{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 565,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.3933267489075661,
      "epoch": 0.08888888888888889,
      "grad_norm": 19.091772079467773,
      "learning_rate": 2e-05,
      "loss": 4.2829,
      "mean_token_accuracy": 0.5681655570864678,
      "num_tokens": 29409.0,
      "step": 10
    },
    {
      "entropy": 0.7650148034095764,
      "epoch": 0.17777777777777778,
      "grad_norm": 3.8214948177337646,
      "learning_rate": 2e-05,
      "loss": 1.9461,
      "mean_token_accuracy": 0.6879387736320496,
      "num_tokens": 58827.0,
      "step": 20
    },
    {
      "entropy": 1.1828202903270721,
      "epoch": 0.26666666666666666,
      "grad_norm": 2.6364810466766357,
      "learning_rate": 2e-05,
      "loss": 1.4227,
      "mean_token_accuracy": 0.7338245347142219,
      "num_tokens": 88299.0,
      "step": 30
    },
    {
      "entropy": 1.134617891907692,
      "epoch": 0.35555555555555557,
      "grad_norm": 1.9795665740966797,
      "learning_rate": 2e-05,
      "loss": 1.1307,
      "mean_token_accuracy": 0.7887017637491226,
      "num_tokens": 117759.0,
      "step": 40
    },
    {
      "entropy": 0.877768449485302,
      "epoch": 0.4444444444444444,
      "grad_norm": 1.8397494554519653,
      "learning_rate": 2e-05,
      "loss": 0.8535,
      "mean_token_accuracy": 0.8351033940911293,
      "num_tokens": 147140.0,
      "step": 50
    },
    {
      "entropy": 0.587233804166317,
      "epoch": 0.5333333333333333,
      "grad_norm": 1.7626832723617554,
      "learning_rate": 2e-05,
      "loss": 0.5781,
      "mean_token_accuracy": 0.8860435307025909,
      "num_tokens": 176659.0,
      "step": 60
    },
    {
      "entropy": 0.3405880033969879,
      "epoch": 0.6222222222222222,
      "grad_norm": 1.520534634590149,
      "learning_rate": 2e-05,
      "loss": 0.3419,
      "mean_token_accuracy": 0.9315642505884171,
      "num_tokens": 206147.0,
      "step": 70
    },
    {
      "entropy": 0.19235755391418935,
      "epoch": 0.7111111111111111,
      "grad_norm": 1.268977403640747,
      "learning_rate": 2e-05,
      "loss": 0.1858,
      "mean_token_accuracy": 0.9728681713342666,
      "num_tokens": 235603.0,
      "step": 80
    },
    {
      "entropy": 0.11804858762770891,
      "epoch": 0.8,
      "grad_norm": 0.781975269317627,
      "learning_rate": 2e-05,
      "loss": 0.1064,
      "mean_token_accuracy": 0.9887924045324326,
      "num_tokens": 265047.0,
      "step": 90
    },
    {
      "entropy": 0.09983876422047615,
      "epoch": 0.8888888888888888,
      "grad_norm": 0.4874080419540405,
      "learning_rate": 2e-05,
      "loss": 0.0815,
      "mean_token_accuracy": 0.9895498856902123,
      "num_tokens": 294524.0,
      "step": 100
    },
    {
      "entropy": 0.08479245882481337,
      "epoch": 0.9777777777777777,
      "grad_norm": 0.3734425902366638,
      "learning_rate": 2e-05,
      "loss": 0.0768,
      "mean_token_accuracy": 0.9904760375618935,
      "num_tokens": 324032.0,
      "step": 110
    },
    {
      "entropy": 0.07308715611304108,
      "epoch": 1.0622222222222222,
      "grad_norm": 0.37292563915252686,
      "learning_rate": 2e-05,
      "loss": 0.0646,
      "mean_token_accuracy": 0.9918417679636102,
      "num_tokens": 351993.0,
      "step": 120
    },
    {
      "entropy": 0.06726833172142506,
      "epoch": 1.1511111111111112,
      "grad_norm": 0.5837728977203369,
      "learning_rate": 2e-05,
      "loss": 0.0625,
      "mean_token_accuracy": 0.9917700842022896,
      "num_tokens": 381506.0,
      "step": 130
    },
    {
      "entropy": 0.06390567375347019,
      "epoch": 1.24,
      "grad_norm": 0.3549947738647461,
      "learning_rate": 2e-05,
      "loss": 0.0606,
      "mean_token_accuracy": 0.9916005581617355,
      "num_tokens": 411017.0,
      "step": 140
    },
    {
