{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 475,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.8691443920135498,
      "epoch": 0.021085925144965736,
      "grad_norm": 47.5,
      "learning_rate": 7.2e-07,
      "loss": 2.1268569946289064,
      "mean_token_accuracy": 0.5856592752039432,
      "num_tokens": 53385.0,
      "step": 10
    },
    {
      "entropy": 0.9018128983676433,
      "epoch": 0.04217185028993147,
      "grad_norm": 49.25,
      "learning_rate": 1.52e-06,
      "loss": 2.1235822677612304,
      "mean_token_accuracy": 0.5826416034251451,
      "num_tokens": 108949.0,
      "step": 20
    },
    {
      "entropy": 0.9313402399420738,
      "epoch": 0.0632577754348972,
      "grad_norm": 38.25,
      "learning_rate": 2.3200000000000002e-06,
      "loss": 1.9997358322143555,
      "mean_token_accuracy": 0.5817345686256885,
      "num_tokens": 162034.0,
      "step": 30
    },
    {
      "entropy": 0.9819381944835186,
      "epoch": 0.08434370057986294,
      "grad_norm": 28.25,
      "learning_rate": 3.12e-06,
      "loss": 1.8760208129882812,
      "mean_token_accuracy": 0.5874959267675877,
      "num_tokens": 214165.0,
      "step": 40
    },
    {
      "entropy": 1.0624087303876877,
      "epoch": 0.10542962572482868,
      "grad_norm": 14.1875,
      "learning_rate": 3.920000000000001e-06,
      "loss": 1.6169570922851562,
      "mean_token_accuracy": 0.601640059798956,
      "num_tokens": 265527.0,
      "step": 50
    },
    {
      "entropy": 1.1721510097384453,
      "epoch": 0.1265155508697944,
      "grad_norm": 11.3125,
      "learning_rate": 4.7200000000000005e-06,
      "loss": 1.4622878074645995,
      "mean_token_accuracy": 0.6209174185991287,
      "num_tokens": 319341.0,
      "step": 60
    },
    {
      "entropy": 1.2009219758212566,
      "epoch": 0.14760147601476015,
      "grad_norm": 9.0625,
      "learning_rate": 5.5200000000000005e-06,
      "loss": 1.2933337211608886,
      "mean_token_accuracy": 0.6467074818909169,
      "num_tokens": 373709.0,
      "step": 70
    },
    {
      "entropy": 1.1439887225627898,
      "epoch": 0.16868740115972589,
      "grad_norm": 9.3125,
      "learning_rate": 6.3200000000000005e-06,
      "loss": 1.2084498405456543,
      "mean_token_accuracy": 0.6660686552524566,
      "num_tokens": 425923.0,
      "step": 80
    },
    {
      "entropy": 1.0713621236383915,
      "epoch": 0.18977332630469163,
      "grad_norm": 7.84375,
      "learning_rate": 7.1200000000000004e-06,
      "loss": 1.1175435066223145,
      "mean_token_accuracy": 0.6812180563807487,
      "num_tokens": 480012.0,
      "step": 90
    },
    {
      "entropy": 1.0369222231209279,
      "epoch": 0.21085925144965736,
      "grad_norm": 7.09375,
      "learning_rate": 7.92e-06,
      "loss": 1.062850570678711,
      "mean_token_accuracy": 0.6899570092558861,
      "num_tokens": 533890.0,
      "step": 100
    },
    {
      "entropy": 0.9894454136490822,
      "epoch": 0.23194517659462308,
      "grad_norm": 7.75,
      "learning_rate": 8.720000000000001e-06,
      "loss": 1.0150799751281738,
      "mean_token_accuracy": 0.7009239844977856,
      "num_tokens": 590654.0,
      "step": 110
    },
    {
      "entropy": 0.9410129584372043,
      "epoch": 0.2530311017395888,
      "grad_norm": 7.28125,
      "learning_rate": 9.52e-06,
      "loss": 0.9810515403747558,
      "mean_token_accuracy": 0.709618617594242,
