{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.3193361535668373,
      "epoch": 0.051118210862619806,
      "grad_norm": 0.8119011521339417,
      "learning_rate": 0.0,
      "loss": 1.692,
      "mean_token_accuracy": 0.654717817902565,
      "num_tokens": 133947.0,
      "step": 1
    },
    {
      "entropy": 1.3207192197442055,
      "epoch": 0.10223642172523961,
      "grad_norm": 0.8001739382743835,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.6938,
      "mean_token_accuracy": 0.6540814265608788,
      "num_tokens": 267949.0,
      "step": 2
    },
    {
      "entropy": 1.3139144703745842,
      "epoch": 0.15335463258785942,
      "grad_norm": 0.8021382689476013,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.6875,
      "mean_token_accuracy": 0.6542951986193657,
      "num_tokens": 402435.0,
      "step": 3
    },
    {
      "entropy": 1.3238477781414986,
      "epoch": 0.20447284345047922,
      "grad_norm": 0.8046473264694214,
      "learning_rate": 6e-06,
      "loss": 1.6979,
      "mean_token_accuracy": 0.6523041352629662,
      "num_tokens": 536339.0,
      "step": 4
    },
    {
      "entropy": 1.3151762038469315,
      "epoch": 0.25559105431309903,
      "grad_norm": 0.7999162077903748,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.6884,
      "mean_token_accuracy": 0.6541883014142513,
      "num_tokens": 670396.0,
      "step": 5
    },
    {
      "entropy": 1.3171968907117844,
      "epoch": 0.30670926517571884,
      "grad_norm": 0.8070191740989685,
      "learning_rate": 1e-05,
      "loss": 1.6881,
      "mean_token_accuracy": 0.6536356993019581,
      "num_tokens": 804575.0,
      "step": 6
    },
    {
      "entropy": 1.3177252262830734,
      "epoch": 0.35782747603833864,
      "grad_norm": 0.8115559220314026,
      "learning_rate": 1.2e-05,
      "loss": 1.6852,
      "mean_token_accuracy": 0.6539545804262161,
      "num_tokens": 938351.0,
      "step": 7
    },
    {
      "entropy": 1.3142458871006966,
      "epoch": 0.40894568690095845,
      "grad_norm": 0.809145987033844,
      "learning_rate": 1.4e-05,
      "loss": 1.6739,
      "mean_token_accuracy": 0.6544475704431534,
      "num_tokens": 1072655.0,
      "step": 8
    },
    {
      "entropy": 1.3176514655351639,
      "epoch": 0.46006389776357826,
      "grad_norm": 0.8079097270965576,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.6733,
      "mean_token_accuracy": 0.6541223935782909,
      "num_tokens": 1206472.0,
      "step": 9
    },
    {
      "entropy": 1.3175865784287453,
      "epoch": 0.5111821086261981,
      "grad_norm": 0.8009534478187561,
      "learning_rate": 1.8e-05,
      "loss": 1.657,
      "mean_token_accuracy": 0.655005007982254,
      "num_tokens": 1340462.0,
      "step": 10
    },
    {
      "entropy": 1.3152535259723663,
      "epoch": 0.5623003194888179,
      "grad_norm": 0.7782304883003235,
      "learning_rate": 2e-05,
      "loss": 1.6391,
      "mean_token_accuracy": 0.6572432741522789,
      "num_tokens": 1474818.0,
      "step": 11
    },
    {
      "entropy": 1.31825902312994,
      "epoch": 0.6134185303514377,
      "grad_norm": 0.7459490299224854,
      "learning_rate": 1.977777777777778e-05,
      "loss": 1.6261,
      "mean_token_accuracy": 0.6601505167782307,
      "num_tokens": 1608731.0,
      "step": 12
    },
    {
      "entropy": 1.322536252439022,
      "epoch": 0.6645367412140575,
