{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9618467457518436,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.6595744680851064e-05,
      "loss": 2.0914,
      "step": 25
    },
    {
      "epoch": 0.02,
      "learning_rate": 5.319148936170213e-05,
      "loss": 1.8503,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 7.978723404255319e-05,
      "loss": 1.3555,
      "step": 75
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00010638297872340425,
      "loss": 1.0528,
      "step": 100
    },
    {
      "epoch": 0.03,
      "eval_loss": 1.3641345500946045,
      "eval_runtime": 76.7384,
      "eval_samples_per_second": 7.089,
      "eval_steps_per_second": 0.886,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00013297872340425532,
      "loss": 0.9497,
      "step": 125
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00015957446808510637,
      "loss": 1.0081,
      "step": 150
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00018617021276595746,
      "loss": 0.9365,
      "step": 175
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019999805857737166,
      "loss": 0.9364,
      "step": 200
    },
    {
      "epoch": 0.06,
      "eval_loss": 1.2751965522766113,
      "eval_runtime": 76.7874,
      "eval_samples_per_second": 7.084,
      "eval_steps_per_second": 0.886,
      "step": 200
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001999815435109719,
      "loss": 0.9226,
      "step": 225
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019994817911010517,
      "loss": 0.8396,
      "step": 250
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001998979709974887,
      "loss": 0.8687,
      "step": 275
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019983092763441697,
      "loss": 0.8593,
      "step": 300
    },
    {
      "epoch": 0.1,
      "eval_loss": 1.2590458393096924,
      "eval_runtime": 76.8321,
      "eval_samples_per_second": 7.08,
      "eval_steps_per_second": 0.885,
      "step": 300
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001997470603193358,
      "loss": 0.8064,
      "step": 325
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001996463831859381,
      "loss": 0.7851,
      "step": 350
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019952891320078236,
      "loss": 0.8516,
      "step": 375
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001993946701604329,
      "loss": 0.8161,
      "step": 400
    },
    {
      "epoch": 0.13,
      "eval_loss": 1.1581201553344727,
      "eval_runtime": 76.7611,
      "eval_samples_per_second": 7.087,
      "eval_steps_per_second": 0.886,
      "step": 400
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019924367668812417,
      "loss": 0.838,
      "step": 425
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019907595822994773,
      "loss": 0.8271,
      "step": 450
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019889154305056426,
      "loss": 0.8005,
      "step": 475
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019869046222844008,
      "loss": 0.7916,
      "step": 500
    },
    {
      "epoch": 0.16,
      "eval_loss": 1.1132324934005737,
      "eval_runtime": 76.7669,
      "eval_samples_per_second": 7.086,
      "eval_steps_per_second": 0.886,
      "step": 500
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00019847274965060984,
      "loss": 0.7991,
      "step": 525
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019823844200696564,
      "loss": 0.8387,
      "step": 550
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019798757878407386,
      "loss": 0.8196,
      "step": 575
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019772020225852077,
      "loss": 0.7508,
      "step": 600
    },
    {
      "epoch": 0.19,
      "eval_loss": 1.0564008951187134,
      "eval_runtime": 76.7274,
      "eval_samples_per_second": 7.09,
      "eval_steps_per_second": 0.886,
      "step": 600
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001974363574897878,
      "loss": 0.8356,
      "step": 625
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019713609231265805,
      "loss": 0.7036,
      "step": 650
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001968194573291549,
      "loss": 0.8082,
      "step": 675
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019648650590001418,
      "loss": 0.7974,
      "step": 700
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.1469172239303589,
      "eval_runtime": 76.886,
      "eval_samples_per_second": 7.075,
      "eval_steps_per_second": 0.884,
      "step": 700
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00019613729413569194,
      "loss": 0.7804,
      "step": 725
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00019577188088690803,
      "loss": 0.7398,
      "step": 750
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019539032773472866,
      "loss": 0.7178,
      "step": 775
