{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.010782276632785,
  "eval_steps": 3000,
  "global_step": 9000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.691954421997071,
      "epoch": 0.0005615769079575448,
      "grad_norm": 13.75,
      "learning_rate": 2e-06,
      "loss": 10.8494,
      "mean_token_accuracy": 0.0,
      "num_tokens": 10897.0,
      "step": 5
    },
    {
      "entropy": 10.692028713226318,
      "epoch": 0.0011231538159150896,
      "grad_norm": 13.375,
      "learning_rate": 4.5e-06,
      "loss": 10.783,
      "mean_token_accuracy": 7.102272938936949e-05,
      "num_tokens": 23084.0,
      "step": 10
    },
    {
      "entropy": 10.69119052886963,
      "epoch": 0.0016847307238726343,
      "grad_norm": 10.5,
      "learning_rate": 7e-06,
      "loss": 10.5147,
      "mean_token_accuracy": 0.01880119484849274,
      "num_tokens": 35810.0,
      "step": 15
    },
    {
      "entropy": 10.683763885498047,
      "epoch": 0.0022463076318301792,
      "grad_norm": 6.15625,
      "learning_rate": 9.5e-06,
      "loss": 10.1375,
      "mean_token_accuracy": 0.03672755789011717,
      "num_tokens": 46019.0,
      "step": 20
    },
    {
      "entropy": 10.640995311737061,
      "epoch": 0.002807884539787724,
      "grad_norm": 4.375,
      "learning_rate": 1.2e-05,
      "loss": 9.8065,
      "mean_token_accuracy": 0.0391056377440691,
      "num_tokens": 57272.0,
      "step": 25
    },
    {
      "entropy": 10.578758907318115,
      "epoch": 0.0033694614477452686,
      "grad_norm": 3.28125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.6789,
      "mean_token_accuracy": 0.036402178928256035,
      "num_tokens": 68686.0,
      "step": 30
    },
    {
      "entropy": 10.573095703125,
      "epoch": 0.003931038355702813,
      "grad_norm": 2.90625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.5714,
      "mean_token_accuracy": 0.04141928404569626,
      "num_tokens": 79187.0,
      "step": 35
    },
    {
      "entropy": 10.575594806671143,
      "epoch": 0.0044926152636603585,
      "grad_norm": 2.75,
      "learning_rate": 1.95e-05,
      "loss": 9.5576,
      "mean_token_accuracy": 0.044133822247385976,
      "num_tokens": 90293.0,
      "step": 40
    },
    {
      "entropy": 10.556080150604249,
      "epoch": 0.005054192171617903,
      "grad_norm": 2.578125,
      "learning_rate": 2.2e-05,
      "loss": 9.4673,
      "mean_token_accuracy": 0.04080899097025394,
      "num_tokens": 100420.0,
      "step": 45
    },
    {
      "entropy": 10.538790798187256,
      "epoch": 0.005615769079575448,
      "grad_norm": 2.515625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.4858,
      "mean_token_accuracy": 0.03646625243127346,
      "num_tokens": 111198.0,
      "step": 50
    },
    {
      "entropy": 10.547248840332031,
      "epoch": 0.006177345987532993,
      "grad_norm": 2.34375,
      "learning_rate": 2.7e-05,
      "loss": 9.3988,
      "mean_token_accuracy": 0.05780095718801022,
      "num_tokens": 124723.0,
      "step": 55
    },
    {
      "entropy": 10.461988258361817,
      "epoch": 0.006738922895490537,
      "grad_norm": 2.46875,
      "learning_rate": 2.95e-05,
      "loss": 9.2587,
      "mean_token_accuracy": 0.061974577233195306,
      "num_tokens": 136366.0,
      "step": 60
    },
    {
      "entropy": 10.363299369812012,
      "epoch": 0.007300499803448083,
      "grad_norm": 2.4375,
      "learning_rate": 3.2e-05,
      "loss": 9.1817,
      "mean_token_accuracy": 0.06283607967197895,
      "num_tokens": 146536.0,
      "step": 65
    },
    {
      "entropy": 10.396018791198731,
      "epoch": 0.007862076711405626,
      "grad_norm": 2.21875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.097,
      "mean_token_accuracy": 0.0707736562937498,
      "num_tokens": 157005.0,
      "step": 70
    },
    {
      "entropy": 10.386055850982666,
      "epoch": 0.008423653619363172,
      "grad_norm": 2.328125,
      "learning_rate": 3.7e-05,
      "loss": 9.017,
      "mean_token_accuracy": 0.07136933580040931,
      "num_tokens": 167864.0,
      "step": 75
    },
    {
      "entropy": 10.352816581726074,
      "epoch": 0.008985230527320717,
      "grad_norm": 2.5625,
      "learning_rate": 3.95e-05,
      "loss": 8.9484,
      "mean_token_accuracy": 0.06433885768055916,
      "num_tokens": 179635.0,
      "step": 80
    },
    {
      "entropy": 10.24338674545288,
      "epoch": 0.00954680743527826,
      "grad_norm": 2.671875,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 8.8572,
      "mean_token_accuracy": 0.07211834266781807,
      "num_tokens": 190614.0,
      "step": 85
    },
    {
      "entropy": 10.245021152496339,
      "epoch": 0.010108384343235806,
      "grad_norm": 2.21875,
      "learning_rate": 4.45e-05,
      "loss": 8.7198,
      "mean_token_accuracy": 0.07616456523537636,
      "num_tokens": 201496.0,
      "step": 90
    },
    {
      "entropy": 10.094779968261719,
      "epoch": 0.010669961251193351,
      "grad_norm": 2.46875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.6055,
      "mean_token_accuracy": 0.0780682846903801,
      "num_tokens": 211606.0,
      "step": 95
    },
    {
      "entropy": 10.115865230560303,
      "epoch": 0.011231538159150897,
      "grad_norm": 2.625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.4856,
      "mean_token_accuracy": 0.07667104676365852,
      "num_tokens": 223503.0,
      "step": 100
    },
    {
      "entropy": 10.032721042633057,
      "epoch": 0.01179311506710844,
      "grad_norm": 2.421875,
      "learning_rate": 5.2e-05,
      "loss": 8.4196,
      "mean_token_accuracy": 0.0808894768357277,
      "num_tokens": 234744.0,
      "step": 105
    },
    {
      "entropy": 9.90083703994751,
      "epoch": 0.012354691975065986,
      "grad_norm": 1.7890625,
      "learning_rate": 5.45e-05,
      "loss": 8.3419,
      "mean_token_accuracy": 0.08203728273510932,
      "num_tokens": 247663.0,
      "step": 110
    },
    {
      "entropy": 9.945332717895507,
      "epoch": 0.012916268883023531,
      "grad_norm": 1.96875,
      "learning_rate": 5.7e-05,
      "loss": 8.2013,
      "mean_token_accuracy": 0.08452324643731117,
      "num_tokens": 260326.0,
      "step": 115
    },
    {
      "entropy": 9.736694145202637,
      "epoch": 0.013477845790981074,
      "grad_norm": 2.1875,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.0563,
      "mean_token_accuracy": 0.0793469488620758,
      "num_tokens": 271568.0,
      "step": 120
    },
    {