      "grad_norm": 0.7278594970703125,
      "learning_rate": 1.9555555555555557e-05,
      "loss": 1.6056,
      "mean_token_accuracy": 0.6621886678040028,
      "num_tokens": 1742229.0,
      "step": 13
    },
    {
      "entropy": 1.3205925300717354,
      "epoch": 0.7156549520766773,
      "grad_norm": 0.6732656359672546,
      "learning_rate": 1.9333333333333333e-05,
      "loss": 1.587,
      "mean_token_accuracy": 0.6625417172908783,
      "num_tokens": 1875890.0,
      "step": 14
    },
    {
      "entropy": 1.3151337951421738,
      "epoch": 0.7667731629392971,
      "grad_norm": 0.6385083794593811,
      "learning_rate": 1.9111111111111113e-05,
      "loss": 1.5551,
      "mean_token_accuracy": 0.6706915572285652,
      "num_tokens": 2009719.0,
      "step": 15
    },
    {
      "entropy": 1.3125323951244354,
      "epoch": 0.8178913738019169,
      "grad_norm": 0.6251479387283325,
      "learning_rate": 1.888888888888889e-05,
      "loss": 1.5312,
      "mean_token_accuracy": 0.6725872829556465,
      "num_tokens": 2143710.0,
      "step": 16
    },
    {
      "entropy": 1.3094838485121727,
      "epoch": 0.8690095846645367,
      "grad_norm": 0.6229560375213623,
      "learning_rate": 1.866666666666667e-05,
      "loss": 1.5063,
      "mean_token_accuracy": 0.6759799160063267,
      "num_tokens": 2277556.0,
      "step": 17
    },
    {
      "entropy": 1.3096359893679619,
      "epoch": 0.9201277955271565,
      "grad_norm": 0.6263618469238281,
      "learning_rate": 1.8444444444444448e-05,
      "loss": 1.4846,
      "mean_token_accuracy": 0.6817599721252918,
      "num_tokens": 2411448.0,
      "step": 18
    },
    {
      "entropy": 1.2960194796323776,
      "epoch": 0.9712460063897763,
      "grad_norm": 0.6250044703483582,
      "learning_rate": 1.8222222222222224e-05,
      "loss": 1.4525,
      "mean_token_accuracy": 0.6899448521435261,
      "num_tokens": 2545738.0,
      "step": 19
    },
    {
      "entropy": 1.2956058846579657,
      "epoch": 1.0,
      "grad_norm": 0.6314756870269775,
      "learning_rate": 1.8e-05,
      "loss": 1.4391,
      "mean_token_accuracy": 0.6940541995896233,
      "num_tokens": 2616926.0,
      "step": 20
    },
    {
      "entropy": 1.2914148643612862,
      "epoch": 1.0511182108626198,
      "grad_norm": 0.631912887096405,
      "learning_rate": 1.7777777777777777e-05,
      "loss": 1.4127,
      "mean_token_accuracy": 0.6994642727077007,
      "num_tokens": 2750943.0,
      "step": 21
    },
    {
      "entropy": 1.2763815149664879,
      "epoch": 1.1022364217252396,
      "grad_norm": 0.6287756562232971,
      "learning_rate": 1.7555555555555556e-05,
      "loss": 1.3806,
      "mean_token_accuracy": 0.7029485926032066,
      "num_tokens": 2885149.0,
      "step": 22
    },
    {
      "entropy": 1.2685835510492325,
      "epoch": 1.1533546325878594,
      "grad_norm": 0.6242936253547668,
      "learning_rate": 1.7333333333333336e-05,
      "loss": 1.3606,
      "mean_token_accuracy": 0.7043894305825233,
      "num_tokens": 3019433.0,
      "step": 23
    },
    {
      "entropy": 1.263872005045414,
      "epoch": 1.2044728434504792,
      "grad_norm": 0.6192987561225891,
      "learning_rate": 1.7111111111111112e-05,
      "loss": 1.337,
      "mean_token_accuracy": 0.709864255040884,
      "num_tokens": 3153309.0,
      "step": 24
    },
    {
      "entropy": 1.2505493015050888,
      "epoch": 1.255591054313099,