      "num_tokens": 643080.0,
      "step": 120
    },
    {
      "entropy": 0.9265115469694137,
      "epoch": 0.2741170268845546,
      "grad_norm": 7.46875,
      "learning_rate": 1.0320000000000001e-05,
      "loss": 0.9485825538635254,
      "mean_token_accuracy": 0.7162063598632813,
      "num_tokens": 697254.0,
      "step": 130
    },
    {
      "entropy": 0.9009825974702835,
      "epoch": 0.2952029520295203,
      "grad_norm": 6.96875,
      "learning_rate": 1.1120000000000002e-05,
      "loss": 0.9119878768920898,
      "mean_token_accuracy": 0.7238413639366627,
      "num_tokens": 751774.0,
      "step": 140
    },
    {
      "entropy": 0.9040703341364861,
      "epoch": 0.316288877174486,
      "grad_norm": 7.21875,
      "learning_rate": 1.1920000000000001e-05,
      "loss": 0.9283308029174805,
      "mean_token_accuracy": 0.7166316144168376,
      "num_tokens": 808927.0,
      "step": 150
    },
    {
      "entropy": 0.8878034949302673,
      "epoch": 0.33737480231945177,
      "grad_norm": 7.375,
      "learning_rate": 1.2720000000000002e-05,
      "loss": 0.9068665504455566,
      "mean_token_accuracy": 0.7254431843757629,
      "num_tokens": 858559.0,
      "step": 160
    },
    {
      "entropy": 0.887480016052723,
      "epoch": 0.3584607274644175,
      "grad_norm": 7.65625,
      "learning_rate": 1.3520000000000003e-05,
      "loss": 0.8971159934997559,
      "mean_token_accuracy": 0.7258390121161937,
      "num_tokens": 912285.0,
      "step": 170
    },
    {
      "entropy": 0.8527634374797344,
      "epoch": 0.37954665260938325,
      "grad_norm": 7.6875,
      "learning_rate": 1.432e-05,
      "loss": 0.8586221694946289,
      "mean_token_accuracy": 0.7359919100999832,
      "num_tokens": 969385.0,
      "step": 180
    },
    {
      "entropy": 0.8445627830922604,
      "epoch": 0.40063257775434896,
      "grad_norm": 7.03125,
      "learning_rate": 1.5120000000000001e-05,
      "loss": 0.855714225769043,
      "mean_token_accuracy": 0.7341729424893856,
      "num_tokens": 1026106.0,
      "step": 190
    },
    {
      "entropy": 0.841130618005991,
      "epoch": 0.42171850289931473,
      "grad_norm": 7.21875,
      "learning_rate": 1.5920000000000003e-05,
      "loss": 0.8620488166809082,
      "mean_token_accuracy": 0.7361006513237953,
      "num_tokens": 1079743.0,
      "step": 200
    },
    {
      "entropy": 0.831258487701416,
      "epoch": 0.44280442804428044,
      "grad_norm": 7.0,
      "learning_rate": 1.672e-05,
      "loss": 0.8441590309143067,
      "mean_token_accuracy": 0.7379129216074943,
      "num_tokens": 1131453.0,
      "step": 210
    },
    {
      "entropy": 0.79731200709939,
      "epoch": 0.46389035318924615,
      "grad_norm": 7.875,
      "learning_rate": 1.752e-05,
      "loss": 0.8135190963745117,
      "mean_token_accuracy": 0.7497627973556519,
      "num_tokens": 1182486.0,
      "step": 220
    },
    {
      "entropy": 0.8269097700715065,
      "epoch": 0.4849762783342119,
      "grad_norm": 7.5625,
      "learning_rate": 1.832e-05,
      "loss": 0.8399410247802734,
      "mean_token_accuracy": 0.7389348305761814,
      "num_tokens": 1238136.0,
      "step": 230
    },
    {
      "entropy": 0.8093138128519058,
      "epoch": 0.5060622034791776,
      "grad_norm": 7.40625,
      "learning_rate": 1.912e-05,
      "loss": 0.810025691986084,
      "mean_token_accuracy": 0.7460841409862041,