      "entropy": 9.62724199295044,
      "epoch": 0.01403942269893862,
      "grad_norm": 2.359375,
      "learning_rate": 6.2e-05,
      "loss": 7.8825,
      "mean_token_accuracy": 0.08820232152938842,
      "num_tokens": 282651.0,
      "step": 125
    },
    {
      "entropy": 9.494030666351318,
      "epoch": 0.014600999606896165,
      "grad_norm": 1.96875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.719,
      "mean_token_accuracy": 0.08837165534496308,
      "num_tokens": 292753.0,
      "step": 130
    },
    {
      "entropy": 9.272327136993407,
      "epoch": 0.015162576514853709,
      "grad_norm": 1.609375,
      "learning_rate": 6.7e-05,
      "loss": 7.7167,
      "mean_token_accuracy": 0.0898565836250782,
      "num_tokens": 302981.0,
      "step": 135
    },
    {
      "entropy": 9.16934871673584,
      "epoch": 0.015724153422811252,
      "grad_norm": 1.5,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.6775,
      "mean_token_accuracy": 0.08807501792907715,
      "num_tokens": 314461.0,
      "step": 140
    },
    {
      "entropy": 8.914916038513184,
      "epoch": 0.0162857303307688,
      "grad_norm": 1.6796875,
      "learning_rate": 7.2e-05,
      "loss": 7.6052,
      "mean_token_accuracy": 0.09032118767499923,
      "num_tokens": 326877.0,
      "step": 145
    },
    {
      "entropy": 8.733859825134278,
      "epoch": 0.016847307238726343,
      "grad_norm": 1.5546875,
      "learning_rate": 7.45e-05,
      "loss": 7.3998,
      "mean_token_accuracy": 0.09276190958917141,
      "num_tokens": 337831.0,
      "step": 150
    },
    {
      "entropy": 8.528983402252198,
      "epoch": 0.017408884146683887,
      "grad_norm": 1.5703125,
      "learning_rate": 7.7e-05,
      "loss": 7.337,
      "mean_token_accuracy": 0.09177347123622895,
      "num_tokens": 348233.0,
      "step": 155
    },
    {
      "entropy": 8.35361089706421,
      "epoch": 0.017970461054641434,
      "grad_norm": 1.4609375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.2213,
      "mean_token_accuracy": 0.09552499204874039,
      "num_tokens": 359113.0,
      "step": 160
    },
    {
      "entropy": 8.23711051940918,
      "epoch": 0.018532037962598977,
      "grad_norm": 2.703125,
      "learning_rate": 8.2e-05,
      "loss": 7.336,
      "mean_token_accuracy": 0.10098176226019859,
      "num_tokens": 370284.0,
      "step": 165
    },
    {
      "entropy": 8.123040294647216,
      "epoch": 0.01909361487055652,
      "grad_norm": 1.7734375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.3538,
      "mean_token_accuracy": 0.08904306292533874,
      "num_tokens": 381415.0,
      "step": 170
    },
    {
      "entropy": 7.9991672992706295,
      "epoch": 0.019655191778514068,
      "grad_norm": 1.34375,
      "learning_rate": 8.7e-05,
      "loss": 7.2764,
      "mean_token_accuracy": 0.09914608523249627,
      "num_tokens": 392080.0,
      "step": 175
    },
    {
      "entropy": 7.944419240951538,
      "epoch": 0.020216768686471612,
      "grad_norm": 1.3359375,
      "learning_rate": 8.95e-05,
      "loss": 7.2597,
      "mean_token_accuracy": 0.09534113705158234,
      "num_tokens": 402551.0,
      "step": 180
    },
    {
      "entropy": 7.984531497955322,
      "epoch": 0.020778345594429155,
      "grad_norm": 1.296875,
      "learning_rate": 9.2e-05,
      "loss": 7.2418,
      "mean_token_accuracy": 0.09288135915994644,
      "num_tokens": 414670.0,
      "step": 185
    },
    {
      "entropy": 7.863457107543946,
      "epoch": 0.021339922502386702,
      "grad_norm": 1.1484375,
      "learning_rate": 9.45e-05,
      "loss": 7.2269,
      "mean_token_accuracy": 0.09570453688502312,
      "num_tokens": 427162.0,
      "step": 190
    },
    {
      "entropy": 7.844120883941651,
      "epoch": 0.021901499410344246,
      "grad_norm": 1.4296875,
      "learning_rate": 9.7e-05,
      "loss": 7.125,
      "mean_token_accuracy": 0.09662900492548943,
      "num_tokens": 438628.0,
      "step": 195
    },
    {
      "entropy": 7.827740716934204,
      "epoch": 0.022463076318301793,
      "grad_norm": 1.3984375,
      "learning_rate": 9.95e-05,
      "loss": 7.0534,
      "mean_token_accuracy": 0.10476038753986358,
      "num_tokens": 449248.0,
      "step": 200
    },
    {
      "entropy": 7.613120603561401,
      "epoch": 0.023024653226259337,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000102,
      "loss": 7.0291,
      "mean_token_accuracy": 0.10421562120318413,
      "num_tokens": 460815.0,
      "step": 205
    },
    {
      "entropy": 7.653258895874023,
      "epoch": 0.02358623013421688,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.1939,
      "mean_token_accuracy": 0.09638625457882881,
      "num_tokens": 471972.0,
      "step": 210
    },
    {
      "entropy": 7.694521474838257,
      "epoch": 0.024147807042174427,
      "grad_norm": 1.265625,
      "learning_rate": 0.000107,
      "loss": 7.1589,
      "mean_token_accuracy": 0.1014483891427517,
      "num_tokens": 482775.0,
      "step": 215
    },
    {
      "entropy": 7.650409984588623,
      "epoch": 0.02470938395013197,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001095,
      "loss": 7.087,
      "mean_token_accuracy": 0.10187960937619209,
      "num_tokens": 495151.0,
      "step": 220
    },
    {
      "entropy": 7.605823087692261,
      "epoch": 0.025270960858089515,
      "grad_norm": 1.328125,
      "learning_rate": 0.000112,
      "loss": 7.0608,
      "mean_token_accuracy": 0.10176030248403549,
      "num_tokens": 506947.0,
      "step": 225
    },
    {
      "entropy": 7.604726552963257,
      "epoch": 0.025832537766047062,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001145,
      "loss": 7.0892,
      "mean_token_accuracy": 0.10170613378286361,
      "num_tokens": 519153.0,
      "step": 230
    },
    {
      "entropy": 7.570750951766968,
      "epoch": 0.026394114674004605,
      "grad_norm": 1.21875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.07,
      "mean_token_accuracy": 0.10081414431333542,
      "num_tokens": 531050.0,
      "step": 235
    },
    {
      "entropy": 7.4688334465026855,
      "epoch": 0.02695569158196215,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.006,
      "mean_token_accuracy": 0.10289591774344445,
      "num_tokens": 542744.0,
      "step": 240
    },
    {
      "entropy": 7.563732862472534,
      "epoch": 0.027517268489919696,
      "grad_norm": 1.453125,
      "learning_rate": 0.000122,
      "loss": 7.0923,
