{
  "best_global_step": 300,
  "best_metric": 0.16315852105617523,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 50,
  "global_step": 338,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 3.8512803614139557,
      "epoch": 0.002962962962962963,
      "grad_norm": 30.730857849121094,
      "learning_rate": 0.0,
      "loss": 2.569148302078247,
      "mean_token_accuracy": 0.6431840881705284,
      "num_tokens": 16384.0,
      "step": 1
    },
    {
      "entropy": 4.0684830248355865,
      "epoch": 0.005925925925925926,
      "grad_norm": 24.129222869873047,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 2.6050946712493896,
      "mean_token_accuracy": 0.6232862882316113,
      "num_tokens": 32768.0,
      "step": 2
    },
    {
      "entropy": 3.7736930549144745,
      "epoch": 0.008888888888888889,
      "grad_norm": 26.998125076293945,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.283066987991333,
      "mean_token_accuracy": 0.6235363148152828,
      "num_tokens": 49152.0,
      "step": 3
    },
    {
      "entropy": 3.407482296228409,
      "epoch": 0.011851851851851851,
      "grad_norm": 30.29498863220215,
      "learning_rate": 6e-06,
      "loss": 2.198716878890991,
      "mean_token_accuracy": 0.6538878083229065,
      "num_tokens": 65536.0,
      "step": 4
    },
    {
      "entropy": 3.503322556614876,
      "epoch": 0.014814814814814815,
      "grad_norm": 30.397808074951172,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.8544838428497314,
      "mean_token_accuracy": 0.7068968862295151,
      "num_tokens": 81920.0,
      "step": 5
    },
    {
      "entropy": 3.8109894394874573,
      "epoch": 0.017777777777777778,
      "grad_norm": 19.67120933532715,
      "learning_rate": 1e-05,
      "loss": 2.1221282482147217,
      "mean_token_accuracy": 0.6750926822423935,
      "num_tokens": 98304.0,
      "step": 6
    },
    {
      "entropy": 3.518632858991623,
      "epoch": 0.02074074074074074,
      "grad_norm": 12.859330177307129,
      "learning_rate": 9.96996996996997e-06,
      "loss": 1.4550659656524658,
      "mean_token_accuracy": 0.7407641783356667,
      "num_tokens": 114688.0,
      "step": 7
    },
    {
      "entropy": 3.280194103717804,
      "epoch": 0.023703703703703703,
      "grad_norm": 9.05553913116455,
      "learning_rate": 9.93993993993994e-06,
      "loss": 0.9593539237976074,
      "mean_token_accuracy": 0.8175147324800491,
      "num_tokens": 131072.0,
      "step": 8
    },
    {
      "entropy": 3.20133575797081,
      "epoch": 0.02666666666666667,
      "grad_norm": 6.0671844482421875,
      "learning_rate": 9.90990990990991e-06,
      "loss": 0.9521955251693726,
      "mean_token_accuracy": 0.7981280162930489,
      "num_tokens": 147456.0,
      "step": 9
    },
    {
      "entropy": 3.4409832060337067,
      "epoch": 0.02962962962962963,
      "grad_norm": 5.146065711975098,
      "learning_rate": 9.879879879879881e-06,
      "loss": 0.8910360932350159,
      "mean_token_accuracy": 0.813666582107544,
      "num_tokens": 163840.0,
      "step": 10
    },
    {
      "entropy": 2.9835513681173325,
      "epoch": 0.03259259259259259,
      "grad_norm": 5.887986660003662,
      "learning_rate": 9.849849849849851e-06,
      "loss": 1.1130229234695435,
      "mean_token_accuracy": 0.8104487508535385,
      "num_tokens": 180224.0,
      "step": 11
    },
    {
      "entropy": 3.2429793179035187,
      "epoch": 0.035555555555555556,
      "grad_norm": 4.49108362197876,
      "learning_rate": 9.81981981981982e-06,
      "loss": 1.0250123739242554,
      "mean_token_accuracy": 0.7772897556424141,
      "num_tokens": 196608.0,
      "step": 12
    },
    {
      "entropy": 3.020932376384735,
      "epoch": 0.03851851851851852,
      "grad_norm": 2.5949156284332275,
      "learning_rate": 9.78978978978979e-06,
      "loss": 0.4803178310394287,
      "mean_token_accuracy": 0.8894667029380798,
      "num_tokens": 212992.0,
      "step": 13
    },
    {
      "entropy": 3.337664783000946,
      "epoch": 0.04148148148148148,
      "grad_norm": 3.608029842376709,
      "learning_rate": 9.75975975975976e-06,
      "loss": 0.8188823461532593,
      "mean_token_accuracy": 0.837014727294445,
      "num_tokens": 229376.0,
      "step": 14
    },
    {
      "entropy": 3.0280968248844147,
      "epoch": 0.044444444444444446,
      "grad_norm": 3.1399412155151367,
      "learning_rate": 9.729729729729732e-06,
      "loss": 0.5812732577323914,
      "mean_token_accuracy": 0.865282416343689,
      "num_tokens": 245760.0,
      "step": 15
    },
    {
      "entropy": 3.096962332725525,
      "epoch": 0.047407407407407405,
      "grad_norm": 3.0975067615509033,
      "learning_rate": 9.699699699699701e-06,
      "loss": 0.6786628365516663,
      "mean_token_accuracy": 0.8422383815050125,
      "num_tokens": 262144.0,
      "step": 16
    },
    {
      "entropy": 2.9266377389431,
      "epoch": 0.05037037037037037,
      "grad_norm": 4.297585964202881,
      "learning_rate": 9.669669669669671e-06,
      "loss": 0.854254424571991,
      "mean_token_accuracy": 0.8154165521264076,
      "num_tokens": 278528.0,
      "step": 17
    },
    {
      "entropy": 3.087370753288269,
      "epoch": 0.05333333333333334,
      "grad_norm": 2.101407766342163,
      "learning_rate": 9.63963963963964e-06,
      "loss": 0.3959370255470276,
      "mean_token_accuracy": 0.8854451179504395,
      "num_tokens": 294912.0,
      "step": 18
    },
    {
      "entropy": 3.048688620328903,
      "epoch": 0.056296296296296296,
      "grad_norm": 2.8039212226867676,
      "learning_rate": 9.60960960960961e-06,
      "loss": 0.5604370832443237,
      "mean_token_accuracy": 0.8462841138243675,
      "num_tokens": 311296.0,
      "step": 19
    },
    {
      "entropy": 2.923619657754898,
      "epoch": 0.05925925925925926,
      "grad_norm": 2.9450409412384033,
      "learning_rate": 9.57957957957958e-06,
      "loss": 0.6463603973388672,
      "mean_token_accuracy": 0.844881497323513,
      "num_tokens": 327680.0,
      "step": 20
    },
    {
      "entropy": 2.6990766674280167,
      "epoch": 0.06222222222222222,
      "grad_norm": 1.9361509084701538,
      "learning_rate": 9.54954954954955e-06,
      "loss": 0.35202786326408386,
      "mean_token_accuracy": 0.9147720187902451,
      "num_tokens": 344064.0,
      "step": 21
    },
    {
      "entropy": 3.31175696849823,
      "epoch": 0.06518518518518518,
      "grad_norm": 2.9889137744903564,
      "learning_rate": 9.51951951951952e-06,
      "loss": 0.6314672231674194,
      "mean_token_accuracy": 0.8451128154993057,
      "num_tokens": 360448.0,
      "step": 22
    },
    {
      "entropy": 3.0947949588298798,
      "epoch": 0.06814814814814815,
      "grad_norm": 2.546428680419922,
      "learning_rate": 9.489489489489491e-06,
      "loss": 0.6163879632949829,
      "mean_token_accuracy": 0.8414732441306114,
      "num_tokens": 376832.0,
      "step": 23
    },
    {
      "entropy": 3.0936408191919327,
      "epoch": 0.07111111111111111,
      "grad_norm": 2.3489434719085693,
      "learning_rate": 9.45945945945946e-06,
      "loss": 0.58915114402771,
      "mean_token_accuracy": 0.8334908187389374,
      "num_tokens": 393216.0,
      "step": 24
    },
    {
      "entropy": 2.962358444929123,
      "epoch": 0.07407407407407407,
      "grad_norm": 1.507115364074707,
      "learning_rate": 9.42942942942943e-06,
      "loss": 0.34845682978630066,
      "mean_token_accuracy": 0.9165022820234299,
      "num_tokens": 409600.0,
      "step": 25
    },
    {
      "entropy": 2.9361921548843384,
      "epoch": 0.07703703703703704,
      "grad_norm": 1.9893490076065063,
      "learning_rate": 9.3993993993994e-06,
      "loss": 0.468715637922287,
      "mean_token_accuracy": 0.892615795135498,
      "num_tokens": 425984.0,
      "step": 26
    },
    {
      "entropy": 3.4704517126083374,
      "epoch": 0.08,
      "grad_norm": 2.070949077606201,
      "learning_rate": 9.36936936936937e-06,
      "loss": 0.5496008396148682,
      "mean_token_accuracy": 0.888049341738224,
      "num_tokens": 442368.0,
      "step": 27
    },
    {
      "entropy": 3.0023911595344543,
      "epoch": 0.08296296296296296,
      "grad_norm": 1.8988116979599,
      "learning_rate": 9.339339339339341e-06,
      "loss": 0.5505701303482056,
      "mean_token_accuracy": 0.8772491812705994,
      "num_tokens": 458752.0,
      "step": 28
    },
    {
      "entropy": 2.8431918621063232,
      "epoch": 0.08592592592592592,
      "grad_norm": 1.559670329093933,
      "learning_rate": 9.30930930930931e-06,
      "loss": 0.4001576006412506,
      "mean_token_accuracy": 0.8993216007947922,
      "num_tokens": 475136.0,
      "step": 29
    },
    {
      "entropy": 2.6845883429050446,
      "epoch": 0.08888888888888889,
      "grad_norm": 1.3400381803512573,
      "learning_rate": 9.27927927927928e-06,
      "loss": 0.31123340129852295,
      "mean_token_accuracy": 0.9135517030954361,
      "num_tokens": 491520.0,
      "step": 30
    },
    {
      "entropy": 2.7273563742637634,
      "epoch": 0.09185185185185185,
      "grad_norm": 1.3198124170303345,
      "learning_rate": 9.24924924924925e-06,
      "loss": 0.2704215943813324,
      "mean_token_accuracy": 0.9170433357357979,
      "num_tokens": 507904.0,
      "step": 31
    },
    {
      "entropy": 2.3541511446237564,
      "epoch": 0.09481481481481481,
      "grad_norm": 1.110152244567871,
      "learning_rate": 9.21921921921922e-06,
      "loss": 0.19951550662517548,
      "mean_token_accuracy": 0.9389949440956116,
      "num_tokens": 524288.0,
      "step": 32
    },
    {
      "entropy": 2.436706930398941,
      "epoch": 0.09777777777777778,
      "grad_norm": 1.0588327646255493,
      "learning_rate": 9.189189189189191e-06,
      "loss": 0.23761028051376343,
      "mean_token_accuracy": 0.9429646283388138,
      "num_tokens": 540672.0,
      "step": 33
    },
    {
      "entropy": 2.6311395913362503,
      "epoch": 0.10074074074074074,
      "grad_norm": 1.230381965637207,
      "learning_rate": 9.15915915915916e-06,
      "loss": 0.26695674657821655,
      "mean_token_accuracy": 0.913018099963665,
      "num_tokens": 557056.0,
      "step": 34
    },
    {
      "entropy": 2.6350721567869186,
      "epoch": 0.1037037037037037,
      "grad_norm": 1.3151986598968506,
      "learning_rate": 9.129129129129129e-06,
      "loss": 0.3762889504432678,
      "mean_token_accuracy": 0.9002421647310257,
      "num_tokens": 573440.0,
      "step": 35
    },
    {
      "entropy": 2.707155853509903,
      "epoch": 0.10666666666666667,
      "grad_norm": 1.3434219360351562,
      "learning_rate": 9.0990990990991e-06,
      "loss": 0.28340280055999756,
      "mean_token_accuracy": 0.9321780577301979,
      "num_tokens": 589824.0,
      "step": 36
    },
    {
      "entropy": 2.6908050775527954,
      "epoch": 0.10962962962962963,
      "grad_norm": 1.3811731338500977,
      "learning_rate": 9.06906906906907e-06,
      "loss": 0.3659118413925171,
      "mean_token_accuracy": 0.902461551129818,
      "num_tokens": 606208.0,
      "step": 37
    },
    {
      "entropy": 2.597966179251671,
      "epoch": 0.11259259259259259,
      "grad_norm": 1.218999981880188,
      "learning_rate": 9.03903903903904e-06,
      "loss": 0.33045729994773865,
      "mean_token_accuracy": 0.9161138013005257,
      "num_tokens": 622592.0,
      "step": 38
    },
    {
      "entropy": 2.3637285828590393,
      "epoch": 0.11555555555555555,
      "grad_norm": 1.121229887008667,
      "learning_rate": 9.00900900900901e-06,
      "loss": 0.22515107691287994,
      "mean_token_accuracy": 0.9431316033005714,
      "num_tokens": 638976.0,
      "step": 39
    },
    {
      "entropy": 2.365230195224285,
      "epoch": 0.11851851851851852,
      "grad_norm": 1.1362413167953491,
      "learning_rate": 8.97897897897898e-06,
      "loss": 0.23698697984218597,
      "mean_token_accuracy": 0.9379475712776184,
      "num_tokens": 655360.0,
      "step": 40
    },
    {
      "entropy": 2.5145474672317505,
      "epoch": 0.12148148148148148,
      "grad_norm": 1.3671321868896484,
      "learning_rate": 8.94894894894895e-06,
      "loss": 0.33862820267677307,
      "mean_token_accuracy": 0.9252145886421204,
      "num_tokens": 671744.0,
      "step": 41
    },
    {
      "entropy": 2.575432315468788,
      "epoch": 0.12444444444444444,
      "grad_norm": 1.561908483505249,
      "learning_rate": 8.91891891891892e-06,
      "loss": 0.40326887369155884,
      "mean_token_accuracy": 0.902219258248806,
      "num_tokens": 688128.0,
      "step": 42
    },
    {
      "entropy": 2.499177247285843,
      "epoch": 0.1274074074074074,
      "grad_norm": 1.0121257305145264,
      "learning_rate": 8.888888888888888e-06,
      "loss": 0.20011436939239502,
      "mean_token_accuracy": 0.9460029155015945,
      "num_tokens": 704512.0,
      "step": 43
    },
    {
      "entropy": 2.2834786027669907,
      "epoch": 0.13037037037037036,
