{
  "best_global_step": 416,
  "best_metric": 2.039247751235962,
  "best_model_checkpoint": "/kaggle/working/exp_b/checkpoint-416",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 520,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.561516413092613,
      "epoch": 0.04819277108433735,
      "grad_norm": 5.225845813751221,
      "learning_rate": 8e-05,
      "loss": 3.611944580078125,
      "mean_token_accuracy": 0.34577792752534153,
      "num_tokens": 3464.0,
      "step": 5
    },
    {
      "entropy": 2.5212378710508347,
      "epoch": 0.0963855421686747,
      "grad_norm": 4.040648460388184,
      "learning_rate": 0.00018,
      "loss": 3.441745376586914,
      "mean_token_accuracy": 0.3602590151131153,
      "num_tokens": 6761.0,
      "step": 10
    },
    {
      "entropy": 3.019538700580597,
      "epoch": 0.14457831325301204,
      "grad_norm": 1.5266512632369995,
      "learning_rate": 0.00019843137254901963,
      "loss": 3.1468936920166017,
      "mean_token_accuracy": 0.38241153322160243,
      "num_tokens": 9657.0,
      "step": 15
    },
    {
      "entropy": 2.973094937205315,
      "epoch": 0.1927710843373494,
      "grad_norm": 1.641349196434021,
      "learning_rate": 0.00019647058823529413,
      "loss": 2.735249137878418,
      "mean_token_accuracy": 0.4360183320939541,
      "num_tokens": 13066.0,
      "step": 20
    },
    {
      "entropy": 2.939211678504944,
      "epoch": 0.24096385542168675,
      "grad_norm": 1.7345398664474487,
      "learning_rate": 0.00019450980392156863,
      "loss": 2.6320152282714844,
      "mean_token_accuracy": 0.47086589187383654,
      "num_tokens": 16370.0,
      "step": 25
    },
    {
      "entropy": 2.681175285577774,
      "epoch": 0.2891566265060241,
      "grad_norm": 1.5084830522537231,
      "learning_rate": 0.00019254901960784316,
      "loss": 2.4271326065063477,
      "mean_token_accuracy": 0.465889797359705,
      "num_tokens": 19335.0,
      "step": 30
    },
    {
      "entropy": 2.602067893743515,
      "epoch": 0.3373493975903614,
      "grad_norm": 1.9293867349624634,
      "learning_rate": 0.00019058823529411766,
      "loss": 2.5105060577392577,
      "mean_token_accuracy": 0.47650288939476015,
      "num_tokens": 22396.0,
      "step": 35
    },
    {
      "entropy": 2.4558135360479354,
      "epoch": 0.3855421686746988,
      "grad_norm": 1.9167650938034058,
      "learning_rate": 0.00018862745098039216,
      "loss": 2.3286008834838867,
      "mean_token_accuracy": 0.4834901452064514,
      "num_tokens": 25507.0,
      "step": 40
    },
    {
      "entropy": 2.405803582072258,
      "epoch": 0.43373493975903615,
      "grad_norm": 2.2135415077209473,
      "learning_rate": 0.0001866666666666667,
      "loss": 2.3707206726074217,
      "mean_token_accuracy": 0.4752545841038227,
      "num_tokens": 28570.0,
      "step": 45
    },
    {
      "entropy": 2.4526087760925295,
      "epoch": 0.4819277108433735,
      "grad_norm": 1.6238704919815063,
      "learning_rate": 0.0001847058823529412,
      "loss": 2.311956214904785,
      "mean_token_accuracy": 0.48496985137462617,
      "num_tokens": 31974.0,
      "step": 50
    },
    {
      "entropy": 2.397727423906326,
      "epoch": 0.5301204819277109,
      "grad_norm": 1.6552271842956543,
      "learning_rate": 0.0001827450980392157,
      "loss": 2.2754472732543944,
      "mean_token_accuracy": 0.49604480862617495,
      "num_tokens": 35481.0,
      "step": 55
    },
    {
      "entropy": 2.3205750346183778,
      "epoch": 0.5783132530120482,
      "grad_norm": 2.311979293823242,
      "learning_rate": 0.00018078431372549022,
      "loss": 2.311609077453613,
      "mean_token_accuracy": 0.4907081700861454,
      "num_tokens": 38923.0,
      "step": 60
    },
    {
      "entropy": 2.327582710981369,
      "epoch": 0.6265060240963856,
      "grad_norm": 2.1570351123809814,
      "learning_rate": 0.00017882352941176472,
      "loss": 2.27120304107666,
      "mean_token_accuracy": 0.5065677210688591,
      "num_tokens": 42217.0,
      "step": 65
    },
    {
      "entropy": 2.400317168235779,
      "epoch": 0.6746987951807228,