      "grad_norm": 0.6113152503967285,
      "learning_rate": 1.688888888888889e-05,
      "loss": 1.3113,
      "mean_token_accuracy": 0.7119522020220757,
      "num_tokens": 3287322.0,
      "step": 25
    },
    {
      "entropy": 1.242555320262909,
      "epoch": 1.3067092651757188,
      "grad_norm": 0.597195029258728,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.2892,
      "mean_token_accuracy": 0.7140648253262043,
      "num_tokens": 3421258.0,
      "step": 26
    },
    {
      "entropy": 1.229688823223114,
      "epoch": 1.3578274760383386,
      "grad_norm": 0.586483359336853,
      "learning_rate": 1.6444444444444444e-05,
      "loss": 1.264,
      "mean_token_accuracy": 0.7178861573338509,
      "num_tokens": 3555933.0,
      "step": 27
    },
    {
      "entropy": 1.2219947651028633,
      "epoch": 1.4089456869009584,
      "grad_norm": 0.5772027373313904,
      "learning_rate": 1.6222222222222223e-05,
      "loss": 1.2416,
      "mean_token_accuracy": 0.7191276662051678,
      "num_tokens": 3689685.0,
      "step": 28
    },
    {
      "entropy": 1.2158958613872528,
      "epoch": 1.4600638977635783,
      "grad_norm": 0.5700508952140808,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.2201,
      "mean_token_accuracy": 0.7224024310708046,
      "num_tokens": 3823781.0,
      "step": 29
    },
    {
      "entropy": 1.2152001112699509,
      "epoch": 1.511182108626198,
      "grad_norm": 0.5722755789756775,
      "learning_rate": 1.577777777777778e-05,
      "loss": 1.2091,
      "mean_token_accuracy": 0.722760371863842,
      "num_tokens": 3956942.0,
      "step": 30
    },
    {
      "entropy": 1.1953495219349861,
      "epoch": 1.5623003194888179,
      "grad_norm": 0.5692858695983887,
      "learning_rate": 1.555555555555556e-05,
      "loss": 1.1714,
      "mean_token_accuracy": 0.7302578240633011,
      "num_tokens": 4091313.0,
      "step": 31
    },
    {
      "entropy": 1.1933885142207146,
      "epoch": 1.6134185303514377,
      "grad_norm": 0.5751745700836182,
      "learning_rate": 1.5333333333333334e-05,
      "loss": 1.1615,
      "mean_token_accuracy": 0.7318685166537762,
      "num_tokens": 4225481.0,
      "step": 32
    },
    {
      "entropy": 1.1871799379587173,
      "epoch": 1.6645367412140575,
      "grad_norm": 0.5843569040298462,
      "learning_rate": 1.5111111111111112e-05,
      "loss": 1.1378,
      "mean_token_accuracy": 0.7380619496107101,
      "num_tokens": 4359158.0,
      "step": 33
    },
    {
      "entropy": 1.1805067732930183,
      "epoch": 1.7156549520766773,
      "grad_norm": 0.5917448997497559,
      "learning_rate": 1.488888888888889e-05,
      "loss": 1.1218,
      "mean_token_accuracy": 0.7423498816788197,
      "num_tokens": 4493077.0,
      "step": 34
    },
    {
      "entropy": 1.170977495610714,
      "epoch": 1.766773162939297,
      "grad_norm": 0.59839928150177,
      "learning_rate": 1.4666666666666666e-05,
      "loss": 1.096,
      "mean_token_accuracy": 0.7493030689656734,
      "num_tokens": 4627251.0,
      "step": 35
    },
    {
      "entropy": 1.1678270995616913,
      "epoch": 1.817891373801917,
      "grad_norm": 0.6055343151092529,
      "learning_rate": 1.4444444444444446e-05,
      "loss": 1.0802,
      "mean_token_accuracy": 0.7507753595709801,
      "num_tokens": 4761020.0,
      "step": 36
    },
    {
      "entropy": 1.158110834658146,
      "epoch": 1.8690095846645367,
