{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0583941605839415,
  "eval_steps": 500,
  "global_step": 55,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.1336327642202377,
      "epoch": 0.058394160583941604,
      "grad_norm": 16.25,
      "learning_rate": 0.0,
      "loss": 2.4507,
      "mean_token_accuracy": 0.4276521671563387,
      "num_tokens": 3890.0,
      "step": 1
    },
    {
      "entropy": 2.222498059272766,
      "epoch": 0.11678832116788321,
      "grad_norm": 17.875,
      "learning_rate": 2e-06,
      "loss": 2.6879,
      "mean_token_accuracy": 0.4140724800527096,
      "num_tokens": 6825.0,
      "step": 2
    },
    {
      "entropy": 2.262238770723343,
      "epoch": 0.17518248175182483,
      "grad_norm": 14.25,
      "learning_rate": 4e-06,
      "loss": 2.4973,
      "mean_token_accuracy": 0.42698577605187893,
      "num_tokens": 10473.0,
      "step": 3
    },
    {
      "entropy": 2.1316296458244324,
      "epoch": 0.23357664233576642,
      "grad_norm": 10.0,
      "learning_rate": 6e-06,
      "loss": 2.0863,
      "mean_token_accuracy": 0.4788516294211149,
      "num_tokens": 15657.0,
      "step": 4
    },
    {
      "entropy": 2.3611446991562843,
      "epoch": 0.291970802919708,
      "grad_norm": 10.1875,
      "learning_rate": 8e-06,
      "loss": 2.1751,
      "mean_token_accuracy": 0.4456866979598999,
      "num_tokens": 20159.0,
      "step": 5
    },
    {
      "entropy": 2.460206426680088,
      "epoch": 0.35036496350364965,
      "grad_norm": 8.875,
      "learning_rate": 9.999999999999999e-06,
      "loss": 2.2655,
      "mean_token_accuracy": 0.4509607646614313,
      "num_tokens": 23949.0,
      "step": 6
    },
    {
      "entropy": 2.321817234158516,
      "epoch": 0.40875912408759124,
      "grad_norm": 7.125,
      "learning_rate": 1.2e-05,
      "loss": 2.0123,
      "mean_token_accuracy": 0.5055391453206539,
      "num_tokens": 27703.0,
      "step": 7
    },
    {
      "entropy": 2.2407592684030533,
      "epoch": 0.46715328467153283,
      "grad_norm": 5.4375,
      "learning_rate": 1.4e-05,
      "loss": 1.8516,
      "mean_token_accuracy": 0.5130146574229002,
      "num_tokens": 32243.0,
      "step": 8
    },
    {
      "entropy": 2.46332585811615,
      "epoch": 0.5255474452554745,
      "grad_norm": 7.09375,
      "learning_rate": 1.6e-05,
      "loss": 2.0974,
      "mean_token_accuracy": 0.5035313870757818,
      "num_tokens": 35222.0,
      "step": 9
    },
    {
      "entropy": 2.237804166972637,
      "epoch": 0.583941605839416,
      "grad_norm": 5.65625,
      "learning_rate": 1.8e-05,
      "loss": 1.7838,
      "mean_token_accuracy": 0.5259560514241457,
      "num_tokens": 39208.0,
      "step": 10
    },
    {
      "entropy": 2.352365091443062,
      "epoch": 0.6423357664233577,
      "grad_norm": 5.84375,
      "learning_rate": 1.9999999999999998e-05,
      "loss": 2.0078,
      "mean_token_accuracy": 0.5078456345945597,
      "num_tokens": 42447.0,
      "step": 11
    },
    {
      "entropy": 2.1229992732405663,
      "epoch": 0.7007299270072993,
      "grad_norm": 4.5625,
      "learning_rate": 2.2e-05,
      "loss": 1.7155,
      "mean_token_accuracy": 0.5374241229146719,
      "num_tokens": 47138.0,
      "step": 12
    },
    {
      "entropy": 2.121931955218315,
      "epoch": 0.7591240875912408,
      "grad_norm": 4.625,
      "learning_rate": 2.4e-05,
      "loss": 1.7379,
      "mean_token_accuracy": 0.5694513749331236,
      "num_tokens": 51009.0,
      "step": 13
    },
    {
      "entropy": 2.085137240588665,
      "epoch": 0.8175182481751825,
      "grad_norm": 4.25,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 1.6524,
      "mean_token_accuracy": 0.5468352809548378,
      "num_tokens": 55235.0,
      "step": 14
    },