      "grad_norm": 1.0577738285064697,
      "learning_rate": 8.85885885885886e-06,
      "loss": 0.20965611934661865,
      "mean_token_accuracy": 0.9443583115935326,
      "num_tokens": 720896.0,
      "step": 44
    },
    {
      "entropy": 2.4016591608524323,
      "epoch": 0.13333333333333333,
      "grad_norm": 1.517271637916565,
      "learning_rate": 8.82882882882883e-06,
      "loss": 0.3337487578392029,
      "mean_token_accuracy": 0.9169893488287926,
      "num_tokens": 737280.0,
      "step": 45
    },
    {
      "entropy": 2.703525424003601,
      "epoch": 0.1362962962962963,
      "grad_norm": 1.4941990375518799,
      "learning_rate": 8.798798798798799e-06,
      "loss": 0.3472306728363037,
      "mean_token_accuracy": 0.9080394953489304,
      "num_tokens": 753664.0,
      "step": 46
    },
    {
      "entropy": 2.75506529211998,
      "epoch": 0.13925925925925925,
      "grad_norm": 1.6088497638702393,
      "learning_rate": 8.768768768768769e-06,
      "loss": 0.43476995825767517,
      "mean_token_accuracy": 0.8839521333575249,
      "num_tokens": 770048.0,
      "step": 47
    },
    {
      "entropy": 2.332780599594116,
      "epoch": 0.14222222222222222,
      "grad_norm": 0.9841637015342712,
      "learning_rate": 8.738738738738739e-06,
      "loss": 0.1702219545841217,
      "mean_token_accuracy": 0.9549055695533752,
      "num_tokens": 786432.0,
      "step": 48
    },
    {
      "entropy": 2.141946002840996,
      "epoch": 0.1451851851851852,
      "grad_norm": 1.1044976711273193,
      "learning_rate": 8.70870870870871e-06,
      "loss": 0.2461618334054947,
      "mean_token_accuracy": 0.9294649511575699,
      "num_tokens": 802816.0,
      "step": 49
    },
    {
      "entropy": 2.1164865493774414,
      "epoch": 0.14814814814814814,
      "grad_norm": 0.8976758122444153,
      "learning_rate": 8.67867867867868e-06,
      "loss": 0.21264278888702393,
      "mean_token_accuracy": 0.9377186372876167,
      "num_tokens": 819200.0,
      "step": 50
    },
    {
      "epoch": 0.14814814814814814,
      "eval_entropy": 2.2788925981521606,
      "eval_loss": 0.23260752856731415,
      "eval_mean_token_accuracy": 0.9388575092951457,
      "eval_num_tokens": 819200.0,
      "eval_runtime": 46.456,
      "eval_samples_per_second": 12.915,
      "eval_steps_per_second": 1.614,
      "step": 50
    },
    {
      "entropy": 2.3432202637195587,
      "epoch": 0.1511111111111111,
      "grad_norm": 1.2383582592010498,
      "learning_rate": 8.64864864864865e-06,
      "loss": 0.32435837388038635,
      "mean_token_accuracy": 0.9200343042612076,
      "num_tokens": 835584.0,
      "step": 51
    },
    {
      "entropy": 2.4892737865448,
      "epoch": 0.15407407407407409,
      "grad_norm": 1.1927871704101562,
      "learning_rate": 8.618618618618619e-06,
      "loss": 0.25952398777008057,
      "mean_token_accuracy": 0.9298947528004646,
      "num_tokens": 851968.0,
      "step": 52
    },
    {
      "entropy": 2.1942788660526276,
      "epoch": 0.15703703703703703,
      "grad_norm": 1.392685055732727,
      "learning_rate": 8.588588588588589e-06,
      "loss": 0.3469521403312683,
      "mean_token_accuracy": 0.9202883318066597,
      "num_tokens": 868352.0,
      "step": 53
    },
    {
      "entropy": 2.505737394094467,
      "epoch": 0.16,
      "grad_norm": 1.295953631401062,
      "learning_rate": 8.55855855855856e-06,
      "loss": 0.29852601885795593,
      "mean_token_accuracy": 0.9220015630126,
      "num_tokens": 884736.0,
      "step": 54
    },
    {
      "entropy": 2.115240439772606,
      "epoch": 0.16296296296296298,
      "grad_norm": 0.7726749777793884,
      "learning_rate": 8.52852852852853e-06,
      "loss": 0.15434573590755463,
      "mean_token_accuracy": 0.963710330426693,
      "num_tokens": 901120.0,
      "step": 55
    },
    {
      "entropy": 2.6289254426956177,
      "epoch": 0.16592592592592592,
      "grad_norm": 1.2951736450195312,
      "learning_rate": 8.4984984984985e-06,
      "loss": 0.26316919922828674,
      "mean_token_accuracy": 0.9279258102178574,
      "num_tokens": 917504.0,
      "step": 56
    },
    {
      "entropy": 2.3661544919013977,
      "epoch": 0.1688888888888889,
      "grad_norm": 1.0332659482955933,
      "learning_rate": 8.46846846846847e-06,
      "loss": 0.2150796353816986,
      "mean_token_accuracy": 0.943064346909523,
      "num_tokens": 933888.0,
      "step": 57
    },
    {
      "entropy": 2.29600927233696,
      "epoch": 0.17185185185185184,
      "grad_norm": 0.9310147762298584,
      "learning_rate": 8.438438438438439e-06,
      "loss": 0.18033596873283386,
      "mean_token_accuracy": 0.9520642012357712,
      "num_tokens": 950272.0,
      "step": 58
    },
    {
      "entropy": 2.40557961165905,
      "epoch": 0.1748148148148148,
      "grad_norm": 1.2708395719528198,
      "learning_rate": 8.408408408408409e-06,
      "loss": 0.31112906336784363,
      "mean_token_accuracy": 0.912487804889679,
      "num_tokens": 966656.0,
      "step": 59
    },
    {
      "entropy": 2.43192321062088,
      "epoch": 0.17777777777777778,
      "grad_norm": 1.1282551288604736,
      "learning_rate": 8.378378378378378e-06,
      "loss": 0.25797221064567566,
      "mean_token_accuracy": 0.9267487153410912,
      "num_tokens": 983040.0,
      "step": 60
    },
    {
      "entropy": 2.471033439040184,
      "epoch": 0.18074074074074073,
      "grad_norm": 0.9804869890213013,
      "learning_rate": 8.348348348348348e-06,
      "loss": 0.21752172708511353,
      "mean_token_accuracy": 0.9400404766201973,
      "num_tokens": 999424.0,
      "step": 61
    },
    {
      "entropy": 2.110298752784729,
      "epoch": 0.1837037037037037,
      "grad_norm": 0.9955158829689026,
      "learning_rate": 8.31831831831832e-06,
      "loss": 0.18774758279323578,
      "mean_token_accuracy": 0.9517924040555954,
      "num_tokens": 1015808.0,
      "step": 62
    },
    {
      "entropy": 2.416291266679764,
      "epoch": 0.18666666666666668,
      "grad_norm": 1.2231297492980957,
      "learning_rate": 8.288288288288289e-06,
      "loss": 0.30843546986579895,
      "mean_token_accuracy": 0.9213244393467903,
      "num_tokens": 1032192.0,
      "step": 63
    },
    {
      "entropy": 2.3274537920951843,
      "epoch": 0.18962962962962962,
      "grad_norm": 0.9842973947525024,
      "learning_rate": 8.258258258258259e-06,
      "loss": 0.22026892006397247,
      "mean_token_accuracy": 0.9371362924575806,
      "num_tokens": 1048576.0,
      "step": 64
    },
    {
      "entropy": 2.1601486206054688,
      "epoch": 0.1925925925925926,
      "grad_norm": 1.1512540578842163,
      "learning_rate": 8.228228228228229e-06,
      "loss": 0.26338401436805725,
      "mean_token_accuracy": 0.9269109815359116,
      "num_tokens": 1064960.0,
      "step": 65
    },
    {
      "entropy": 2.321208655834198,
      "epoch": 0.19555555555555557,
      "grad_norm": 1.1600278615951538,
      "learning_rate": 8.198198198198198e-06,
      "loss": 0.22572128474712372,
      "mean_token_accuracy": 0.9418924897909164,
      "num_tokens": 1081344.0,
      "step": 66
    },
    {
      "entropy": 2.2064977884292603,
      "epoch": 0.1985185185185185,
      "grad_norm": 1.249359130859375,
      "learning_rate": 8.16816816816817e-06,
      "loss": 0.3008494973182678,
      "mean_token_accuracy": 0.9249227941036224,
      "num_tokens": 1097728.0,
      "step": 67
    },
    {
      "entropy": 2.5402083545923233,
      "epoch": 0.20148148148148148,
      "grad_norm": 1.7166448831558228,
      "learning_rate": 8.13813813813814e-06,
      "loss": 0.36760830879211426,
      "mean_token_accuracy": 0.9052070081233978,
      "num_tokens": 1114112.0,
      "step": 68
    },
    {
      "entropy": 2.5308094322681427,
      "epoch": 0.20444444444444446,
      "grad_norm": 1.4214552640914917,
      "learning_rate": 8.108108108108109e-06,
      "loss": 0.27426525950431824,
      "mean_token_accuracy": 0.9390730857849121,
      "num_tokens": 1130496.0,
      "step": 69
    },
    {
      "entropy": 2.1990968585014343,
      "epoch": 0.2074074074074074,
      "grad_norm": 0.8868707418441772,
      "learning_rate": 8.078078078078079e-06,
      "loss": 0.18394234776496887,
      "mean_token_accuracy": 0.9577051475644112,
      "num_tokens": 1146880.0,
      "step": 70
    },
    {
      "entropy": 2.7524623572826385,
      "epoch": 0.21037037037037037,
      "grad_norm": 1.220381498336792,
      "learning_rate": 8.048048048048048e-06,
      "loss": 0.25256532430648804,
      "mean_token_accuracy": 0.9326800480484962,
      "num_tokens": 1163264.0,
      "step": 71
    },
    {
      "entropy": 2.6490699648857117,
      "epoch": 0.21333333333333335,
      "grad_norm": 1.284562110900879,
      "learning_rate": 8.018018018018018e-06,
      "loss": 0.27179843187332153,
      "mean_token_accuracy": 0.9425966218113899,
      "num_tokens": 1179648.0,
      "step": 72
    },
    {
      "entropy": 2.456472247838974,
      "epoch": 0.2162962962962963,
      "grad_norm": 0.9432771801948547,
      "learning_rate": 7.987987987987988e-06,
      "loss": 0.17216819524765015,
      "mean_token_accuracy": 0.949951246380806,
      "num_tokens": 1196032.0,
      "step": 73
    },
    {
      "entropy": 2.6553167700767517,
      "epoch": 0.21925925925925926,
      "grad_norm": 0.9341099262237549,
      "learning_rate": 7.95795795795796e-06,
      "loss": 0.1365150362253189,
      "mean_token_accuracy": 0.9531153365969658,
      "num_tokens": 1212416.0,
      "step": 74
    },
    {
      "entropy": 2.7954386472702026,
      "epoch": 0.2222222222222222,
      "grad_norm": 1.0198298692703247,
      "learning_rate": 7.927927927927929e-06,
      "loss": 0.18447062373161316,
      "mean_token_accuracy": 0.9517708420753479,
      "num_tokens": 1228800.0,
      "step": 75
    },
    {
      "entropy": 2.7095753848552704,
      "epoch": 0.22518518518518518,
      "grad_norm": 1.2833144664764404,
      "learning_rate": 7.897897897897899e-06,
      "loss": 0.21448825299739838,
      "mean_token_accuracy": 0.9327085986733437,
      "num_tokens": 1245184.0,
      "step": 76
    },
    {
      "entropy": 2.6067453622817993,
      "epoch": 0.22814814814814816,
      "grad_norm": 1.2703853845596313,
      "learning_rate": 7.867867867867868e-06,
      "loss": 0.30238083004951477,
      "mean_token_accuracy": 0.9226602166891098,
      "num_tokens": 1261568.0,
      "step": 77
    },
    {
      "entropy": 2.600490003824234,
      "epoch": 0.2311111111111111,
      "grad_norm": 1.1312607526779175,
      "learning_rate": 7.837837837837838e-06,
      "loss": 0.22982332110404968,
      "mean_token_accuracy": 0.9376447200775146,
      "num_tokens": 1277952.0,
      "step": 78
    },
    {
      "entropy": 2.4881619811058044,
      "epoch": 0.23407407407407407,
      "grad_norm": 0.871771514415741,
      "learning_rate": 7.807807807807808e-06,
      "loss": 0.16425110399723053,
      "mean_token_accuracy": 0.9514938965439796,
      "num_tokens": 1294336.0,
      "step": 79
    },
    {
      "entropy": 2.2865116894245148,
      "epoch": 0.23703703703703705,
      "grad_norm": 1.1134834289550781,
      "learning_rate": 7.77777777777778e-06,
      "loss": 0.21876588463783264,
      "mean_token_accuracy": 0.9456857740879059,
      "num_tokens": 1310720.0,
      "step": 80
    },
    {
      "entropy": 2.511552333831787,
      "epoch": 0.24,
      "grad_norm": 1.0024129152297974,
      "learning_rate": 7.747747747747749e-06,
      "loss": 0.22860540449619293,
      "mean_token_accuracy": 0.9404076114296913,
      "num_tokens": 1327104.0,
      "step": 81
    },
    {
      "entropy": 2.5314147919416428,
      "epoch": 0.24296296296296296,
      "grad_norm": 1.1556036472320557,
      "learning_rate": 7.717717717717719e-06,
      "loss": 0.18219460546970367,
      "mean_token_accuracy": 0.9528129249811172,
      "num_tokens": 1343488.0,
      "step": 82
    },
    {
      "entropy": 2.424291893839836,
      "epoch": 0.24592592592592594,
      "grad_norm": 0.8692384362220764,
      "learning_rate": 7.687687687687688e-06,
      "loss": 0.16242916882038116,
      "mean_token_accuracy": 0.9525845125317574,
      "num_tokens": 1359872.0,
      "step": 83
    },
    {
      "entropy": 2.6161986738443375,
      "epoch": 0.24888888888888888,
      "grad_norm": 1.1681318283081055,
      "learning_rate": 7.657657657657658e-06,
      "loss": 0.22786562144756317,
      "mean_token_accuracy": 0.9332282468676567,
      "num_tokens": 1376256.0,
      "step": 84
    },
    {
      "entropy": 2.547512024641037,
      "epoch": 0.2518518518518518,
      "grad_norm": 1.185215950012207,
      "learning_rate": 7.6276276276276285e-06,
      "loss": 0.23456929624080658,
      "mean_token_accuracy": 0.9403537809848785,
      "num_tokens": 1392640.0,
      "step": 85
    },
    {
      "entropy": 2.2024350464344025,
      "epoch": 0.2548148148148148,
      "grad_norm": 0.8941919803619385,