      "mean_token_accuracy": 0.0983924463391304,
      "num_tokens": 554748.0,
      "step": 245
    },
    {
      "entropy": 7.534366416931152,
      "epoch": 0.02807884539787724,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001245,
      "loss": 7.0106,
      "mean_token_accuracy": 0.10552474185824394,
      "num_tokens": 564868.0,
      "step": 250
    },
    {
      "entropy": 7.530099248886108,
      "epoch": 0.028640422305834783,
      "grad_norm": 1.375,
      "learning_rate": 0.000127,
      "loss": 7.0091,
      "mean_token_accuracy": 0.11061361208558082,
      "num_tokens": 575668.0,
      "step": 255
    },
    {
      "entropy": 7.455753898620605,
      "epoch": 0.02920199921379233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001295,
      "loss": 7.1352,
      "mean_token_accuracy": 0.10184136554598808,
      "num_tokens": 587956.0,
      "step": 260
    },
    {
      "entropy": 7.4932270526885985,
      "epoch": 0.029763576121749874,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000132,
      "loss": 6.9785,
      "mean_token_accuracy": 0.10903615280985832,
      "num_tokens": 598246.0,
      "step": 265
    },
    {
      "entropy": 7.416919708251953,
      "epoch": 0.030325153029707418,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.1108,
      "mean_token_accuracy": 0.10414381325244904,
      "num_tokens": 610628.0,
      "step": 270
    },
    {
      "entropy": 7.3690831661224365,
      "epoch": 0.030886729937664965,
      "grad_norm": 1.453125,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.8616,
      "mean_token_accuracy": 0.1097691647708416,
      "num_tokens": 621214.0,
      "step": 275
    },
    {
      "entropy": 7.360501384735107,
      "epoch": 0.031448306845622505,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001395,
      "loss": 6.957,
      "mean_token_accuracy": 0.1151120163500309,
      "num_tokens": 631123.0,
      "step": 280
    },
    {
      "entropy": 7.397417736053467,
      "epoch": 0.032009883753580055,
      "grad_norm": 1.296875,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.9474,
      "mean_token_accuracy": 0.10703005790710449,
      "num_tokens": 643434.0,
      "step": 285
    },
    {
      "entropy": 7.342480707168579,
      "epoch": 0.0325714606615376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001445,
      "loss": 6.9321,
      "mean_token_accuracy": 0.11211270317435265,
      "num_tokens": 655336.0,
      "step": 290
    },
    {
      "entropy": 7.372163486480713,
      "epoch": 0.03313303756949514,
      "grad_norm": 1.5,
      "learning_rate": 0.000147,
      "loss": 6.9245,
      "mean_token_accuracy": 0.10963715985417366,
      "num_tokens": 666591.0,
      "step": 295
    },
    {
      "entropy": 7.286645698547363,
      "epoch": 0.033694614477452686,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001495,
      "loss": 6.9458,
      "mean_token_accuracy": 0.10590762570500374,
      "num_tokens": 678288.0,
      "step": 300
    },
    {
      "entropy": 7.292938613891602,
      "epoch": 0.03425619138541023,
      "grad_norm": 1.21875,
      "learning_rate": 0.000152,
      "loss": 6.8788,
      "mean_token_accuracy": 0.11080723032355308,
      "num_tokens": 688526.0,
      "step": 305
    },
    {
      "entropy": 7.20158314704895,
      "epoch": 0.03481776829336777,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.7955,
      "mean_token_accuracy": 0.12063507586717606,
      "num_tokens": 698727.0,
      "step": 310
    },
    {
      "entropy": 7.281469917297363,
      "epoch": 0.035379345201325324,
      "grad_norm": 1.359375,
      "learning_rate": 0.000157,
      "loss": 6.8411,
      "mean_token_accuracy": 0.1111941859126091,
      "num_tokens": 710042.0,
      "step": 315
    },
    {
      "entropy": 7.2135881900787355,
      "epoch": 0.03594092210928287,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001595,
      "loss": 6.7838,
      "mean_token_accuracy": 0.11598038524389268,
      "num_tokens": 721390.0,
      "step": 320
    },
    {
      "entropy": 7.166607904434204,
      "epoch": 0.03650249901724041,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000162,
      "loss": 6.9676,
      "mean_token_accuracy": 0.11324010118842125,
      "num_tokens": 732949.0,
      "step": 325
    },
    {
      "entropy": 7.382533311843872,
      "epoch": 0.037064075925197955,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.983,
      "mean_token_accuracy": 0.11112697571516036,
      "num_tokens": 744943.0,
      "step": 330
    },
    {
      "entropy": 7.1175624370574955,
      "epoch": 0.0376256528331555,
      "grad_norm": 1.265625,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.7549,
      "mean_token_accuracy": 0.11743907704949379,
      "num_tokens": 756579.0,
      "step": 335
    },
    {
      "entropy": 7.059293413162232,
      "epoch": 0.03818722974111304,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.7064,
      "mean_token_accuracy": 0.11693573296070099,
      "num_tokens": 767260.0,
      "step": 340
    },
    {
      "entropy": 7.129382991790772,
      "epoch": 0.03874880664907059,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.8055,
      "mean_token_accuracy": 0.11503136828541756,
      "num_tokens": 778163.0,
      "step": 345
    },
    {
      "entropy": 7.21411828994751,
      "epoch": 0.039310383557028136,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.841,
      "mean_token_accuracy": 0.11127241551876069,
      "num_tokens": 788912.0,
      "step": 350
    },
    {
      "entropy": 7.170050191879272,
      "epoch": 0.03987196046498568,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000177,
      "loss": 6.9508,
      "mean_token_accuracy": 0.1135722354054451,
      "num_tokens": 801405.0,
      "step": 355
    },
    {
      "entropy": 7.058458709716797,
      "epoch": 0.040433537372943223,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001795,
      "loss": 6.7437,
      "mean_token_accuracy": 0.11591695919632912,
      "num_tokens": 813023.0,
      "step": 360
    },
    {
      "entropy": 7.132532978057862,
      "epoch": 0.04099511428090077,
      "grad_norm": 1.15625,
      "learning_rate": 0.000182,
      "loss": 6.7692,
      "mean_token_accuracy": 0.1168873094022274,
      "num_tokens": 823993.0,
      "step": 365
    },
    {
      "entropy": 7.123906469345092,
      "epoch": 0.04155669118885831,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001845,
      "loss": 6.8202,
      "mean_token_accuracy": 0.1134017400443554,