    {
      "entropy": 2.1976606771349907,
      "epoch": 0.8759124087591241,
      "grad_norm": 5.625,
      "learning_rate": 2.8e-05,
      "loss": 1.8096,
      "mean_token_accuracy": 0.5231170020997524,
      "num_tokens": 58219.0,
      "step": 15
    },
    {
      "entropy": 1.9179195016622543,
      "epoch": 0.9343065693430657,
      "grad_norm": 3.84375,
      "learning_rate": 3e-05,
      "loss": 1.5974,
      "mean_token_accuracy": 0.5759452320635319,
      "num_tokens": 63057.0,
      "step": 16
    },
    {
      "entropy": 2.0428223088383675,
      "epoch": 0.9927007299270073,
      "grad_norm": 4.53125,
      "learning_rate": 2.9986842451482876e-05,
      "loss": 1.7372,
      "mean_token_accuracy": 0.5385774970054626,
      "num_tokens": 66564.0,
      "step": 17
    },
    {
      "entropy": 1.9300671219825745,
      "epoch": 1.0,
      "grad_norm": 12.9375,
      "learning_rate": 2.9947392888742566e-05,
      "loss": 1.7476,
      "mean_token_accuracy": 0.5453733801841736,
      "num_tokens": 66897.0,
      "step": 18
    },
    {
      "entropy": 1.935000792145729,
      "epoch": 1.0583941605839415,
      "grad_norm": 3.484375,
      "learning_rate": 2.988172051971717e-05,
      "loss": 1.4249,
      "mean_token_accuracy": 0.6101336404681206,
      "num_tokens": 71403.0,
      "step": 19
    },
    {
      "entropy": 2.0335680916905403,
      "epoch": 1.1167883211678833,
      "grad_norm": 3.84375,
      "learning_rate": 2.9789940556057574e-05,
      "loss": 1.5345,
      "mean_token_accuracy": 0.5629026051610708,
      "num_tokens": 75484.0,
      "step": 20
    },
    {
      "entropy": 2.10165449231863,
      "epoch": 1.1751824817518248,
      "grad_norm": 4.0625,
      "learning_rate": 2.9672214011007087e-05,
      "loss": 1.4949,
      "mean_token_accuracy": 0.5799959097057581,
      "num_tokens": 79113.0,
      "step": 21
    },
    {
      "entropy": 1.992341309785843,
      "epoch": 1.2335766423357664,
      "grad_norm": 3.59375,
      "learning_rate": 2.9528747416929467e-05,
      "loss": 1.4678,
      "mean_token_accuracy": 0.5918100215494633,
      "num_tokens": 83095.0,
      "step": 22
    },
    {
      "entropy": 1.9140778183937073,
      "epoch": 1.2919708029197081,
      "grad_norm": 3.375,
      "learning_rate": 2.9359792462981007e-05,
      "loss": 1.4038,
      "mean_token_accuracy": 0.6022733096033335,
      "num_tokens": 87754.0,
      "step": 23
    },
    {
      "entropy": 1.8838416188955307,
      "epoch": 1.3503649635036497,
      "grad_norm": 3.8125,
      "learning_rate": 2.9165645553562215e-05,
      "loss": 1.4554,
      "mean_token_accuracy": 0.6133127138018608,
      "num_tokens": 91666.0,
      "step": 24
    },
    {
      "entropy": 1.816191054880619,
      "epoch": 1.4087591240875912,
      "grad_norm": 3.859375,
      "learning_rate": 2.894664728832377e-05,
      "loss": 1.3643,
      "mean_token_accuracy": 0.6147295907139778,
      "num_tokens": 95819.0,
      "step": 25
    },
    {
      "entropy": 1.7681904509663582,
      "epoch": 1.4671532846715327,
      "grad_norm": 3.609375,
      "learning_rate": 2.8703181864639013e-05,
      "loss": 1.3711,
      "mean_token_accuracy": 0.6297403201460838,
      "num_tokens": 99865.0,
      "step": 26
    },
    {
      "entropy": 1.7096636295318604,
      "epoch": 1.5255474452554745,
      "grad_norm": 3.390625,
      "learning_rate": 2.8435676403591193e-05,
      "loss": 1.3362,
      "mean_token_accuracy": 0.6145001202821732,
      "num_tokens": 104135.0,
      "step": 27
    },
    {
      "entropy": 1.828701414167881,
      "epoch": 1.583941605839416,
      "grad_norm": 4.3125,
      "learning_rate": 2.8144600200657953e-05,
      "loss": 1.4266,
      "mean_token_accuracy": 0.5893764644861221,
      "num_tokens": 107324.0,
      "step": 28