      "learning_rate": 7.597597597597598e-06,
      "loss": 0.20554354786872864,
      "mean_token_accuracy": 0.9462639912962914,
      "num_tokens": 1409024.0,
      "step": 86
    },
    {
      "entropy": 2.441995397210121,
      "epoch": 0.2577777777777778,
      "grad_norm": 1.4575777053833008,
      "learning_rate": 7.567567567567569e-06,
      "loss": 0.3014245927333832,
      "mean_token_accuracy": 0.9144093245267868,
      "num_tokens": 1425408.0,
      "step": 87
    },
    {
      "entropy": 2.480868488550186,
      "epoch": 0.2607407407407407,
      "grad_norm": 1.260637640953064,
      "learning_rate": 7.5375375375375385e-06,
      "loss": 0.27455228567123413,
      "mean_token_accuracy": 0.9261277168989182,
      "num_tokens": 1441792.0,
      "step": 88
    },
    {
      "entropy": 2.4851946234703064,
      "epoch": 0.2637037037037037,
      "grad_norm": 0.9694555401802063,
      "learning_rate": 7.507507507507507e-06,
      "loss": 0.17230632901191711,
      "mean_token_accuracy": 0.9571247696876526,
      "num_tokens": 1458176.0,
      "step": 89
    },
    {
      "entropy": 2.219466745853424,
      "epoch": 0.26666666666666666,
      "grad_norm": 1.0834273099899292,
      "learning_rate": 7.477477477477479e-06,
      "loss": 0.22051620483398438,
      "mean_token_accuracy": 0.9419185891747475,
      "num_tokens": 1474560.0,
      "step": 90
    },
    {
      "entropy": 2.3855587989091873,
      "epoch": 0.2696296296296296,
      "grad_norm": 1.016973853111267,
      "learning_rate": 7.447447447447448e-06,
      "loss": 0.2431911677122116,
      "mean_token_accuracy": 0.9292957335710526,
      "num_tokens": 1490944.0,
      "step": 91
    },
    {
      "entropy": 2.4104628562927246,
      "epoch": 0.2725925925925926,
      "grad_norm": 0.9278807044029236,
      "learning_rate": 7.417417417417418e-06,
      "loss": 0.17476391792297363,
      "mean_token_accuracy": 0.9608084857463837,
      "num_tokens": 1507328.0,
      "step": 92
    },
    {
      "entropy": 2.3218894600868225,
      "epoch": 0.27555555555555555,
      "grad_norm": 0.8093988299369812,
      "learning_rate": 7.387387387387388e-06,
      "loss": 0.13575376570224762,
      "mean_token_accuracy": 0.9647113382816315,
      "num_tokens": 1523712.0,
      "step": 93
    },
    {
      "entropy": 2.7123343348503113,
      "epoch": 0.2785185185185185,
      "grad_norm": 1.3076725006103516,
      "learning_rate": 7.3573573573573575e-06,
      "loss": 0.33463215827941895,
      "mean_token_accuracy": 0.9083034172654152,
      "num_tokens": 1540096.0,
      "step": 94
    },
    {
      "entropy": 2.6522365659475327,
      "epoch": 0.2814814814814815,
      "grad_norm": 1.0030099153518677,
      "learning_rate": 7.327327327327328e-06,
      "loss": 0.222689688205719,
      "mean_token_accuracy": 0.9446021169424057,
      "num_tokens": 1556480.0,
      "step": 95
    },
    {
      "entropy": 2.4811577796936035,
      "epoch": 0.28444444444444444,
      "grad_norm": 1.2005258798599243,
      "learning_rate": 7.297297297297298e-06,
      "loss": 0.2586423456668854,
      "mean_token_accuracy": 0.9401304647326469,
      "num_tokens": 1572864.0,
      "step": 96
    },
    {
      "entropy": 2.6387641429901123,
      "epoch": 0.2874074074074074,
      "grad_norm": 1.1932110786437988,
      "learning_rate": 7.267267267267268e-06,
      "loss": 0.2620566189289093,
      "mean_token_accuracy": 0.9247189462184906,
      "num_tokens": 1589248.0,
      "step": 97
    },
    {
      "entropy": 2.5274933576583862,
      "epoch": 0.2903703703703704,
      "grad_norm": 1.050783395767212,
      "learning_rate": 7.237237237237238e-06,
      "loss": 0.21884086728096008,
      "mean_token_accuracy": 0.944552093744278,
      "num_tokens": 1605632.0,
      "step": 98
    },
    {
      "entropy": 2.489010214805603,
      "epoch": 0.29333333333333333,
      "grad_norm": 1.1955422163009644,
      "learning_rate": 7.207207207207208e-06,
      "loss": 0.33184245228767395,
      "mean_token_accuracy": 0.9049209803342819,
      "num_tokens": 1622016.0,
      "step": 99
    },
    {
      "entropy": 2.314561605453491,
      "epoch": 0.2962962962962963,
      "grad_norm": 0.956768810749054,
      "learning_rate": 7.177177177177178e-06,
      "loss": 0.22311189770698547,
      "mean_token_accuracy": 0.9367908239364624,
      "num_tokens": 1638400.0,
      "step": 100
    },
    {
      "epoch": 0.2962962962962963,
      "eval_entropy": 2.517024199167887,
      "eval_loss": 0.1889665275812149,
      "eval_mean_token_accuracy": 0.9483304850260417,
      "eval_num_tokens": 1638400.0,
      "eval_runtime": 46.0152,
      "eval_samples_per_second": 13.039,
      "eval_steps_per_second": 1.63,
      "step": 100
    },
    {
      "entropy": 2.688704162836075,
      "epoch": 0.2992592592592593,
      "grad_norm": 0.9534380435943604,
      "learning_rate": 7.147147147147148e-06,
      "loss": 0.14701221883296967,
      "mean_token_accuracy": 0.961182214319706,
      "num_tokens": 1654784.0,
      "step": 101
    },
    {
      "entropy": 2.872332900762558,
      "epoch": 0.3022222222222222,
      "grad_norm": 1.1050176620483398,
      "learning_rate": 7.117117117117117e-06,
      "loss": 0.21337822079658508,
      "mean_token_accuracy": 0.9484675973653793,
      "num_tokens": 1671168.0,
      "step": 102
    },
    {
      "entropy": 2.5930943489074707,
      "epoch": 0.30518518518518517,
      "grad_norm": 1.3288185596466064,
      "learning_rate": 7.087087087087087e-06,
      "loss": 0.32390618324279785,
      "mean_token_accuracy": 0.918847531080246,
      "num_tokens": 1687552.0,
      "step": 103
    },
    {
      "entropy": 2.6078614592552185,
      "epoch": 0.30814814814814817,
      "grad_norm": 0.9944614768028259,
      "learning_rate": 7.057057057057057e-06,
      "loss": 0.19208504259586334,
      "mean_token_accuracy": 0.9433681443333626,
      "num_tokens": 1703936.0,
      "step": 104
    },
    {
      "entropy": 2.6580496728420258,
      "epoch": 0.3111111111111111,
      "grad_norm": 1.0881201028823853,
      "learning_rate": 7.027027027027028e-06,
      "loss": 0.28542912006378174,
      "mean_token_accuracy": 0.9226915240287781,
      "num_tokens": 1720320.0,
      "step": 105
    },
    {
      "entropy": 2.476440891623497,
      "epoch": 0.31407407407407406,
      "grad_norm": 0.7635496258735657,
      "learning_rate": 6.996996996996997e-06,
      "loss": 0.15295691788196564,
      "mean_token_accuracy": 0.9588403776288033,
      "num_tokens": 1736704.0,
      "step": 106
    },
    {
      "entropy": 2.9398592710494995,
      "epoch": 0.31703703703703706,
      "grad_norm": 1.1687581539154053,
      "learning_rate": 6.966966966966967e-06,
      "loss": 0.18745118379592896,
      "mean_token_accuracy": 0.9476362094283104,
      "num_tokens": 1753088.0,
      "step": 107
    },
    {
      "entropy": 2.383945018053055,
      "epoch": 0.32,
      "grad_norm": 0.9587693810462952,
      "learning_rate": 6.936936936936938e-06,
      "loss": 0.20505432784557343,
      "mean_token_accuracy": 0.9416079446673393,
      "num_tokens": 1769472.0,
      "step": 108
    },
    {
      "entropy": 2.54445344209671,
      "epoch": 0.32296296296296295,
      "grad_norm": 1.2000741958618164,
      "learning_rate": 6.906906906906907e-06,
      "loss": 0.24609866738319397,
      "mean_token_accuracy": 0.9358780607581139,
      "num_tokens": 1785856.0,
      "step": 109
    },
    {
      "entropy": 2.5802957117557526,
      "epoch": 0.32592592592592595,
      "grad_norm": 1.0026284456253052,
      "learning_rate": 6.876876876876878e-06,
      "loss": 0.21341544389724731,
      "mean_token_accuracy": 0.9375114217400551,
      "num_tokens": 1802240.0,
      "step": 110
    },
    {
      "entropy": 2.755191922187805,
      "epoch": 0.3288888888888889,
      "grad_norm": 1.2425813674926758,
      "learning_rate": 6.846846846846848e-06,
      "loss": 0.2857995331287384,
      "mean_token_accuracy": 0.9357155188918114,
      "num_tokens": 1818624.0,
      "step": 111
    },
    {
      "entropy": 2.5264368653297424,
      "epoch": 0.33185185185185184,
      "grad_norm": 1.0797715187072754,
      "learning_rate": 6.816816816816817e-06,
      "loss": 0.22755734622478485,
      "mean_token_accuracy": 0.9356759935617447,
      "num_tokens": 1835008.0,
      "step": 112
    },
    {
      "entropy": 2.3567262440919876,
      "epoch": 0.3348148148148148,
      "grad_norm": 1.0413719415664673,
      "learning_rate": 6.786786786786788e-06,
      "loss": 0.18666744232177734,
      "mean_token_accuracy": 0.9502386301755905,
      "num_tokens": 1851392.0,
      "step": 113
    },
    {
      "entropy": 2.660577952861786,
      "epoch": 0.3377777777777778,
      "grad_norm": 1.207374095916748,
      "learning_rate": 6.7567567567567575e-06,
      "loss": 0.2654038071632385,
      "mean_token_accuracy": 0.9348957762122154,
      "num_tokens": 1867776.0,
      "step": 114
    },
    {
      "entropy": 2.5813668370246887,
      "epoch": 0.34074074074074073,
      "grad_norm": 1.0566004514694214,
      "learning_rate": 6.726726726726728e-06,
      "loss": 0.19535279273986816,
      "mean_token_accuracy": 0.9428145363926888,
      "num_tokens": 1884160.0,
      "step": 115
    },
    {
      "entropy": 2.8093952238559723,
      "epoch": 0.3437037037037037,
      "grad_norm": 1.7544386386871338,
      "learning_rate": 6.696696696696697e-06,
      "loss": 0.3484290540218353,
      "mean_token_accuracy": 0.9055285677313805,
      "num_tokens": 1900544.0,
      "step": 116
    },
    {
      "entropy": 2.5164105892181396,
      "epoch": 0.3466666666666667,
      "grad_norm": 1.038743495941162,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.1875537633895874,
      "mean_token_accuracy": 0.948108434677124,
      "num_tokens": 1916928.0,
      "step": 117
    },
    {
      "entropy": 2.763442724943161,
      "epoch": 0.3496296296296296,
      "grad_norm": 1.1303430795669556,
      "learning_rate": 6.636636636636637e-06,
      "loss": 0.2767588496208191,
      "mean_token_accuracy": 0.9210386201739311,
      "num_tokens": 1933312.0,
      "step": 118
    },
    {
      "entropy": 2.562091127038002,
      "epoch": 0.35259259259259257,
      "grad_norm": 1.188616394996643,
      "learning_rate": 6.606606606606607e-06,
      "loss": 0.2767156660556793,
      "mean_token_accuracy": 0.9297419562935829,
      "num_tokens": 1949696.0,
      "step": 119
    },
    {
      "entropy": 2.4443581104278564,
      "epoch": 0.35555555555555557,
      "grad_norm": 0.9797606468200684,
      "learning_rate": 6.5765765765765775e-06,
      "loss": 0.1993044912815094,
      "mean_token_accuracy": 0.9437215775251389,
      "num_tokens": 1966080.0,
      "step": 120
    },
    {
      "entropy": 2.517397552728653,
      "epoch": 0.3585185185185185,
      "grad_norm": 0.9872597455978394,
      "learning_rate": 6.546546546546547e-06,
      "loss": 0.22233453392982483,
      "mean_token_accuracy": 0.9273230358958244,
      "num_tokens": 1982464.0,
      "step": 121
    },
    {
      "entropy": 2.591138929128647,
      "epoch": 0.36148148148148146,
      "grad_norm": 0.9894529581069946,
      "learning_rate": 6.516516516516517e-06,
      "loss": 0.19580335915088654,
      "mean_token_accuracy": 0.9563343375921249,
      "num_tokens": 1998848.0,
      "step": 122
    },
    {
      "entropy": 2.7103771567344666,
      "epoch": 0.36444444444444446,
      "grad_norm": 1.0713337659835815,
      "learning_rate": 6.486486486486487e-06,
      "loss": 0.1986352503299713,
      "mean_token_accuracy": 0.9485766142606735,
      "num_tokens": 2015232.0,
      "step": 123
    },
    {
      "entropy": 3.0009114146232605,
      "epoch": 0.3674074074074074,
      "grad_norm": 1.3179116249084473,
      "learning_rate": 6.456456456456457e-06,
      "loss": 0.3161464333534241,
      "mean_token_accuracy": 0.9274860098958015,
      "num_tokens": 2031616.0,
      "step": 124
    },
    {
      "entropy": 2.6024164855480194,
      "epoch": 0.37037037037037035,
      "grad_norm": 1.175925612449646,
      "learning_rate": 6.426426426426427e-06,
      "loss": 0.21718275547027588,
      "mean_token_accuracy": 0.939902774989605,
      "num_tokens": 2048000.0,
      "step": 125
    },
    {
      "entropy": 2.8810903429985046,
      "epoch": 0.37333333333333335,
      "grad_norm": 1.25202476978302,
      "learning_rate": 6.396396396396397e-06,
      "loss": 0.26412585377693176,
      "mean_token_accuracy": 0.9326214715838432,
      "num_tokens": 2064384.0,
      "step": 126
    },
    {
      "entropy": 2.5949351489543915,
      "epoch": 0.3762962962962963,
      "grad_norm": 0.94535893201828,
      "learning_rate": 6.366366366366366e-06,
      "loss": 0.1769305318593979,
      "mean_token_accuracy": 0.9426699578762054,
      "num_tokens": 2080768.0,
      "step": 127
    },
    {
      "entropy": 2.4384057819843292,
      "epoch": 0.37925925925925924,
      "grad_norm": 0.8953754305839539,
