
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.3171876668930054,
      "learning_rate": 5.3818181818181827e-05,
      "loss": 1.4783,
      "step": 75
    },
    {
      "epoch": 0.04,
      "grad_norm": 1.141636610031128,
      "learning_rate": 7.2e-05,
      "loss": 1.2943,
      "step": 100
    },
    {
      "epoch": 0.05,
      "grad_norm": 1.0257848501205444,
      "learning_rate": 9.018181818181819e-05,
      "loss": 1.2977,
      "step": 125
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.816966712474823,
      "learning_rate": 0.00010836363636363638,
      "loss": 1.1967,
      "step": 150
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.92842036485672,
      "learning_rate": 0.00012654545454545454,
      "loss": 1.1982,
      "step": 175
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.8091335892677307,
      "learning_rate": 0.00014472727272727274,
      "loss": 1.2558,
      "step": 200
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.6926524043083191,
      "learning_rate": 0.00016290909090909092,
      "loss": 1.2308,
      "step": 225
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.8470534086227417,
      "learning_rate": 0.0001810909090909091,
      "loss": 1.1464,
      "step": 250
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.9637011289596558,
      "learning_rate": 0.00019927272727272727,
      "loss": 1.2155,
      "step": 275
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.799651026725769,
      "learning_rate": 0.00019805825242718447,
      "loss": 1.1846,
      "step": 300
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.7091360092163086,
      "learning_rate": 0.00019603559870550163,
      "loss": 1.1741,
      "step": 325
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.7197450995445251,
      "learning_rate": 0.0001940129449838188,
      "loss": 1.1863,
      "step": 350
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.6201531291007996,
      "learning_rate": 0.00019199029126213592,
      "loss": 1.1242,
      "step": 375
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.7758994698524475,
      "learning_rate": 0.00018996763754045308,
      "loss": 1.1845,
      "step": 400
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.676243245601654,
      "learning_rate": 0.00018794498381877024,
      "loss": 1.1862,
      "step": 425
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6000869274139404,
      "learning_rate": 0.0001859223300970874,
      "loss": 1.1684,
      "step": 450
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.5969446897506714,
      "learning_rate": 0.00018389967637540456,
      "loss": 1.0784,
      "step": 475
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.6639478206634521,
      "learning_rate": 0.0001818770226537217,
      "loss": 1.2557,
      "step": 500
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.7261694669723511,
      "learning_rate": 0.00017985436893203885,
      "loss": 1.1934,
      "step": 525
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.77945876121521,
      "learning_rate": 0.00017783171521035599,
      "loss": 1.153,
      "step": 550
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.9188840985298157,
      "learning_rate": 0.00017580906148867315,
      "loss": 1.2041,
      "step": 575
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.7868269681930542,
      "learning_rate": 0.00017378640776699028,
      "loss": 1.0461,
      "step": 600
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.6817301511764526,
      "learning_rate": 0.00017176375404530744,
      "loss": 1.134,
      "step": 625
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.8629248142242432,
      "learning_rate": 0.0001697411003236246,
      "loss": 1.1121,
      "step": 650
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.8092464208602905,
      "learning_rate": 0.00016771844660194176,
      "loss": 1.0449,
      "step": 675
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.8356920480728149,
      "learning_rate": 0.00016569579288025892,
      "loss": 1.1207,
      "step": 700
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5683802962303162,
      "learning_rate": 0.00016367313915857605,
      "loss": 1.1117,
      "step": 725
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.6121156215667725,
      "learning_rate": 0.0001616504854368932,
      "loss": 1.0837,
      "step": 750
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.5916587710380554,
      "learning_rate": 0.00015962783171521037,
      "loss": 1.1184,
      "step": 775
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.798072874546051,
      "learning_rate": 0.00015760517799352753,
      "loss": 1.1487,
      "step": 800
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.5768774747848511,
      "learning_rate": 0.0001555825242718447,
      "loss": 1.1142,
      "step": 825
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.1297376155853271,
      "learning_rate": 0.00015355987055016182,
      "loss": 1.0542,
      "step": 850
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.6882333755493164,
      "learning_rate": 0.00015153721682847896,
      "loss": 1.1733,
      "step": 875
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.6317344903945923,
      "learning_rate": 0.00014951456310679611,
      "loss": 1.0417,
      "step": 900
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.6320317387580872,
      "learning_rate": 0.00014749190938511327,
      "loss": 1.1673,
      "step": 925
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.6424859762191772,
      "learning_rate": 0.00014546925566343043,
      "loss": 1.0318,
      "step": 950