      "entropy": 0.06265801694244147,
      "epoch": 1.3288888888888888,
      "grad_norm": 0.2998465597629547,
      "learning_rate": 2e-05,
      "loss": 0.0564,
      "mean_token_accuracy": 0.9918477952480316,
      "num_tokens": 440460.0,
      "step": 150
    },
    {
      "entropy": 0.060036468971520665,
      "epoch": 1.4177777777777778,
      "grad_norm": 0.5888973474502563,
      "learning_rate": 2e-05,
      "loss": 0.0557,
      "mean_token_accuracy": 0.9919296875596046,
      "num_tokens": 469970.0,
      "step": 160
    },
    {
      "entropy": 0.05825678938999772,
      "epoch": 1.5066666666666668,
      "grad_norm": 0.27591630816459656,
      "learning_rate": 2e-05,
      "loss": 0.0557,
      "mean_token_accuracy": 0.991797935962677,
      "num_tokens": 499471.0,
      "step": 170
    },
    {
      "entropy": 0.058804288040846586,
      "epoch": 1.5955555555555554,
      "grad_norm": 0.3655984401702881,
      "learning_rate": 2e-05,
      "loss": 0.0537,
      "mean_token_accuracy": 0.9921618834137916,
      "num_tokens": 528959.0,
      "step": 180
    },
    {
      "entropy": 0.05707454737275839,
      "epoch": 1.6844444444444444,
      "grad_norm": 0.3361314535140991,
      "learning_rate": 2e-05,
      "loss": 0.0514,
      "mean_token_accuracy": 0.9922538578510285,
      "num_tokens": 558384.0,
      "step": 190
    },
    {
      "entropy": 0.05885109649971128,
      "epoch": 1.7733333333333334,
      "grad_norm": 0.2926768958568573,
      "learning_rate": 2e-05,
      "loss": 0.0565,
      "mean_token_accuracy": 0.9916261032223701,
      "num_tokens": 587873.0,
      "step": 200
    },
    {
      "entropy": 0.05613070921972394,
      "epoch": 1.8622222222222222,
      "grad_norm": 0.2952570617198944,
      "learning_rate": 2e-05,
      "loss": 0.0495,
      "mean_token_accuracy": 0.9927162423729896,
      "num_tokens": 617263.0,
      "step": 210
    },
    {
      "entropy": 0.05452471813187003,
      "epoch": 1.951111111111111,
      "grad_norm": 0.2781422734260559,
      "learning_rate": 2e-05,
      "loss": 0.0503,
      "mean_token_accuracy": 0.9923128932714462,
      "num_tokens": 646673.0,
      "step": 220
    },
    {
      "entropy": 0.05329967241146063,
      "epoch": 2.0355555555555553,
      "grad_norm": 0.3413642942905426,
      "learning_rate": 2e-05,
      "loss": 0.0495,
      "mean_token_accuracy": 0.9925352836910047,
      "num_tokens": 674550.0,
      "step": 230
    },
    {
      "entropy": 0.055740222427994014,
      "epoch": 2.1244444444444444,
      "grad_norm": 0.3916967213153839,
      "learning_rate": 2e-05,
      "loss": 0.0494,
      "mean_token_accuracy": 0.992403993010521,
      "num_tokens": 704049.0,
      "step": 240
    },
    {
      "entropy": 0.05344773568212986,
      "epoch": 2.2133333333333334,
      "grad_norm": 0.41811126470565796,
      "learning_rate": 2e-05,
      "loss": 0.0458,
      "mean_token_accuracy": 0.9928064867854118,
      "num_tokens": 733527.0,
      "step": 250
    },
    {
      "entropy": 0.05095162307843566,
      "epoch": 2.3022222222222224,
      "grad_norm": 0.5080037117004395,
      "learning_rate": 2e-05,
      "loss": 0.0462,
      "mean_token_accuracy": 0.9926309958100319,
      "num_tokens": 762980.0,
      "step": 260
    },
    {
      "entropy": 0.05230198642238974,
      "epoch": 2.391111111111111,
      "grad_norm": 0.34135064482688904,
      "learning_rate": 2e-05,
      "loss": 0.047,
      "mean_token_accuracy": 0.9925719112157821,
      "num_tokens": 792457.0,
      "step": 270
    },
    {
      "entropy": 0.053009994141757485,
      "epoch": 2.48,
      "grad_norm": 0.261165976524353,
      "learning_rate": 2e-05,
      "loss": 0.044,
      "mean_token_accuracy": 0.9927724987268448,
      "num_tokens": 821928.0,
      "step": 280
    },
    {
      "entropy": 0.052336765173822644,
      "epoch": 2.568888888888889,
