{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.890224225226729,
  "eval_steps": 100,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.6993201158940794,
      "epoch": 0.00890224225226729,
      "grad_norm": 264.0,
      "learning_rate": 1.8e-07,
      "loss": 6.8158,
      "mean_token_accuracy": 0.21653764923103153,
      "num_tokens": 129942.0,
      "step": 10
    },
    {
      "entropy": 2.748416267335415,
      "epoch": 0.01780448450453458,
      "grad_norm": 342.0,
      "learning_rate": 3.8e-07,
      "loss": 6.842,
      "mean_token_accuracy": 0.21704859719611705,
      "num_tokens": 259048.0,
      "step": 20
    },
    {
      "entropy": 2.856283613294363,
      "epoch": 0.026706726756801868,
      "grad_norm": 260.0,
      "learning_rate": 5.800000000000001e-07,
      "loss": 6.5796,
      "mean_token_accuracy": 0.21546755363233389,
      "num_tokens": 397518.0,
      "step": 30
    },
    {
      "entropy": 3.1255638763308524,
      "epoch": 0.03560896900906916,
      "grad_norm": 235.0,
      "learning_rate": 7.8e-07,
      "loss": 6.0702,
      "mean_token_accuracy": 0.22630739556625484,
      "num_tokens": 532019.0,
      "step": 40
    },
    {
      "entropy": 3.437747097015381,
      "epoch": 0.04451121126133645,
      "grad_norm": 120.0,
      "learning_rate": 9.800000000000001e-07,
      "loss": 5.3567,
      "mean_token_accuracy": 0.24060304677113892,
      "num_tokens": 667753.0,
      "step": 50
    },
    {
      "entropy": 3.8268849313259126,
      "epoch": 0.053413453513603736,
      "grad_norm": 69.0,
      "learning_rate": 1.1800000000000001e-06,
      "loss": 4.7458,
      "mean_token_accuracy": 0.25712613044306637,
      "num_tokens": 807220.0,
      "step": 60
    },
    {
      "entropy": 4.098993194103241,
      "epoch": 0.06231569576587103,
      "grad_norm": 44.0,
      "learning_rate": 1.3800000000000001e-06,
      "loss": 4.1873,
      "mean_token_accuracy": 0.2863635550253093,
      "num_tokens": 938024.0,
      "step": 70
    },
    {
      "entropy": 3.7632771015167235,
      "epoch": 0.07121793801813832,
      "grad_norm": 39.75,
      "learning_rate": 1.5800000000000001e-06,
      "loss": 3.8191,
      "mean_token_accuracy": 0.3185409564524889,
      "num_tokens": 1068429.0,
      "step": 80
    },
    {
      "entropy": 3.4629461631178855,
      "epoch": 0.08012018027040561,
      "grad_norm": 32.0,
      "learning_rate": 1.7800000000000001e-06,
      "loss": 3.5033,
      "mean_token_accuracy": 0.3425675223581493,
      "num_tokens": 1209109.0,
      "step": 90
    },
    {
      "entropy": 3.3570878595113753,
      "epoch": 0.0890224225226729,
      "grad_norm": 22.375,
      "learning_rate": 1.98e-06,
      "loss": 3.3283,
      "mean_token_accuracy": 0.36780647775158287,
      "num_tokens": 1336698.0,
      "step": 100
    },
    {
      "epoch": 0.0890224225226729,
      "eval_coding_entropy": 2.025357067108154,
      "eval_coding_loss": 2.231159210205078,
      "eval_coding_mean_token_accuracy": 0.49899543941020963,
      "eval_coding_num_tokens": 1336698.0,
      "eval_coding_runtime": 81.3311,
      "eval_coding_samples_per_second": 6.148,
      "eval_coding_steps_per_second": 3.074,
      "step": 100
    },
    {
      "epoch": 0.0890224225226729,
      "eval_chemistry_entropy": 3.242468201637268,
      "eval_chemistry_loss": 3.3159916400909424,
      "eval_chemistry_mean_token_accuracy": 0.3768811701536179,
      "eval_chemistry_num_tokens": 1336698.0,
      "eval_chemistry_runtime": 44.3638,
      "eval_chemistry_samples_per_second": 11.27,
      "eval_chemistry_steps_per_second": 5.635,
      "step": 100
    },
    {
      "entropy": 3.208672270178795,
      "epoch": 0.09792466477494019,
      "grad_norm": 26.625,
      "learning_rate": 2.1800000000000003e-06,
      "loss": 3.2262,
      "mean_token_accuracy": 0.38037093859165905,
      "num_tokens": 1470326.0,
      "step": 110
    },
    {
      "entropy": 3.073743937909603,
      "epoch": 0.10682690702720747,
      "grad_norm": 27.75,
      "learning_rate": 2.38e-06,
      "loss": 3.0418,
      "mean_token_accuracy": 0.4051793536171317,
      "num_tokens": 1608574.0,
      "step": 120
    },
    {
      "entropy": 3.0620595172047613,
      "epoch": 0.11572914927947477,
      "grad_norm": 23.0,
      "learning_rate": 2.5800000000000003e-06,
      "loss": 3.0619,
      "mean_token_accuracy": 0.407612837664783,
      "num_tokens": 1740382.0,