      "grad_norm": 1.7414659261703491,
      "learning_rate": 0.00017686274509803922,
      "loss": 2.2884599685668947,
      "mean_token_accuracy": 0.4961890891194344,
      "num_tokens": 45084.0,
      "step": 70
    },
    {
      "entropy": 2.2782546311616896,
      "epoch": 0.7228915662650602,
      "grad_norm": 1.7846909761428833,
      "learning_rate": 0.00017490196078431375,
      "loss": 2.2046764373779295,
      "mean_token_accuracy": 0.5048216477036476,
      "num_tokens": 48593.0,
      "step": 75
    },
    {
      "entropy": 2.340757152438164,
      "epoch": 0.7710843373493976,
      "grad_norm": 2.5892934799194336,
      "learning_rate": 0.00017294117647058825,
      "loss": 2.205982780456543,
      "mean_token_accuracy": 0.52031245008111,
      "num_tokens": 51824.0,
      "step": 80
    },
    {
      "entropy": 2.2771576404571534,
      "epoch": 0.8192771084337349,
      "grad_norm": 2.445878744125366,
      "learning_rate": 0.00017098039215686275,
      "loss": 2.1431158065795897,
      "mean_token_accuracy": 0.5121273666620254,
      "num_tokens": 54702.0,
      "step": 85
    },
    {
      "entropy": 2.3005059540271757,
      "epoch": 0.8674698795180723,
      "grad_norm": 1.692912220954895,
      "learning_rate": 0.00016901960784313725,
      "loss": 2.2247289657592773,
      "mean_token_accuracy": 0.5257683046162128,
      "num_tokens": 57721.0,
      "step": 90
    },
    {
      "entropy": 2.280690374970436,
      "epoch": 0.9156626506024096,
      "grad_norm": 1.8366621732711792,
      "learning_rate": 0.00016705882352941178,
      "loss": 2.1423772811889648,
      "mean_token_accuracy": 0.5117545984685421,
      "num_tokens": 60906.0,
      "step": 95
    },
    {
      "entropy": 2.2130038917064665,
      "epoch": 0.963855421686747,
      "grad_norm": 2.175920009613037,
      "learning_rate": 0.00016509803921568628,
      "loss": 2.08164005279541,
      "mean_token_accuracy": 0.5335044473409652,
      "num_tokens": 63973.0,
      "step": 100
    },
    {
      "epoch": 1.0,
      "eval_entropy": 2.2261579770308275,
      "eval_loss": 2.2078733444213867,
      "eval_mean_token_accuracy": 0.5120092527224467,
      "eval_num_tokens": 66341.0,
      "eval_runtime": 14.48,
      "eval_samples_per_second": 14.365,
      "eval_steps_per_second": 1.796,
      "step": 104
    },
    {
      "entropy": 2.1521650395895304,
      "epoch": 1.0096385542168675,
      "grad_norm": 2.3988542556762695,
      "learning_rate": 0.00016313725490196078,
      "loss": 2.144173431396484,
      "mean_token_accuracy": 0.5304121116274282,
      "num_tokens": 66937.0,
      "step": 105
    },
    {
      "entropy": 2.319179782271385,
      "epoch": 1.0578313253012048,
      "grad_norm": 1.800117015838623,
      "learning_rate": 0.0001611764705882353,
      "loss": 2.145477867126465,
      "mean_token_accuracy": 0.5247455418109894,
      "num_tokens": 70623.0,
      "step": 110
    },
    {
      "entropy": 2.214820149540901,
      "epoch": 1.106024096385542,
      "grad_norm": 2.066795825958252,
      "learning_rate": 0.0001592156862745098,
      "loss": 2.1166513442993162,
      "mean_token_accuracy": 0.5487547442317009,
      "num_tokens": 73494.0,
      "step": 115
    },
    {
      "entropy": 2.166776442527771,
      "epoch": 1.1542168674698796,
      "grad_norm": 2.3190572261810303,
      "learning_rate": 0.0001572549019607843,
      "loss": 2.0480432510375977,
      "mean_token_accuracy": 0.5457861229777337,
      "num_tokens": 76636.0,
      "step": 120
    },
    {
      "entropy": 2.107932832837105,
      "epoch": 1.202409638554217,
      "grad_norm": 2.302002191543579,
      "learning_rate": 0.00015529411764705884,
      "loss": 2.0083555221557616,
      "mean_token_accuracy": 0.5434873178601265,
      "num_tokens": 79802.0,
      "step": 125
    },
    {
      "entropy": 2.0425092548131945,
      "epoch": 1.2506024096385542,
      "grad_norm": 2.0986533164978027,
      "learning_rate": 0.00015333333333333334,
      "loss": 1.9207183837890625,
      "mean_token_accuracy": 0.5640334501862526,
      "num_tokens": 83041.0,
      "step": 130