      "learning_rate": 6.336336336336338e-06,
      "loss": 0.15506362915039062,
      "mean_token_accuracy": 0.9541881009936333,
      "num_tokens": 2097152.0,
      "step": 128
    },
    {
      "entropy": 2.5724623650312424,
      "epoch": 0.38222222222222224,
      "grad_norm": 1.0130114555358887,
      "learning_rate": 6.3063063063063065e-06,
      "loss": 0.16999563574790955,
      "mean_token_accuracy": 0.9517735466361046,
      "num_tokens": 2113536.0,
      "step": 129
    },
    {
      "entropy": 2.7234296202659607,
      "epoch": 0.3851851851851852,
      "grad_norm": 1.0108418464660645,
      "learning_rate": 6.276276276276276e-06,
      "loss": 0.15174464881420135,
      "mean_token_accuracy": 0.9553050771355629,
      "num_tokens": 2129920.0,
      "step": 130
    },
    {
      "entropy": 2.599887192249298,
      "epoch": 0.38814814814814813,
      "grad_norm": 1.166524052619934,
      "learning_rate": 6.246246246246247e-06,
      "loss": 0.23332904279232025,
      "mean_token_accuracy": 0.9338735938072205,
      "num_tokens": 2146304.0,
      "step": 131
    },
    {
      "entropy": 2.3881106972694397,
      "epoch": 0.39111111111111113,
      "grad_norm": 0.9948994517326355,
      "learning_rate": 6.2162162162162164e-06,
      "loss": 0.18515516817569733,
      "mean_token_accuracy": 0.9488838911056519,
      "num_tokens": 2162688.0,
      "step": 132
    },
    {
      "entropy": 2.7394265830516815,
      "epoch": 0.3940740740740741,
      "grad_norm": 1.248687505722046,
      "learning_rate": 6.186186186186187e-06,
      "loss": 0.2647465169429779,
      "mean_token_accuracy": 0.9202074483036995,
      "num_tokens": 2179072.0,
      "step": 133
    },
    {
      "entropy": 2.674542635679245,
      "epoch": 0.397037037037037,
      "grad_norm": 1.3193742036819458,
      "learning_rate": 6.156156156156157e-06,
      "loss": 0.27039027214050293,
      "mean_token_accuracy": 0.9321836233139038,
      "num_tokens": 2195456.0,
      "step": 134
    },
    {
      "entropy": 2.585142433643341,
      "epoch": 0.4,
      "grad_norm": 1.0456122159957886,
      "learning_rate": 6.126126126126126e-06,
      "loss": 0.19117112457752228,
      "mean_token_accuracy": 0.9458754733204842,
      "num_tokens": 2211840.0,
      "step": 135
    },
    {
      "entropy": 2.6550826132297516,
      "epoch": 0.40296296296296297,
      "grad_norm": 1.0033628940582275,
      "learning_rate": 6.096096096096097e-06,
      "loss": 0.16141018271446228,
      "mean_token_accuracy": 0.9520677551627159,
      "num_tokens": 2228224.0,
      "step": 136
    },
    {
      "entropy": 2.407243072986603,
      "epoch": 0.4059259259259259,
      "grad_norm": 0.823492169380188,
      "learning_rate": 6.066066066066067e-06,
      "loss": 0.11283991485834122,
      "mean_token_accuracy": 0.9654894545674324,
      "num_tokens": 2244608.0,
      "step": 137
    },
    {
      "entropy": 2.6922158002853394,
      "epoch": 0.4088888888888889,
      "grad_norm": 1.164628028869629,
      "learning_rate": 6.036036036036037e-06,
      "loss": 0.21780996024608612,
      "mean_token_accuracy": 0.9432759061455727,
      "num_tokens": 2260992.0,
      "step": 138
    },
    {
      "entropy": 2.5815781950950623,
      "epoch": 0.41185185185185186,
      "grad_norm": 1.1495994329452515,
      "learning_rate": 6.006006006006007e-06,
      "loss": 0.23245574533939362,
      "mean_token_accuracy": 0.9351079985499382,
      "num_tokens": 2277376.0,
      "step": 139
    },
    {
      "entropy": 2.7152522206306458,
      "epoch": 0.4148148148148148,
      "grad_norm": 1.4183478355407715,
      "learning_rate": 5.975975975975976e-06,
      "loss": 0.3127621114253998,
      "mean_token_accuracy": 0.9116102382540703,
      "num_tokens": 2293760.0,
      "step": 140
    },
    {
      "entropy": 2.483951151371002,
      "epoch": 0.4177777777777778,
      "grad_norm": 1.069791316986084,
      "learning_rate": 5.945945945945947e-06,
      "loss": 0.2473391443490982,
      "mean_token_accuracy": 0.9309801012277603,
      "num_tokens": 2310144.0,
      "step": 141
    },
    {
      "entropy": 2.5401608645915985,
      "epoch": 0.42074074074074075,
      "grad_norm": 0.9594464302062988,
      "learning_rate": 5.915915915915916e-06,
      "loss": 0.15854710340499878,
      "mean_token_accuracy": 0.957264818251133,
      "num_tokens": 2326528.0,
      "step": 142
    },
    {
      "entropy": 2.564148470759392,
      "epoch": 0.4237037037037037,
      "grad_norm": 1.3057323694229126,
      "learning_rate": 5.885885885885886e-06,
      "loss": 0.29683682322502136,
      "mean_token_accuracy": 0.9266512393951416,
      "num_tokens": 2342912.0,
      "step": 143
    },
    {
      "entropy": 2.4974671006202698,
      "epoch": 0.4266666666666667,
      "grad_norm": 1.146114468574524,
      "learning_rate": 5.855855855855856e-06,
      "loss": 0.21560530364513397,
      "mean_token_accuracy": 0.9404177963733673,
      "num_tokens": 2359296.0,
      "step": 144
    },
    {
      "entropy": 2.683354616165161,
      "epoch": 0.42962962962962964,
      "grad_norm": 0.9514636993408203,
      "learning_rate": 5.825825825825826e-06,
      "loss": 0.14973832666873932,
      "mean_token_accuracy": 0.9553343281149864,
      "num_tokens": 2375680.0,
      "step": 145
    },
    {
      "entropy": 2.158581957221031,
      "epoch": 0.4325925925925926,
      "grad_norm": 0.7359983325004578,
      "learning_rate": 5.7957957957957965e-06,
      "loss": 0.1089167669415474,
      "mean_token_accuracy": 0.9682858139276505,
      "num_tokens": 2392064.0,
      "step": 146
    },
    {
      "entropy": 2.2121075242757797,
      "epoch": 0.43555555555555553,
      "grad_norm": 1.0066356658935547,
      "learning_rate": 5.765765765765766e-06,
      "loss": 0.16243846714496613,
      "mean_token_accuracy": 0.9607509821653366,
      "num_tokens": 2408448.0,
      "step": 147
    },
    {
      "entropy": 2.2374942302703857,
      "epoch": 0.43851851851851853,
      "grad_norm": 1.0947750806808472,
      "learning_rate": 5.735735735735736e-06,
      "loss": 0.2027081847190857,
      "mean_token_accuracy": 0.9401649683713913,
      "num_tokens": 2424832.0,
      "step": 148
    },
    {
      "entropy": 2.378810226917267,
      "epoch": 0.4414814814814815,
      "grad_norm": 1.0474845170974731,
      "learning_rate": 5.7057057057057065e-06,
      "loss": 0.21840409934520721,
      "mean_token_accuracy": 0.9281701818108559,
      "num_tokens": 2441216.0,
      "step": 149
    },
    {
      "entropy": 2.2447919696569443,
      "epoch": 0.4444444444444444,
      "grad_norm": 1.2253620624542236,
      "learning_rate": 5.675675675675676e-06,
      "loss": 0.26433685421943665,
      "mean_token_accuracy": 0.9277924746274948,
      "num_tokens": 2457600.0,
      "step": 150
    },
    {
      "epoch": 0.4444444444444444,
      "eval_entropy": 2.2217272885640464,
      "eval_loss": 0.176072359085083,
      "eval_mean_token_accuracy": 0.950895783106486,
      "eval_num_tokens": 2457600.0,
      "eval_runtime": 46.0502,
      "eval_samples_per_second": 13.029,
      "eval_steps_per_second": 1.629,
      "step": 150
    },
    {
      "entropy": 2.5773930847644806,
      "epoch": 0.4474074074074074,
      "grad_norm": 1.3818382024765015,
      "learning_rate": 5.645645645645647e-06,
      "loss": 0.2721666395664215,
      "mean_token_accuracy": 0.931941531598568,
      "num_tokens": 2473984.0,
      "step": 151
    },
    {
      "entropy": 2.2613002955913544,
      "epoch": 0.45037037037037037,
      "grad_norm": 0.8969493508338928,
      "learning_rate": 5.615615615615616e-06,
      "loss": 0.15523913502693176,
      "mean_token_accuracy": 0.9584640488028526,
      "num_tokens": 2490368.0,
      "step": 152
    },
    {
      "entropy": 2.3443939834833145,
      "epoch": 0.4533333333333333,
      "grad_norm": 1.13895845413208,
      "learning_rate": 5.585585585585585e-06,
      "loss": 0.20162707567214966,
      "mean_token_accuracy": 0.9339290484786034,
      "num_tokens": 2506752.0,
      "step": 153
    },
    {
      "entropy": 2.3639166355133057,
      "epoch": 0.4562962962962963,
      "grad_norm": 1.1336380243301392,
      "learning_rate": 5.555555555555557e-06,
      "loss": 0.197519451379776,
      "mean_token_accuracy": 0.9421832039952278,
      "num_tokens": 2523136.0,
      "step": 154
    },
    {
      "entropy": 2.2084722816944122,
      "epoch": 0.45925925925925926,
      "grad_norm": 0.8601342439651489,
      "learning_rate": 5.5255255255255255e-06,
      "loss": 0.12722940742969513,
      "mean_token_accuracy": 0.9639387279748917,
      "num_tokens": 2539520.0,
      "step": 155
    },
    {
      "entropy": 2.1656414270401,
      "epoch": 0.4622222222222222,
      "grad_norm": 0.710282564163208,
      "learning_rate": 5.495495495495496e-06,
      "loss": 0.06730201095342636,
      "mean_token_accuracy": 0.9817825853824615,
      "num_tokens": 2555904.0,
      "step": 156
    },
    {
      "entropy": 2.2940256893634796,
      "epoch": 0.4651851851851852,
      "grad_norm": 1.3642774820327759,
      "learning_rate": 5.465465465465466e-06,
      "loss": 0.24089768528938293,
      "mean_token_accuracy": 0.9300526157021523,
      "num_tokens": 2572288.0,
      "step": 157
    },
    {
      "entropy": 2.309079870581627,
      "epoch": 0.46814814814814815,
      "grad_norm": 0.8081016540527344,
      "learning_rate": 5.4354354354354355e-06,
      "loss": 0.15388454496860504,
      "mean_token_accuracy": 0.957360602915287,
      "num_tokens": 2588672.0,
      "step": 158
    },
    {
      "entropy": 2.13990081846714,
      "epoch": 0.4711111111111111,
      "grad_norm": 1.099754810333252,
      "learning_rate": 5.405405405405406e-06,
      "loss": 0.25663745403289795,
      "mean_token_accuracy": 0.9273329675197601,
      "num_tokens": 2605056.0,
      "step": 159
    },
    {
      "entropy": 2.409682944417,
      "epoch": 0.4740740740740741,
      "grad_norm": 1.0948880910873413,
      "learning_rate": 5.375375375375376e-06,
      "loss": 0.16294530034065247,
      "mean_token_accuracy": 0.960837721824646,
      "num_tokens": 2621440.0,
      "step": 160
    },
    {
      "entropy": 2.3806820809841156,
      "epoch": 0.47703703703703704,
      "grad_norm": 1.1878008842468262,
      "learning_rate": 5.345345345345346e-06,
      "loss": 0.23541851341724396,
      "mean_token_accuracy": 0.9291239827871323,
      "num_tokens": 2637824.0,
      "step": 161
    },
    {
      "entropy": 2.371952071785927,
      "epoch": 0.48,
      "grad_norm": 0.8815953731536865,
      "learning_rate": 5.315315315315316e-06,
      "loss": 0.13257887959480286,
      "mean_token_accuracy": 0.9596656411886215,
      "num_tokens": 2654208.0,
      "step": 162
    },
    {
      "entropy": 2.2399575412273407,
      "epoch": 0.482962962962963,
      "grad_norm": 1.053903341293335,
      "learning_rate": 5.285285285285286e-06,
      "loss": 0.19735129177570343,
      "mean_token_accuracy": 0.946366935968399,
      "num_tokens": 2670592.0,
      "step": 163
    },
    {
      "entropy": 2.518914133310318,
      "epoch": 0.48592592592592593,
      "grad_norm": 1.053481936454773,
      "learning_rate": 5.255255255255256e-06,
      "loss": 0.1289147138595581,
      "mean_token_accuracy": 0.9557346403598785,
      "num_tokens": 2686976.0,
      "step": 164
    },
    {
      "entropy": 2.2350364327430725,
      "epoch": 0.4888888888888889,
      "grad_norm": 1.091158390045166,
      "learning_rate": 5.225225225225226e-06,
      "loss": 0.20302288234233856,
      "mean_token_accuracy": 0.9473936185240746,
      "num_tokens": 2703360.0,
      "step": 165
    },
    {
      "entropy": 2.365757405757904,
      "epoch": 0.4918518518518519,
      "grad_norm": 0.873606264591217,
      "learning_rate": 5.195195195195195e-06,
      "loss": 0.14717966318130493,
      "mean_token_accuracy": 0.9562769159674644,
      "num_tokens": 2719744.0,
      "step": 166
    },
    {
      "entropy": 2.349894866347313,
      "epoch": 0.4948148148148148,
      "grad_norm": 1.0036927461624146,
      "learning_rate": 5.165165165165165e-06,
      "loss": 0.1657121628522873,
      "mean_token_accuracy": 0.9512332528829575,
      "num_tokens": 2736128.0,
      "step": 167
    },
    {
      "entropy": 2.3541809022426605,
      "epoch": 0.49777777777777776,
      "grad_norm": 1.3941274881362915,
      "learning_rate": 5.135135135135135e-06,
      "loss": 0.33090633153915405,
      "mean_token_accuracy": 0.9235115647315979,
      "num_tokens": 2752512.0,
      "step": 168
    },
    {
      "entropy": 2.113956794142723,
      "epoch": 0.5007407407407407,
      "grad_norm": 1.1116021871566772,
      "learning_rate": 5.105105105105106e-06,
      "loss": 0.20843583345413208,
      "mean_token_accuracy": 0.9469147622585297,
      "num_tokens": 2768896.0,
      "step": 169
    },
    {
      "entropy": 2.2251565158367157,
      "epoch": 0.5037037037037037,
      "grad_norm": 1.1401128768920898,