      "num_tokens": 1294939.0,
      "step": 240
    },
    {
      "entropy": 0.8069386176764966,
      "epoch": 0.5271481286241434,
      "grad_norm": 6.84375,
      "learning_rate": 1.9920000000000002e-05,
      "loss": 0.8215888023376465,
      "mean_token_accuracy": 0.7438369721174241,
      "num_tokens": 1346462.0,
      "step": 250
    },
    {
      "entropy": 0.8088659413158894,
      "epoch": 0.5482340537691092,
      "grad_norm": 6.75,
      "learning_rate": 1.9920000000000002e-05,
      "loss": 0.8201449394226075,
      "mean_token_accuracy": 0.7393839374184609,
      "num_tokens": 1397444.0,
      "step": 260
    },
    {
      "entropy": 0.7969542622566224,
      "epoch": 0.5693199789140748,
      "grad_norm": 6.59375,
      "learning_rate": 1.9831111111111113e-05,
      "loss": 0.8009759902954101,
      "mean_token_accuracy": 0.7512488305568695,
      "num_tokens": 1453900.0,
      "step": 270
    },
    {
      "entropy": 0.780416651815176,
      "epoch": 0.5904059040590406,
      "grad_norm": 7.21875,
      "learning_rate": 1.9742222222222223e-05,
      "loss": 0.7874263763427735,
      "mean_token_accuracy": 0.7526524804532528,
      "num_tokens": 1505529.0,
      "step": 280
    },
    {
      "entropy": 0.7752278268337249,
      "epoch": 0.6114918292040064,
      "grad_norm": 6.875,
      "learning_rate": 1.9653333333333334e-05,
      "loss": 0.8075678825378418,
      "mean_token_accuracy": 0.746569037437439,
      "num_tokens": 1562473.0,
      "step": 290
    },
    {
      "entropy": 0.7895354829728604,
      "epoch": 0.632577754348972,
      "grad_norm": 6.6875,
      "learning_rate": 1.9564444444444444e-05,
      "loss": 0.7880948543548584,
      "mean_token_accuracy": 0.7494045957922936,
      "num_tokens": 1613320.0,
      "step": 300
    },
    {
      "entropy": 0.7653828673064709,
      "epoch": 0.6536636794939378,
      "grad_norm": 6.78125,
      "learning_rate": 1.9475555555555558e-05,
      "loss": 0.7780828475952148,
      "mean_token_accuracy": 0.7556171745061875,
      "num_tokens": 1662847.0,
      "step": 310
    },
    {
      "entropy": 0.7652887552976608,
      "epoch": 0.6747496046389035,
      "grad_norm": 6.375,
      "learning_rate": 1.938666666666667e-05,
      "loss": 0.7539787292480469,
      "mean_token_accuracy": 0.7582654722034932,
      "num_tokens": 1718553.0,
      "step": 320
    },
    {
      "entropy": 0.7076033115386963,
      "epoch": 0.6958355297838693,
      "grad_norm": 7.15625,
      "learning_rate": 1.929777777777778e-05,
      "loss": 0.7396605491638184,
      "mean_token_accuracy": 0.7659220136702061,
      "num_tokens": 1771024.0,
      "step": 330
    },
    {
      "entropy": 0.7509547933936119,
      "epoch": 0.716921454928835,
      "grad_norm": 6.15625,
      "learning_rate": 1.920888888888889e-05,
      "loss": 0.7459388256072998,
      "mean_token_accuracy": 0.7644543245434761,
      "num_tokens": 1821303.0,
      "step": 340
    },
    {
      "entropy": 0.7591827407479286,
      "epoch": 0.7380073800738007,
      "grad_norm": 7.03125,
      "learning_rate": 1.912e-05,
      "loss": 0.7627717971801757,
      "mean_token_accuracy": 0.7567056350409984,
      "num_tokens": 1868674.0,
      "step": 350
    },
    {
      "entropy": 0.7316406652331352,
      "epoch": 0.7590933052187665,
      "grad_norm": 6.59375,
      "learning_rate": 1.9031111111111114e-05,
      "loss": 0.7425707340240478,