      "num_tokens": 836642.0,
      "step": 370
    },
    {
      "entropy": 7.020151281356812,
      "epoch": 0.04211826809681586,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000187,
      "loss": 6.6249,
      "mean_token_accuracy": 0.12558069303631783,
      "num_tokens": 847475.0,
      "step": 375
    },
    {
      "entropy": 7.032984447479248,
      "epoch": 0.042679845004773405,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001895,
      "loss": 6.6947,
      "mean_token_accuracy": 0.12115299850702285,
      "num_tokens": 857949.0,
      "step": 380
    },
    {
      "entropy": 7.027848196029663,
      "epoch": 0.04324142191273095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000192,
      "loss": 6.7528,
      "mean_token_accuracy": 0.11466690823435784,
      "num_tokens": 868661.0,
      "step": 385
    },
    {
      "entropy": 7.095096111297607,
      "epoch": 0.04380299882068849,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0001945,
      "loss": 6.7906,
      "mean_token_accuracy": 0.11758541241288185,
      "num_tokens": 879651.0,
      "step": 390
    },
    {
      "entropy": 7.0685296058654785,
      "epoch": 0.044364575728646036,
      "grad_norm": 1.046875,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.7722,
      "mean_token_accuracy": 0.10885974839329719,
      "num_tokens": 891189.0,
      "step": 395
    },
    {
      "entropy": 6.994202566146851,
      "epoch": 0.044926152636603586,
      "grad_norm": 1.125,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.6748,
      "mean_token_accuracy": 0.119440957903862,
      "num_tokens": 902915.0,
      "step": 400
    },
    {
      "entropy": 7.045473241806031,
      "epoch": 0.04548772954456113,
      "grad_norm": 1.453125,
      "learning_rate": 0.000202,
      "loss": 6.7649,
      "mean_token_accuracy": 0.11777753084897995,
      "num_tokens": 914108.0,
      "step": 405
    },
    {
      "entropy": 7.049249458312988,
      "epoch": 0.046049306452518673,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.7918,
      "mean_token_accuracy": 0.11295270174741745,
      "num_tokens": 925729.0,
      "step": 410
    },
    {
      "entropy": 7.0732354640960695,
      "epoch": 0.04661088336047622,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000207,
      "loss": 6.7993,
      "mean_token_accuracy": 0.11720411777496338,
      "num_tokens": 936912.0,
      "step": 415
    },
    {
      "entropy": 6.98739104270935,
      "epoch": 0.04717246026843376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002095,
      "loss": 6.7739,
      "mean_token_accuracy": 0.11668220460414887,
      "num_tokens": 948576.0,
      "step": 420
    },
    {
      "entropy": 6.989003896713257,
      "epoch": 0.047734037176391304,
      "grad_norm": 1.15625,
      "learning_rate": 0.000212,
      "loss": 6.6228,
      "mean_token_accuracy": 0.12309176921844482,
      "num_tokens": 959346.0,
      "step": 425
    },
    {
      "entropy": 6.867386531829834,
      "epoch": 0.048295614084348855,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002145,
      "loss": 6.6393,
      "mean_token_accuracy": 0.12302001640200615,
      "num_tokens": 970124.0,
      "step": 430
    },
    {
      "entropy": 7.04110198020935,
      "epoch": 0.0488571909923064,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.7167,
      "mean_token_accuracy": 0.12065900415182114,
      "num_tokens": 981628.0,
      "step": 435
    },
    {
      "entropy": 6.839612150192261,
      "epoch": 0.04941876790026394,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002195,
      "loss": 6.6603,
      "mean_token_accuracy": 0.12483339831233024,
      "num_tokens": 992423.0,
      "step": 440
    },
    {
      "entropy": 7.010668039321899,
      "epoch": 0.049980344808221486,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000222,
      "loss": 6.6703,
      "mean_token_accuracy": 0.11580908223986626,
      "num_tokens": 1002954.0,
      "step": 445
    },
    {
      "entropy": 6.892636823654175,
      "epoch": 0.05054192171617903,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002245,
      "loss": 6.5524,
      "mean_token_accuracy": 0.12802066281437874,
      "num_tokens": 1013351.0,
      "step": 450
    },
    {
      "entropy": 6.97490382194519,
      "epoch": 0.05110349862413657,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.6699,
      "mean_token_accuracy": 0.11892755851149558,
      "num_tokens": 1023781.0,
      "step": 455
    },
    {
      "entropy": 7.029914331436157,
      "epoch": 0.051665075532094124,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.7291,
      "mean_token_accuracy": 0.12224347591400146,
      "num_tokens": 1034565.0,
      "step": 460
    },
    {
      "entropy": 6.768797492980957,
      "epoch": 0.05222665244005167,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.5403,
      "mean_token_accuracy": 0.12486013248562813,
      "num_tokens": 1045595.0,
      "step": 465
    },
    {
      "entropy": 6.811560869216919,
      "epoch": 0.05278822934800921,
      "grad_norm": 1.25,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.5365,
      "mean_token_accuracy": 0.12281112596392632,
      "num_tokens": 1056084.0,
      "step": 470
    },
    {
      "entropy": 6.891090345382691,
      "epoch": 0.053349806255966754,
      "grad_norm": 1.125,
      "learning_rate": 0.000237,
      "loss": 6.5832,
      "mean_token_accuracy": 0.12444683387875558,
      "num_tokens": 1067014.0,
      "step": 475
    },
    {
      "entropy": 6.7909644603729244,
      "epoch": 0.0539113831639243,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002395,
      "loss": 6.549,
      "mean_token_accuracy": 0.1283904306590557,
      "num_tokens": 1078557.0,
      "step": 480
    },
    {
      "entropy": 6.747442579269409,
      "epoch": 0.05447296007188184,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000242,
      "loss": 6.6193,
      "mean_token_accuracy": 0.1250327579677105,
      "num_tokens": 1090541.0,
      "step": 485
    },
    {
      "entropy": 6.929131269454956,
      "epoch": 0.05503453697983939,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002445,
      "loss": 6.6901,
      "mean_token_accuracy": 0.12543862983584403,
      "num_tokens": 1101851.0,
      "step": 490
    },
    {
      "entropy": 6.778433561325073,
      "epoch": 0.055596113887796936,
      "grad_norm": 1.421875,
      "learning_rate": 0.000247,
      "loss": 6.5898,
      "mean_token_accuracy": 0.12135263979434967,