    },
    {
      "entropy": 2.0928288742899896,
      "epoch": 1.2987951807228915,
      "grad_norm": 1.8585127592086792,
      "learning_rate": 0.00015137254901960784,
      "loss": 2.0677839279174806,
      "mean_token_accuracy": 0.553177984803915,
      "num_tokens": 86586.0,
      "step": 135
    },
    {
      "entropy": 2.1725668996572494,
      "epoch": 1.346987951807229,
      "grad_norm": 2.228135824203491,
      "learning_rate": 0.00014941176470588237,
      "loss": 2.0742416381835938,
      "mean_token_accuracy": 0.5617913477122783,
      "num_tokens": 89974.0,
      "step": 140
    },
    {
      "entropy": 2.060410702228546,
      "epoch": 1.3951807228915662,
      "grad_norm": 2.812140941619873,
      "learning_rate": 0.00014745098039215687,
      "loss": 1.9109399795532227,
      "mean_token_accuracy": 0.5670658618211746,
      "num_tokens": 92987.0,
      "step": 145
    },
    {
      "entropy": 2.060049146413803,
      "epoch": 1.4433734939759035,
      "grad_norm": 2.627701759338379,
      "learning_rate": 0.00014549019607843137,
      "loss": 1.9155263900756836,
      "mean_token_accuracy": 0.5867504954338074,
      "num_tokens": 95988.0,
      "step": 150
    },
    {
      "entropy": 2.0021081775426866,
      "epoch": 1.491566265060241,
      "grad_norm": 2.964092493057251,
      "learning_rate": 0.0001435294117647059,
      "loss": 1.9390331268310548,
      "mean_token_accuracy": 0.5690168596804142,
      "num_tokens": 99014.0,
      "step": 155
    },
    {
      "entropy": 2.0013501018285753,
      "epoch": 1.5397590361445783,
      "grad_norm": 2.574568510055542,
      "learning_rate": 0.0001415686274509804,
      "loss": 1.920250129699707,
      "mean_token_accuracy": 0.5765035949647427,
      "num_tokens": 101635.0,
      "step": 160
    },
    {
      "entropy": 2.014915320277214,
      "epoch": 1.5879518072289156,
      "grad_norm": 2.3566172122955322,
      "learning_rate": 0.0001396078431372549,
      "loss": 2.008928680419922,
      "mean_token_accuracy": 0.5551697753369809,
      "num_tokens": 104625.0,
      "step": 165
    },
    {
      "entropy": 2.0899668782949448,
      "epoch": 1.636144578313253,
      "grad_norm": 2.6340577602386475,
      "learning_rate": 0.00013764705882352943,
      "loss": 2.0793405532836915,
      "mean_token_accuracy": 0.5444259770214558,
      "num_tokens": 107892.0,
      "step": 170
    },
    {
      "entropy": 2.042271265387535,
      "epoch": 1.6843373493975904,
      "grad_norm": 2.3889081478118896,
      "learning_rate": 0.00013568627450980393,
      "loss": 1.9268774032592773,
      "mean_token_accuracy": 0.5699412971735001,
      "num_tokens": 111085.0,
      "step": 175
    },
    {
      "entropy": 2.1282086342573168,
      "epoch": 1.7325301204819277,
      "grad_norm": 2.5944511890411377,
      "learning_rate": 0.00013372549019607843,
      "loss": 2.0320133209228515,
      "mean_token_accuracy": 0.5444036960601807,
      "num_tokens": 114168.0,
      "step": 180
    },
    {
      "entropy": 2.113537332415581,
      "epoch": 1.7807228915662652,
      "grad_norm": 2.2976858615875244,
      "learning_rate": 0.00013176470588235296,
      "loss": 1.9620359420776368,
      "mean_token_accuracy": 0.5624074958264828,
      "num_tokens": 117650.0,
      "step": 185
    },
    {
      "entropy": 2.016071784496307,
      "epoch": 1.8289156626506025,
      "grad_norm": 2.0810470581054688,
      "learning_rate": 0.00012980392156862746,
      "loss": 1.9303123474121093,
      "mean_token_accuracy": 0.5449617192149162,
      "num_tokens": 120863.0,
      "step": 190
    },
    {
      "entropy": 1.978957825899124,
      "epoch": 1.8771084337349397,
      "grad_norm": 2.611105442047119,
      "learning_rate": 0.00012784313725490196,
      "loss": 1.9237918853759766,
      "mean_token_accuracy": 0.5678628019988536,
      "num_tokens": 124357.0,
      "step": 195
    },
    {
      "entropy": 2.024098601937294,
      "epoch": 1.9253012048192772,
      "grad_norm": 2.551222562789917,
      "learning_rate": 0.0001258823529411765,
      "loss": 1.9830829620361328,
      "mean_token_accuracy": 0.5428289316594601,
