{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 626,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.6765631157904863,
      "epoch": 0.032,
      "grad_norm": 2.875,
      "learning_rate": 9.473684210526316e-05,
      "loss": 3.33,
      "mean_token_accuracy": 0.5463392566889524,
      "num_tokens": 32197.0,
      "step": 10
    },
    {
      "entropy": 1.1614421313628553,
      "epoch": 0.064,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0002,
      "loss": 1.6539,
      "mean_token_accuracy": 0.6597556989639998,
      "num_tokens": 64756.0,
      "step": 20
    },
    {
      "entropy": 0.939408042281866,
      "epoch": 0.096,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0001998660954234636,
      "loss": 0.7808,
      "mean_token_accuracy": 0.781595965847373,
      "num_tokens": 95410.0,
      "step": 30
    },
    {
      "entropy": 0.9574495874345302,
      "epoch": 0.128,
      "grad_norm": 0.55078125,
      "learning_rate": 0.00019946474030256675,
      "loss": 0.859,
      "mean_token_accuracy": 0.7595490481704473,
      "num_tokens": 123619.0,
      "step": 40
    },
    {
      "entropy": 0.8130528301000595,
      "epoch": 0.16,
      "grad_norm": 0.498046875,
      "learning_rate": 0.00019879700950305957,
      "loss": 0.7411,
      "mean_token_accuracy": 0.8014852415770293,
      "num_tokens": 156066.0,
      "step": 50
    },
    {
      "entropy": 0.8076014310121536,
      "epoch": 0.192,
      "grad_norm": 0.67578125,
      "learning_rate": 0.00019786469126914098,
      "loss": 0.8028,
      "mean_token_accuracy": 0.7921789325773716,
      "num_tokens": 186345.0,
      "step": 60
    },
    {
      "entropy": 0.7697046212852001,
      "epoch": 0.224,
      "grad_norm": 0.58984375,
      "learning_rate": 0.0001966702824343772,
      "loss": 0.7332,
      "mean_token_accuracy": 0.7981064647436142,
      "num_tokens": 219449.0,
      "step": 70
    },
    {
      "entropy": 0.7365534473210573,
      "epoch": 0.256,
      "grad_norm": 0.66015625,
      "learning_rate": 0.0001952169817349528,
      "loss": 0.6818,
      "mean_token_accuracy": 0.8138540856540203,
      "num_tokens": 252398.0,
      "step": 80
    },
    {
      "entropy": 0.8327326748520136,
      "epoch": 0.288,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0001935086812431626,
      "loss": 0.7288,
      "mean_token_accuracy": 0.7888761077076196,
      "num_tokens": 281833.0,
      "step": 90
    },
    {
      "entropy": 0.8304124176502228,
      "epoch": 0.32,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0001915499559440855,
      "loss": 0.7666,
      "mean_token_accuracy": 0.7884823590517044,
      "num_tokens": 313194.0,
      "step": 100
    },
    {
      "entropy": 0.7173997653648257,
      "epoch": 0.352,
      "grad_norm": 0.609375,
      "learning_rate": 0.00018934605148335608,
      "loss": 0.6474,
      "mean_token_accuracy": 0.8159435257315636,
      "num_tokens": 345629.0,
      "step": 110
    },
    {
      "entropy": 0.7204551868140697,
      "epoch": 0.384,
      "grad_norm": 0.6640625,
      "learning_rate": 0.00018690287011884508,
      "loss": 0.674,
      "mean_token_accuracy": 0.8059780243784189,
      "num_tokens": 379281.0,
      "step": 120
    },
    {
      "entropy": 0.7626429490745068,
      "epoch": 0.416,
      "grad_norm": 0.470703125,
      "learning_rate": 0.00018422695491387284,