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.5994564294815063,
      "learning_rate": 0.00014344660194174757,
      "loss": 1.0861,
      "step": 975
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.7013177871704102,
      "learning_rate": 0.00014142394822006473,
      "loss": 1.1187,
      "step": 1000
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.7184430360794067,
      "learning_rate": 0.0001394012944983819,
      "loss": 1.0835,
      "step": 1025
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.7505134344100952,
      "learning_rate": 0.00013737864077669905,
      "loss": 1.0466,
      "step": 1050
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.8915770053863525,
      "learning_rate": 0.0001353559870550162,
      "loss": 1.0865,
      "step": 1075
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.5262236595153809,
      "learning_rate": 0.00013333333333333334,
      "loss": 1.124,
      "step": 1100
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.7065467238426208,
      "learning_rate": 0.0001313106796116505,
      "loss": 1.1761,
      "step": 1125
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.634323000907898,
      "learning_rate": 0.00012928802588996763,
      "loss": 1.1285,
      "step": 1150
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.6720498204231262,
      "learning_rate": 0.0001272653721682848,
      "loss": 1.0726,
      "step": 1175
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.5689863562583923,
      "learning_rate": 0.00012524271844660195,
      "loss": 1.003,
      "step": 1200
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.6395437717437744,
      "learning_rate": 0.00012322006472491908,
      "loss": 1.0641,
      "step": 1225
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.5762475728988647,
      "learning_rate": 0.00012119741100323624,
      "loss": 1.0343,
      "step": 1250
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.7222539782524109,
      "learning_rate": 0.0001191747572815534,
      "loss": 1.0858,
      "step": 1275
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.7421953678131104,
      "learning_rate": 0.00011715210355987056,
      "loss": 1.0676,
      "step": 1300
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.5756516456604004,
      "learning_rate": 0.0001151294498381877,
      "loss": 1.1356,
      "step": 1325
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.8439106345176697,
      "learning_rate": 0.00011310679611650486,
      "loss": 1.0783,
      "step": 1350
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.6689855456352234,
      "learning_rate": 0.00011108414239482202,
      "loss": 1.063,
      "step": 1375
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.6357136368751526,
      "learning_rate": 0.00010906148867313916,
      "loss": 1.1426,
      "step": 1400
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.5917367935180664,
      "learning_rate": 0.00010703883495145632,
      "loss": 1.0824,
      "step": 1425
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.7326861619949341,
      "learning_rate": 0.00010501618122977346,
      "loss": 1.062,
      "step": 1450
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.7226691246032715,
      "learning_rate": 0.00010299352750809062,
      "loss": 1.0033,
      "step": 1475
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.0422271490097046,
      "learning_rate": 0.00010097087378640778,
      "loss": 1.0165,
      "step": 1500
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.7214282155036926,
      "learning_rate": 9.894822006472492e-05,
      "loss": 1.1014,
      "step": 1525
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.6132606267929077,
      "learning_rate": 9.692556634304207e-05,
      "loss": 1.085,
      "step": 1550
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.6944576501846313,
      "learning_rate": 9.490291262135923e-05,
      "loss": 1.1325,
      "step": 1575
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.5832145810127258,
      "learning_rate": 9.288025889967637e-05,
      "loss": 1.0155,
      "step": 1600
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.8507530689239502,
      "learning_rate": 9.085760517799353e-05,
      "loss": 1.0128,
      "step": 1625
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.6273424625396729,
      "learning_rate": 8.88349514563107e-05,
      "loss": 1.1325,
      "step": 1650
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.7286422848701477,
      "learning_rate": 8.681229773462784e-05,
      "loss": 1.0364,
      "step": 1675
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.7322813272476196,
      "learning_rate": 8.478964401294499e-05,
      "loss": 1.0556,
      "step": 1700
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.8728362917900085,
      "learning_rate": 8.276699029126213e-05,
      "loss": 1.0334,
      "step": 1725
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.621295154094696,
      "learning_rate": 8.074433656957929e-05,
      "loss": 1.0005,
      "step": 1750
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.7512532472610474,
      "learning_rate": 7.872168284789644e-05,
      "loss": 1.06,
      "step": 1775
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.605369508266449,
      "learning_rate": 7.66990291262136e-05,
      "loss": 0.9775,
      "step": 1800
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.6124484539031982,
      "learning_rate": 7.467637540453075e-05,
      "loss": 1.0157,
      "step": 1825
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.7696976661682129,
      "learning_rate": 7.265372168284789e-05,
      "loss": 1.0534,
      "step": 1850