      "step": 130
    },
    {
      "entropy": 2.9336021929979323,
      "epoch": 0.12463139153174206,
      "grad_norm": 24.0,
      "learning_rate": 2.7800000000000005e-06,
      "loss": 2.8918,
      "mean_token_accuracy": 0.4315342627465725,
      "num_tokens": 1872747.0,
      "step": 140
    },
    {
      "entropy": 2.933651026338339,
      "epoch": 0.13353363378400934,
      "grad_norm": 21.125,
      "learning_rate": 2.9800000000000003e-06,
      "loss": 2.8977,
      "mean_token_accuracy": 0.4332870215177536,
      "num_tokens": 1999786.0,
      "step": 150
    },
    {
      "entropy": 2.78707619830966,
      "epoch": 0.14243587603627664,
      "grad_norm": 26.75,
      "learning_rate": 3.1800000000000005e-06,
      "loss": 2.713,
      "mean_token_accuracy": 0.45073309261351824,
      "num_tokens": 2133173.0,
      "step": 160
    },
    {
      "entropy": 2.686560308933258,
      "epoch": 0.15133811828854393,
      "grad_norm": 21.0,
      "learning_rate": 3.3800000000000007e-06,
      "loss": 2.6608,
      "mean_token_accuracy": 0.45698707569390534,
      "num_tokens": 2268889.0,
      "step": 170
    },
    {
      "entropy": 2.688250133395195,
      "epoch": 0.16024036054081123,
      "grad_norm": 23.0,
      "learning_rate": 3.58e-06,
      "loss": 2.6085,
      "mean_token_accuracy": 0.4693053809925914,
      "num_tokens": 2406145.0,
      "step": 180
    },
    {
      "entropy": 2.606059934198856,
      "epoch": 0.1691426027930785,
      "grad_norm": 19.0,
      "learning_rate": 3.7800000000000002e-06,
      "loss": 2.5776,
      "mean_token_accuracy": 0.4732652172446251,
      "num_tokens": 2533703.0,
      "step": 190
    },
    {
      "entropy": 2.6039498060941697,
      "epoch": 0.1780448450453458,
      "grad_norm": 20.625,
      "learning_rate": 3.980000000000001e-06,
      "loss": 2.5669,
      "mean_token_accuracy": 0.4830343261361122,
      "num_tokens": 2663709.0,
      "step": 200
    },
    {
      "epoch": 0.1780448450453458,
      "eval_coding_entropy": 1.940526816368103,
      "eval_coding_loss": 2.1769461631774902,
      "eval_coding_mean_token_accuracy": 0.5133515602350235,
      "eval_coding_num_tokens": 2663709.0,
      "eval_coding_runtime": 80.7684,
      "eval_coding_samples_per_second": 6.191,
      "eval_coding_steps_per_second": 3.095,
      "step": 200
    },
    {
      "epoch": 0.1780448450453458,
      "eval_chemistry_entropy": 2.574181836128235,
      "eval_chemistry_loss": 2.5678696632385254,
      "eval_chemistry_mean_token_accuracy": 0.4837975095510483,
      "eval_chemistry_num_tokens": 2663709.0,
      "eval_chemistry_runtime": 44.3338,
      "eval_chemistry_samples_per_second": 11.278,
      "eval_chemistry_steps_per_second": 5.639,
      "step": 200
    },
    {
      "entropy": 2.517921609431505,
      "epoch": 0.18694708729761308,
      "grad_norm": 21.875,
      "learning_rate": 4.18e-06,
      "loss": 2.4289,
      "mean_token_accuracy": 0.4946174016222358,
      "num_tokens": 2803638.0,
      "step": 210
    },
    {
      "entropy": 2.4712012231349947,
      "epoch": 0.19584932954988038,
      "grad_norm": 20.625,
      "learning_rate": 4.38e-06,
      "loss": 2.4073,
      "mean_token_accuracy": 0.4968462478369474,
      "num_tokens": 2934980.0,
      "step": 220
    },
    {
      "entropy": 2.402497890591621,
      "epoch": 0.20475157180214767,
      "grad_norm": 19.375,
      "learning_rate": 4.58e-06,
      "loss": 2.3435,
      "mean_token_accuracy": 0.507471177726984,
      "num_tokens": 3071339.0,
      "step": 230
    },
    {
      "entropy": 2.4326496072113515,
      "epoch": 0.21365381405441494,
      "grad_norm": 19.875,
      "learning_rate": 4.78e-06,
      "loss": 2.3438,
      "mean_token_accuracy": 0.5086988668888808,
      "num_tokens": 3208702.0,
      "step": 240
    },
    {
      "entropy": 2.321472028642893,
      "epoch": 0.22255605630668224,
      "grad_norm": 20.0,
      "learning_rate": 4.980000000000001e-06,
      "loss": 2.2871,
      "mean_token_accuracy": 0.5164231477305293,
      "num_tokens": 3347298.0,
      "step": 250
    },
    {
      "entropy": 2.342081367969513,
      "epoch": 0.23145829855894953,
      "grad_norm": 20.375,
      "learning_rate": 5.18e-06,
      "loss": 2.3225,
      "mean_token_accuracy": 0.5174826141446829,
      "num_tokens": 3480119.0,
      "step": 260
    },
    {
      "entropy": 2.30421741977334,
      "epoch": 0.24036054081121683,