      "grad_norm": 0.2925412356853485,
      "learning_rate": 2e-05,
      "loss": 0.0459,
      "mean_token_accuracy": 0.9917180150747299,
      "num_tokens": 851523.0,
      "step": 290
    },
    {
      "entropy": 0.04876706637442112,
      "epoch": 2.6577777777777776,
      "grad_norm": 0.37404191493988037,
      "learning_rate": 2e-05,
      "loss": 0.0406,
      "mean_token_accuracy": 0.9927994713187218,
      "num_tokens": 880960.0,
      "step": 300
    },
    {
      "entropy": 0.051569243893027306,
      "epoch": 2.7466666666666666,
      "grad_norm": 0.33673104643821716,
      "learning_rate": 2e-05,
      "loss": 0.0408,
      "mean_token_accuracy": 0.9931992888450623,
      "num_tokens": 910363.0,
      "step": 310
    },
    {
      "entropy": 0.049042111821472646,
      "epoch": 2.8355555555555556,
      "grad_norm": 0.3380878269672394,
      "learning_rate": 2e-05,
      "loss": 0.0405,
      "mean_token_accuracy": 0.9932568341493606,
      "num_tokens": 939743.0,
      "step": 320
    },
    {
      "entropy": 0.04809595588594675,
      "epoch": 2.924444444444444,
      "grad_norm": 0.27717074751853943,
      "learning_rate": 2e-05,
      "loss": 0.0401,
      "mean_token_accuracy": 0.9928347066044807,
      "num_tokens": 969188.0,
      "step": 330
    },
    {
      "entropy": 0.047154336873638,
      "epoch": 3.008888888888889,
      "grad_norm": 0.26735636591911316,
      "learning_rate": 2e-05,
      "loss": 0.0388,
      "mean_token_accuracy": 0.9933515975349828,
      "num_tokens": 997173.0,
      "step": 340
    },
    {
      "entropy": 0.0476642238907516,
      "epoch": 3.097777777777778,
      "grad_norm": 0.3359168767929077,
      "learning_rate": 2e-05,
      "loss": 0.0405,
      "mean_token_accuracy": 0.9925565898418427,
      "num_tokens": 1026726.0,
      "step": 350
    },
    {
      "entropy": 0.04855702333152294,
      "epoch": 3.1866666666666665,
      "grad_norm": 0.29510918259620667,
      "learning_rate": 2e-05,
      "loss": 0.0364,
      "mean_token_accuracy": 0.993024954199791,
      "num_tokens": 1056268.0,
      "step": 360
    },
    {
      "entropy": 0.044791849609464404,
      "epoch": 3.2755555555555556,
      "grad_norm": 0.27017977833747864,
      "learning_rate": 2e-05,
      "loss": 0.0359,
      "mean_token_accuracy": 0.9939059600234031,
      "num_tokens": 1085652.0,
      "step": 370
    },
    {
      "entropy": 0.045806001592427495,
      "epoch": 3.3644444444444446,
      "grad_norm": 0.2858346700668335,
      "learning_rate": 2e-05,
      "loss": 0.036,
      "mean_token_accuracy": 0.9928093075752258,
      "num_tokens": 1115142.0,
      "step": 380
    },
    {
      "entropy": 0.04463256490416825,
      "epoch": 3.453333333333333,
      "grad_norm": 0.3520112931728363,
      "learning_rate": 2e-05,
      "loss": 0.0362,
      "mean_token_accuracy": 0.9932136535644531,
      "num_tokens": 1144607.0,
      "step": 390
    },
    {
      "entropy": 0.04314003074541688,
      "epoch": 3.542222222222222,
      "grad_norm": 0.3692231774330139,
      "learning_rate": 2e-05,
      "loss": 0.0353,
      "mean_token_accuracy": 0.9933312207460403,
      "num_tokens": 1173996.0,
      "step": 400
    },
    {
      "entropy": 0.0454170742072165,
      "epoch": 3.631111111111111,
      "grad_norm": 0.23268748819828033,
      "learning_rate": 2e-05,
      "loss": 0.0342,
      "mean_token_accuracy": 0.9934337288141251,
      "num_tokens": 1203400.0,
      "step": 410
    },
    {
      "entropy": 0.043088483065366744,
      "epoch": 3.7199999999999998,
      "grad_norm": 0.40086743235588074,
      "learning_rate": 2e-05,
      "loss": 0.0363,
      "mean_token_accuracy": 0.9924963280558586,
      "num_tokens": 1232863.0,
      "step": 420
    },
    {
      "entropy": 0.042456195782870056,
      "epoch": 3.8088888888888888,
      "grad_norm": 0.30352523922920227,