      "num_tokens": 127575.0,
      "step": 200
    },
    {
      "entropy": 2.0877326667308806,
      "epoch": 1.9734939759036143,
      "grad_norm": 2.405113697052002,
      "learning_rate": 0.000123921568627451,
      "loss": 2.041797637939453,
      "mean_token_accuracy": 0.5457921862602234,
      "num_tokens": 130849.0,
      "step": 205
    },
    {
      "epoch": 2.0,
      "eval_entropy": 2.156877522285168,
      "eval_loss": 2.0858378410339355,
      "eval_mean_token_accuracy": 0.5350976666578879,
      "eval_num_tokens": 132682.0,
      "eval_runtime": 14.4815,
      "eval_samples_per_second": 14.363,
      "eval_steps_per_second": 1.795,
      "step": 208
    },
    {
      "entropy": 2.066597398958708,
      "epoch": 2.019277108433735,
      "grad_norm": 2.3556201457977295,
      "learning_rate": 0.0001219607843137255,
      "loss": 1.9169322967529296,
      "mean_token_accuracy": 0.5674474333461962,
      "num_tokens": 134025.0,
      "step": 210
    },
    {
      "entropy": 2.017352694272995,
      "epoch": 2.067469879518072,
      "grad_norm": 2.575573682785034,
      "learning_rate": 0.00012,
      "loss": 1.8107269287109375,
      "mean_token_accuracy": 0.5836580470204353,
      "num_tokens": 137677.0,
      "step": 215
    },
    {
      "entropy": 1.9778071194887161,
      "epoch": 2.1156626506024097,
      "grad_norm": 3.0942084789276123,
      "learning_rate": 0.00011803921568627452,
      "loss": 1.7721063613891601,
      "mean_token_accuracy": 0.5669537462294102,
      "num_tokens": 140588.0,
      "step": 220
    },
    {
      "entropy": 1.8797006398439406,
      "epoch": 2.163855421686747,
      "grad_norm": 2.7767162322998047,
      "learning_rate": 0.00011607843137254903,
      "loss": 1.8987955093383788,
      "mean_token_accuracy": 0.5647506386041641,
      "num_tokens": 143992.0,
      "step": 225
    },
    {
      "entropy": 1.8839523375034333,
      "epoch": 2.212048192771084,
      "grad_norm": 2.7695963382720947,
      "learning_rate": 0.00011411764705882353,
      "loss": 1.9240623474121095,
      "mean_token_accuracy": 0.587371251732111,
      "num_tokens": 146839.0,
      "step": 230
    },
    {
      "entropy": 1.9874276012182235,
      "epoch": 2.2602409638554217,
      "grad_norm": 2.6279964447021484,
      "learning_rate": 0.00011215686274509805,
      "loss": 1.821434211730957,
      "mean_token_accuracy": 0.5931334540247917,
      "num_tokens": 150301.0,
      "step": 235
    },
    {
      "entropy": 1.9438043296337129,
      "epoch": 2.3084337349397592,
      "grad_norm": 2.8450863361358643,
      "learning_rate": 0.00011019607843137256,
      "loss": 1.8939037322998047,
      "mean_token_accuracy": 0.5910127840936183,
      "num_tokens": 153306.0,
      "step": 240
    },
    {
      "entropy": 1.950391983985901,
      "epoch": 2.3566265060240963,
      "grad_norm": 3.17146635055542,
      "learning_rate": 0.00010823529411764706,
      "loss": 1.784164047241211,
      "mean_token_accuracy": 0.6021963611245156,
      "num_tokens": 156524.0,
      "step": 245
    },
    {
      "entropy": 1.9486233979463576,
      "epoch": 2.404819277108434,
      "grad_norm": 2.3902153968811035,
      "learning_rate": 0.00010627450980392158,
      "loss": 1.8011253356933594,
      "mean_token_accuracy": 0.5998784765601158,
      "num_tokens": 159769.0,
      "step": 250
    },
    {
      "entropy": 1.9773405611515045,
      "epoch": 2.453012048192771,
      "grad_norm": 2.7645199298858643,
      "learning_rate": 0.00010431372549019609,
      "loss": 1.8103084564208984,
      "mean_token_accuracy": 0.5776617653667927,
      "num_tokens": 162826.0,
      "step": 255
    },
    {
      "entropy": 1.8726800799369812,
      "epoch": 2.5012048192771084,
      "grad_norm": 3.1880316734313965,
      "learning_rate": 0.00010235294117647058,
      "loss": 1.8403133392333983,
      "mean_token_accuracy": 0.5980828732252121,
      "num_tokens": 165757.0,
      "step": 260
    },
    {
      "entropy": 1.9151667952537537,
      "epoch": 2.549397590361446,
      "grad_norm": 2.401385545730591,
      "learning_rate": 0.00010039215686274511,