      "grad_norm": 25.125,
      "learning_rate": 5.380000000000001e-06,
      "loss": 2.2163,
      "mean_token_accuracy": 0.5236617928370834,
      "num_tokens": 3615777.0,
      "step": 270
    },
    {
      "entropy": 2.3181479632854463,
      "epoch": 0.24926278306348412,
      "grad_norm": 19.625,
      "learning_rate": 5.580000000000001e-06,
      "loss": 2.2445,
      "mean_token_accuracy": 0.5264638463035226,
      "num_tokens": 3746392.0,
      "step": 280
    },
    {
      "entropy": 2.3754931248724462,
      "epoch": 0.2581650253157514,
      "grad_norm": 19.5,
      "learning_rate": 5.78e-06,
      "loss": 2.3122,
      "mean_token_accuracy": 0.5208129966631532,
      "num_tokens": 3880425.0,
      "step": 290
    },
    {
      "entropy": 2.2281612612307073,
      "epoch": 0.2670672675680187,
      "grad_norm": 21.5,
      "learning_rate": 5.98e-06,
      "loss": 2.1504,
      "mean_token_accuracy": 0.5389395957812667,
      "num_tokens": 4019487.0,
      "step": 300
    },
    {
      "epoch": 0.2670672675680187,
      "eval_coding_entropy": 1.8798885374069214,
      "eval_coding_loss": 2.1364073753356934,
      "eval_coding_mean_token_accuracy": 0.5203545212745666,
      "eval_coding_num_tokens": 4019487.0,
      "eval_coding_runtime": 80.6635,
      "eval_coding_samples_per_second": 6.199,
      "eval_coding_steps_per_second": 3.099,
      "step": 300
    },
    {
      "epoch": 0.2670672675680187,
      "eval_chemistry_entropy": 2.227835373401642,
      "eval_chemistry_loss": 2.2248311042785645,
      "eval_chemistry_mean_token_accuracy": 0.5376261820793152,
      "eval_chemistry_num_tokens": 4019487.0,
      "eval_chemistry_runtime": 44.3841,
      "eval_chemistry_samples_per_second": 11.265,
      "eval_chemistry_steps_per_second": 5.633,
      "step": 300
    },
    {
      "entropy": 2.2015350952744486,
      "epoch": 0.275969509820286,
      "grad_norm": 21.0,
      "learning_rate": 6.18e-06,
      "loss": 2.1543,
      "mean_token_accuracy": 0.5409426406025887,
      "num_tokens": 4162935.0,
      "step": 310
    },
    {
      "entropy": 2.148462438583374,
      "epoch": 0.2848717520725533,
      "grad_norm": 23.5,
      "learning_rate": 6.380000000000001e-06,
      "loss": 2.0957,
      "mean_token_accuracy": 0.5452342376112937,
      "num_tokens": 4298100.0,
      "step": 320
    },
    {
      "entropy": 2.0984656289219856,
      "epoch": 0.29377399432482054,
      "grad_norm": 19.5,
      "learning_rate": 6.5800000000000005e-06,
      "loss": 2.0404,
      "mean_token_accuracy": 0.5558066863566637,
      "num_tokens": 4431990.0,
      "step": 330
    },
    {
      "entropy": 2.1295887492597103,
      "epoch": 0.30267623657708786,
      "grad_norm": 17.625,
      "learning_rate": 6.780000000000001e-06,
      "loss": 2.0432,
      "mean_token_accuracy": 0.5554485151544213,
      "num_tokens": 4568116.0,
      "step": 340
    },
    {
      "entropy": 2.0176764823496343,
      "epoch": 0.31157847882935513,
      "grad_norm": 15.6875,
      "learning_rate": 6.98e-06,
      "loss": 1.9609,
      "mean_token_accuracy": 0.5690282396972179,
      "num_tokens": 4696575.0,
      "step": 350
    },
    {
      "entropy": 2.138897743076086,
      "epoch": 0.32048072108162245,
      "grad_norm": 20.0,
      "learning_rate": 7.180000000000001e-06,
      "loss": 2.0699,
      "mean_token_accuracy": 0.5525342913344502,
      "num_tokens": 4831334.0,
      "step": 360
    },
    {
      "entropy": 1.9979716561734677,
      "epoch": 0.3293829633338897,
      "grad_norm": 19.875,
      "learning_rate": 7.3800000000000005e-06,
      "loss": 1.9417,
      "mean_token_accuracy": 0.5726395657286047,
      "num_tokens": 4968132.0,
      "step": 370
    },
    {
      "entropy": 2.0710657715797423,
      "epoch": 0.338285205586157,
      "grad_norm": 23.125,
      "learning_rate": 7.58e-06,
      "loss": 1.9983,
      "mean_token_accuracy": 0.5640093663707375,
      "num_tokens": 5103301.0,
      "step": 380
    },
    {
      "entropy": 2.0690263971686362,
      "epoch": 0.3471874478384243,
      "grad_norm": 19.125,
      "learning_rate": 7.78e-06,
      "loss": 2.0099,
      "mean_token_accuracy": 0.5661305790767074,
      "num_tokens": 5230700.0,
      "step": 390
    },
    {
      "entropy": 2.0010731495916843,
      "epoch": 0.3560896900906916,
      "grad_norm": 15.625,
      "learning_rate": 7.980000000000002e-06,
      "loss": 1.9302,