      "learning_rate": 2e-05,
      "loss": 0.035,
      "mean_token_accuracy": 0.9930051982402801,
      "num_tokens": 1262316.0,
      "step": 430
    },
    {
      "entropy": 0.04498578486964107,
      "epoch": 3.897777777777778,
      "grad_norm": 0.2815539836883545,
      "learning_rate": 2e-05,
      "loss": 0.0337,
      "mean_token_accuracy": 0.9933026045560837,
      "num_tokens": 1291738.0,
      "step": 440
    },
    {
      "entropy": 0.04180623982101679,
      "epoch": 3.986666666666667,
      "grad_norm": 0.4016154706478119,
      "learning_rate": 2e-05,
      "loss": 0.0335,
      "mean_token_accuracy": 0.9940232038497925,
      "num_tokens": 1321198.0,
      "step": 450
    },
    {
      "entropy": 0.040797956140809936,
      "epoch": 4.071111111111111,
      "grad_norm": 0.3967713415622711,
      "learning_rate": 2e-05,
      "loss": 0.0303,
      "mean_token_accuracy": 0.9949048961463728,
      "num_tokens": 1349259.0,
      "step": 460
    },
    {
      "entropy": 0.041362932836636904,
      "epoch": 4.16,
      "grad_norm": 0.3272438943386078,
      "learning_rate": 2e-05,
      "loss": 0.0313,
      "mean_token_accuracy": 0.9941667526960373,
      "num_tokens": 1378731.0,
      "step": 470
    },
    {
      "entropy": 0.042792328353971246,
      "epoch": 4.248888888888889,
      "grad_norm": 0.34763824939727783,
      "learning_rate": 2e-05,
      "loss": 0.0321,
      "mean_token_accuracy": 0.9943796172738075,
      "num_tokens": 1408249.0,
      "step": 480
    },
    {
      "entropy": 0.042336183181032536,
      "epoch": 4.337777777777778,
      "grad_norm": 0.2802582383155823,
      "learning_rate": 2e-05,
      "loss": 0.0317,
      "mean_token_accuracy": 0.9944384515285491,
      "num_tokens": 1437770.0,
      "step": 490
    },
    {
      "entropy": 0.04013953167013824,
      "epoch": 4.426666666666667,
      "grad_norm": 0.31569746136665344,
      "learning_rate": 2e-05,
      "loss": 0.028,
      "mean_token_accuracy": 0.9952725186944008,
      "num_tokens": 1467204.0,
      "step": 500
    },
    {
      "entropy": 0.039097797218710185,
      "epoch": 4.515555555555555,
      "grad_norm": 0.33069005608558655,
      "learning_rate": 2e-05,
      "loss": 0.0312,
      "mean_token_accuracy": 0.9946726128458977,
      "num_tokens": 1496671.0,
      "step": 510
    },
    {
      "entropy": 0.03892837380990386,
      "epoch": 4.604444444444445,
      "grad_norm": 0.24601057171821594,
      "learning_rate": 2e-05,
      "loss": 0.0288,
      "mean_token_accuracy": 0.9946699410676956,
      "num_tokens": 1526129.0,
      "step": 520
    },
    {
      "entropy": 0.039517662627622484,
      "epoch": 4.693333333333333,
      "grad_norm": 0.2797907292842865,
      "learning_rate": 2e-05,
      "loss": 0.0294,
      "mean_token_accuracy": 0.9946090206503868,
      "num_tokens": 1555591.0,
      "step": 530
    },
    {
      "entropy": 0.039668824058026075,
      "epoch": 4.782222222222222,
      "grad_norm": 0.2804340422153473,
      "learning_rate": 2e-05,
      "loss": 0.0286,
      "mean_token_accuracy": 0.9947286590933799,
      "num_tokens": 1584969.0,
      "step": 540
    },
    {
      "entropy": 0.038332745106890796,
      "epoch": 4.871111111111111,
      "grad_norm": 0.22508393228054047,
      "learning_rate": 2e-05,
      "loss": 0.0298,
      "mean_token_accuracy": 0.9944633066654205,
      "num_tokens": 1614402.0,
      "step": 550
    },
    {
      "entropy": 0.03850612454116344,
      "epoch": 4.96,
      "grad_norm": 0.29442259669303894,
      "learning_rate": 2e-05,
      "loss": 0.0263,
      "mean_token_accuracy": 0.9948647573590279,
      "num_tokens": 1643824.0,
      "step": 560
    }
  ],
  "logging_steps": 10,
  "max_steps": 904,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1035206563987200.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}