      "loss": 1.7347957611083984,
      "mean_token_accuracy": 0.5913618542253971,
      "num_tokens": 169133.0,
      "step": 265
    },
    {
      "entropy": 1.9463086932897569,
      "epoch": 2.597590361445783,
      "grad_norm": 3.136993646621704,
      "learning_rate": 9.843137254901961e-05,
      "loss": 1.9120018005371093,
      "mean_token_accuracy": 0.5739120878279209,
      "num_tokens": 172522.0,
      "step": 270
    },
    {
      "entropy": 1.8978543788194657,
      "epoch": 2.6457831325301204,
      "grad_norm": 2.5544567108154297,
      "learning_rate": 9.647058823529412e-05,
      "loss": 1.7139780044555664,
      "mean_token_accuracy": 0.6048392586410045,
      "num_tokens": 175761.0,
      "step": 275
    },
    {
      "entropy": 2.0296180874109266,
      "epoch": 2.693975903614458,
      "grad_norm": 2.2905755043029785,
      "learning_rate": 9.450980392156862e-05,
      "loss": 1.8753133773803712,
      "mean_token_accuracy": 0.5943016424775124,
      "num_tokens": 179040.0,
      "step": 280
    },
    {
      "entropy": 1.8527490228414536,
      "epoch": 2.742168674698795,
      "grad_norm": 3.0274784564971924,
      "learning_rate": 9.254901960784315e-05,
      "loss": 1.7792593002319337,
      "mean_token_accuracy": 0.5924512811005116,
      "num_tokens": 181920.0,
      "step": 285
    },
    {
      "entropy": 1.8110746875405312,
      "epoch": 2.7903614457831325,
      "grad_norm": 3.0517375469207764,
      "learning_rate": 9.058823529411765e-05,
      "loss": 1.7169677734375,
      "mean_token_accuracy": 0.6138487733900547,
      "num_tokens": 184885.0,
      "step": 290
    },
    {
      "entropy": 1.8668894231319428,
      "epoch": 2.83855421686747,
      "grad_norm": 3.392165184020996,
      "learning_rate": 8.862745098039215e-05,
      "loss": 1.805967140197754,
      "mean_token_accuracy": 0.606683998554945,
      "num_tokens": 187940.0,
      "step": 295
    },
    {
      "entropy": 1.863291582465172,
      "epoch": 2.886746987951807,
      "grad_norm": 2.9377148151397705,
      "learning_rate": 8.666666666666667e-05,
      "loss": 1.81500186920166,
      "mean_token_accuracy": 0.6028773806989193,
      "num_tokens": 190965.0,
      "step": 300
    },
    {
      "entropy": 1.8498487681150437,
      "epoch": 2.9349397590361446,
      "grad_norm": 2.664236068725586,
      "learning_rate": 8.470588235294118e-05,
      "loss": 1.7328081130981445,
      "mean_token_accuracy": 0.5906364403665065,
      "num_tokens": 193935.0,
      "step": 305
    },
    {
      "entropy": 1.8544571667909622,
      "epoch": 2.983132530120482,
      "grad_norm": 2.7755041122436523,
      "learning_rate": 8.274509803921568e-05,
      "loss": 1.7405027389526366,
      "mean_token_accuracy": 0.5939664505422115,
      "num_tokens": 197548.0,
      "step": 310
    },
    {
      "epoch": 3.0,
      "eval_entropy": 1.985469556771792,
      "eval_loss": 2.0493173599243164,
      "eval_mean_token_accuracy": 0.5406613980348294,
      "eval_num_tokens": 199023.0,
      "eval_runtime": 14.4807,
      "eval_samples_per_second": 14.364,
      "eval_steps_per_second": 1.795,
      "step": 312
    },
    {
      "entropy": 1.930559371647082,
      "epoch": 3.0289156626506024,
      "grad_norm": 2.4079020023345947,
      "learning_rate": 8.07843137254902e-05,
      "loss": 1.8143762588500976,
      "mean_token_accuracy": 0.5763831867983467,
      "num_tokens": 200654.0,
      "step": 315
    },
    {
      "entropy": 1.8832986682653428,
      "epoch": 3.07710843373494,
      "grad_norm": 3.238736152648926,
      "learning_rate": 7.882352941176471e-05,
      "loss": 1.642791748046875,
      "mean_token_accuracy": 0.6261811457574368,
      "num_tokens": 203959.0,
      "step": 320
    },
    {
      "entropy": 1.8605275303125381,
      "epoch": 3.125301204819277,
      "grad_norm": 2.613211154937744,
      "learning_rate": 7.686274509803923e-05,
      "loss": 1.6668607711791992,
      "mean_token_accuracy": 0.6219084404408932,
      "num_tokens": 207164.0,
      "step": 325
    },
    {
      "entropy": 1.7501476407051086,
      "epoch": 3.1734939759036145,