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.8271819353103638,
      "learning_rate": 7.063106796116505e-05,
      "loss": 1.0123,
      "step": 1875
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.6692514419555664,
      "learning_rate": 6.86084142394822e-05,
      "loss": 1.0332,
      "step": 1900
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.6449939012527466,
      "learning_rate": 6.658576051779936e-05,
      "loss": 1.1004,
      "step": 1925
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.7441306114196777,
      "learning_rate": 6.456310679611652e-05,
      "loss": 1.0315,
      "step": 1950
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.5857362747192383,
      "learning_rate": 6.254045307443366e-05,
      "loss": 1.1277,
      "step": 1975
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.49737441539764404,
      "learning_rate": 6.051779935275082e-05,
      "loss": 1.017,
      "step": 2000
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.7484114766120911,
      "learning_rate": 5.8495145631067963e-05,
      "loss": 1.0246,
      "step": 2025
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.8260968923568726,
      "learning_rate": 5.6472491909385117e-05,
      "loss": 1.0697,
      "step": 2050
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.697239100933075,
      "learning_rate": 5.4449838187702276e-05,
      "loss": 1.0593,
      "step": 2075
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.7635684609413147,
      "learning_rate": 5.2427184466019416e-05,
      "loss": 1.0806,
      "step": 2100
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.8260409832000732,
      "learning_rate": 5.0404530744336576e-05,
      "loss": 1.0663,
      "step": 2125
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.6073005199432373,
      "learning_rate": 4.838187702265373e-05,
      "loss": 1.0867,
      "step": 2150
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.5522834658622742,
      "learning_rate": 4.6359223300970875e-05,
      "loss": 1.0729,
      "step": 2175
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.8805132508277893,
      "learning_rate": 4.433656957928803e-05,
      "loss": 1.0101,
      "step": 2200
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.8581610321998596,
      "learning_rate": 4.231391585760518e-05,
      "loss": 1.0228,
      "step": 2225
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.726747989654541,
      "learning_rate": 4.029126213592233e-05,
      "loss": 1.0266,
      "step": 2250
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.5809868574142456,
      "learning_rate": 3.826860841423948e-05,
      "loss": 1.0265,
      "step": 2275
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.6280074715614319,
      "learning_rate": 3.624595469255664e-05,
      "loss": 1.0451,
      "step": 2300
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.8596406579017639,
      "learning_rate": 3.422330097087379e-05,
      "loss": 1.0799,
      "step": 2325
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.7058184742927551,
      "learning_rate": 3.220064724919094e-05,
      "loss": 0.9396,
      "step": 2350
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.7207942605018616,
      "learning_rate": 3.0177993527508093e-05,
      "loss": 1.054,
      "step": 2375
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.91398686170578,
      "learning_rate": 2.8155339805825243e-05,
      "loss": 1.013,
      "step": 2400
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.7698450684547424,
      "learning_rate": 2.6132686084142393e-05,
      "loss": 1.0448,
      "step": 2425
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.7863798141479492,
      "learning_rate": 2.411003236245955e-05,
      "loss": 1.0062,
      "step": 2450
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.6553971767425537,
      "learning_rate": 2.2087378640776702e-05,
      "loss": 0.9851,
      "step": 2475
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.7182110548019409,
      "learning_rate": 2.0064724919093852e-05,
      "loss": 1.004,
      "step": 2500
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.6455408930778503,
      "learning_rate": 1.8042071197411005e-05,
      "loss": 1.0134,
      "step": 2525
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.7096872329711914,
      "learning_rate": 1.6019417475728158e-05,
      "loss": 0.9618,
      "step": 2550
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.5774602890014648,
      "learning_rate": 1.3996763754045308e-05,
      "loss": 1.0738,
      "step": 2575
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.6721768975257874,
      "learning_rate": 1.197411003236246e-05,
      "loss": 1.0201,
      "step": 2600
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5367263555526733,
      "learning_rate": 9.951456310679612e-06,
      "loss": 1.0792,
      "step": 2625
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.7015663385391235,
      "learning_rate": 7.928802588996764e-06,
      "loss": 0.9896,
      "step": 2650
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.5989572405815125,
      "learning_rate": 5.906148867313916e-06,
      "loss": 1.0487,
      "step": 2675
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.9555190801620483,
      "learning_rate": 3.883495145631068e-06,
      "loss": 1.0667,
      "step": 2700
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.6964536905288696,
      "learning_rate": 1.8608414239482202e-06,
      "loss": 1.0087,
      "step": 2725
    }
  ],
  "logging_steps": 25,
  "max_steps": 2747,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 4.805027299118285e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}