      "grad_norm": 0.6079038381576538,
      "learning_rate": 1.4222222222222224e-05,
      "loss": 1.0628,
      "mean_token_accuracy": 0.755638737231493,
      "num_tokens": 4895183.0,
      "step": 37
    },
    {
      "entropy": 1.154124453663826,
      "epoch": 1.9201277955271565,
      "grad_norm": 0.6134995222091675,
      "learning_rate": 1.4e-05,
      "loss": 1.043,
      "mean_token_accuracy": 0.7614487372338772,
      "num_tokens": 5028509.0,
      "step": 38
    },
    {
      "entropy": 1.1415963619947433,
      "epoch": 1.9712460063897763,
      "grad_norm": 0.608974277973175,
      "learning_rate": 1.377777777777778e-05,
      "loss": 1.0203,
      "mean_token_accuracy": 0.7697952277958393,
      "num_tokens": 5162590.0,
      "step": 39
    },
    {
      "entropy": 1.1324340105056763,
      "epoch": 2.0,
      "grad_norm": 0.6109018325805664,
      "learning_rate": 1.3555555555555557e-05,
      "loss": 0.9992,
      "mean_token_accuracy": 0.7730923626157973,
      "num_tokens": 5233852.0,
      "step": 40
    },
    {
      "entropy": 1.1270885691046715,
      "epoch": 2.0511182108626196,
      "grad_norm": 0.6178512573242188,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.982,
      "mean_token_accuracy": 0.7760093286633492,
      "num_tokens": 5367686.0,
      "step": 41
    },
    {
      "entropy": 1.1163304820656776,
      "epoch": 2.1022364217252396,
      "grad_norm": 0.6236265301704407,
      "learning_rate": 1.3111111111111113e-05,
      "loss": 0.9606,
      "mean_token_accuracy": 0.7832612432539463,
      "num_tokens": 5501101.0,
      "step": 42
    },
    {
      "entropy": 1.1050259843468666,
      "epoch": 2.1533546325878596,
      "grad_norm": 0.6330907940864563,
      "learning_rate": 1.288888888888889e-05,
      "loss": 0.9442,
      "mean_token_accuracy": 0.7889900915324688,
      "num_tokens": 5634943.0,
      "step": 43
    },
    {
      "entropy": 1.0991561263799667,
      "epoch": 2.2044728434504792,
      "grad_norm": 0.6459551453590393,
      "learning_rate": 1.2666666666666667e-05,
      "loss": 0.9323,
      "mean_token_accuracy": 0.7888863421976566,
      "num_tokens": 5768035.0,
      "step": 44
    },
    {
      "entropy": 1.080874651670456,
      "epoch": 2.255591054313099,
      "grad_norm": 0.6497413516044617,
      "learning_rate": 1.2444444444444446e-05,
      "loss": 0.9089,
      "mean_token_accuracy": 0.7905767410993576,
      "num_tokens": 5901673.0,
      "step": 45
    },
    {
      "entropy": 1.0577788427472115,
      "epoch": 2.306709265175719,
      "grad_norm": 0.6514442563056946,
      "learning_rate": 1.2222222222222224e-05,
      "loss": 0.8845,
      "mean_token_accuracy": 0.7969931028783321,
      "num_tokens": 6036094.0,
      "step": 46
    },
    {
      "entropy": 1.0467759743332863,
      "epoch": 2.357827476038339,
      "grad_norm": 0.6560313105583191,
      "learning_rate": 1.2e-05,
      "loss": 0.8659,
      "mean_token_accuracy": 0.802506472915411,
      "num_tokens": 6170134.0,
      "step": 47
    },
    {
      "entropy": 1.0260686576366425,
      "epoch": 2.4089456869009584,
      "grad_norm": 0.6551167368888855,
      "learning_rate": 1.177777777777778e-05,
      "loss": 0.8486,
      "mean_token_accuracy": 0.8058239929378033,
      "num_tokens": 6304397.0,
      "step": 48
    },
    {
      "entropy": 1.0083096772432327,
      "epoch": 2.460063897763578,
      "grad_norm": 0.6541892290115356,