      "grad_norm": 3.0143356323242188,
      "learning_rate": 7.490196078431373e-05,
      "loss": 1.6125011444091797,
      "mean_token_accuracy": 0.6130474418401718,
      "num_tokens": 210414.0,
      "step": 330
    },
    {
      "entropy": 1.760798415541649,
      "epoch": 3.221686746987952,
      "grad_norm": 2.8369691371917725,
      "learning_rate": 7.294117647058823e-05,
      "loss": 1.634211540222168,
      "mean_token_accuracy": 0.6276122540235519,
      "num_tokens": 213751.0,
      "step": 335
    },
    {
      "entropy": 1.6883425265550613,
      "epoch": 3.269879518072289,
      "grad_norm": 2.9251861572265625,
      "learning_rate": 7.098039215686276e-05,
      "loss": 1.690659523010254,
      "mean_token_accuracy": 0.6411514416337013,
      "num_tokens": 216613.0,
      "step": 340
    },
    {
      "entropy": 1.8279808849096297,
      "epoch": 3.3180722891566266,
      "grad_norm": 3.7624499797821045,
      "learning_rate": 6.901960784313726e-05,
      "loss": 1.754872703552246,
      "mean_token_accuracy": 0.6131856985390186,
      "num_tokens": 219787.0,
      "step": 345
    },
    {
      "entropy": 1.756396821141243,
      "epoch": 3.3662650602409636,
      "grad_norm": 3.5243866443634033,
      "learning_rate": 6.705882352941176e-05,
      "loss": 1.655768585205078,
      "mean_token_accuracy": 0.6300212450325489,
      "num_tokens": 222932.0,
      "step": 350
    },
    {
      "entropy": 1.874008411169052,
      "epoch": 3.414457831325301,
      "grad_norm": 3.3146021366119385,
      "learning_rate": 6.509803921568627e-05,
      "loss": 1.7400325775146483,
      "mean_token_accuracy": 0.6154260173439979,
      "num_tokens": 226276.0,
      "step": 355
    },
    {
      "entropy": 1.8046866059303284,
      "epoch": 3.4626506024096386,
      "grad_norm": 2.6372153759002686,
      "learning_rate": 6.313725490196079e-05,
      "loss": 1.6389144897460937,
      "mean_token_accuracy": 0.6158021651208401,
      "num_tokens": 229590.0,
      "step": 360
    },
    {
      "entropy": 1.866220897436142,
      "epoch": 3.5108433734939757,
      "grad_norm": 3.475135087966919,
      "learning_rate": 6.11764705882353e-05,
      "loss": 1.7979347229003906,
      "mean_token_accuracy": 0.5996182210743427,
      "num_tokens": 232687.0,
      "step": 365
    },
    {
      "entropy": 1.7321909636259079,
      "epoch": 3.559036144578313,
      "grad_norm": 3.6927313804626465,
      "learning_rate": 5.921568627450981e-05,
      "loss": 1.5469629287719726,
      "mean_token_accuracy": 0.6493223175406456,
      "num_tokens": 235499.0,
      "step": 370
    },
    {
      "entropy": 1.830334907770157,
      "epoch": 3.6072289156626507,
      "grad_norm": 3.4615859985351562,
      "learning_rate": 5.725490196078431e-05,
      "loss": 1.7406286239624023,
      "mean_token_accuracy": 0.5963820047676563,
      "num_tokens": 238822.0,
      "step": 375
    },
    {
      "entropy": 1.79691823720932,
      "epoch": 3.6554216867469878,
      "grad_norm": 3.1364524364471436,
      "learning_rate": 5.529411764705883e-05,
      "loss": 1.6433067321777344,
      "mean_token_accuracy": 0.6264043740928174,
      "num_tokens": 241849.0,
      "step": 380
    },
    {
      "entropy": 1.7372142761945724,
      "epoch": 3.7036144578313253,
      "grad_norm": 3.6655216217041016,
      "learning_rate": 5.333333333333333e-05,
      "loss": 1.5511286735534668,
      "mean_token_accuracy": 0.640411450713873,
      "num_tokens": 244887.0,
      "step": 385
    },
    {
      "entropy": 1.6415932849049568,
      "epoch": 3.7518072289156628,
      "grad_norm": 3.589327335357666,
      "learning_rate": 5.137254901960784e-05,
      "loss": 1.6273122787475587,
      "mean_token_accuracy": 0.6260945774614811,
      "num_tokens": 248336.0,
      "step": 390
    },
    {
      "entropy": 1.697372230887413,
      "epoch": 3.8,
      "grad_norm": 3.1451239585876465,
      "learning_rate": 4.9411764705882355e-05,
      "loss": 1.6096899032592773,
      "mean_token_accuracy": 0.6343577854335308,
      "num_tokens": 251693.0,
      "step": 395
    },
    {