      "mean_token_accuracy": 0.7653280735015869,
      "num_tokens": 1924619.0,
      "step": 360
    },
    {
      "entropy": 0.7539067789912224,
      "epoch": 0.7801792303637322,
      "grad_norm": 7.0625,
      "learning_rate": 1.8942222222222224e-05,
      "loss": 0.7716987609863282,
      "mean_token_accuracy": 0.7524559155106545,
      "num_tokens": 1978795.0,
      "step": 370
    },
    {
      "entropy": 0.7687754213809967,
      "epoch": 0.8012651555086979,
      "grad_norm": 7.1875,
      "learning_rate": 1.8853333333333335e-05,
      "loss": 0.7734373569488525,
      "mean_token_accuracy": 0.7497503250837326,
      "num_tokens": 2029994.0,
      "step": 380
    },
    {
      "entropy": 0.7508607231080532,
      "epoch": 0.8223510806536637,
      "grad_norm": 7.09375,
      "learning_rate": 1.8764444444444445e-05,
      "loss": 0.7575721740722656,
      "mean_token_accuracy": 0.7572125047445297,
      "num_tokens": 2082653.0,
      "step": 390
    },
    {
      "entropy": 0.7252861328423024,
      "epoch": 0.8434370057986295,
      "grad_norm": 6.5625,
      "learning_rate": 1.867555555555556e-05,
      "loss": 0.7381186485290527,
      "mean_token_accuracy": 0.7612978041172027,
      "num_tokens": 2135276.0,
      "step": 400
    },
    {
      "entropy": 0.7339485622942448,
      "epoch": 0.8645229309435951,
      "grad_norm": 6.625,
      "learning_rate": 1.858666666666667e-05,
      "loss": 0.7348906993865967,
      "mean_token_accuracy": 0.7663561321794987,
      "num_tokens": 2187466.0,
      "step": 410
    },
    {
      "entropy": 0.7272624362260103,
      "epoch": 0.8856088560885609,
      "grad_norm": 6.90625,
      "learning_rate": 1.849777777777778e-05,
      "loss": 0.749025821685791,
      "mean_token_accuracy": 0.7617379955947399,
      "num_tokens": 2237434.0,
      "step": 420
    },
    {
      "entropy": 0.7477380312979222,
      "epoch": 0.9066947812335266,
      "grad_norm": 6.34375,
      "learning_rate": 1.840888888888889e-05,
      "loss": 0.7498275756835937,
      "mean_token_accuracy": 0.7604753144085408,
      "num_tokens": 2288985.0,
      "step": 430
    },
    {
      "entropy": 0.7346535548567772,
      "epoch": 0.9277807063784923,
      "grad_norm": 6.9375,
      "learning_rate": 1.832e-05,
      "loss": 0.739972448348999,
      "mean_token_accuracy": 0.7658773727715016,
      "num_tokens": 2341187.0,
      "step": 440
    },
    {
      "entropy": 0.7167556572705507,
      "epoch": 0.9488666315234581,
      "grad_norm": 6.4375,
      "learning_rate": 1.823111111111111e-05,
      "loss": 0.7350784778594971,
      "mean_token_accuracy": 0.7644179269671441,
      "num_tokens": 2393385.0,
      "step": 450
    },
    {
      "entropy": 0.7537519961595536,
      "epoch": 0.9699525566684238,
      "grad_norm": 7.28125,
      "learning_rate": 1.814222222222222e-05,
      "loss": 0.757652759552002,
      "mean_token_accuracy": 0.7587068513035774,
      "num_tokens": 2442248.0,
      "step": 460
    },
    {
      "entropy": 0.72038439437747,
      "epoch": 0.9910384818133896,
      "grad_norm": 7.40625,
      "learning_rate": 1.8053333333333332e-05,
      "loss": 0.7361824035644531,
      "mean_token_accuracy": 0.7609050884842873,
      "num_tokens": 2498394.0,
      "step": 470
    }
  ],
  "logging_steps": 10,
  "max_steps": 2500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.5798444746092544e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}