      "num_tokens": 1112728.0,
      "step": 495
    },
    {
      "entropy": 6.916776704788208,
      "epoch": 0.05615769079575448,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002495,
      "loss": 6.7185,
      "mean_token_accuracy": 0.12112508118152618,
      "num_tokens": 1124537.0,
      "step": 500
    },
    {
      "entropy": 6.7057990550994875,
      "epoch": 0.05671926770371202,
      "grad_norm": 1.3125,
      "learning_rate": 0.000252,
      "loss": 6.6177,
      "mean_token_accuracy": 0.11861735060811043,
      "num_tokens": 1135749.0,
      "step": 505
    },
    {
      "entropy": 6.887033128738404,
      "epoch": 0.05728084461166957,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002545,
      "loss": 6.523,
      "mean_token_accuracy": 0.12177179381251335,
      "num_tokens": 1147597.0,
      "step": 510
    },
    {
      "entropy": 6.736017560958862,
      "epoch": 0.05784242151962711,
      "grad_norm": 1.125,
      "learning_rate": 0.000257,
      "loss": 6.483,
      "mean_token_accuracy": 0.13104297444224358,
      "num_tokens": 1158975.0,
      "step": 515
    },
    {
      "entropy": 6.74561619758606,
      "epoch": 0.05840399842758466,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002595,
      "loss": 6.6127,
      "mean_token_accuracy": 0.12473157271742821,
      "num_tokens": 1170556.0,
      "step": 520
    },
    {
      "entropy": 6.734094333648682,
      "epoch": 0.058965575335542204,
      "grad_norm": 1.25,
      "learning_rate": 0.000262,
      "loss": 6.4825,
      "mean_token_accuracy": 0.12628087177872657,
      "num_tokens": 1180757.0,
      "step": 525
    },
    {
      "entropy": 6.786943149566651,
      "epoch": 0.05952715224349975,
      "grad_norm": 1.25,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.5112,
      "mean_token_accuracy": 0.12826752737164498,
      "num_tokens": 1192428.0,
      "step": 530
    },
    {
      "entropy": 6.773551893234253,
      "epoch": 0.06008872915145729,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.5812,
      "mean_token_accuracy": 0.1194170169532299,
      "num_tokens": 1203867.0,
      "step": 535
    },
    {
      "entropy": 6.7591596126556395,
      "epoch": 0.060650306059414835,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.5734,
      "mean_token_accuracy": 0.1305850028991699,
      "num_tokens": 1214059.0,
      "step": 540
    },
    {
      "entropy": 6.792087364196777,
      "epoch": 0.06121188296737238,
      "grad_norm": 1.375,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6338,
      "mean_token_accuracy": 0.13109224960207938,
      "num_tokens": 1224750.0,
      "step": 545
    },
    {
      "entropy": 6.816738271713257,
      "epoch": 0.06177345987532993,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002745,
      "loss": 6.7109,
      "mean_token_accuracy": 0.11931365355849266,
      "num_tokens": 1235936.0,
      "step": 550
    },
    {
      "entropy": 6.697342395782471,
      "epoch": 0.06233503678328747,
      "grad_norm": 1.0625,
      "learning_rate": 0.000277,
      "loss": 6.5229,
      "mean_token_accuracy": 0.12435328289866447,
      "num_tokens": 1247619.0,
      "step": 555
    },
    {
      "entropy": 6.797955465316773,
      "epoch": 0.06289661369124501,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002795,
      "loss": 6.5973,
      "mean_token_accuracy": 0.12734515964984894,
      "num_tokens": 1258089.0,
      "step": 560
    },
    {
      "entropy": 6.7907819747924805,
      "epoch": 0.06345819059920256,
      "grad_norm": 0.984375,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.6714,
      "mean_token_accuracy": 0.12250098884105683,
      "num_tokens": 1270727.0,
      "step": 565
    },
    {
      "entropy": 6.718038558959961,
      "epoch": 0.06401976750716011,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002845,
      "loss": 6.5278,
      "mean_token_accuracy": 0.12575457990169525,
      "num_tokens": 1280886.0,
      "step": 570
    },
    {
      "entropy": 6.816065120697021,
      "epoch": 0.06458134441511765,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000287,
      "loss": 6.5532,
      "mean_token_accuracy": 0.13024218007922173,
      "num_tokens": 1292205.0,
      "step": 575
    },
    {
      "entropy": 6.724169111251831,
      "epoch": 0.0651429213230752,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002895,
      "loss": 6.5387,
      "mean_token_accuracy": 0.12204434722661972,
      "num_tokens": 1304621.0,
      "step": 580
    },
    {
      "entropy": 6.828542423248291,
      "epoch": 0.06570449823103273,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000292,
      "loss": 6.6198,
      "mean_token_accuracy": 0.12465316653251649,
      "num_tokens": 1315481.0,
      "step": 585
    },
    {
      "entropy": 6.641292333602905,
      "epoch": 0.06626607513899029,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002945,
      "loss": 6.6021,
      "mean_token_accuracy": 0.12428974732756615,
      "num_tokens": 1328726.0,
      "step": 590
    },
    {
      "entropy": 6.790183639526367,
      "epoch": 0.06682765204694784,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000297,
      "loss": 6.6603,
      "mean_token_accuracy": 0.12199008911848068,
      "num_tokens": 1340151.0,
      "step": 595
    },
    {
      "entropy": 6.6274354457855225,
      "epoch": 0.06738922895490537,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002995,
      "loss": 6.4411,
      "mean_token_accuracy": 0.12323693186044693,
      "num_tokens": 1350833.0,
      "step": 600
    },
    {
      "entropy": 6.738854312896729,
      "epoch": 0.06795080586286292,
      "grad_norm": 1.234375,
      "learning_rate": 0.000302,
      "loss": 6.6111,
      "mean_token_accuracy": 0.12676038667559625,
      "num_tokens": 1362332.0,
      "step": 605
    },
    {
      "entropy": 6.6895301818847654,
      "epoch": 0.06851238277082046,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003045,
      "loss": 6.5002,
      "mean_token_accuracy": 0.13042355701327324,
      "num_tokens": 1373696.0,
      "step": 610
    },
    {
      "entropy": 6.526501417160034,
      "epoch": 0.06907395967877801,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000307,
      "loss": 6.4218,
      "mean_token_accuracy": 0.1356612041592598,
      "num_tokens": 1383553.0,
      "step": 615
    },
    {
      "entropy": 6.660169887542724,
      "epoch": 0.06963553658673555,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003095,