      "entropy": 1.842886620759964,
      "epoch": 3.8481927710843373,
      "grad_norm": 3.4734911918640137,
      "learning_rate": 4.745098039215686e-05,
      "loss": 1.627140998840332,
      "mean_token_accuracy": 0.6251780308783055,
      "num_tokens": 254586.0,
      "step": 400
    },
    {
      "entropy": 1.7731292784214019,
      "epoch": 3.896385542168675,
      "grad_norm": 3.6335442066192627,
      "learning_rate": 4.549019607843137e-05,
      "loss": 1.6728458404541016,
      "mean_token_accuracy": 0.6188982456922532,
      "num_tokens": 257947.0,
      "step": 405
    },
    {
      "entropy": 1.8043945997953414,
      "epoch": 3.944578313253012,
      "grad_norm": 3.6249866485595703,
      "learning_rate": 4.3529411764705885e-05,
      "loss": 1.7272470474243165,
      "mean_token_accuracy": 0.6200564041733742,
      "num_tokens": 261397.0,
      "step": 410
    },
    {
      "entropy": 1.7803307950496674,
      "epoch": 3.9927710843373494,
      "grad_norm": 3.3378946781158447,
      "learning_rate": 4.156862745098039e-05,
      "loss": 1.714345359802246,
      "mean_token_accuracy": 0.6034644410014153,
      "num_tokens": 264865.0,
      "step": 415
    },
    {
      "epoch": 4.0,
      "eval_entropy": 1.9170196056365967,
      "eval_loss": 2.039247751235962,
      "eval_mean_token_accuracy": 0.5426948953133363,
      "eval_num_tokens": 265364.0,
      "eval_runtime": 14.4907,
      "eval_samples_per_second": 14.354,
      "eval_steps_per_second": 1.794,
      "step": 416
    },
    {
      "entropy": 1.7436758593509072,
      "epoch": 4.03855421686747,
      "grad_norm": 3.224684000015259,
      "learning_rate": 3.96078431372549e-05,
      "loss": 1.6162948608398438,
      "mean_token_accuracy": 0.6365054469359549,
      "num_tokens": 268139.0,
      "step": 420
    },
    {
      "entropy": 1.7693778246641159,
      "epoch": 4.086746987951807,
      "grad_norm": 3.8523001670837402,
      "learning_rate": 3.7647058823529415e-05,
      "loss": 1.4568367004394531,
      "mean_token_accuracy": 0.6637644946575165,
      "num_tokens": 270838.0,
      "step": 425
    },
    {
      "entropy": 1.7599476888775825,
      "epoch": 4.134939759036144,
      "grad_norm": 3.5239920616149902,
      "learning_rate": 3.568627450980392e-05,
      "loss": 1.536618423461914,
      "mean_token_accuracy": 0.6365332305431366,
      "num_tokens": 274251.0,
      "step": 430
    },
    {
      "entropy": 1.7379733711481093,
      "epoch": 4.183132530120482,
      "grad_norm": 2.5421361923217773,
      "learning_rate": 3.372549019607844e-05,
      "loss": 1.6236806869506837,
      "mean_token_accuracy": 0.6300823666155339,
      "num_tokens": 277840.0,
      "step": 435
    },
    {
      "entropy": 1.6128298133611678,
      "epoch": 4.231325301204819,
      "grad_norm": 2.651508331298828,
      "learning_rate": 3.176470588235294e-05,
      "loss": 1.6208284378051758,
      "mean_token_accuracy": 0.6503724060952664,
      "num_tokens": 281232.0,
      "step": 440
    },
    {
      "entropy": 1.6323140501976012,
      "epoch": 4.279518072289156,
      "grad_norm": 3.4628922939300537,
      "learning_rate": 2.9803921568627453e-05,
      "loss": 1.4836601257324218,
      "mean_token_accuracy": 0.6427287131547927,
      "num_tokens": 284608.0,
      "step": 445
    },
    {
      "entropy": 1.734370046854019,
      "epoch": 4.327710843373494,
      "grad_norm": 3.549942970275879,
      "learning_rate": 2.7843137254901964e-05,
      "loss": 1.5159525871276855,
      "mean_token_accuracy": 0.6537479400634766,
      "num_tokens": 287633.0,
      "step": 450
    },
    {
      "entropy": 1.7052687168121339,
      "epoch": 4.375903614457831,
      "grad_norm": 3.043501138687134,
      "learning_rate": 2.5882352941176475e-05,
      "loss": 1.5797093391418457,
      "mean_token_accuracy": 0.634084390848875,
      "num_tokens": 290871.0,
      "step": 455
    },
    {
      "entropy": 1.6739964008331298,
      "epoch": 4.424096385542168,
      "grad_norm": 3.586318254470825,
      "learning_rate": 2.3921568627450983e-05,
      "loss": 1.4949226379394531,
      "mean_token_accuracy": 0.6558637455105781,