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00019499269898018832,
      "loss": 0.7303,
      "step": 800
    },
    {
      "epoch": 0.26,
      "eval_loss": 1.1135411262512207,
      "eval_runtime": 76.7509,
      "eval_samples_per_second": 7.088,
      "eval_steps_per_second": 0.886,
      "step": 800
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00019457906163345353,
      "loss": 0.7133,
      "step": 825
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00019414948540253,
      "loss": 0.744,
      "step": 850
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.000193704042681515,
      "loss": 0.7195,
      "step": 875
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019324280853839744,
      "loss": 0.6909,
      "step": 900
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.1239628791809082,
      "eval_runtime": 76.7964,
      "eval_samples_per_second": 7.084,
      "eval_steps_per_second": 0.885,
      "step": 900
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019276586070240682,
      "loss": 0.7061,
      "step": 925
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019227327955091412,
      "loss": 0.694,
      "step": 950
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019176514809588616,
      "loss": 0.707,
      "step": 975
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001912415519698961,
      "loss": 0.7267,
      "step": 1000
    },
    {
      "epoch": 0.32,
      "eval_loss": 1.0490331649780273,
      "eval_runtime": 76.7671,
      "eval_samples_per_second": 7.086,
      "eval_steps_per_second": 0.886,
      "step": 1000
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019070257941169224,
      "loss": 0.694,
      "step": 1025
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00019014832125132763,
      "loss": 0.638,
      "step": 1050
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.000189578870894853,
      "loss": 0.6951,
      "step": 1075
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00018899432430857552,
      "loss": 0.6649,
      "step": 1100
    },
    {
      "epoch": 0.35,
      "eval_loss": 1.127469539642334,
      "eval_runtime": 76.7816,
      "eval_samples_per_second": 7.085,
      "eval_steps_per_second": 0.886,
      "step": 1100
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00018839478000288627,
      "loss": 0.7284,
      "step": 1125
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00018778033901565856,
      "loss": 0.6459,
      "step": 1150
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0001871511048952208,
      "loss": 0.6616,
      "step": 1175
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0001865071836829061,
      "loss": 0.6306,
      "step": 1200
    },
    {
      "epoch": 0.38,
      "eval_loss": 1.0801633596420288,
      "eval_runtime": 76.7811,
      "eval_samples_per_second": 7.085,
      "eval_steps_per_second": 0.886,
      "step": 1200
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00018584868389518152,
      "loss": 0.7158,
      "step": 1225
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00018517571650536056,
      "loss": 0.601,
      "step": 1250
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001844883949249013,
      "loss": 0.6362,
      "step": 1275
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00018378683498429402,
      "loss": 0.5795,
      "step": 1300
    },
    {
      "epoch": 0.42,
      "eval_loss": 1.1198811531066895,
      "eval_runtime": 76.7858,
      "eval_samples_per_second": 7.085,
      "eval_steps_per_second": 0.886,
      "step": 1300
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00018307115491354064,
      "loss": 0.6037,
      "step": 1325
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001823414753222303,
      "loss": 0.5866,
      "step": 1350
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00018159791917921362,
      "loss": 0.6389,
      "step": 1375
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00018084061179187946,
      "loss": 0.6039,
      "step": 1400
    },
    {
      "epoch": 0.45,
      "eval_loss": 1.1011444330215454,
      "eval_runtime": 76.8374,
      "eval_samples_per_second": 7.08,
      "eval_steps_per_second": 0.885,
      "step": 1400
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00018006968078503746,
      "loss": 0.5734,
      "step": 1425
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001792852560794103,
      "loss": 0.565,
      "step": 1450
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017848746986973883,
      "loss": 0.6221,
      "step": 1475
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00017767645660250384,
      "loss": 0.6266,
      "step": 1500
    },
    {
      "epoch": 0.48,
      "eval_loss": 1.0841057300567627,
      "eval_runtime": 76.7968,
      "eval_samples_per_second": 7.084,
      "eval_steps_per_second": 0.885,
      "step": 1500
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017685235295326893,
      "loss": 0.5798,