      "loss": 0.6938,
      "mean_token_accuracy": 0.8045202873647213,
      "num_tokens": 410187.0,
      "step": 130
    },
    {
      "entropy": 0.7266103811562061,
      "epoch": 0.448,
      "grad_norm": 0.484375,
      "learning_rate": 0.00018132547221428675,
      "loss": 0.7094,
      "mean_token_accuracy": 0.8129661556333303,
      "num_tokens": 442476.0,
      "step": 140
    },
    {
      "entropy": 0.7414866480976343,
      "epoch": 0.48,
      "grad_norm": 0.6875,
      "learning_rate": 0.00017820619245633115,
      "loss": 0.6545,
      "mean_token_accuracy": 0.8073488272726536,
      "num_tokens": 472224.0,
      "step": 150
    },
    {
      "entropy": 0.7822258766740561,
      "epoch": 0.512,
      "grad_norm": 0.54296875,
      "learning_rate": 0.00017487746935670776,
      "loss": 0.7011,
      "mean_token_accuracy": 0.795686560869217,
      "num_tokens": 501232.0,
      "step": 160
    },
    {
      "entropy": 0.745093465037644,
      "epoch": 0.544,
      "grad_norm": 0.55078125,
      "learning_rate": 0.00017134821754055782,
      "loss": 0.7028,
      "mean_token_accuracy": 0.8048942796885967,
      "num_tokens": 531476.0,
      "step": 170
    },
    {
      "entropy": 0.7522185202687979,
      "epoch": 0.576,
      "grad_norm": 0.55078125,
      "learning_rate": 0.00016762788866728,
      "loss": 0.7073,
      "mean_token_accuracy": 0.8059838317334652,
      "num_tokens": 564807.0,
      "step": 180
    },
    {
      "entropy": 0.7345199611037969,
      "epoch": 0.608,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0001637264461181213,
      "loss": 0.6915,
      "mean_token_accuracy": 0.80965431407094,
      "num_tokens": 597440.0,
      "step": 190
    },
    {
      "entropy": 0.8049266938120127,
      "epoch": 0.64,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0001596543383133303,
      "loss": 0.7237,
      "mean_token_accuracy": 0.7935581188648939,
      "num_tokens": 630050.0,
      "step": 200
    },
    {
      "entropy": 0.6652943708002568,
      "epoch": 0.672,
      "grad_norm": 0.52734375,
      "learning_rate": 0.00015542247073033116,
      "loss": 0.6553,
      "mean_token_accuracy": 0.8136730916798115,
      "num_tokens": 660867.0,
      "step": 210
    },
    {
      "entropy": 0.6704742344096303,
      "epoch": 0.704,
      "grad_norm": 0.55078125,
      "learning_rate": 0.00015104217669785713,
      "loss": 0.6572,
      "mean_token_accuracy": 0.8189906783401966,
      "num_tokens": 694781.0,
      "step": 220
    },
    {
      "entropy": 0.8146899126470089,
      "epoch": 0.736,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0001465251870442586,
      "loss": 0.7881,
      "mean_token_accuracy": 0.7911321885883809,
      "num_tokens": 727207.0,
      "step": 230
    },
    {
      "entropy": 0.6836955199018121,
      "epoch": 0.768,
      "grad_norm": 0.53515625,
      "learning_rate": 0.00014188359868127134,
      "loss": 0.6743,
      "mean_token_accuracy": 0.8147166304290294,
      "num_tokens": 760836.0,
      "step": 240
    },
    {
      "entropy": 0.8043613195419311,
      "epoch": 0.8,
      "grad_norm": 0.48828125,
      "learning_rate": 0.00013712984220737938,
      "loss": 0.7463,
      "mean_token_accuracy": 0.7957966670393943,
      "num_tokens": 794412.0,
      "step": 250
    },
    {
      "entropy": 0.6497382273897528,