      "mean_token_accuracy": 0.5734745427966118,
      "num_tokens": 5362915.0,
      "step": 400
    },
    {
      "epoch": 0.3560896900906916,
      "eval_coding_entropy": 1.9175666098594666,
      "eval_coding_loss": 2.1198952198028564,
      "eval_coding_mean_token_accuracy": 0.5218169560432434,
      "eval_coding_num_tokens": 5362915.0,
      "eval_coding_runtime": 80.6411,
      "eval_coding_samples_per_second": 6.2,
      "eval_coding_steps_per_second": 3.1,
      "step": 400
    },
    {
      "epoch": 0.3560896900906916,
      "eval_chemistry_entropy": 2.0164283666610716,
      "eval_chemistry_loss": 1.9814609289169312,
      "eval_chemistry_mean_token_accuracy": 0.5761862490177154,
      "eval_chemistry_num_tokens": 5362915.0,
      "eval_chemistry_runtime": 44.318,
      "eval_chemistry_samples_per_second": 11.282,
      "eval_chemistry_steps_per_second": 5.641,
      "step": 400
    },
    {
      "entropy": 1.9029628101736307,
      "epoch": 0.3649919323429589,
      "grad_norm": 20.625,
      "learning_rate": 8.18e-06,
      "loss": 1.861,
      "mean_token_accuracy": 0.5885544015094638,
      "num_tokens": 5497551.0,
      "step": 410
    },
    {
      "entropy": 1.8858691919595003,
      "epoch": 0.37389417459522617,
      "grad_norm": 18.75,
      "learning_rate": 8.380000000000001e-06,
      "loss": 1.8473,
      "mean_token_accuracy": 0.5886353930458427,
      "num_tokens": 5625455.0,
      "step": 420
    },
    {
      "entropy": 1.917143502831459,
      "epoch": 0.38279641684749344,
      "grad_norm": 18.0,
      "learning_rate": 8.580000000000001e-06,
      "loss": 1.8792,
      "mean_token_accuracy": 0.5864967351779342,
      "num_tokens": 5757241.0,
      "step": 430
    },
    {
      "entropy": 1.9342061571776867,
      "epoch": 0.39169865909976076,
      "grad_norm": 18.25,
      "learning_rate": 8.78e-06,
      "loss": 1.8538,
      "mean_token_accuracy": 0.5882492190226912,
      "num_tokens": 5890806.0,
      "step": 440
    },
    {
      "entropy": 1.8766419854015113,
      "epoch": 0.400600901352028,
      "grad_norm": 19.25,
      "learning_rate": 8.98e-06,
      "loss": 1.8162,
      "mean_token_accuracy": 0.5928706232458353,
      "num_tokens": 6023088.0,
      "step": 450
    },
    {
      "entropy": 1.8140752121806145,
      "epoch": 0.40950314360429535,
      "grad_norm": 17.375,
      "learning_rate": 9.180000000000002e-06,
      "loss": 1.7678,
      "mean_token_accuracy": 0.6031433455646038,
      "num_tokens": 6158751.0,
      "step": 460
    },
    {
      "entropy": 1.8589354373514653,
      "epoch": 0.4184053858565626,
      "grad_norm": 17.25,
      "learning_rate": 9.38e-06,
      "loss": 1.8104,
      "mean_token_accuracy": 0.5982734605669975,
      "num_tokens": 6289699.0,
      "step": 470
    },
    {
      "entropy": 1.8204911682754754,
      "epoch": 0.4273076281088299,
      "grad_norm": 18.125,
      "learning_rate": 9.58e-06,
      "loss": 1.7427,
      "mean_token_accuracy": 0.6014625774696469,
      "num_tokens": 6423747.0,
      "step": 480
    },
    {
      "entropy": 1.8097763508558273,
      "epoch": 0.4362098703610972,
      "grad_norm": 18.0,
      "learning_rate": 9.780000000000001e-06,
      "loss": 1.7556,
      "mean_token_accuracy": 0.606043404340744,
      "num_tokens": 6556651.0,
      "step": 490
    },
    {
      "entropy": 1.7938172847032547,
      "epoch": 0.4451121126133645,
      "grad_norm": 16.125,
      "learning_rate": 9.980000000000001e-06,
      "loss": 1.7349,
      "mean_token_accuracy": 0.6063602799549699,
      "num_tokens": 6684517.0,
      "step": 500
    },
    {
      "epoch": 0.4451121126133645,
      "eval_coding_entropy": 1.9086761941909791,
      "eval_coding_loss": 2.1499152183532715,
      "eval_coding_mean_token_accuracy": 0.5191502428054809,
      "eval_coding_num_tokens": 6684517.0,
      "eval_coding_runtime": 80.761,
      "eval_coding_samples_per_second": 6.191,
      "eval_coding_steps_per_second": 3.096,
      "step": 500
    },
    {
      "epoch": 0.4451121126133645,
      "eval_chemistry_entropy": 1.8382492520809173,
      "eval_chemistry_loss": 1.8092410564422607,
      "eval_chemistry_mean_token_accuracy": 0.6044963012933731,
      "eval_chemistry_num_tokens": 6684517.0,
      "eval_chemistry_runtime": 43.9817,
      "eval_chemistry_samples_per_second": 11.368,
      "eval_chemistry_steps_per_second": 5.684,
      "step": 500
    },
    {
      "entropy": 1.7506890654563905,