      "step": 1525
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00017601529780364688,
      "loss": 0.5628,
      "step": 1550
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001751654322178951,
      "loss": 0.5943,
      "step": 1575
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017430289941914258,
      "loss": 0.57,
      "step": 1600
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.067173957824707,
      "eval_runtime": 76.7443,
      "eval_samples_per_second": 7.088,
      "eval_steps_per_second": 0.886,
      "step": 1600
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00017342784476525347,
      "loss": 0.6021,
      "step": 1625
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017254041572433058,
      "loss": 0.5314,
      "step": 1650
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00017164076184986345,
      "loss": 0.5162,
      "step": 1675
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00017072903475552503,
      "loss": 0.5693,
      "step": 1700
    },
    {
      "epoch": 0.55,
      "eval_loss": 1.059133768081665,
      "eval_runtime": 76.8125,
      "eval_samples_per_second": 7.082,
      "eval_steps_per_second": 0.885,
      "step": 1700
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00016980538808962102,
      "loss": 0.5466,
      "step": 1725
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00016886997750919619,
      "loss": 0.4934,
      "step": 1750
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001679229606538026,
      "loss": 0.5666,
      "step": 1775
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00016696449711893344,
      "loss": 0.4821,
      "step": 1800
    },
    {
      "epoch": 0.58,
      "eval_loss": 1.0875073671340942,
      "eval_runtime": 76.8354,
      "eval_samples_per_second": 7.08,
      "eval_steps_per_second": 0.885,
      "step": 1800
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00016599474842912722,
      "loss": 0.5816,
      "step": 1825
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001650138780107469,
      "loss": 0.5616,
      "step": 1850
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00016402205116443868,
      "loss": 0.5239,
      "step": 1875
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00016301943503727468,
      "loss": 0.4953,
      "step": 1900
    },
    {
      "epoch": 0.61,
      "eval_loss": 1.0953130722045898,
      "eval_runtime": 76.7571,
      "eval_samples_per_second": 7.087,
      "eval_steps_per_second": 0.886,
      "step": 1900
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00016200619859458467,
      "loss": 0.5118,
      "step": 1925
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00016098251259148125,
      "loss": 0.5194,
      "step": 1950
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00015994854954408337,
      "loss": 0.5503,
      "step": 1975
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00015890448370044317,
      "loss": 0.4673,
      "step": 2000
    },
    {
      "epoch": 0.64,
      "eval_loss": 1.1145678758621216,
      "eval_runtime": 76.7971,
      "eval_samples_per_second": 7.084,
      "eval_steps_per_second": 0.885,
      "step": 2000
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001578504910111811,
      "loss": 0.5086,
      "step": 2025
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00015678674909983364,
      "loss": 0.4686,
      "step": 2050
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00015571343723291963,
      "loss": 0.5025,
      "step": 2075
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00015463073628972936,
      "loss": 0.5223,
      "step": 2100
    },
    {
      "epoch": 0.67,
      "eval_loss": 1.075910210609436,
      "eval_runtime": 76.8054,
      "eval_samples_per_second": 7.083,
      "eval_steps_per_second": 0.885,
      "step": 2100
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.000153538828731842,
      "loss": 0.4457,
      "step": 2125
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00015243789857237645,
      "loss": 0.471,
      "step": 2150
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00015132813134498054,
      "loss": 0.4692,
      "step": 2175
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.000150209714072564,
      "loss": 0.4661,
      "step": 2200
    },
    {
      "epoch": 0.71,
      "eval_loss": 1.0707464218139648,
      "eval_runtime": 76.7556,
      "eval_samples_per_second": 7.087,
      "eval_steps_per_second": 0.886,
      "step": 2200
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00014908283523578077,
      "loss": 0.442,
      "step": 2225
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00014794768474126518,
      "loss": 0.5075,
      "step": 2250
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00014680445388962814,
      "loss": 0.4969,
      "step": 2275
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00014565333534321826,
      "loss": 0.4642,