      "loss": 6.5499,
      "mean_token_accuracy": 0.1267920345067978,
      "num_tokens": 1394487.0,
      "step": 620
    },
    {
      "entropy": 6.846566820144654,
      "epoch": 0.0701971134946931,
      "grad_norm": 1.015625,
      "learning_rate": 0.000312,
      "loss": 6.5606,
      "mean_token_accuracy": 0.13110520616173743,
      "num_tokens": 1404928.0,
      "step": 625
    },
    {
      "entropy": 6.503131103515625,
      "epoch": 0.07075869040265065,
      "grad_norm": 1.25,
      "learning_rate": 0.0003145,
      "loss": 6.4294,
      "mean_token_accuracy": 0.1311935856938362,
      "num_tokens": 1415854.0,
      "step": 630
    },
    {
      "entropy": 6.592182874679565,
      "epoch": 0.07132026731060818,
      "grad_norm": 1.171875,
      "learning_rate": 0.000317,
      "loss": 6.4816,
      "mean_token_accuracy": 0.1329976461827755,
      "num_tokens": 1427153.0,
      "step": 635
    },
    {
      "entropy": 6.661157464981079,
      "epoch": 0.07188184421856574,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003195,
      "loss": 6.5885,
      "mean_token_accuracy": 0.12770819365978242,
      "num_tokens": 1438907.0,
      "step": 640
    },
    {
      "entropy": 6.7143834114074705,
      "epoch": 0.07244342112652327,
      "grad_norm": 1.046875,
      "learning_rate": 0.000322,
      "loss": 6.4669,
      "mean_token_accuracy": 0.12574334740638732,
      "num_tokens": 1449975.0,
      "step": 645
    },
    {
      "entropy": 6.589789390563965,
      "epoch": 0.07300499803448082,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.4557,
      "mean_token_accuracy": 0.12991306632757188,
      "num_tokens": 1460815.0,
      "step": 650
    },
    {
      "entropy": 6.6092170715332035,
      "epoch": 0.07356657494243837,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.5044,
      "mean_token_accuracy": 0.1279452659189701,
      "num_tokens": 1472403.0,
      "step": 655
    },
    {
      "entropy": 6.639557504653931,
      "epoch": 0.07412815185039591,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5141,
      "mean_token_accuracy": 0.12161320745944977,
      "num_tokens": 1484015.0,
      "step": 660
    },
    {
      "entropy": 6.653767776489258,
      "epoch": 0.07468972875835346,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5425,
      "mean_token_accuracy": 0.12279097139835357,
      "num_tokens": 1495383.0,
      "step": 665
    },
    {
      "entropy": 6.608673715591431,
      "epoch": 0.075251305666311,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.468,
      "mean_token_accuracy": 0.1286981873214245,
      "num_tokens": 1505893.0,
      "step": 670
    },
    {
      "entropy": 6.724495363235474,
      "epoch": 0.07581288257426855,
      "grad_norm": 1.15625,
      "learning_rate": 0.000337,
      "loss": 6.6061,
      "mean_token_accuracy": 0.12201442718505859,
      "num_tokens": 1518933.0,
      "step": 675
    },
    {
      "entropy": 6.7100588321685795,
      "epoch": 0.07637445948222608,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003395,
      "loss": 6.6149,
      "mean_token_accuracy": 0.1266277812421322,
      "num_tokens": 1530603.0,
      "step": 680
    },
    {
      "entropy": 6.643559837341309,
      "epoch": 0.07693603639018363,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000342,
      "loss": 6.4551,
      "mean_token_accuracy": 0.131996688246727,
      "num_tokens": 1540977.0,
      "step": 685
    },
    {
      "entropy": 6.6613860607147215,
      "epoch": 0.07749761329814119,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.5621,
      "mean_token_accuracy": 0.12197319865226745,
      "num_tokens": 1553004.0,
      "step": 690
    },
    {
      "entropy": 6.604646968841553,
      "epoch": 0.07805919020609872,
      "grad_norm": 1.140625,
      "learning_rate": 0.000347,
      "loss": 6.4726,
      "mean_token_accuracy": 0.12990233227610587,
      "num_tokens": 1563446.0,
      "step": 695
    },
    {
      "entropy": 6.59793119430542,
      "epoch": 0.07862076711405627,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003495,
      "loss": 6.5153,
      "mean_token_accuracy": 0.1260698601603508,
      "num_tokens": 1573884.0,
      "step": 700
    },
    {
      "entropy": 6.554755544662475,
      "epoch": 0.07918234402201381,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000352,
      "loss": 6.3739,
      "mean_token_accuracy": 0.13849396333098413,
      "num_tokens": 1585396.0,
      "step": 705
    },
    {
      "entropy": 6.4631202697753904,
      "epoch": 0.07974392092997136,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003545,
      "loss": 6.3757,
      "mean_token_accuracy": 0.1374616429209709,
      "num_tokens": 1596934.0,
      "step": 710
    },
    {
      "entropy": 6.508552122116089,
      "epoch": 0.08030549783792891,
      "grad_norm": 1.0625,
      "learning_rate": 0.000357,
      "loss": 6.4243,
      "mean_token_accuracy": 0.13476341739296913,
      "num_tokens": 1608069.0,
      "step": 715
    },
    {
      "entropy": 6.504626750946045,
      "epoch": 0.08086707474588645,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003595,
      "loss": 6.345,
      "mean_token_accuracy": 0.1367340750992298,
      "num_tokens": 1619532.0,
      "step": 720
    },
    {
      "entropy": 6.577574253082275,
      "epoch": 0.081428651653844,
      "grad_norm": 1.15625,
      "learning_rate": 0.000362,
      "loss": 6.5057,
      "mean_token_accuracy": 0.12562305480241776,
      "num_tokens": 1630697.0,
      "step": 725
    },
    {
      "entropy": 6.569473791122436,
      "epoch": 0.08199022856180153,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003645,
      "loss": 6.483,
      "mean_token_accuracy": 0.12953646928071977,
      "num_tokens": 1641854.0,
      "step": 730
    },
    {
      "entropy": 6.676255798339843,
      "epoch": 0.08255180546975908,
      "grad_norm": 1.078125,
      "learning_rate": 0.000367,
      "loss": 6.6127,
      "mean_token_accuracy": 0.1251482017338276,
      "num_tokens": 1653036.0,
      "step": 735
    },
    {
      "entropy": 6.500600099563599,
      "epoch": 0.08311338237771662,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003695,
      "loss": 6.3686,
      "mean_token_accuracy": 0.12898936495184898,
      "num_tokens": 1664096.0,
      "step": 740
    },
    {
      "entropy": 6.575688219070434,
      "epoch": 0.08367495928567417,
      "grad_norm": 0.96875,
      "learning_rate": 0.000372,
      "loss": 6.5405,