      "learning_rate": 5.075075075075075e-06,
      "loss": 0.24853655695915222,
      "mean_token_accuracy": 0.9191123321652412,
      "num_tokens": 2785280.0,
      "step": 170
    },
    {
      "entropy": 2.312959998846054,
      "epoch": 0.5066666666666667,
      "grad_norm": 1.1748965978622437,
      "learning_rate": 5.045045045045045e-06,
      "loss": 0.2280878722667694,
      "mean_token_accuracy": 0.9393928498029709,
      "num_tokens": 2801664.0,
      "step": 171
    },
    {
      "entropy": 2.2148368656635284,
      "epoch": 0.5096296296296297,
      "grad_norm": 1.2531650066375732,
      "learning_rate": 5.0150150150150156e-06,
      "loss": 0.25916075706481934,
      "mean_token_accuracy": 0.9258779436349869,
      "num_tokens": 2818048.0,
      "step": 172
    },
    {
      "entropy": 2.3727904856204987,
      "epoch": 0.5125925925925926,
      "grad_norm": 1.1567667722702026,
      "learning_rate": 4.984984984984985e-06,
      "loss": 0.2302863448858261,
      "mean_token_accuracy": 0.9301356449723244,
      "num_tokens": 2834432.0,
      "step": 173
    },
    {
      "entropy": 2.331768572330475,
      "epoch": 0.5155555555555555,
      "grad_norm": 1.143294095993042,
      "learning_rate": 4.954954954954955e-06,
      "loss": 0.22374312579631805,
      "mean_token_accuracy": 0.9277942329645157,
      "num_tokens": 2850816.0,
      "step": 174
    },
    {
      "entropy": 2.1124974340200424,
      "epoch": 0.5185185185185185,
      "grad_norm": 1.1063649654388428,
      "learning_rate": 4.9249249249249255e-06,
      "loss": 0.1937328726053238,
      "mean_token_accuracy": 0.9410960748791695,
      "num_tokens": 2867200.0,
      "step": 175
    },
    {
      "entropy": 2.1501243710517883,
      "epoch": 0.5214814814814814,
      "grad_norm": 1.2337923049926758,
      "learning_rate": 4.894894894894895e-06,
      "loss": 0.23605607450008392,
      "mean_token_accuracy": 0.9322064965963364,
      "num_tokens": 2883584.0,
      "step": 176
    },
    {
      "entropy": 2.2725354731082916,
      "epoch": 0.5244444444444445,
      "grad_norm": 0.8507163524627686,
      "learning_rate": 4.864864864864866e-06,
      "loss": 0.12817996740341187,
      "mean_token_accuracy": 0.9600297063589096,
      "num_tokens": 2899968.0,
      "step": 177
    },
    {
      "entropy": 2.2773052901029587,
      "epoch": 0.5274074074074074,
      "grad_norm": 1.0258901119232178,
      "learning_rate": 4.8348348348348355e-06,
      "loss": 0.19734807312488556,
      "mean_token_accuracy": 0.933582991361618,
      "num_tokens": 2916352.0,
      "step": 178
    },
    {
      "entropy": 2.1450360268354416,
      "epoch": 0.5303703703703704,
      "grad_norm": 1.1847776174545288,
      "learning_rate": 4.804804804804805e-06,
      "loss": 0.22736720740795135,
      "mean_token_accuracy": 0.9363805949687958,
      "num_tokens": 2932736.0,
      "step": 179
    },
    {
      "entropy": 2.12079681456089,
      "epoch": 0.5333333333333333,
      "grad_norm": 1.0377146005630493,
      "learning_rate": 4.774774774774775e-06,
      "loss": 0.18095552921295166,
      "mean_token_accuracy": 0.945849597454071,
      "num_tokens": 2949120.0,
      "step": 180
    },
    {
      "entropy": 1.9365530014038086,
      "epoch": 0.5362962962962963,
      "grad_norm": 0.9215973019599915,
      "learning_rate": 4.7447447447447454e-06,
      "loss": 0.17776966094970703,
      "mean_token_accuracy": 0.9480276927351952,
      "num_tokens": 2965504.0,
      "step": 181
    },
    {
      "entropy": 2.5179311633110046,
      "epoch": 0.5392592592592592,
      "grad_norm": 1.2804752588272095,
      "learning_rate": 4.714714714714715e-06,
      "loss": 0.1974414885044098,
      "mean_token_accuracy": 0.9477459937334061,
      "num_tokens": 2981888.0,
      "step": 182
    },
    {
      "entropy": 2.268339142203331,
      "epoch": 0.5422222222222223,
      "grad_norm": 0.9747254252433777,
      "learning_rate": 4.684684684684685e-06,
      "loss": 0.14459143579006195,
      "mean_token_accuracy": 0.9605521634221077,
      "num_tokens": 2998272.0,
      "step": 183
    },
    {
      "entropy": 2.3601040840148926,
      "epoch": 0.5451851851851852,
      "grad_norm": 1.1514772176742554,
      "learning_rate": 4.654654654654655e-06,
      "loss": 0.25473877787590027,
      "mean_token_accuracy": 0.9344798102974892,
      "num_tokens": 3014656.0,
      "step": 184
    },
    {
      "entropy": 2.413866639137268,
      "epoch": 0.5481481481481482,
      "grad_norm": 0.8757903575897217,
      "learning_rate": 4.624624624624625e-06,
      "loss": 0.13507026433944702,
      "mean_token_accuracy": 0.9587804228067398,
      "num_tokens": 3031040.0,
      "step": 185
    },
    {
      "entropy": 2.507603168487549,
      "epoch": 0.5511111111111111,
      "grad_norm": 1.2733794450759888,
      "learning_rate": 4.594594594594596e-06,
      "loss": 0.2414797842502594,
      "mean_token_accuracy": 0.9326403141021729,
      "num_tokens": 3047424.0,
      "step": 186
    },
    {
      "entropy": 2.4590699672698975,
      "epoch": 0.554074074074074,
      "grad_norm": 1.409479022026062,
      "learning_rate": 4.5645645645645645e-06,
      "loss": 0.2938889265060425,
      "mean_token_accuracy": 0.9220807403326035,
      "num_tokens": 3063808.0,
      "step": 187
    },
    {
      "entropy": 2.1663295924663544,
      "epoch": 0.557037037037037,
      "grad_norm": 0.9258942008018494,
      "learning_rate": 4.534534534534535e-06,
      "loss": 0.18513718247413635,
      "mean_token_accuracy": 0.9526167958974838,
      "num_tokens": 3080192.0,
      "step": 188
    },
    {
      "entropy": 2.511041134595871,
      "epoch": 0.56,
      "grad_norm": 1.2542656660079956,
      "learning_rate": 4.504504504504505e-06,
      "loss": 0.2336530238389969,
      "mean_token_accuracy": 0.9392572045326233,
      "num_tokens": 3096576.0,
      "step": 189
    },
    {
      "entropy": 2.3700854629278183,
      "epoch": 0.562962962962963,
      "grad_norm": 1.2292245626449585,
      "learning_rate": 4.474474474474475e-06,
      "loss": 0.21047058701515198,
      "mean_token_accuracy": 0.950310468673706,
      "num_tokens": 3112960.0,
      "step": 190
    },
    {
      "entropy": 2.24889375269413,
      "epoch": 0.5659259259259259,
      "grad_norm": 0.9944830536842346,
      "learning_rate": 4.444444444444444e-06,
      "loss": 0.2177523672580719,
      "mean_token_accuracy": 0.9447462782263756,
      "num_tokens": 3129344.0,
      "step": 191
    },
    {
      "entropy": 2.3181798458099365,
      "epoch": 0.5688888888888889,
      "grad_norm": 0.8251722455024719,
      "learning_rate": 4.414414414414415e-06,
      "loss": 0.1356397569179535,
      "mean_token_accuracy": 0.9578035846352577,
      "num_tokens": 3145728.0,
      "step": 192
    },
    {
      "entropy": 2.184191331267357,
      "epoch": 0.5718518518518518,
      "grad_norm": 0.9963244795799255,
      "learning_rate": 4.384384384384384e-06,
      "loss": 0.2417733073234558,
      "mean_token_accuracy": 0.9288726896047592,
      "num_tokens": 3162112.0,
      "step": 193
    },
    {
      "entropy": 2.1568250358104706,
      "epoch": 0.5748148148148148,
      "grad_norm": 1.0079498291015625,
      "learning_rate": 4.354354354354355e-06,
      "loss": 0.2385617345571518,
      "mean_token_accuracy": 0.9309723824262619,
      "num_tokens": 3178496.0,
      "step": 194
    },
    {
      "entropy": 2.4423921406269073,
      "epoch": 0.5777777777777777,
      "grad_norm": 0.9145395755767822,
      "learning_rate": 4.324324324324325e-06,
      "loss": 0.14828522503376007,
      "mean_token_accuracy": 0.9523493349552155,
      "num_tokens": 3194880.0,
      "step": 195
    },
    {
      "entropy": 2.072684645652771,
      "epoch": 0.5807407407407408,
      "grad_norm": 0.8940830230712891,
      "learning_rate": 4.294294294294294e-06,
      "loss": 0.17374826967716217,
      "mean_token_accuracy": 0.9539390578866005,
      "num_tokens": 3211264.0,
      "step": 196
    },
    {
      "entropy": 2.1707707941532135,
      "epoch": 0.5837037037037037,
      "grad_norm": 1.0081565380096436,
      "learning_rate": 4.264264264264265e-06,
      "loss": 0.218732550740242,
      "mean_token_accuracy": 0.941913515329361,
      "num_tokens": 3227648.0,
      "step": 197
    },
    {
      "entropy": 2.1819057166576385,
      "epoch": 0.5866666666666667,
      "grad_norm": 0.9585004448890686,
      "learning_rate": 4.234234234234235e-06,
      "loss": 0.19803617894649506,
      "mean_token_accuracy": 0.9306176900863647,
      "num_tokens": 3244032.0,
      "step": 198
    },
    {
      "entropy": 2.0674026161432266,
      "epoch": 0.5896296296296296,
      "grad_norm": 1.064789891242981,
      "learning_rate": 4.204204204204204e-06,
      "loss": 0.17672395706176758,
      "mean_token_accuracy": 0.9564284309744835,
      "num_tokens": 3260416.0,
      "step": 199
    },
    {
      "entropy": 2.0122371912002563,
      "epoch": 0.5925925925925926,
      "grad_norm": 0.8654584288597107,
      "learning_rate": 4.174174174174174e-06,
      "loss": 0.14349326491355896,
      "mean_token_accuracy": 0.9576803296804428,
      "num_tokens": 3276800.0,
      "step": 200
    },
    {
      "epoch": 0.5925925925925926,
      "eval_entropy": 2.135404682159424,
      "eval_loss": 0.16948683559894562,
      "eval_mean_token_accuracy": 0.9527613043785095,
      "eval_num_tokens": 3276800.0,
      "eval_runtime": 46.0394,
      "eval_samples_per_second": 13.032,
      "eval_steps_per_second": 1.629,
      "step": 200
    },
    {
      "entropy": 2.095166191458702,
      "epoch": 0.5955555555555555,
      "grad_norm": 1.1691136360168457,
      "learning_rate": 4.1441441441441446e-06,
      "loss": 0.2649495601654053,
      "mean_token_accuracy": 0.9213642030954361,
      "num_tokens": 3293184.0,
      "step": 201
    },
    {
      "entropy": 2.120430037379265,
      "epoch": 0.5985185185185186,
      "grad_norm": 0.9470794796943665,
      "learning_rate": 4.114114114114114e-06,
      "loss": 0.19404877722263336,
      "mean_token_accuracy": 0.9462184235453606,
      "num_tokens": 3309568.0,
      "step": 202
    },
    {
      "entropy": 2.0398283153772354,
      "epoch": 0.6014814814814815,
      "grad_norm": 0.8615677356719971,
      "learning_rate": 4.084084084084085e-06,
      "loss": 0.14602459967136383,
      "mean_token_accuracy": 0.9589569196105003,
      "num_tokens": 3325952.0,
      "step": 203
    },
    {
      "entropy": 2.2172525823116302,
      "epoch": 0.6044444444444445,
      "grad_norm": 1.3697327375411987,
      "learning_rate": 4.0540540540540545e-06,
      "loss": 0.2774120569229126,
      "mean_token_accuracy": 0.9347071945667267,
      "num_tokens": 3342336.0,
      "step": 204
    },
    {
      "entropy": 2.1588571965694427,
      "epoch": 0.6074074074074074,
      "grad_norm": 1.145128607749939,
      "learning_rate": 4.024024024024024e-06,
      "loss": 0.22687923908233643,
      "mean_token_accuracy": 0.9330701977014542,
      "num_tokens": 3358720.0,
      "step": 205
    },
    {
      "entropy": 1.83328016102314,
      "epoch": 0.6103703703703703,
      "grad_norm": 0.8561811447143555,
      "learning_rate": 3.993993993993994e-06,
      "loss": 0.16048912703990936,
      "mean_token_accuracy": 0.9440218880772591,
      "num_tokens": 3375104.0,
      "step": 206
    },
    {
      "entropy": 2.3320832699537277,
      "epoch": 0.6133333333333333,
      "grad_norm": 0.8710148930549622,
      "learning_rate": 3.9639639639639645e-06,
      "loss": 0.11265282332897186,
      "mean_token_accuracy": 0.9709779843688011,
      "num_tokens": 3391488.0,
      "step": 207
    },
    {
      "entropy": 2.091501921415329,
      "epoch": 0.6162962962962963,
      "grad_norm": 1.1158099174499512,
      "learning_rate": 3.933933933933934e-06,
      "loss": 0.18026238679885864,
      "mean_token_accuracy": 0.9497736319899559,
      "num_tokens": 3407872.0,
      "step": 208
    },
    {
      "entropy": 2.1043567657470703,
      "epoch": 0.6192592592592593,
      "grad_norm": 0.9538904428482056,
      "learning_rate": 3.903903903903904e-06,
      "loss": 0.21173931658267975,
      "mean_token_accuracy": 0.9397457838058472,
      "num_tokens": 3424256.0,
      "step": 209
    },
    {
      "entropy": 2.1878659576177597,
      "epoch": 0.6222222222222222,
      "grad_norm": 1.0738528966903687,
      "learning_rate": 3.8738738738738744e-06,
      "loss": 0.21238672733306885,
      "mean_token_accuracy": 0.9378648400306702,
      "num_tokens": 3440640.0,
      "step": 210
    },
    {
      "entropy": 2.134036511182785,
      "epoch": 0.6251851851851852,
      "grad_norm": 1.0524516105651855,
      "learning_rate": 3.843843843843844e-06,
      "loss": 0.20693530142307281,
      "mean_token_accuracy": 0.9373673647642136,
      "num_tokens": 3457024.0,
      "step": 211
    },
    {
      "entropy": 2.088997006416321,
      "epoch": 0.6281481481481481,
      "grad_norm": 0.8862026929855347,