      "epoch": 0.4540143548656318,
      "grad_norm": 17.0,
      "learning_rate": 1.018e-05,
      "loss": 1.6717,
      "mean_token_accuracy": 0.6144311860203743,
      "num_tokens": 6814065.0,
      "step": 510
    },
    {
      "entropy": 1.752192370966077,
      "epoch": 0.46291659711789906,
      "grad_norm": 18.5,
      "learning_rate": 1.038e-05,
      "loss": 1.6926,
      "mean_token_accuracy": 0.6115166075527668,
      "num_tokens": 6945659.0,
      "step": 520
    },
    {
      "entropy": 1.7476748272776603,
      "epoch": 0.4718188393701664,
      "grad_norm": 18.875,
      "learning_rate": 1.0580000000000002e-05,
      "loss": 1.6791,
      "mean_token_accuracy": 0.614359668083489,
      "num_tokens": 7083465.0,
      "step": 530
    },
    {
      "entropy": 1.7340088631957769,
      "epoch": 0.48072108162243365,
      "grad_norm": 14.125,
      "learning_rate": 1.0780000000000002e-05,
      "loss": 1.7051,
      "mean_token_accuracy": 0.6148024540394544,
      "num_tokens": 7217797.0,
      "step": 540
    },
    {
      "entropy": 1.7364882446825505,
      "epoch": 0.4896233238747009,
      "grad_norm": 16.25,
      "learning_rate": 1.0980000000000002e-05,
      "loss": 1.6523,
      "mean_token_accuracy": 0.6168902778998018,
      "num_tokens": 7364316.0,
      "step": 550
    },
    {
      "entropy": 1.6127678520977498,
      "epoch": 0.49852556612696824,
      "grad_norm": 20.5,
      "learning_rate": 1.1180000000000001e-05,
      "loss": 1.5842,
      "mean_token_accuracy": 0.6330170270055533,
      "num_tokens": 7498182.0,
      "step": 560
    },
    {
      "entropy": 1.7071593157947063,
      "epoch": 0.5074278083792355,
      "grad_norm": 15.1875,
      "learning_rate": 1.138e-05,
      "loss": 1.6117,
      "mean_token_accuracy": 0.6257967110723257,
      "num_tokens": 7630581.0,
      "step": 570
    },
    {
      "entropy": 1.6894463405013085,
      "epoch": 0.5163300506315028,
      "grad_norm": 17.875,
      "learning_rate": 1.1580000000000001e-05,
      "loss": 1.6253,
      "mean_token_accuracy": 0.6228752240538598,
      "num_tokens": 7761686.0,
      "step": 580
    },
    {
      "entropy": 1.7127487953752278,
      "epoch": 0.52523229288377,
      "grad_norm": 14.8125,
      "learning_rate": 1.178e-05,
      "loss": 1.6567,
      "mean_token_accuracy": 0.6184359768405556,
      "num_tokens": 7894733.0,
      "step": 590
    },
    {
      "entropy": 1.7086713213473559,
      "epoch": 0.5341345351360374,
      "grad_norm": 15.75,
      "learning_rate": 1.198e-05,
      "loss": 1.6548,
      "mean_token_accuracy": 0.6228462293744087,
      "num_tokens": 8026923.0,
      "step": 600
    },
    {
      "epoch": 0.5341345351360374,
      "eval_coding_entropy": 1.9360160393714905,
      "eval_coding_loss": 2.150575876235962,
      "eval_coding_mean_token_accuracy": 0.5199909027814865,
      "eval_coding_num_tokens": 8026923.0,
      "eval_coding_runtime": 80.6899,
      "eval_coding_samples_per_second": 6.197,
      "eval_coding_steps_per_second": 3.098,
      "step": 600
    },
    {
      "epoch": 0.5341345351360374,
      "eval_chemistry_entropy": 1.7303061096668244,
      "eval_chemistry_loss": 1.6744147539138794,
      "eval_chemistry_mean_token_accuracy": 0.6269070360660552,
      "eval_chemistry_num_tokens": 8026923.0,
      "eval_chemistry_runtime": 44.5775,
      "eval_chemistry_samples_per_second": 11.216,
      "eval_chemistry_steps_per_second": 5.608,
      "step": 600
    },
    {
      "entropy": 1.692430131137371,
      "epoch": 0.5430367773883047,
      "grad_norm": 14.75,
      "learning_rate": 1.218e-05,
      "loss": 1.6309,
      "mean_token_accuracy": 0.6261043848469854,
      "num_tokens": 8162179.0,
      "step": 610
    },
    {
      "entropy": 1.6273927468806506,
      "epoch": 0.551939019640572,
      "grad_norm": 17.875,
      "learning_rate": 1.2380000000000002e-05,
      "loss": 1.6052,
      "mean_token_accuracy": 0.6333680363371968,
      "num_tokens": 8301506.0,
      "step": 620
    },
    {
      "entropy": 1.6359726019203662,
      "epoch": 0.5608412618928392,
      "grad_norm": 16.625,
      "learning_rate": 1.2580000000000002e-05,
      "loss": 1.574,
      "mean_token_accuracy": 0.6349722359329462,
      "num_tokens": 8425846.0,
      "step": 630
    },
    {
      "entropy": 1.6046722147613763,
      "epoch": 0.5697435041451065,
      "grad_norm": 16.125,
      "learning_rate": 1.2780000000000001e-05,