      "learning_rate": 1.1555555555555556e-05,
      "loss": 0.8356,
      "mean_token_accuracy": 0.8059861660003662,
      "num_tokens": 6438340.0,
      "step": 49
    },
    {
      "entropy": 0.9794742912054062,
      "epoch": 2.511182108626198,
      "grad_norm": 0.6508305668830872,
      "learning_rate": 1.1333333333333334e-05,
      "loss": 0.8151,
      "mean_token_accuracy": 0.8144995309412479,
      "num_tokens": 6573003.0,
      "step": 50
    },
    {
      "entropy": 0.9591087996959686,
      "epoch": 2.562300319488818,
      "grad_norm": 0.6544970273971558,
      "learning_rate": 1.1111111111111113e-05,
      "loss": 0.7982,
      "mean_token_accuracy": 0.821755301207304,
      "num_tokens": 6707094.0,
      "step": 51
    },
    {
      "entropy": 0.9311400800943375,
      "epoch": 2.6134185303514377,
      "grad_norm": 0.661201536655426,
      "learning_rate": 1.088888888888889e-05,
      "loss": 0.7748,
      "mean_token_accuracy": 0.8267807699739933,
      "num_tokens": 6841315.0,
      "step": 52
    },
    {
      "entropy": 0.9006332121789455,
      "epoch": 2.6645367412140573,
      "grad_norm": 0.6626005172729492,
      "learning_rate": 1.0666666666666667e-05,
      "loss": 0.7543,
      "mean_token_accuracy": 0.8362897895276546,
      "num_tokens": 6976249.0,
      "step": 53
    },
    {
      "entropy": 0.8812699876725674,
      "epoch": 2.7156549520766773,
      "grad_norm": 0.674384355545044,
      "learning_rate": 1.0444444444444445e-05,
      "loss": 0.7417,
      "mean_token_accuracy": 0.8424257524311543,
      "num_tokens": 7110449.0,
      "step": 54
    },
    {
      "entropy": 0.8507667072117329,
      "epoch": 2.7667731629392973,
      "grad_norm": 0.695296049118042,
      "learning_rate": 1.0222222222222223e-05,
      "loss": 0.7201,
      "mean_token_accuracy": 0.8484714813530445,
      "num_tokens": 7244307.0,
      "step": 55
    },
    {
      "entropy": 0.8221894763410091,
      "epoch": 2.817891373801917,
      "grad_norm": 0.7484252452850342,
      "learning_rate": 1e-05,
      "loss": 0.7067,
      "mean_token_accuracy": 0.8503611832857132,
      "num_tokens": 7378331.0,
      "step": 56
    },
    {
      "entropy": 0.7954868413507938,
      "epoch": 2.8690095846645365,
      "grad_norm": 0.8117406368255615,
      "learning_rate": 9.777777777777779e-06,
      "loss": 0.6898,
      "mean_token_accuracy": 0.8546305038034916,
      "num_tokens": 7511657.0,
      "step": 57
    },
    {
      "entropy": 0.7787146084010601,
      "epoch": 2.9201277955271565,
      "grad_norm": 0.7788737416267395,
      "learning_rate": 9.555555555555556e-06,
      "loss": 0.6761,
      "mean_token_accuracy": 0.8574383407831192,
      "num_tokens": 7645680.0,
      "step": 58
    },
    {
      "entropy": 0.7654511369764805,
      "epoch": 2.9712460063897765,
      "grad_norm": 0.6763613820075989,
      "learning_rate": 9.333333333333334e-06,
      "loss": 0.6651,
      "mean_token_accuracy": 0.8578773178160191,
      "num_tokens": 7779761.0,
      "step": 59
    },
    {
      "entropy": 0.7472146418359544,
      "epoch": 3.0,
      "grad_norm": 0.6285676956176758,
      "learning_rate": 9.111111111111112e-06,
      "loss": 0.6435,
      "mean_token_accuracy": 0.86120914750629,
      "num_tokens": 7850778.0,
      "step": 60
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.847758104931533e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}