      "epoch": 0.832,
      "grad_norm": 0.5703125,
      "learning_rate": 0.00013227664861753454,
      "loss": 0.6431,
      "mean_token_accuracy": 0.8244783975183964,
      "num_tokens": 827099.0,
      "step": 260
    },
    {
      "entropy": 0.8308647315949201,
      "epoch": 0.864,
      "grad_norm": 0.609375,
      "learning_rate": 0.00012733701520838681,
      "loss": 0.7665,
      "mean_token_accuracy": 0.7912245698273181,
      "num_tokens": 858402.0,
      "step": 270
    },
    {
      "entropy": 0.6155038874596357,
      "epoch": 0.896,
      "grad_norm": 0.54296875,
      "learning_rate": 0.00012232417077033414,
      "loss": 0.6085,
      "mean_token_accuracy": 0.8311799973249435,
      "num_tokens": 887820.0,
      "step": 280
    },
    {
      "entropy": 0.7490381442010403,
      "epoch": 0.928,
      "grad_norm": 0.4921875,
      "learning_rate": 0.00011725154015961088,
      "loss": 0.6872,
      "mean_token_accuracy": 0.8019387792795897,
      "num_tokens": 918679.0,
      "step": 290
    },
    {
      "entropy": 0.7262276472523809,
      "epoch": 0.96,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0001121327083452942,
      "loss": 0.7096,
      "mean_token_accuracy": 0.8102037977427244,
      "num_tokens": 951444.0,
      "step": 300
    },
    {
      "entropy": 0.6726594571024179,
      "epoch": 0.992,
      "grad_norm": 0.5625,
      "learning_rate": 0.00010698138402751314,
      "loss": 0.6115,
      "mean_token_accuracy": 0.8189524404704571,
      "num_tokens": 983476.0,
      "step": 310
    },
    {
      "entropy": 0.6149586082289094,
      "epoch": 1.0224,
      "grad_norm": 0.515625,
      "learning_rate": 0.00010181136292429528,
      "loss": 0.5614,
      "mean_token_accuracy": 0.840733151686819,
      "num_tokens": 1014089.0,
      "step": 320
    },
    {
      "entropy": 0.6150874266400933,
      "epoch": 1.0544,
      "grad_norm": 0.6328125,
      "learning_rate": 9.663649082537075e-05,
      "loss": 0.5826,
      "mean_token_accuracy": 0.8344197846949101,
      "num_tokens": 1048175.0,
      "step": 330
    },
    {
      "entropy": 0.688701249845326,
      "epoch": 1.0864,
      "grad_norm": 0.63671875,
      "learning_rate": 9.147062651188034e-05,
      "loss": 0.582,
      "mean_token_accuracy": 0.822768672183156,
      "num_tokens": 1077167.0,
      "step": 340
    },
    {
      "entropy": 0.6307795088738203,
      "epoch": 1.1184,
      "grad_norm": 0.7890625,
      "learning_rate": 8.632760464129085e-05,
      "loss": 0.6128,
      "mean_token_accuracy": 0.8314752034842968,
      "num_tokens": 1106931.0,
      "step": 350
    },
    {
      "entropy": 0.5494781306013465,
      "epoch": 1.1504,
      "grad_norm": 0.69921875,
      "learning_rate": 8.122119869691636e-05,
      "loss": 0.5115,
      "mean_token_accuracy": 0.8432619094848632,
      "num_tokens": 1140267.0,
      "step": 360
    },
    {
      "entropy": 0.5744671380147338,
      "epoch": 1.1824,
      "grad_norm": 0.703125,
      "learning_rate": 7.616508410126881e-05,
      "loss": 0.5383,
      "mean_token_accuracy": 0.8436760693788529,
      "num_tokens": 1174016.0,
      "step": 370
    },
    {
      "entropy": 0.7073690647259354,
      "epoch": 1.2144,
      "grad_norm": 0.68359375,
      "learning_rate": 7.11728015920253e-05,
      "loss": 0.6142,
      "mean_token_accuracy": 0.815117885172367,