    },
    {
      "entropy": 1.8768919259309769,
      "epoch": 1.6423357664233578,
      "grad_norm": 4.65625,
      "learning_rate": 2.78304639024076e-05,
      "loss": 1.5031,
      "mean_token_accuracy": 0.5983850117772818,
      "num_tokens": 110263.0,
      "step": 29
    },
    {
      "entropy": 1.7338064908981323,
      "epoch": 1.7007299270072993,
      "grad_norm": 4.34375,
      "learning_rate": 2.7493818610651493e-05,
      "loss": 1.4431,
      "mean_token_accuracy": 0.5914898477494717,
      "num_tokens": 113911.0,
      "step": 30
    },
    {
      "entropy": 1.7540361359715462,
      "epoch": 1.7591240875912408,
      "grad_norm": 3.734375,
      "learning_rate": 2.7135254915624213e-05,
      "loss": 1.3489,
      "mean_token_accuracy": 0.6010549142956734,
      "num_tokens": 118007.0,
      "step": 31
    },
    {
      "entropy": 1.8890240713953972,
      "epoch": 1.8175182481751824,
      "grad_norm": 4.65625,
      "learning_rate": 2.6755401859887598e-05,
      "loss": 1.4448,
      "mean_token_accuracy": 0.6083299573510885,
      "num_tokens": 120725.0,
      "step": 32
    },
    {
      "entropy": 1.850830078125,
      "epoch": 1.8759124087591241,
      "grad_norm": 4.28125,
      "learning_rate": 2.6354925834776346e-05,
      "loss": 1.502,
      "mean_token_accuracy": 0.6061263754963875,
      "num_tokens": 124333.0,
      "step": 33
    },
    {
      "entropy": 1.7397000417113304,
      "epoch": 1.9343065693430657,
      "grad_norm": 3.671875,
      "learning_rate": 2.5934529411321174e-05,
      "loss": 1.2539,
      "mean_token_accuracy": 0.6317082159221172,
      "num_tokens": 128615.0,
      "step": 34
    },
    {
      "entropy": 1.813131682574749,
      "epoch": 1.9927007299270074,
      "grad_norm": 3.5625,
      "learning_rate": 2.5494950107700482e-05,
      "loss": 1.3284,
      "mean_token_accuracy": 0.6140319798141718,
      "num_tokens": 132847.0,
      "step": 35
    },
    {
      "entropy": 1.5973615646362305,
      "epoch": 2.0,
      "grad_norm": 7.46875,
      "learning_rate": 2.5036959095382875e-05,
      "loss": 1.2697,
      "mean_token_accuracy": 0.6285321712493896,
      "num_tokens": 133794.0,
      "step": 36
    },
    {
      "entropy": 1.7645720839500427,
      "epoch": 2.0583941605839415,
      "grad_norm": 3.859375,
      "learning_rate": 2.4561359846230346e-05,
      "loss": 1.0785,
      "mean_token_accuracy": 0.6664150357246399,
      "num_tokens": 137554.0,
      "step": 37
    },
    {
      "entropy": 1.753688521683216,
      "epoch": 2.116788321167883,
      "grad_norm": 3.3125,
      "learning_rate": 2.4068986722935625e-05,
      "loss": 1.0716,
      "mean_token_accuracy": 0.6744864694774151,
      "num_tokens": 141721.0,
      "step": 38
    },
    {
      "entropy": 1.6263050064444542,
      "epoch": 2.1751824817518246,
      "grad_norm": 4.3125,
      "learning_rate": 2.356070351526648e-05,
      "loss": 1.0687,
      "mean_token_accuracy": 0.6837072521448135,
      "num_tokens": 146069.0,
      "step": 39
    },
    {
      "entropy": 1.8026663437485695,
      "epoch": 2.2335766423357666,
      "grad_norm": 3.84375,
      "learning_rate": 2.303740192468495e-05,
      "loss": 1.1566,
      "mean_token_accuracy": 0.6734990328550339,
      "num_tokens": 149664.0,
      "step": 40
    },
    {
      "entropy": 1.5874073877930641,
      "epoch": 2.291970802919708,
      "grad_norm": 3.53125,
      "learning_rate": 2.25e-05,
      "loss": 1.0591,
      "mean_token_accuracy": 0.6754884608089924,
      "num_tokens": 153330.0,
      "step": 41
    },
    {
      "entropy": 1.4746350944042206,
      "epoch": 2.3503649635036497,
      "grad_norm": 3.453125,
      "learning_rate": 2.1949440526797928e-05,
      "loss": 0.9312,
      "mean_token_accuracy": 0.7215368486940861,