      "loss": 1.5789,
      "mean_token_accuracy": 0.6387555615976452,
      "num_tokens": 8559126.0,
      "step": 640
    },
    {
      "entropy": 1.641243514046073,
      "epoch": 0.5786457463973739,
      "grad_norm": 18.375,
      "learning_rate": 1.2980000000000001e-05,
      "loss": 1.5626,
      "mean_token_accuracy": 0.6389757834374905,
      "num_tokens": 8685256.0,
      "step": 650
    },
    {
      "entropy": 1.6162871330976487,
      "epoch": 0.5875479886496411,
      "grad_norm": 13.375,
      "learning_rate": 1.3180000000000001e-05,
      "loss": 1.585,
      "mean_token_accuracy": 0.6322070205584168,
      "num_tokens": 8821088.0,
      "step": 660
    },
    {
      "entropy": 1.5711755152791738,
      "epoch": 0.5964502309019084,
      "grad_norm": 16.125,
      "learning_rate": 1.3380000000000002e-05,
      "loss": 1.5036,
      "mean_token_accuracy": 0.6462705478072166,
      "num_tokens": 8950665.0,
      "step": 670
    },
    {
      "entropy": 1.5002975206822158,
      "epoch": 0.6053524731541757,
      "grad_norm": 11.875,
      "learning_rate": 1.3580000000000002e-05,
      "loss": 1.455,
      "mean_token_accuracy": 0.6536407358944416,
      "num_tokens": 9089866.0,
      "step": 680
    },
    {
      "entropy": 1.5762819994241,
      "epoch": 0.614254715406443,
      "grad_norm": 15.3125,
      "learning_rate": 1.378e-05,
      "loss": 1.5317,
      "mean_token_accuracy": 0.6418447192758322,
      "num_tokens": 9219941.0,
      "step": 690
    },
    {
      "entropy": 1.5155543368309736,
      "epoch": 0.6231569576587103,
      "grad_norm": 14.1875,
      "learning_rate": 1.398e-05,
      "loss": 1.4688,
      "mean_token_accuracy": 0.6533356573432684,
      "num_tokens": 9358866.0,
      "step": 700
    },
    {
      "epoch": 0.6231569576587103,
      "eval_coding_entropy": 1.9042859573364257,
      "eval_coding_loss": 2.178054094314575,
      "eval_coding_mean_token_accuracy": 0.517359386563301,
      "eval_coding_num_tokens": 9358866.0,
      "eval_coding_runtime": 80.7138,
      "eval_coding_samples_per_second": 6.195,
      "eval_coding_steps_per_second": 3.097,
      "step": 700
    },
    {
      "epoch": 0.6231569576587103,
      "eval_chemistry_entropy": 1.56386332988739,
      "eval_chemistry_loss": 1.572709560394287,
      "eval_chemistry_mean_token_accuracy": 0.6433606895208359,
      "eval_chemistry_num_tokens": 9358866.0,
      "eval_chemistry_runtime": 44.295,
      "eval_chemistry_samples_per_second": 11.288,
      "eval_chemistry_steps_per_second": 5.644,
      "step": 700
    },
    {
      "entropy": 1.5002719465643168,
      "epoch": 0.6320591999109776,
      "grad_norm": 15.75,
      "learning_rate": 1.418e-05,
      "loss": 1.4804,
      "mean_token_accuracy": 0.6501881824806333,
      "num_tokens": 9493065.0,
      "step": 710
    },
    {
      "entropy": 1.522989520803094,
      "epoch": 0.6409614421632449,
      "grad_norm": 16.75,
      "learning_rate": 1.4380000000000001e-05,
      "loss": 1.4427,
      "mean_token_accuracy": 0.6548429084941745,
      "num_tokens": 9629977.0,
      "step": 720
    },
    {
      "entropy": 1.4675043471157552,
      "epoch": 0.6498636844155121,
      "grad_norm": 11.9375,
      "learning_rate": 1.4580000000000001e-05,
      "loss": 1.4134,
      "mean_token_accuracy": 0.6610828962177038,
      "num_tokens": 9765477.0,
      "step": 730
    },
    {
      "entropy": 1.491576162353158,
      "epoch": 0.6587659266677794,
      "grad_norm": 14.625,
      "learning_rate": 1.478e-05,
      "loss": 1.4587,
      "mean_token_accuracy": 0.6557513184845447,
      "num_tokens": 9897757.0,
      "step": 740
    },
    {
      "entropy": 1.4832498595118522,
      "epoch": 0.6676681689200468,
      "grad_norm": 13.0625,
      "learning_rate": 1.498e-05,
      "loss": 1.4184,
      "mean_token_accuracy": 0.6580261953175068,
      "num_tokens": 10026698.0,
      "step": 750
    },
    {
      "entropy": 1.4681301727890967,
      "epoch": 0.676570411172314,
      "grad_norm": 12.6875,
      "learning_rate": 1.5180000000000002e-05,
      "loss": 1.4207,
      "mean_token_accuracy": 0.6607962453737855,
      "num_tokens": 10168367.0,
      "step": 760
    },
    {
      "entropy": 1.431799278780818,
      "epoch": 0.6854726534245813,
      "grad_norm": 12.375,
      "learning_rate": 1.5380000000000002e-05,
      "loss": 1.4078,
      "mean_token_accuracy": 0.6646994914859533,