      "num_tokens": 1203362.0,
      "step": 380
    },
    {
      "entropy": 0.6476075932383537,
      "epoch": 1.2464,
      "grad_norm": 0.59375,
      "learning_rate": 6.625772095869278e-05,
      "loss": 0.576,
      "mean_token_accuracy": 0.8274026662111282,
      "num_tokens": 1233733.0,
      "step": 390
    },
    {
      "entropy": 0.6079700440168381,
      "epoch": 1.2784,
      "grad_norm": 0.75390625,
      "learning_rate": 6.143300523708828e-05,
      "loss": 0.5926,
      "mean_token_accuracy": 0.8361447125673294,
      "num_tokens": 1266408.0,
      "step": 400
    },
    {
      "entropy": 0.710476384870708,
      "epoch": 1.3104,
      "grad_norm": 0.62890625,
      "learning_rate": 5.6711575457523904e-05,
      "loss": 0.6673,
      "mean_token_accuracy": 0.8070973232388496,
      "num_tokens": 1297692.0,
      "step": 410
    },
    {
      "entropy": 0.6220435939729214,
      "epoch": 1.3424,
      "grad_norm": 0.58984375,
      "learning_rate": 5.210607604110537e-05,
      "loss": 0.535,
      "mean_token_accuracy": 0.8349618587642909,
      "num_tokens": 1328647.0,
      "step": 420
    },
    {
      "entropy": 0.636304665543139,
      "epoch": 1.3744,
      "grad_norm": 0.71875,
      "learning_rate": 4.762884093681439e-05,
      "loss": 0.5681,
      "mean_token_accuracy": 0.8337738730013371,
      "num_tokens": 1360726.0,
      "step": 430
    },
    {
      "entropy": 0.5927035486325621,
      "epoch": 1.4064,
      "grad_norm": 0.640625,
      "learning_rate": 4.329186059006498e-05,
      "loss": 0.5686,
      "mean_token_accuracy": 0.843344721570611,
      "num_tokens": 1392566.0,
      "step": 440
    },
    {
      "entropy": 0.5271363547071815,
      "epoch": 1.4384000000000001,
      "grad_norm": 0.78515625,
      "learning_rate": 3.9106749831192655e-05,
      "loss": 0.4787,
      "mean_token_accuracy": 0.8514938339591026,
      "num_tokens": 1423983.0,
      "step": 450
    },
    {
      "entropy": 0.5616443024948239,
      "epoch": 1.4704,
      "grad_norm": 0.609375,
      "learning_rate": 3.50847167698759e-05,
      "loss": 0.562,
      "mean_token_accuracy": 0.8439631327986717,
      "num_tokens": 1457630.0,
      "step": 460
    },
    {
      "entropy": 0.6092825196683407,
      "epoch": 1.5024,
      "grad_norm": 0.55078125,
      "learning_rate": 3.123653277879256e-05,
      "loss": 0.5474,
      "mean_token_accuracy": 0.8354076690971851,
      "num_tokens": 1487297.0,
      "step": 470
    },
    {
      "entropy": 0.6257630828768015,
      "epoch": 1.5344,
      "grad_norm": 0.7265625,
      "learning_rate": 2.7572503646897817e-05,
      "loss": 0.5658,
      "mean_token_accuracy": 0.8328265182673931,
      "num_tokens": 1520243.0,
      "step": 480
    },
    {
      "entropy": 0.6298206310719252,
      "epoch": 1.5664,
      "grad_norm": 0.65625,
      "learning_rate": 2.410244197957815e-05,
      "loss": 0.5689,
      "mean_token_accuracy": 0.8329555809497833,
      "num_tokens": 1552070.0,
      "step": 490
    },
    {
      "entropy": 0.639691986143589,
      "epoch": 1.5984,
      "grad_norm": 0.81640625,
      "learning_rate": 2.0835640919595878e-05,
      "loss": 0.5927,
      "mean_token_accuracy": 0.834860210865736,
      "num_tokens": 1583470.0,
      "step": 500
    },
    {
      "entropy": 0.6124603090807795,
      "epoch": 1.6303999999999998,
      "grad_norm": 0.6875,