      "learning_rate": 3.8138138138138143e-06,
      "loss": 0.17873947322368622,
      "mean_token_accuracy": 0.9474726691842079,
      "num_tokens": 3473408.0,
      "step": 212
    },
    {
      "entropy": 2.2946579456329346,
      "epoch": 0.6311111111111111,
      "grad_norm": 1.1507551670074463,
      "learning_rate": 3.7837837837837844e-06,
      "loss": 0.21955841779708862,
      "mean_token_accuracy": 0.9343761205673218,
      "num_tokens": 3489792.0,
      "step": 213
    },
    {
      "entropy": 2.1536125987768173,
      "epoch": 0.6340740740740741,
      "grad_norm": 0.9619444012641907,
      "learning_rate": 3.7537537537537537e-06,
      "loss": 0.18027596175670624,
      "mean_token_accuracy": 0.9448429644107819,
      "num_tokens": 3506176.0,
      "step": 214
    },
    {
      "entropy": 2.387341320514679,
      "epoch": 0.6370370370370371,
      "grad_norm": 0.9111767411231995,
      "learning_rate": 3.723723723723724e-06,
      "loss": 0.14665061235427856,
      "mean_token_accuracy": 0.9585870578885078,
      "num_tokens": 3522560.0,
      "step": 215
    },
    {
      "entropy": 2.2919792979955673,
      "epoch": 0.64,
      "grad_norm": 1.0334508419036865,
      "learning_rate": 3.693693693693694e-06,
      "loss": 0.18953123688697815,
      "mean_token_accuracy": 0.9423899203538895,
      "num_tokens": 3538944.0,
      "step": 216
    },
    {
      "entropy": 2.339949756860733,
      "epoch": 0.642962962962963,
      "grad_norm": 1.151109218597412,
      "learning_rate": 3.663663663663664e-06,
      "loss": 0.19529879093170166,
      "mean_token_accuracy": 0.9467525854706764,
      "num_tokens": 3555328.0,
      "step": 217
    },
    {
      "entropy": 2.212970107793808,
      "epoch": 0.6459259259259259,
      "grad_norm": 0.9182158708572388,
      "learning_rate": 3.633633633633634e-06,
      "loss": 0.1373278945684433,
      "mean_token_accuracy": 0.955323226749897,
      "num_tokens": 3571712.0,
      "step": 218
    },
    {
      "entropy": 2.4035662710666656,
      "epoch": 0.6488888888888888,
      "grad_norm": 1.3012800216674805,
      "learning_rate": 3.603603603603604e-06,
      "loss": 0.23834392428398132,
      "mean_token_accuracy": 0.9361925646662712,
      "num_tokens": 3588096.0,
      "step": 219
    },
    {
      "entropy": 2.2333020716905594,
      "epoch": 0.6518518518518519,
      "grad_norm": 0.9093357920646667,
      "learning_rate": 3.573573573573574e-06,
      "loss": 0.165394589304924,
      "mean_token_accuracy": 0.9541044980287552,
      "num_tokens": 3604480.0,
      "step": 220
    },
    {
      "entropy": 2.0754988938570023,
      "epoch": 0.6548148148148148,
      "grad_norm": 0.8467314839363098,
      "learning_rate": 3.5435435435435437e-06,
      "loss": 0.17804819345474243,
      "mean_token_accuracy": 0.9425498619675636,
      "num_tokens": 3620864.0,
      "step": 221
    },
    {
      "entropy": 2.220280945301056,
      "epoch": 0.6577777777777778,
      "grad_norm": 1.0632840394973755,
      "learning_rate": 3.513513513513514e-06,
      "loss": 0.18905292451381683,
      "mean_token_accuracy": 0.9376566633582115,
      "num_tokens": 3637248.0,
      "step": 222
    },
    {
      "entropy": 2.625906676054001,
      "epoch": 0.6607407407407407,
      "grad_norm": 1.0544439554214478,
      "learning_rate": 3.4834834834834835e-06,
      "loss": 0.1657475382089615,
      "mean_token_accuracy": 0.9552930593490601,
      "num_tokens": 3653632.0,
      "step": 223
    },
    {
      "entropy": 2.533998668193817,
      "epoch": 0.6637037037037037,
      "grad_norm": 0.9885019659996033,
      "learning_rate": 3.4534534534534537e-06,
      "loss": 0.11697129160165787,
      "mean_token_accuracy": 0.9653114303946495,
      "num_tokens": 3670016.0,
      "step": 224
    },
    {
      "entropy": 2.3260088562965393,
      "epoch": 0.6666666666666666,
      "grad_norm": 0.9760135412216187,
      "learning_rate": 3.423423423423424e-06,
      "loss": 0.1582515835762024,
      "mean_token_accuracy": 0.954810656607151,
      "num_tokens": 3686400.0,
      "step": 225
    },
    {
      "entropy": 2.3039702475070953,
      "epoch": 0.6696296296296296,
      "grad_norm": 1.2249606847763062,
      "learning_rate": 3.393393393393394e-06,
      "loss": 0.18487505614757538,
      "mean_token_accuracy": 0.9549200311303139,
      "num_tokens": 3702784.0,
      "step": 226
    },
    {
      "entropy": 2.149510011076927,
      "epoch": 0.6725925925925926,
      "grad_norm": 1.0753812789916992,
      "learning_rate": 3.363363363363364e-06,
      "loss": 0.24741613864898682,
      "mean_token_accuracy": 0.9362531825900078,
      "num_tokens": 3719168.0,
      "step": 227
    },
    {
      "entropy": 2.2836580872535706,
      "epoch": 0.6755555555555556,
      "grad_norm": 0.7547692656517029,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.11351179331541061,
      "mean_token_accuracy": 0.9703884795308113,
      "num_tokens": 3735552.0,
      "step": 228
    },
    {
      "entropy": 2.181907206773758,
      "epoch": 0.6785185185185185,
      "grad_norm": 0.8284390568733215,
      "learning_rate": 3.3033033033033035e-06,
      "loss": 0.10280636698007584,
      "mean_token_accuracy": 0.9684972018003464,
      "num_tokens": 3751936.0,
      "step": 229
    },
    {
      "entropy": 1.9404956251382828,
      "epoch": 0.6814814814814815,
      "grad_norm": 0.9197994470596313,
      "learning_rate": 3.2732732732732736e-06,
      "loss": 0.15574616193771362,
      "mean_token_accuracy": 0.9558748230338097,
      "num_tokens": 3768320.0,
      "step": 230
    },
    {
      "entropy": 2.2773773968219757,
      "epoch": 0.6844444444444444,
      "grad_norm": 1.2663758993148804,
      "learning_rate": 3.2432432432432437e-06,
      "loss": 0.2824876308441162,
      "mean_token_accuracy": 0.929023265838623,
      "num_tokens": 3784704.0,
      "step": 231
    },
    {
      "entropy": 2.1691990345716476,
      "epoch": 0.6874074074074074,
      "grad_norm": 1.029785394668579,
      "learning_rate": 3.2132132132132134e-06,
      "loss": 0.17473608255386353,
      "mean_token_accuracy": 0.9522128999233246,
      "num_tokens": 3801088.0,
      "step": 232
    },
    {
      "entropy": 2.3093802481889725,
      "epoch": 0.6903703703703704,
      "grad_norm": 0.9596537947654724,
      "learning_rate": 3.183183183183183e-06,
      "loss": 0.14627918601036072,
      "mean_token_accuracy": 0.954444482922554,
      "num_tokens": 3817472.0,
      "step": 233
    },
    {
      "entropy": 2.3403891175985336,
      "epoch": 0.6933333333333334,
      "grad_norm": 0.8634820580482483,
      "learning_rate": 3.1531531531531532e-06,
      "loss": 0.12733465433120728,
      "mean_token_accuracy": 0.9697809740900993,
      "num_tokens": 3833856.0,
      "step": 234
    },
    {
      "entropy": 2.366154581308365,
      "epoch": 0.6962962962962963,
      "grad_norm": 1.3474041223526,
      "learning_rate": 3.1231231231231234e-06,
      "loss": 0.20135396718978882,
      "mean_token_accuracy": 0.9478387236595154,
      "num_tokens": 3850240.0,
      "step": 235
    },
    {
      "entropy": 2.3419956266880035,
      "epoch": 0.6992592592592592,
      "grad_norm": 1.3067741394042969,
      "learning_rate": 3.0930930930930935e-06,
      "loss": 0.23931923508644104,
      "mean_token_accuracy": 0.9419998005032539,
      "num_tokens": 3866624.0,
      "step": 236
    },
    {
      "entropy": 2.3942779302597046,
      "epoch": 0.7022222222222222,
      "grad_norm": 1.084053635597229,
      "learning_rate": 3.063063063063063e-06,
      "loss": 0.20017841458320618,
      "mean_token_accuracy": 0.9434688687324524,
      "num_tokens": 3883008.0,
      "step": 237
    },
    {
      "entropy": 2.487148553133011,
      "epoch": 0.7051851851851851,
      "grad_norm": 1.0191210508346558,
      "learning_rate": 3.0330330330330333e-06,
      "loss": 0.18339340388774872,
      "mean_token_accuracy": 0.9428181573748589,
      "num_tokens": 3899392.0,
      "step": 238
    },
    {
      "entropy": 2.383811265230179,
      "epoch": 0.7081481481481482,
      "grad_norm": 0.9769102931022644,
      "learning_rate": 3.0030030030030034e-06,
      "loss": 0.13787123560905457,
      "mean_token_accuracy": 0.9573397561907768,
      "num_tokens": 3915776.0,
      "step": 239
    },
    {
      "entropy": 2.4776239097118378,
      "epoch": 0.7111111111111111,
      "grad_norm": 1.1633875370025635,
      "learning_rate": 2.9729729729729736e-06,
      "loss": 0.22200757265090942,
      "mean_token_accuracy": 0.9228844493627548,
      "num_tokens": 3932160.0,
      "step": 240
    },
    {
      "entropy": 2.3706377893686295,
      "epoch": 0.7140740740740741,
      "grad_norm": 0.906835675239563,
      "learning_rate": 2.942942942942943e-06,
      "loss": 0.1858144998550415,
      "mean_token_accuracy": 0.945995032787323,
      "num_tokens": 3948544.0,
      "step": 241
    },
    {
      "entropy": 2.2541957199573517,
      "epoch": 0.717037037037037,
      "grad_norm": 1.116713523864746,
      "learning_rate": 2.912912912912913e-06,
      "loss": 0.21036140620708466,
      "mean_token_accuracy": 0.9408295825123787,
      "num_tokens": 3964928.0,
      "step": 242
    },
    {
      "entropy": 2.371729791164398,
      "epoch": 0.72,
      "grad_norm": 1.1916154623031616,
      "learning_rate": 2.882882882882883e-06,
      "loss": 0.27101314067840576,
      "mean_token_accuracy": 0.9270181208848953,
      "num_tokens": 3981312.0,
      "step": 243
    },
    {
      "entropy": 2.4406691640615463,
      "epoch": 0.7229629629629629,
      "grad_norm": 0.8946484923362732,
      "learning_rate": 2.8528528528528532e-06,
      "loss": 0.13982370495796204,
      "mean_token_accuracy": 0.954224206507206,
      "num_tokens": 3997696.0,
      "step": 244
    },
    {
      "entropy": 2.288629859685898,
      "epoch": 0.725925925925926,
      "grad_norm": 0.8298711776733398,
      "learning_rate": 2.8228228228228234e-06,
      "loss": 0.10307710617780685,
      "mean_token_accuracy": 0.9729797169566154,
      "num_tokens": 4014080.0,
      "step": 245
    },
    {
      "entropy": 2.4576481729745865,
      "epoch": 0.7288888888888889,
      "grad_norm": 1.1461255550384521,
      "learning_rate": 2.7927927927927926e-06,
      "loss": 0.22268827259540558,
      "mean_token_accuracy": 0.9335534200072289,
      "num_tokens": 4030464.0,
      "step": 246
    },
    {
      "entropy": 2.3772233724594116,
      "epoch": 0.7318518518518519,
      "grad_norm": 0.8248373866081238,
      "learning_rate": 2.7627627627627628e-06,
      "loss": 0.11924245953559875,
      "mean_token_accuracy": 0.9646312370896339,
      "num_tokens": 4046848.0,
      "step": 247
    },
    {
      "entropy": 2.323305293917656,
      "epoch": 0.7348148148148148,
      "grad_norm": 0.9537495970726013,
      "learning_rate": 2.732732732732733e-06,
      "loss": 0.13450981676578522,
      "mean_token_accuracy": 0.9690083563327789,
      "num_tokens": 4063232.0,
      "step": 248
    },
    {
      "entropy": 2.0796739161014557,
      "epoch": 0.7377777777777778,
      "grad_norm": 1.079800009727478,
      "learning_rate": 2.702702702702703e-06,
      "loss": 0.22776569426059723,
      "mean_token_accuracy": 0.9393508806824684,
      "num_tokens": 4079616.0,
      "step": 249
    },
    {
      "entropy": 2.260486900806427,
      "epoch": 0.7407407407407407,
      "grad_norm": 1.0986967086791992,
      "learning_rate": 2.672672672672673e-06,
      "loss": 0.21423429250717163,
      "mean_token_accuracy": 0.9410804063081741,
      "num_tokens": 4096000.0,
      "step": 250
    },
    {
      "epoch": 0.7407407407407407,
      "eval_entropy": 2.2857923634847004,
      "eval_loss": 0.16553135216236115,
      "eval_mean_token_accuracy": 0.9536256949106853,
      "eval_num_tokens": 4096000.0,
      "eval_runtime": 46.046,
      "eval_samples_per_second": 13.03,
      "eval_steps_per_second": 1.629,
      "step": 250
    },
    {
      "entropy": 2.4411289393901825,
      "epoch": 0.7437037037037038,
      "grad_norm": 1.1856727600097656,
      "learning_rate": 2.642642642642643e-06,
      "loss": 0.1928912103176117,
      "mean_token_accuracy": 0.9504735544323921,
      "num_tokens": 4112384.0,
      "step": 251
    },
    {
      "entropy": 2.207347810268402,
      "epoch": 0.7466666666666667,
      "grad_norm": 0.8472045063972473,
      "learning_rate": 2.612612612612613e-06,
      "loss": 0.14188183844089508,
      "mean_token_accuracy": 0.9627470448613167,
      "num_tokens": 4128768.0,
      "step": 252
    },
    {
      "entropy": 2.323253870010376,
      "epoch": 0.7496296296296296,
      "grad_norm": 1.235590934753418,
      "learning_rate": 2.5825825825825827e-06,
      "loss": 0.21392673254013062,
      "mean_token_accuracy": 0.9414556175470352,
      "num_tokens": 4145152.0,
      "step": 253
    },
    {
      "entropy": 2.381735548377037,
      "epoch": 0.7525925925925926,
      "grad_norm": 1.038611650466919,