      "num_tokens": 10307268.0,
      "step": 770
    },
    {
      "entropy": 1.4440745569765567,
      "epoch": 0.6943748956768486,
      "grad_norm": 11.4375,
      "learning_rate": 1.5580000000000003e-05,
      "loss": 1.3797,
      "mean_token_accuracy": 0.666821115091443,
      "num_tokens": 10439039.0,
      "step": 780
    },
    {
      "entropy": 1.4467374693602324,
      "epoch": 0.703277137929116,
      "grad_norm": 12.3125,
      "learning_rate": 1.578e-05,
      "loss": 1.4202,
      "mean_token_accuracy": 0.6628865763545037,
      "num_tokens": 10576127.0,
      "step": 790
    },
    {
      "entropy": 1.476112574338913,
      "epoch": 0.7121793801813832,
      "grad_norm": 12.375,
      "learning_rate": 1.5980000000000003e-05,
      "loss": 1.397,
      "mean_token_accuracy": 0.663382687792182,
      "num_tokens": 10709549.0,
      "step": 800
    },
    {
      "epoch": 0.7121793801813832,
      "eval_coding_entropy": 1.8322501955032349,
      "eval_coding_loss": 2.2156379222869873,
      "eval_coding_mean_token_accuracy": 0.5140866179466248,
      "eval_coding_num_tokens": 10709549.0,
      "eval_coding_runtime": 80.6982,
      "eval_coding_samples_per_second": 6.196,
      "eval_coding_steps_per_second": 3.098,
      "step": 800
    },
    {
      "epoch": 0.7121793801813832,
      "eval_chemistry_entropy": 1.456686591386795,
      "eval_chemistry_loss": 1.4780359268188477,
      "eval_chemistry_mean_token_accuracy": 0.6602535841464996,
      "eval_chemistry_num_tokens": 10709549.0,
      "eval_chemistry_runtime": 44.3155,
      "eval_chemistry_samples_per_second": 11.283,
      "eval_chemistry_steps_per_second": 5.641,
      "step": 800
    },
    {
      "entropy": 1.3352404598146677,
      "epoch": 0.7210816224336505,
      "grad_norm": 13.75,
      "learning_rate": 1.618e-05,
      "loss": 1.3165,
      "mean_token_accuracy": 0.6821591448038816,
      "num_tokens": 10841474.0,
      "step": 810
    },
    {
      "entropy": 1.4511364437639713,
      "epoch": 0.7299838646859178,
      "grad_norm": 11.75,
      "learning_rate": 1.638e-05,
      "loss": 1.4194,
      "mean_token_accuracy": 0.6614615445956588,
      "num_tokens": 10976011.0,
      "step": 820
    },
    {
      "entropy": 1.4821706857532262,
      "epoch": 0.738886106938185,
      "grad_norm": 10.875,
      "learning_rate": 1.658e-05,
      "loss": 1.4022,
      "mean_token_accuracy": 0.662100600078702,
      "num_tokens": 11109618.0,
      "step": 830
    },
    {
      "entropy": 1.3720035433769227,
      "epoch": 0.7477883491904523,
      "grad_norm": 13.0625,
      "learning_rate": 1.6780000000000002e-05,
      "loss": 1.3532,
      "mean_token_accuracy": 0.6731239786371589,
      "num_tokens": 11245251.0,
      "step": 840
    },
    {
      "entropy": 1.428163754194975,
      "epoch": 0.7566905914427197,
      "grad_norm": 12.625,
      "learning_rate": 1.698e-05,
      "loss": 1.384,
      "mean_token_accuracy": 0.6671924598515033,
      "num_tokens": 11379950.0,
      "step": 850
    },
    {
      "entropy": 1.425218838453293,
      "epoch": 0.7655928336949869,
      "grad_norm": 12.4375,
      "learning_rate": 1.718e-05,
      "loss": 1.3805,
      "mean_token_accuracy": 0.6696572992950678,
      "num_tokens": 11518514.0,
      "step": 860
    },
    {
      "entropy": 1.3892541889101266,
      "epoch": 0.7744950759472542,
      "grad_norm": 10.0,
      "learning_rate": 1.7380000000000003e-05,
      "loss": 1.3583,
      "mean_token_accuracy": 0.6718392688781023,
      "num_tokens": 11650235.0,
      "step": 870
    },
    {
      "entropy": 1.409494410455227,
      "epoch": 0.7833973181995215,
      "grad_norm": 9.6875,
      "learning_rate": 1.758e-05,
      "loss": 1.3757,
      "mean_token_accuracy": 0.6687737111002207,
      "num_tokens": 11787117.0,
      "step": 880
    },
    {
      "entropy": 1.3409042585641147,
      "epoch": 0.7922995604517888,
      "grad_norm": 13.625,
      "learning_rate": 1.7780000000000003e-05,
      "loss": 1.3062,
      "mean_token_accuracy": 0.6828067794442176,
      "num_tokens": 11924960.0,
      "step": 890
    },
    {
      "entropy": 1.461272732913494,
      "epoch": 0.801201802704056,
      "grad_norm": 12.8125,
      "learning_rate": 1.798e-05,
      "loss": 1.4331,
      "mean_token_accuracy": 0.6593699859455228,
      "num_tokens": 12050993.0,
      "step": 900
    },
    {
      "epoch": 0.801201802704056,