      "step": 2300
    },
    {
      "epoch": 0.74,
      "eval_loss": 1.0463252067565918,
      "eval_runtime": 76.7954,
      "eval_samples_per_second": 7.084,
      "eval_steps_per_second": 0.885,
      "step": 2300
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001444945230936536,
      "loss": 0.4412,
      "step": 2325
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00014332821242912927,
      "loss": 0.4657,
      "step": 2350
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001421545999015066,
      "loss": 0.4311,
      "step": 2375
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00014097388329318943,
      "loss": 0.45,
      "step": 2400
    },
    {
      "epoch": 0.77,
      "eval_loss": 1.076710820198059,
      "eval_runtime": 76.7708,
      "eval_samples_per_second": 7.086,
      "eval_steps_per_second": 0.886,
      "step": 2400
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00013978626158379285,
      "loss": 0.4336,
      "step": 2425
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00013859193491661036,
      "loss": 0.4002,
      "step": 2450
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001373911045648846,
      "loss": 0.4141,
      "step": 2475
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00013618397289788832,
      "loss": 0.452,
      "step": 2500
    },
    {
      "epoch": 0.8,
      "eval_loss": 1.085994005203247,
      "eval_runtime": 76.8556,
      "eval_samples_per_second": 7.078,
      "eval_steps_per_second": 0.885,
      "step": 2500
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00013497074334681966,
      "loss": 0.3574,
      "step": 2525
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00013375162037051954,
      "loss": 0.4107,
      "step": 2550
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.000132526809421015,
      "loss": 0.4616,
      "step": 2575
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001312965169088957,
      "loss": 0.4397,
      "step": 2600
    },
    {
      "epoch": 0.83,
      "eval_loss": 1.088114619255066,
      "eval_runtime": 76.7923,
      "eval_samples_per_second": 7.084,
      "eval_steps_per_second": 0.886,
      "step": 2600
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00013006095016852848,
      "loss": 0.3993,
      "step": 2625
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0001288203174231168,
      "loss": 0.3973,
      "step": 2650
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00012757482774960964,
      "loss": 0.3843,
      "step": 2675
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00012632469104346722,
      "loss": 0.3932,
      "step": 2700
    },
    {
      "epoch": 0.87,
      "eval_loss": 1.0900908708572388,
      "eval_runtime": 76.8201,
      "eval_samples_per_second": 7.081,
      "eval_steps_per_second": 0.885,
      "step": 2700
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00012507011798328823,
      "loss": 0.367,
      "step": 2725
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00012381131999530563,
      "loss": 0.4149,
      "step": 2750
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.000122548509217756,
      "loss": 0.3864,
      "step": 2775
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00012128189846512894,
      "loss": 0.3654,
      "step": 2800
    },
    {
      "epoch": 0.9,
      "eval_loss": 1.0753364562988281,
      "eval_runtime": 76.8365,
      "eval_samples_per_second": 7.08,
      "eval_steps_per_second": 0.885,
      "step": 2800
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00012001170119230286,
      "loss": 0.387,
      "step": 2825
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00011873813145857249,
      "loss": 0.3551,
      "step": 2850
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00011746140389157462,
      "loss": 0.385,
      "step": 2875
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00011618173365111824,
      "loss": 0.3227,
      "step": 2900
    },
    {
      "epoch": 0.93,
      "eval_loss": 1.0998799800872803,
      "eval_runtime": 76.6275,
      "eval_samples_per_second": 7.099,
      "eval_steps_per_second": 0.887,
      "step": 2900
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00011489933639292469,
      "loss": 0.3459,
      "step": 2925
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00011361442823228453,
      "loss": 0.3505,
      "step": 2950
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00011232722570763667,
      "loss": 0.3644,
      "step": 2975
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00011103794574407657,
      "loss": 0.3238,
      "step": 3000
    },
    {
      "epoch": 0.96,
      "eval_loss": 1.1243908405303955,
      "eval_runtime": 76.7936,
      "eval_samples_per_second": 7.084,
      "eval_steps_per_second": 0.885,
      "step": 3000
    }
  ],
  "max_steps": 6238,
  "num_train_epochs": 2,
  "total_flos": 6.802205860248576e+16,
  "trial_name": null,
  "trial_params": null
}