      "learning_rate": 1.7780849259202394e-05,
      "loss": 0.5623,
      "mean_token_accuracy": 0.8385609880089759,
      "num_tokens": 1615277.0,
      "step": 510
    },
    {
      "entropy": 0.5668893689289689,
      "epoch": 1.6623999999999999,
      "grad_norm": 0.625,
      "learning_rate": 1.494624801007144e-05,
      "loss": 0.5323,
      "mean_token_accuracy": 0.8450584053993225,
      "num_tokens": 1648050.0,
      "step": 520
    },
    {
      "entropy": 0.6057403447106481,
      "epoch": 1.6944,
      "grad_norm": 0.63671875,
      "learning_rate": 1.2339428493801386e-05,
      "loss": 0.5769,
      "mean_token_accuracy": 0.835369348526001,
      "num_tokens": 1680674.0,
      "step": 530
    },
    {
      "entropy": 0.6277687439695001,
      "epoch": 1.7264,
      "grad_norm": 0.78125,
      "learning_rate": 9.967372011660813e-06,
      "loss": 0.5623,
      "mean_token_accuracy": 0.8294578112661839,
      "num_tokens": 1710429.0,
      "step": 540
    },
    {
      "entropy": 0.6762315159663558,
      "epoch": 1.7584,
      "grad_norm": 0.6796875,
      "learning_rate": 7.836431148025003e-06,
      "loss": 0.5577,
      "mean_token_accuracy": 0.8203907363116741,
      "num_tokens": 1739709.0,
      "step": 550
    },
    {
      "entropy": 0.6461215754970908,
      "epoch": 1.7904,
      "grad_norm": 0.71875,
      "learning_rate": 5.9523127575733415e-06,
      "loss": 0.5774,
      "mean_token_accuracy": 0.8332192912697792,
      "num_tokens": 1770745.0,
      "step": 560
    },
    {
      "entropy": 0.6354038525372744,
      "epoch": 1.8224,
      "grad_norm": 0.73046875,
      "learning_rate": 4.320062681810167e-06,
      "loss": 0.6049,
      "mean_token_accuracy": 0.8269469507038594,
      "num_tokens": 1801462.0,
      "step": 570
    },
    {
      "entropy": 0.6516931420192122,
      "epoch": 1.8544,
      "grad_norm": 0.6796875,
      "learning_rate": 2.9440522358394227e-06,
      "loss": 0.5778,
      "mean_token_accuracy": 0.8247631914913655,
      "num_tokens": 1834635.0,
      "step": 580
    },
    {
      "entropy": 0.6212354129180312,
      "epoch": 1.8864,
      "grad_norm": 0.76171875,
      "learning_rate": 1.8279665015826342e-06,
      "loss": 0.546,
      "mean_token_accuracy": 0.8374237649142742,
      "num_tokens": 1865151.0,
      "step": 590
    },
    {
      "entropy": 0.6352507598698139,
      "epoch": 1.9184,
      "grad_norm": 0.70703125,
      "learning_rate": 9.747944587923163e-07,
      "loss": 0.6198,
      "mean_token_accuracy": 0.8281367566436529,
      "num_tokens": 1897057.0,
      "step": 600
    },
    {
      "entropy": 0.5361566325649619,
      "epoch": 1.9504000000000001,
      "grad_norm": 0.84375,
      "learning_rate": 3.86820980290481e-07,
      "loss": 0.5049,
      "mean_token_accuracy": 0.8522287331521511,
      "num_tokens": 1930326.0,
      "step": 610
    },
    {
      "entropy": 0.5473375553265214,
      "epoch": 1.9824000000000002,
      "grad_norm": 0.65625,
      "learning_rate": 6.562071287022908e-08,
      "loss": 0.5049,
      "mean_token_accuracy": 0.8522140845656395,
      "num_tokens": 1964962.0,
      "step": 620
    }
  ],
  "logging_steps": 10,
  "max_steps": 626,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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": 4.345811216383795e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}