      "mean_token_accuracy": 0.12776357308030128,
      "num_tokens": 1677890.0,
      "step": 745
    },
    {
      "entropy": 6.513050270080567,
      "epoch": 0.08423653619363172,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003745,
      "loss": 6.4039,
      "mean_token_accuracy": 0.13092774003744126,
      "num_tokens": 1687627.0,
      "step": 750
    },
    {
      "entropy": 6.6496001243591305,
      "epoch": 0.08479811310158926,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000377,
      "loss": 6.6299,
      "mean_token_accuracy": 0.12669486477971076,
      "num_tokens": 1700843.0,
      "step": 755
    },
    {
      "entropy": 6.54553918838501,
      "epoch": 0.08535969000954681,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003795,
      "loss": 6.4485,
      "mean_token_accuracy": 0.12898636609315872,
      "num_tokens": 1712356.0,
      "step": 760
    },
    {
      "entropy": 6.5033485889434814,
      "epoch": 0.08592126691750435,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000382,
      "loss": 6.44,
      "mean_token_accuracy": 0.1329922027885914,
      "num_tokens": 1722041.0,
      "step": 765
    },
    {
      "entropy": 6.540292739868164,
      "epoch": 0.0864828438254619,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003845,
      "loss": 6.5334,
      "mean_token_accuracy": 0.12715438082814218,
      "num_tokens": 1733628.0,
      "step": 770
    },
    {
      "entropy": 6.52957124710083,
      "epoch": 0.08704442073341945,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.3951,
      "mean_token_accuracy": 0.13516492396593094,
      "num_tokens": 1745174.0,
      "step": 775
    },
    {
      "entropy": 6.472682189941406,
      "epoch": 0.08760599764137698,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.3261,
      "mean_token_accuracy": 0.13572794049978257,
      "num_tokens": 1757117.0,
      "step": 780
    },
    {
      "entropy": 6.4541521072387695,
      "epoch": 0.08816757454933453,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.3477,
      "mean_token_accuracy": 0.13643818646669387,
      "num_tokens": 1767343.0,
      "step": 785
    },
    {
      "entropy": 6.52891354560852,
      "epoch": 0.08872915145729207,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.4341,
      "mean_token_accuracy": 0.1295755423605442,
      "num_tokens": 1778500.0,
      "step": 790
    },
    {
      "entropy": 6.495887565612793,
      "epoch": 0.08929072836524962,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.4257,
      "mean_token_accuracy": 0.13391977846622466,
      "num_tokens": 1788474.0,
      "step": 795
    },
    {
      "entropy": 6.669151258468628,
      "epoch": 0.08985230527320717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003995,
      "loss": 6.6066,
      "mean_token_accuracy": 0.12243209406733513,
      "num_tokens": 1800756.0,
      "step": 800
    },
    {
      "entropy": 6.264161491394043,
      "epoch": 0.09041388218116471,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000402,
      "loss": 6.3051,
      "mean_token_accuracy": 0.13255022764205932,
      "num_tokens": 1812172.0,
      "step": 805
    },
    {
      "entropy": 6.581796360015869,
      "epoch": 0.09097545908912226,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004045,
      "loss": 6.4369,
      "mean_token_accuracy": 0.13564810305833816,
      "num_tokens": 1822057.0,
      "step": 810
    },
    {
      "entropy": 6.388931703567505,
      "epoch": 0.0915370359970798,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.3692,
      "mean_token_accuracy": 0.13979063034057618,
      "num_tokens": 1833074.0,
      "step": 815
    },
    {
      "entropy": 6.4439905166625975,
      "epoch": 0.09209861290503735,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004095,
      "loss": 6.3689,
      "mean_token_accuracy": 0.1369178406894207,
      "num_tokens": 1844932.0,
      "step": 820
    },
    {
      "entropy": 6.451093339920044,
      "epoch": 0.09266018981299488,
      "grad_norm": 1.09375,
      "learning_rate": 0.000412,
      "loss": 6.3812,
      "mean_token_accuracy": 0.13171854838728905,
      "num_tokens": 1855591.0,
      "step": 825
    },
    {
      "entropy": 6.412005949020386,
      "epoch": 0.09322176672095243,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004145,
      "loss": 6.3368,
      "mean_token_accuracy": 0.13641790226101874,
      "num_tokens": 1866726.0,
      "step": 830
    },
    {
      "entropy": 6.4959104537963865,
      "epoch": 0.09378334362890998,
      "grad_norm": 1.140625,
      "learning_rate": 0.000417,
      "loss": 6.3763,
      "mean_token_accuracy": 0.13648990392684937,
      "num_tokens": 1877385.0,
      "step": 835
    },
    {
      "entropy": 6.375614738464355,
      "epoch": 0.09434492053686752,
      "grad_norm": 1.125,
      "learning_rate": 0.0004195,
      "loss": 6.2828,
      "mean_token_accuracy": 0.1361901879310608,
      "num_tokens": 1887905.0,
      "step": 840
    },
    {
      "entropy": 6.32490758895874,
      "epoch": 0.09490649744482507,
      "grad_norm": 1.09375,
      "learning_rate": 0.000422,
      "loss": 6.3418,
      "mean_token_accuracy": 0.13226142525672913,
      "num_tokens": 1898649.0,
      "step": 845
    },
    {
      "entropy": 6.440097618103027,
      "epoch": 0.09546807435278261,
      "grad_norm": 1.25,
      "learning_rate": 0.0004245,
      "loss": 6.3581,
      "mean_token_accuracy": 0.12981755211949347,
      "num_tokens": 1908808.0,
      "step": 850
    },
    {
      "entropy": 6.449653148651123,
      "epoch": 0.09602965126074016,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000427,
      "loss": 6.4636,
      "mean_token_accuracy": 0.13088747188448907,
      "num_tokens": 1920360.0,
      "step": 855
    },
    {
      "entropy": 6.425679159164429,
      "epoch": 0.09659122816869771,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004295,
      "loss": 6.3601,
      "mean_token_accuracy": 0.13659372106194495,
      "num_tokens": 1931082.0,
      "step": 860
    },
    {
      "entropy": 6.450051259994507,
      "epoch": 0.09715280507665525,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000432,
      "loss": 6.5104,
      "mean_token_accuracy": 0.13292263224720954,
      "num_tokens": 1942712.0,
      "step": 865
    },
    {
      "entropy": 6.353779935836792,
      "epoch": 0.0977143819846128,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004345,
      "loss": 6.2649,
      "mean_token_accuracy": 0.13690216466784477,