      "learning_rate": 2.552552552552553e-06,
      "loss": 0.16648142039775848,
      "mean_token_accuracy": 0.9451280310750008,
      "num_tokens": 4161536.0,
      "step": 254
    },
    {
      "entropy": 2.2459085881710052,
      "epoch": 0.7555555555555555,
      "grad_norm": 1.0215892791748047,
      "learning_rate": 2.5225225225225225e-06,
      "loss": 0.19291386008262634,
      "mean_token_accuracy": 0.9446487799286842,
      "num_tokens": 4177920.0,
      "step": 255
    },
    {
      "entropy": 2.401110053062439,
      "epoch": 0.7585185185185185,
      "grad_norm": 1.0392042398452759,
      "learning_rate": 2.4924924924924926e-06,
      "loss": 0.18167832493782043,
      "mean_token_accuracy": 0.9427581205964088,
      "num_tokens": 4194304.0,
      "step": 256
    },
    {
      "entropy": 2.1185553520917892,
      "epoch": 0.7614814814814815,
      "grad_norm": 1.0228697061538696,
      "learning_rate": 2.4624624624624628e-06,
      "loss": 0.22100982069969177,
      "mean_token_accuracy": 0.9449682831764221,
      "num_tokens": 4210688.0,
      "step": 257
    },
    {
      "entropy": 2.0173698365688324,
      "epoch": 0.7644444444444445,
      "grad_norm": 1.028183102607727,
      "learning_rate": 2.432432432432433e-06,
      "loss": 0.21454045176506042,
      "mean_token_accuracy": 0.9366749078035355,
      "num_tokens": 4227072.0,
      "step": 258
    },
    {
      "entropy": 2.360345870256424,
      "epoch": 0.7674074074074074,
      "grad_norm": 1.0896662473678589,
      "learning_rate": 2.4024024024024026e-06,
      "loss": 0.21896213293075562,
      "mean_token_accuracy": 0.9268064647912979,
      "num_tokens": 4243456.0,
      "step": 259
    },
    {
      "entropy": 2.266966611146927,
      "epoch": 0.7703703703703704,
      "grad_norm": 1.1020430326461792,
      "learning_rate": 2.3723723723723727e-06,
      "loss": 0.16863444447517395,
      "mean_token_accuracy": 0.9455827325582504,
      "num_tokens": 4259840.0,
      "step": 260
    },
    {
      "entropy": 2.4497701972723007,
      "epoch": 0.7733333333333333,
      "grad_norm": 1.2800923585891724,
      "learning_rate": 2.3423423423423424e-06,
      "loss": 0.2419387847185135,
      "mean_token_accuracy": 0.9233163520693779,
      "num_tokens": 4276224.0,
      "step": 261
    },
    {
      "entropy": 2.4977878034114838,
      "epoch": 0.7762962962962963,
      "grad_norm": 1.3704835176467896,
      "learning_rate": 2.3123123123123125e-06,
      "loss": 0.23335173726081848,
      "mean_token_accuracy": 0.9396708309650421,
      "num_tokens": 4292608.0,
      "step": 262
    },
    {
      "entropy": 2.510371834039688,
      "epoch": 0.7792592592592592,
      "grad_norm": 1.4372755289077759,
      "learning_rate": 2.2822822822822822e-06,
      "loss": 0.27837955951690674,
      "mean_token_accuracy": 0.9243473038077354,
      "num_tokens": 4308992.0,
      "step": 263
    },
    {
      "entropy": 2.327836185693741,
      "epoch": 0.7822222222222223,
      "grad_norm": 0.9518190622329712,
      "learning_rate": 2.2522522522522524e-06,
      "loss": 0.13848723471164703,
      "mean_token_accuracy": 0.9614205211400986,
      "num_tokens": 4325376.0,
      "step": 264
    },
    {
      "entropy": 2.383026659488678,
      "epoch": 0.7851851851851852,
      "grad_norm": 0.9475295543670654,
      "learning_rate": 2.222222222222222e-06,
      "loss": 0.1333014816045761,
      "mean_token_accuracy": 0.9606904834508896,
      "num_tokens": 4341760.0,
      "step": 265
    },
    {
      "entropy": 1.9964693933725357,
      "epoch": 0.7881481481481482,
      "grad_norm": 0.6908922791481018,
      "learning_rate": 2.192192192192192e-06,
      "loss": 0.11969106644392014,
      "mean_token_accuracy": 0.9559045284986496,
      "num_tokens": 4358144.0,
      "step": 266
    },
    {
      "entropy": 2.226464867591858,
      "epoch": 0.7911111111111111,
      "grad_norm": 1.3186053037643433,
      "learning_rate": 2.1621621621621623e-06,
      "loss": 0.24855168163776398,
      "mean_token_accuracy": 0.9359507337212563,
      "num_tokens": 4374528.0,
      "step": 267
    },
    {
      "entropy": 2.2805541902780533,
      "epoch": 0.794074074074074,
      "grad_norm": 1.0862208604812622,
      "learning_rate": 2.1321321321321325e-06,
      "loss": 0.16623716056346893,
      "mean_token_accuracy": 0.9557871297001839,
      "num_tokens": 4390912.0,
      "step": 268
    },
    {
      "entropy": 2.446542590856552,
      "epoch": 0.797037037037037,
      "grad_norm": 1.2609816789627075,
      "learning_rate": 2.102102102102102e-06,
      "loss": 0.2186906933784485,
      "mean_token_accuracy": 0.9219034239649773,
      "num_tokens": 4407296.0,
      "step": 269
    },
    {
      "entropy": 2.5203704237937927,
      "epoch": 0.8,
      "grad_norm": 1.3990377187728882,
      "learning_rate": 2.0720720720720723e-06,
      "loss": 0.280950665473938,
      "mean_token_accuracy": 0.9188291057944298,
      "num_tokens": 4423680.0,
      "step": 270
    },
    {
      "entropy": 2.266892895102501,
      "epoch": 0.802962962962963,
      "grad_norm": 1.0920964479446411,
      "learning_rate": 2.0420420420420424e-06,
      "loss": 0.1616411805152893,
      "mean_token_accuracy": 0.9480260238051414,
      "num_tokens": 4440064.0,
      "step": 271
    },
    {
      "entropy": 2.407806247472763,
      "epoch": 0.8059259259259259,
      "grad_norm": 1.408501386642456,
      "learning_rate": 2.012012012012012e-06,
      "loss": 0.22945785522460938,
      "mean_token_accuracy": 0.9419111758470535,
      "num_tokens": 4456448.0,
      "step": 272
    },
    {
      "entropy": 2.379624664783478,
      "epoch": 0.8088888888888889,
      "grad_norm": 1.3813397884368896,
      "learning_rate": 1.9819819819819822e-06,
      "loss": 0.2909376919269562,
      "mean_token_accuracy": 0.9264494106173515,
      "num_tokens": 4472832.0,
      "step": 273
    },
    {
      "entropy": 2.114370122551918,
      "epoch": 0.8118518518518518,
      "grad_norm": 0.7531031966209412,
      "learning_rate": 1.951951951951952e-06,
      "loss": 0.10272254794836044,
      "mean_token_accuracy": 0.96744055300951,
      "num_tokens": 4489216.0,
      "step": 274
    },
    {
      "entropy": 2.3059495389461517,
      "epoch": 0.8148148148148148,
      "grad_norm": 1.1459184885025024,
      "learning_rate": 1.921921921921922e-06,
      "loss": 0.22518137097358704,
      "mean_token_accuracy": 0.94258102029562,
      "num_tokens": 4505600.0,
      "step": 275
    },
    {
      "entropy": 2.3124334812164307,
      "epoch": 0.8177777777777778,
      "grad_norm": 1.4195786714553833,
      "learning_rate": 1.8918918918918922e-06,
      "loss": 0.2533921003341675,
      "mean_token_accuracy": 0.9260800033807755,
      "num_tokens": 4521984.0,
      "step": 276
    },
    {
      "entropy": 2.171090543270111,
      "epoch": 0.8207407407407408,
      "grad_norm": 1.0381304025650024,
      "learning_rate": 1.861861861861862e-06,
      "loss": 0.19962534308433533,
      "mean_token_accuracy": 0.9385664910078049,
      "num_tokens": 4538368.0,
      "step": 277
    },
    {
      "entropy": 2.361858978867531,
      "epoch": 0.8237037037037037,
      "grad_norm": 0.8469721674919128,
      "learning_rate": 1.831831831831832e-06,
      "loss": 0.11718244105577469,
      "mean_token_accuracy": 0.9603727757930756,
      "num_tokens": 4554752.0,
      "step": 278
    },
    {
      "entropy": 2.2768435925245285,
      "epoch": 0.8266666666666667,
      "grad_norm": 1.0475202798843384,
      "learning_rate": 1.801801801801802e-06,
      "loss": 0.14657334983348846,
      "mean_token_accuracy": 0.9575242176651955,
      "num_tokens": 4571136.0,
      "step": 279
    },
    {
      "entropy": 2.1211865693330765,
      "epoch": 0.8296296296296296,
      "grad_norm": 1.0412694215774536,
      "learning_rate": 1.7717717717717719e-06,
      "loss": 0.2138775885105133,
      "mean_token_accuracy": 0.9455457031726837,
      "num_tokens": 4587520.0,
      "step": 280
    },
    {
      "entropy": 1.8967775255441666,
      "epoch": 0.8325925925925926,
      "grad_norm": 0.7513962984085083,
      "learning_rate": 1.7417417417417418e-06,
      "loss": 0.11539328843355179,
      "mean_token_accuracy": 0.9611426964402199,
      "num_tokens": 4603904.0,
      "step": 281
    },
    {
      "entropy": 2.4314176738262177,
      "epoch": 0.8355555555555556,
      "grad_norm": 1.236851692199707,
      "learning_rate": 1.711711711711712e-06,
      "loss": 0.2368789166212082,
      "mean_token_accuracy": 0.9335590749979019,
      "num_tokens": 4620288.0,
      "step": 282
    },
    {
      "entropy": 1.8090631067752838,
      "epoch": 0.8385185185185186,
      "grad_norm": 0.8214592337608337,
      "learning_rate": 1.681681681681682e-06,
      "loss": 0.14901985228061676,
      "mean_token_accuracy": 0.9552467167377472,
      "num_tokens": 4636672.0,
      "step": 283
    },
    {
      "entropy": 2.4889764189720154,
      "epoch": 0.8414814814814815,
      "grad_norm": 1.0620553493499756,
      "learning_rate": 1.6516516516516517e-06,
      "loss": 0.1789499819278717,
      "mean_token_accuracy": 0.9371338561177254,
      "num_tokens": 4653056.0,
      "step": 284
    },
    {
      "entropy": 2.430898815393448,
      "epoch": 0.8444444444444444,
      "grad_norm": 1.0657001733779907,
      "learning_rate": 1.6216216216216219e-06,
      "loss": 0.1547377109527588,
      "mean_token_accuracy": 0.9515146836638451,
      "num_tokens": 4669440.0,
      "step": 285
    },
    {
      "entropy": 2.0875085443258286,
      "epoch": 0.8474074074074074,
      "grad_norm": 0.7366988658905029,
      "learning_rate": 1.5915915915915916e-06,
      "loss": 0.11972160637378693,
      "mean_token_accuracy": 0.9585642591118813,
      "num_tokens": 4685824.0,
      "step": 286
    },
    {
      "entropy": 1.9925609827041626,
      "epoch": 0.8503703703703703,
      "grad_norm": 0.9413182139396667,
      "learning_rate": 1.5615615615615617e-06,
      "loss": 0.1366596668958664,
      "mean_token_accuracy": 0.9544644504785538,
      "num_tokens": 4702208.0,
      "step": 287
    },
    {
      "entropy": 2.0718739926815033,
      "epoch": 0.8533333333333334,
      "grad_norm": 0.8501741290092468,
      "learning_rate": 1.5315315315315316e-06,
      "loss": 0.12096701562404633,
      "mean_token_accuracy": 0.9625785499811172,
      "num_tokens": 4718592.0,
      "step": 288
    },
    {
      "entropy": 2.1788668632507324,
      "epoch": 0.8562962962962963,
      "grad_norm": 1.0222554206848145,
      "learning_rate": 1.5015015015015017e-06,
      "loss": 0.16477710008621216,
      "mean_token_accuracy": 0.9502250105142593,
      "num_tokens": 4734976.0,
      "step": 289
    },
    {
      "entropy": 2.0906175076961517,
      "epoch": 0.8592592592592593,
      "grad_norm": 0.7250029444694519,
      "learning_rate": 1.4714714714714714e-06,
      "loss": 0.0923469215631485,
      "mean_token_accuracy": 0.9741655886173248,
      "num_tokens": 4751360.0,
      "step": 290
    },
    {
      "entropy": 2.2420231997966766,
      "epoch": 0.8622222222222222,
      "grad_norm": 0.9206843972206116,
      "learning_rate": 1.4414414414414416e-06,
      "loss": 0.14284971356391907,
      "mean_token_accuracy": 0.9554818272590637,
      "num_tokens": 4767744.0,
      "step": 291
    },
    {
      "entropy": 1.9762412011623383,
      "epoch": 0.8651851851851852,
      "grad_norm": 0.9231134653091431,
      "learning_rate": 1.4114114114114117e-06,
      "loss": 0.151216521859169,
      "mean_token_accuracy": 0.9512001723051071,
      "num_tokens": 4784128.0,
      "step": 292
    },
    {
      "entropy": 2.2375262826681137,
      "epoch": 0.8681481481481481,
      "grad_norm": 0.8700419664382935,
      "learning_rate": 1.3813813813813814e-06,
      "loss": 0.10943577438592911,
      "mean_token_accuracy": 0.9611372575163841,
      "num_tokens": 4800512.0,
      "step": 293
    },
    {
      "entropy": 2.307922139763832,
      "epoch": 0.8711111111111111,
      "grad_norm": 1.2652742862701416,
      "learning_rate": 1.3513513513513515e-06,
      "loss": 0.24527373909950256,
      "mean_token_accuracy": 0.9318349733948708,
      "num_tokens": 4816896.0,
      "step": 294
    },
    {
      "entropy": 1.9223668277263641,
      "epoch": 0.8740740740740741,
      "grad_norm": 0.8681498765945435,
      "learning_rate": 1.3213213213213214e-06,
      "loss": 0.1427430659532547,
      "mean_token_accuracy": 0.9587398767471313,
      "num_tokens": 4833280.0,
      "step": 295
    },
    {
      "entropy": 2.2050732374191284,
      "epoch": 0.8770370370370371,
      "grad_norm": 1.1095103025436401,
      "learning_rate": 1.2912912912912913e-06,
      "loss": 0.18677918612957,
      "mean_token_accuracy": 0.9376562461256981,
      "num_tokens": 4849664.0,
      "step": 296
    },
    {
      "entropy": 1.829752430319786,
      "epoch": 0.88,
      "grad_norm": 0.7530157566070557,
      "learning_rate": 1.2612612612612613e-06,