      "num_tokens": 294036.0,
      "step": 460
    },
    {
      "entropy": 1.7056181490421296,
      "epoch": 4.472289156626506,
      "grad_norm": 3.9863288402557373,
      "learning_rate": 2.196078431372549e-05,
      "loss": 1.5965059280395508,
      "mean_token_accuracy": 0.6334700129926205,
      "num_tokens": 297025.0,
      "step": 465
    },
    {
      "entropy": 1.7901102036237717,
      "epoch": 4.5204819277108435,
      "grad_norm": 3.508091688156128,
      "learning_rate": 2e-05,
      "loss": 1.6554153442382813,
      "mean_token_accuracy": 0.620038353651762,
      "num_tokens": 300127.0,
      "step": 470
    },
    {
      "entropy": 1.6874196112155915,
      "epoch": 4.5686746987951805,
      "grad_norm": 3.2619593143463135,
      "learning_rate": 1.803921568627451e-05,
      "loss": 1.5016522407531738,
      "mean_token_accuracy": 0.6612366795539856,
      "num_tokens": 303134.0,
      "step": 475
    },
    {
      "entropy": 1.7402487054467202,
      "epoch": 4.6168674698795185,
      "grad_norm": 3.0993025302886963,
      "learning_rate": 1.607843137254902e-05,
      "loss": 1.6372486114501954,
      "mean_token_accuracy": 0.6456566855311394,
      "num_tokens": 306272.0,
      "step": 480
    },
    {
      "entropy": 1.7473211318254471,
      "epoch": 4.6650602409638555,
      "grad_norm": 3.765566825866699,
      "learning_rate": 1.411764705882353e-05,
      "loss": 1.506948757171631,
      "mean_token_accuracy": 0.642524740844965,
      "num_tokens": 309481.0,
      "step": 485
    },
    {
      "entropy": 1.7598285585641862,
      "epoch": 4.713253012048193,
      "grad_norm": 3.702375650405884,
      "learning_rate": 1.215686274509804e-05,
      "loss": 1.7290857315063477,
      "mean_token_accuracy": 0.6187055550515652,
      "num_tokens": 313174.0,
      "step": 490
    },
    {
      "entropy": 1.7480218917131425,
      "epoch": 4.76144578313253,
      "grad_norm": 3.211798906326294,
      "learning_rate": 1.0196078431372549e-05,
      "loss": 1.5974117279052735,
      "mean_token_accuracy": 0.647635517269373,
      "num_tokens": 316145.0,
      "step": 495
    },
    {
      "entropy": 1.7365855067968368,
      "epoch": 4.809638554216868,
      "grad_norm": 3.4989044666290283,
      "learning_rate": 8.23529411764706e-06,
      "loss": 1.5805817604064942,
      "mean_token_accuracy": 0.6247361660003662,
      "num_tokens": 319594.0,
      "step": 500
    },
    {
      "entropy": 1.7110005378723145,
      "epoch": 4.857831325301205,
      "grad_norm": 4.468890190124512,
      "learning_rate": 6.274509803921569e-06,
      "loss": 1.4910991668701172,
      "mean_token_accuracy": 0.6495989665389061,
      "num_tokens": 322676.0,
      "step": 505
    },
    {
      "entropy": 1.6686782240867615,
      "epoch": 4.906024096385542,
      "grad_norm": 3.632615566253662,
      "learning_rate": 4.313725490196079e-06,
      "loss": 1.4474696159362792,
      "mean_token_accuracy": 0.6759913340210915,
      "num_tokens": 325876.0,
      "step": 510
    },
    {
      "entropy": 1.792736566066742,
      "epoch": 4.95421686746988,
      "grad_norm": 3.7487971782684326,
      "learning_rate": 2.3529411764705885e-06,
      "loss": 1.5956071853637694,
      "mean_token_accuracy": 0.6452264726161957,
      "num_tokens": 328755.0,
      "step": 515
    },
    {
      "entropy": 1.747338687118731,
      "epoch": 5.0,
      "grad_norm": 4.007129192352295,
      "learning_rate": 3.921568627450981e-07,
      "loss": 1.5880197525024413,
      "mean_token_accuracy": 0.6304794377402255,
      "num_tokens": 331705.0,
      "step": 520
    },
    {
      "epoch": 5.0,
      "eval_entropy": 1.8492318941996648,
      "eval_loss": 2.0438461303710938,
      "eval_mean_token_accuracy": 0.5493836334118476,
      "eval_num_tokens": 331705.0,
      "eval_runtime": 14.4967,
      "eval_samples_per_second": 14.348,
      "eval_steps_per_second": 1.794,
      "step": 520
    }
  ],
  "logging_steps": 5,
  "max_steps": 520,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2063401172152320.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}