      "eval_coding_entropy": 1.9756887073516847,
      "eval_coding_loss": 2.2261524200439453,
      "eval_coding_mean_token_accuracy": 0.5107356268167496,
      "eval_coding_num_tokens": 12050993.0,
      "eval_coding_runtime": 80.8151,
      "eval_coding_samples_per_second": 6.187,
      "eval_coding_steps_per_second": 3.093,
      "step": 900
    },
    {
      "epoch": 0.801201802704056,
      "eval_chemistry_entropy": 1.4363152146339417,
      "eval_chemistry_loss": 1.4172585010528564,
      "eval_chemistry_mean_token_accuracy": 0.6695912653207778,
      "eval_chemistry_num_tokens": 12050993.0,
      "eval_chemistry_runtime": 44.4444,
      "eval_chemistry_samples_per_second": 11.25,
      "eval_chemistry_steps_per_second": 5.625,
      "step": 900
    },
    {
      "entropy": 1.409971272200346,
      "epoch": 0.8101040449563234,
      "grad_norm": 9.75,
      "learning_rate": 1.8180000000000002e-05,
      "loss": 1.3519,
      "mean_token_accuracy": 0.6748842839151621,
      "num_tokens": 12184261.0,
      "step": 910
    },
    {
      "entropy": 1.3415631584823131,
      "epoch": 0.8190062872085907,
      "grad_norm": 11.1875,
      "learning_rate": 1.8380000000000004e-05,
      "loss": 1.3134,
      "mean_token_accuracy": 0.6776627466082573,
      "num_tokens": 12311688.0,
      "step": 920
    },
    {
      "entropy": 1.3450417269021273,
      "epoch": 0.8279085294608579,
      "grad_norm": 9.6875,
      "learning_rate": 1.858e-05,
      "loss": 1.2681,
      "mean_token_accuracy": 0.6853212669491768,
      "num_tokens": 12453515.0,
      "step": 930
    },
    {
      "entropy": 1.3162209324538707,
      "epoch": 0.8368107717131252,
      "grad_norm": 9.3125,
      "learning_rate": 1.878e-05,
      "loss": 1.2935,
      "mean_token_accuracy": 0.6805886892601848,
      "num_tokens": 12598408.0,
      "step": 940
    },
    {
      "entropy": 1.3450543098151684,
      "epoch": 0.8457130139653926,
      "grad_norm": 11.125,
      "learning_rate": 1.898e-05,
      "loss": 1.3186,
      "mean_token_accuracy": 0.6790889304131269,
      "num_tokens": 12732830.0,
      "step": 950
    },
    {
      "entropy": 1.3630160838365555,
      "epoch": 0.8546152562176598,
      "grad_norm": 11.0,
      "learning_rate": 1.918e-05,
      "loss": 1.3251,
      "mean_token_accuracy": 0.6795368468388915,
      "num_tokens": 12869642.0,
      "step": 960
    },
    {
      "entropy": 1.2341018915176392,
      "epoch": 0.8635174984699271,
      "grad_norm": 9.4375,
      "learning_rate": 1.938e-05,
      "loss": 1.2047,
      "mean_token_accuracy": 0.697525155171752,
      "num_tokens": 13002795.0,
      "step": 970
    },
    {
      "entropy": 1.3432362783700227,
      "epoch": 0.8724197407221944,
      "grad_norm": 11.9375,
      "learning_rate": 1.9580000000000002e-05,
      "loss": 1.3242,
      "mean_token_accuracy": 0.682424726523459,
      "num_tokens": 13139555.0,
      "step": 980
    },
    {
      "entropy": 1.3661971539258957,
      "epoch": 0.8813219829744617,
      "grad_norm": 11.0625,
      "learning_rate": 1.978e-05,
      "loss": 1.3181,
      "mean_token_accuracy": 0.678005202487111,
      "num_tokens": 13261742.0,
      "step": 990
    },
    {
      "entropy": 1.33810890391469,
      "epoch": 0.890224225226729,
      "grad_norm": 9.3125,
      "learning_rate": 1.9980000000000002e-05,
      "loss": 1.2809,
      "mean_token_accuracy": 0.6845312876626849,
      "num_tokens": 13396545.0,
      "step": 1000
    },
    {
      "epoch": 0.890224225226729,
      "eval_coding_entropy": 1.8324014544487,
      "eval_coding_loss": 2.2719883918762207,
      "eval_coding_mean_token_accuracy": 0.506569258928299,
      "eval_coding_num_tokens": 13396545.0,
      "eval_coding_runtime": 80.808,
      "eval_coding_samples_per_second": 6.188,
      "eval_coding_steps_per_second": 3.094,
      "step": 1000
    },
    {
      "epoch": 0.890224225226729,
      "eval_chemistry_entropy": 1.2783789792060851,
      "eval_chemistry_loss": 1.3424817323684692,
      "eval_chemistry_mean_token_accuracy": 0.6828865656852722,
      "eval_chemistry_num_tokens": 13396545.0,
      "eval_chemistry_runtime": 43.9749,
      "eval_chemistry_samples_per_second": 11.37,
      "eval_chemistry_steps_per_second": 5.685,
      "step": 1000
    }
  ],
  "logging_steps": 10,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 8.606016510296003e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}