      "loss": 0.12429338693618774,
      "mean_token_accuracy": 0.9644976407289505,
      "num_tokens": 4866048.0,
      "step": 297
    },
    {
      "entropy": 2.21427358686924,
      "epoch": 0.882962962962963,
      "grad_norm": 0.9627335071563721,
      "learning_rate": 1.2312312312312314e-06,
      "loss": 0.1912863850593567,
      "mean_token_accuracy": 0.9530238881707191,
      "num_tokens": 4882432.0,
      "step": 298
    },
    {
      "entropy": 1.9351110607385635,
      "epoch": 0.8859259259259259,
      "grad_norm": 0.8564191460609436,
      "learning_rate": 1.2012012012012013e-06,
      "loss": 0.15595872700214386,
      "mean_token_accuracy": 0.9551159590482712,
      "num_tokens": 4898816.0,
      "step": 299
    },
    {
      "entropy": 2.0285956412553787,
      "epoch": 0.8888888888888888,
      "grad_norm": 1.0513765811920166,
      "learning_rate": 1.1711711711711712e-06,
      "loss": 0.13914167881011963,
      "mean_token_accuracy": 0.9650851264595985,
      "num_tokens": 4915200.0,
      "step": 300
    },
    {
      "epoch": 0.8888888888888888,
      "eval_entropy": 2.1289974006017047,
      "eval_loss": 0.16315852105617523,
      "eval_mean_token_accuracy": 0.9539673368136088,
      "eval_num_tokens": 4915200.0,
      "eval_runtime": 46.0435,
      "eval_samples_per_second": 13.031,
      "eval_steps_per_second": 1.629,
      "step": 300
    },
    {
      "entropy": 2.2243823409080505,
      "epoch": 0.8918518518518519,
      "grad_norm": 1.0535805225372314,
      "learning_rate": 1.1411411411411411e-06,
      "loss": 0.1894441843032837,
      "mean_token_accuracy": 0.9421443119645119,
      "num_tokens": 4931584.0,
      "step": 301
    },
    {
      "entropy": 2.28277687728405,
      "epoch": 0.8948148148148148,
      "grad_norm": 1.216139316558838,
      "learning_rate": 1.111111111111111e-06,
      "loss": 0.2096002995967865,
      "mean_token_accuracy": 0.9440338909626007,
      "num_tokens": 4947968.0,
      "step": 302
    },
    {
      "entropy": 1.8862746357917786,
      "epoch": 0.8977777777777778,
      "grad_norm": 0.7725651860237122,
      "learning_rate": 1.0810810810810812e-06,
      "loss": 0.12285682559013367,
      "mean_token_accuracy": 0.9725690931081772,
      "num_tokens": 4964352.0,
      "step": 303
    },
    {
      "entropy": 2.3395698815584183,
      "epoch": 0.9007407407407407,
      "grad_norm": 0.957956850528717,
      "learning_rate": 1.051051051051051e-06,
      "loss": 0.19620835781097412,
      "mean_token_accuracy": 0.9431343004107475,
      "num_tokens": 4980736.0,
      "step": 304
    },
    {
      "entropy": 2.096309944987297,
      "epoch": 0.9037037037037037,
      "grad_norm": 0.9171137809753418,
      "learning_rate": 1.0210210210210212e-06,
      "loss": 0.13535550236701965,
      "mean_token_accuracy": 0.9602317288517952,
      "num_tokens": 4997120.0,
      "step": 305
    },
    {
      "entropy": 2.459953263401985,
      "epoch": 0.9066666666666666,
      "grad_norm": 1.248123049736023,
      "learning_rate": 9.909909909909911e-07,
      "loss": 0.14553408324718475,
      "mean_token_accuracy": 0.9519344419240952,
      "num_tokens": 5013504.0,
      "step": 306
    },
    {
      "entropy": 2.0497602075338364,
      "epoch": 0.9096296296296297,
      "grad_norm": 0.6765176653862,
      "learning_rate": 9.60960960960961e-07,
      "loss": 0.10876749455928802,
      "mean_token_accuracy": 0.9702319800853729,
      "num_tokens": 5029888.0,
      "step": 307
    },
    {
      "entropy": 2.1265750378370285,
      "epoch": 0.9125925925925926,
      "grad_norm": 1.310102105140686,
      "learning_rate": 9.30930930930931e-07,
      "loss": 0.2542022168636322,
      "mean_token_accuracy": 0.9320131316781044,
      "num_tokens": 5046272.0,
      "step": 308
    },
    {
      "entropy": 2.1615616977214813,
      "epoch": 0.9155555555555556,
      "grad_norm": 1.1663780212402344,
      "learning_rate": 9.00900900900901e-07,
      "loss": 0.22287176549434662,
      "mean_token_accuracy": 0.9356193542480469,
      "num_tokens": 5062656.0,
      "step": 309
    },
    {
      "entropy": 2.0678174644708633,
      "epoch": 0.9185185185185185,
      "grad_norm": 1.0281835794448853,
      "learning_rate": 8.708708708708709e-07,
      "loss": 0.1608068346977234,
      "mean_token_accuracy": 0.9438650831580162,
      "num_tokens": 5079040.0,
      "step": 310
    },
    {
      "entropy": 1.9104770570993423,
      "epoch": 0.9214814814814815,
      "grad_norm": 0.7292677760124207,
      "learning_rate": 8.40840840840841e-07,
      "loss": 0.11529237031936646,
      "mean_token_accuracy": 0.9619604721665382,
      "num_tokens": 5095424.0,
      "step": 311
    },
    {
      "entropy": 2.068757176399231,
      "epoch": 0.9244444444444444,
      "grad_norm": 0.9124966263771057,
      "learning_rate": 8.108108108108109e-07,
      "loss": 0.12017247080802917,
      "mean_token_accuracy": 0.9675944149494171,
      "num_tokens": 5111808.0,
      "step": 312
    },
    {
      "entropy": 2.2241465747356415,
      "epoch": 0.9274074074074075,
      "grad_norm": 1.1473603248596191,
      "learning_rate": 7.807807807807808e-07,
      "loss": 0.19720472395420074,
      "mean_token_accuracy": 0.9395891204476357,
      "num_tokens": 5128192.0,
      "step": 313
    },
    {
      "entropy": 2.162980794906616,
      "epoch": 0.9303703703703704,
      "grad_norm": 0.8037374019622803,
      "learning_rate": 7.507507507507509e-07,
      "loss": 0.1078617125749588,
      "mean_token_accuracy": 0.9652191251516342,
      "num_tokens": 5144576.0,
      "step": 314
    },
    {
      "entropy": 2.26632422208786,
      "epoch": 0.9333333333333333,
      "grad_norm": 0.9895154237747192,
      "learning_rate": 7.207207207207208e-07,
      "loss": 0.1747983992099762,
      "mean_token_accuracy": 0.9484885483980179,
      "num_tokens": 5160960.0,
      "step": 315
    },
    {
      "entropy": 1.9516086429357529,
      "epoch": 0.9362962962962963,
      "grad_norm": 1.01240873336792,
      "learning_rate": 6.906906906906907e-07,
      "loss": 0.2307746261358261,
      "mean_token_accuracy": 0.9373140260577202,
      "num_tokens": 5177344.0,
      "step": 316
    },
    {
      "entropy": 2.229855254292488,
      "epoch": 0.9392592592592592,
      "grad_norm": 1.1963154077529907,
      "learning_rate": 6.606606606606607e-07,
      "loss": 0.17771650850772858,
      "mean_token_accuracy": 0.9508583545684814,
      "num_tokens": 5193728.0,
      "step": 317
    },
    {
      "entropy": 2.361253499984741,
      "epoch": 0.9422222222222222,
      "grad_norm": 1.0119667053222656,
      "learning_rate": 6.306306306306306e-07,
      "loss": 0.16228321194648743,
      "mean_token_accuracy": 0.9524679556488991,
      "num_tokens": 5210112.0,
      "step": 318
    },
    {
      "entropy": 2.080986261367798,
      "epoch": 0.9451851851851852,
      "grad_norm": 0.8841698169708252,
      "learning_rate": 6.006006006006006e-07,
      "loss": 0.16172832250595093,
      "mean_token_accuracy": 0.9546033516526222,
      "num_tokens": 5226496.0,
      "step": 319
    },
    {
      "entropy": 2.3255287408828735,
      "epoch": 0.9481481481481482,
      "grad_norm": 0.9623970985412598,
      "learning_rate": 5.705705705705706e-07,
      "loss": 0.14459393918514252,
      "mean_token_accuracy": 0.9667885974049568,
      "num_tokens": 5242880.0,
      "step": 320
    },
    {
      "entropy": 2.230205252766609,
      "epoch": 0.9511111111111111,
      "grad_norm": 0.8823454976081848,
      "learning_rate": 5.405405405405406e-07,
      "loss": 0.15521474182605743,
      "mean_token_accuracy": 0.9544969797134399,
      "num_tokens": 5259264.0,
      "step": 321
    },
    {
      "entropy": 2.3776808977127075,
      "epoch": 0.9540740740740741,
      "grad_norm": 0.9496967196464539,
      "learning_rate": 5.105105105105106e-07,
      "loss": 0.16865761578083038,
      "mean_token_accuracy": 0.9501723572611809,
      "num_tokens": 5275648.0,
      "step": 322
    },
    {
      "entropy": 2.004732206463814,
      "epoch": 0.957037037037037,
      "grad_norm": 0.8436097502708435,
      "learning_rate": 4.804804804804805e-07,
      "loss": 0.15380743145942688,
      "mean_token_accuracy": 0.9575323164463043,
      "num_tokens": 5292032.0,
      "step": 323
    },
    {
      "entropy": 2.209444969892502,
      "epoch": 0.96,
      "grad_norm": 0.9449238181114197,
      "learning_rate": 4.504504504504505e-07,
      "loss": 0.12582997977733612,
      "mean_token_accuracy": 0.9606783166527748,
      "num_tokens": 5308416.0,
      "step": 324
    },
    {
      "entropy": 1.969630315899849,
      "epoch": 0.9629629629629629,
      "grad_norm": 0.7218546867370605,
      "learning_rate": 4.204204204204205e-07,
      "loss": 0.1140923872590065,
      "mean_token_accuracy": 0.9729727059602737,
      "num_tokens": 5324800.0,
      "step": 325
    },
    {
      "entropy": 2.101584643125534,
      "epoch": 0.965925925925926,
      "grad_norm": 1.0329421758651733,
      "learning_rate": 3.903903903903904e-07,
      "loss": 0.19542184472084045,
      "mean_token_accuracy": 0.9496479779481888,
      "num_tokens": 5341184.0,
      "step": 326
    },
    {
      "entropy": 1.9905516058206558,
      "epoch": 0.9688888888888889,
      "grad_norm": 0.7505958080291748,
      "learning_rate": 3.603603603603604e-07,
      "loss": 0.1095927506685257,
      "mean_token_accuracy": 0.9744936227798462,
      "num_tokens": 5357568.0,
      "step": 327
    },
    {
      "entropy": 2.0378862619400024,
      "epoch": 0.9718518518518519,
      "grad_norm": 1.1044785976409912,
      "learning_rate": 3.3033033033033036e-07,
      "loss": 0.20492230355739594,
      "mean_token_accuracy": 0.9419190138578415,
      "num_tokens": 5373952.0,
      "step": 328
    },
    {
      "entropy": 2.187395453453064,
      "epoch": 0.9748148148148148,
      "grad_norm": 1.0248366594314575,
      "learning_rate": 3.003003003003003e-07,
      "loss": 0.182743102312088,
      "mean_token_accuracy": 0.9471024870872498,
      "num_tokens": 5390336.0,
      "step": 329
    },
    {
      "entropy": 2.4062585532665253,
      "epoch": 0.9777777777777777,
      "grad_norm": 0.8741236925125122,
      "learning_rate": 2.702702702702703e-07,
      "loss": 0.1125425323843956,
      "mean_token_accuracy": 0.966920331120491,
      "num_tokens": 5406720.0,
      "step": 330
    },
    {
      "entropy": 2.3566349744796753,
      "epoch": 0.9807407407407407,
      "grad_norm": 1.1644935607910156,
      "learning_rate": 2.4024024024024026e-07,
      "loss": 0.2482391893863678,
      "mean_token_accuracy": 0.9393013268709183,
      "num_tokens": 5423104.0,
      "step": 331
    },
    {
      "entropy": 1.9211042374372482,
      "epoch": 0.9837037037037037,
      "grad_norm": 0.8591116070747375,
      "learning_rate": 2.1021021021021025e-07,
      "loss": 0.1710982322692871,
      "mean_token_accuracy": 0.9416057541966438,
      "num_tokens": 5439488.0,
      "step": 332
    },
    {
      "entropy": 2.103690877556801,
      "epoch": 0.9866666666666667,
      "grad_norm": 1.0648715496063232,
      "learning_rate": 1.801801801801802e-07,
      "loss": 0.14113223552703857,
      "mean_token_accuracy": 0.9545880630612373,
      "num_tokens": 5455872.0,
      "step": 333
    },
    {
      "entropy": 1.92289400100708,
      "epoch": 0.9896296296296296,
      "grad_norm": 0.8167691230773926,
      "learning_rate": 1.5015015015015016e-07,
      "loss": 0.12173149734735489,
      "mean_token_accuracy": 0.9657229334115982,
      "num_tokens": 5472256.0,
      "step": 334
    },
    {
      "entropy": 2.223811000585556,
      "epoch": 0.9925925925925926,
      "grad_norm": 1.20535147190094,
      "learning_rate": 1.2012012012012013e-07,
      "loss": 0.20519784092903137,
      "mean_token_accuracy": 0.9467141255736351,
      "num_tokens": 5488640.0,
      "step": 335
    },
    {
      "entropy": 2.3554239869117737,
      "epoch": 0.9955555555555555,
      "grad_norm": 1.3575146198272705,
      "learning_rate": 9.00900900900901e-08,
      "loss": 0.23333318531513214,
      "mean_token_accuracy": 0.9375893771648407,
      "num_tokens": 5505024.0,
      "step": 336
    },
    {
      "entropy": 2.1689008474349976,
      "epoch": 0.9985185185185185,
      "grad_norm": 1.091620922088623,
      "learning_rate": 6.006006006006006e-08,
      "loss": 0.17526425421237946,
      "mean_token_accuracy": 0.9516578018665314,
      "num_tokens": 5521408.0,
      "step": 337
    },
    {
      "entropy": 2.229114055633545,
      "epoch": 1.0,
      "grad_norm": 1.4628632068634033,
      "learning_rate": 3.003003003003003e-08,
      "loss": 0.1733732372522354,
      "mean_token_accuracy": 0.9446267187595367,
      "num_tokens": 5529600.0,
      "step": 338
    }
  ],
  "logging_steps": 1,
  "max_steps": 338,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.53130594336768e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}