      "num_tokens": 157396.0,
      "step": 42
    },
    {
      "entropy": 1.4334233030676842,
      "epoch": 2.408759124087591,
      "grad_norm": 6.96875,
      "learning_rate": 2.138668937347609e-05,
      "loss": 0.9952,
      "mean_token_accuracy": 0.7047883793711662,
      "num_tokens": 162386.0,
      "step": 43
    },
    {
      "entropy": 1.4815244674682617,
      "epoch": 2.4671532846715327,
      "grad_norm": 3.9375,
      "learning_rate": 2.0812733796781544e-05,
      "loss": 1.0847,
      "mean_token_accuracy": 0.680337205529213,
      "num_tokens": 166308.0,
      "step": 44
    },
    {
      "entropy": 1.4337811917066574,
      "epoch": 2.5255474452554747,
      "grad_norm": 4.21875,
      "learning_rate": 2.022858070982723e-05,
      "loss": 1.0594,
      "mean_token_accuracy": 0.686751551926136,
      "num_tokens": 169979.0,
      "step": 45
    },
    {
      "entropy": 1.380111612379551,
      "epoch": 2.5839416058394162,
      "grad_norm": 3.984375,
      "learning_rate": 1.963525491562421e-05,
      "loss": 0.9718,
      "mean_token_accuracy": 0.7241853773593903,
      "num_tokens": 174586.0,
      "step": 46
    },
    {
      "entropy": 1.339597962796688,
      "epoch": 2.6423357664233578,
      "grad_norm": 4.0625,
      "learning_rate": 1.9033797309228984e-05,
      "loss": 0.9445,
      "mean_token_accuracy": 0.7082682773470879,
      "num_tokens": 178535.0,
      "step": 47
    },
    {
      "entropy": 1.293665699660778,
      "epoch": 2.7007299270072993,
      "grad_norm": 3.765625,
      "learning_rate": 1.8425263051659838e-05,
      "loss": 0.9213,
      "mean_token_accuracy": 0.7238599583506584,
      "num_tokens": 183350.0,
      "step": 48
    },
    {
      "entropy": 1.3446906879544258,
      "epoch": 2.759124087591241,
      "grad_norm": 4.46875,
      "learning_rate": 1.781071971878587e-05,
      "loss": 0.9652,
      "mean_token_accuracy": 0.6951282061636448,
      "num_tokens": 187493.0,
      "step": 49
    },
    {
      "entropy": 1.3415213227272034,
      "epoch": 2.8175182481751824,
      "grad_norm": 4.8125,
      "learning_rate": 1.7191245428436175e-05,
      "loss": 1.0102,
      "mean_token_accuracy": 0.7021605856716633,
      "num_tokens": 190843.0,
      "step": 50
    },
    {
      "entropy": 1.4499380737543106,
      "epoch": 2.875912408759124,
      "grad_norm": 5.71875,
      "learning_rate": 1.6567926949014805e-05,
      "loss": 1.0649,
      "mean_token_accuracy": 0.7037234976887703,
      "num_tokens": 193518.0,
      "step": 51
    },
    {
      "entropy": 1.3929353207349777,
      "epoch": 2.9343065693430654,
      "grad_norm": 4.75,
      "learning_rate": 1.5941857792939702e-05,
      "loss": 1.0284,
      "mean_token_accuracy": 0.6902767680585384,
      "num_tokens": 196895.0,
      "step": 52
    },
    {
      "entropy": 1.4459699764847755,
      "epoch": 2.9927007299270074,
      "grad_norm": 4.75,
      "learning_rate": 1.5314136298250355e-05,
      "loss": 1.013,
      "mean_token_accuracy": 0.6965249925851822,
      "num_tokens": 200296.0,
      "step": 53
    },
    {
      "entropy": 1.399910032749176,
      "epoch": 3.0,
      "grad_norm": 13.0625,
      "learning_rate": 1.4685863701749648e-05,
      "loss": 1.0552,
      "mean_token_accuracy": 0.6890038251876831,
      "num_tokens": 200691.0,
      "step": 54
    },
    {
      "entropy": 1.3579635098576546,
      "epoch": 3.0583941605839415,
      "grad_norm": 4.28125,
      "learning_rate": 1.40581422070603e-05,
      "loss": 0.7865,
      "mean_token_accuracy": 0.765391580760479,
      "num_tokens": 204197.0,
      "step": 55
    }
  ],
  "logging_steps": 1,
  "max_steps": 90,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5469020090400768.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}