      "num_tokens": 1953316.0,
      "step": 870
    },
    {
      "entropy": 6.438309907913208,
      "epoch": 0.09827595889257033,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000437,
      "loss": 6.3389,
      "mean_token_accuracy": 0.13307465687394143,
      "num_tokens": 1963936.0,
      "step": 875
    },
    {
      "entropy": 6.4518403053283695,
      "epoch": 0.09883753580052788,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004395,
      "loss": 6.3619,
      "mean_token_accuracy": 0.1389799363911152,
      "num_tokens": 1974502.0,
      "step": 880
    },
    {
      "entropy": 6.3525231838226315,
      "epoch": 0.09939911270848542,
      "grad_norm": 0.90625,
      "learning_rate": 0.000442,
      "loss": 6.2961,
      "mean_token_accuracy": 0.14234573692083358,
      "num_tokens": 1985746.0,
      "step": 885
    },
    {
      "entropy": 6.370750093460083,
      "epoch": 0.09996068961644297,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004445,
      "loss": 6.3998,
      "mean_token_accuracy": 0.1339979127049446,
      "num_tokens": 1997696.0,
      "step": 890
    },
    {
      "entropy": 6.476981210708618,
      "epoch": 0.10052226652440052,
      "grad_norm": 1.078125,
      "learning_rate": 0.000447,
      "loss": 6.3688,
      "mean_token_accuracy": 0.13148148730397224,
      "num_tokens": 2009462.0,
      "step": 895
    },
    {
      "entropy": 6.306058502197265,
      "epoch": 0.10108384343235806,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3004,
      "mean_token_accuracy": 0.13532110527157784,
      "num_tokens": 2019547.0,
      "step": 900
    },
    {
      "entropy": 6.276200008392334,
      "epoch": 0.10164542034031561,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.3087,
      "mean_token_accuracy": 0.1303063564002514,
      "num_tokens": 2030990.0,
      "step": 905
    },
    {
      "entropy": 6.328003215789795,
      "epoch": 0.10220699724827315,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.379,
      "mean_token_accuracy": 0.13275872617959977,
      "num_tokens": 2042488.0,
      "step": 910
    },
    {
      "entropy": 6.5090795993804935,
      "epoch": 0.1027685741562307,
      "grad_norm": 1.125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.4124,
      "mean_token_accuracy": 0.13137131109833716,
      "num_tokens": 2052822.0,
      "step": 915
    },
    {
      "entropy": 6.463271427154541,
      "epoch": 0.10333015106418825,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.3703,
      "mean_token_accuracy": 0.1341596059501171,
      "num_tokens": 2063274.0,
      "step": 920
    },
    {
      "entropy": 6.4825742721557615,
      "epoch": 0.10389172797214578,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000462,
      "loss": 6.4069,
      "mean_token_accuracy": 0.13037585765123366,
      "num_tokens": 2074424.0,
      "step": 925
    },
    {
      "entropy": 6.314002275466919,
      "epoch": 0.10445330488010333,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004645,
      "loss": 6.4013,
      "mean_token_accuracy": 0.13306034207344056,
      "num_tokens": 2085776.0,
      "step": 930
    },
    {
      "entropy": 6.591546201705933,
      "epoch": 0.10501488178806087,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000467,
      "loss": 6.4721,
      "mean_token_accuracy": 0.13484944850206376,
      "num_tokens": 2096789.0,
      "step": 935
    },
    {
      "entropy": 6.330018711090088,
      "epoch": 0.10557645869601842,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004695,
      "loss": 6.3848,
      "mean_token_accuracy": 0.13674520328640938,
      "num_tokens": 2108335.0,
      "step": 940
    },
    {
      "entropy": 6.46536111831665,
      "epoch": 0.10613803560397596,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000472,
      "loss": 6.4749,
      "mean_token_accuracy": 0.13200692012906073,
      "num_tokens": 2121063.0,
      "step": 945
    },
    {
      "entropy": 6.343993902206421,
      "epoch": 0.10669961251193351,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004745,
      "loss": 6.3351,
      "mean_token_accuracy": 0.1397795021533966,
      "num_tokens": 2132442.0,
      "step": 950
    },
    {
      "entropy": 6.362326002120971,
      "epoch": 0.10726118941989106,
      "grad_norm": 1.0625,
      "learning_rate": 0.000477,
      "loss": 6.2455,
      "mean_token_accuracy": 0.1406833589076996,
      "num_tokens": 2142155.0,
      "step": 955
    },
    {
      "entropy": 6.290420150756836,
      "epoch": 0.1078227663278486,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004795,
      "loss": 6.3118,
      "mean_token_accuracy": 0.13632866069674493,
      "num_tokens": 2152923.0,
      "step": 960
    },
    {
      "entropy": 6.347664928436279,
      "epoch": 0.10838434323580615,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000482,
      "loss": 6.3058,
      "mean_token_accuracy": 0.13932796120643615,
      "num_tokens": 2164640.0,
      "step": 965
    },
    {
      "entropy": 6.229414701461792,
      "epoch": 0.10894592014376368,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004845,
      "loss": 6.2222,
      "mean_token_accuracy": 0.13819631040096284,
      "num_tokens": 2175403.0,
      "step": 970
    },
    {
      "entropy": 6.34880313873291,
      "epoch": 0.10950749705172123,
      "grad_norm": 1.0625,
      "learning_rate": 0.000487,
      "loss": 6.3313,
      "mean_token_accuracy": 0.1353020928800106,
      "num_tokens": 2186031.0,
      "step": 975
    },
    {
      "entropy": 6.380644750595093,
      "epoch": 0.11006907395967878,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004895,
      "loss": 6.3329,
      "mean_token_accuracy": 0.13262429535388948,
      "num_tokens": 2198152.0,
      "step": 980
    },
    {
      "entropy": 6.278173971176147,
      "epoch": 0.11063065086763632,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492,
      "loss": 6.338,
      "mean_token_accuracy": 0.13674701675772666,
      "num_tokens": 2209435.0,
      "step": 985
    },
    {
      "entropy": 6.350253009796143,
      "epoch": 0.11119222777559387,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004945,
      "loss": 6.2994,
      "mean_token_accuracy": 0.14112076461315154,
      "num_tokens": 2219476.0,
      "step": 990
    },
    {
      "entropy": 6.3657073974609375,
      "epoch": 0.11175380468355141,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000497,
      "loss": 6.3837,
      "mean_token_accuracy": 0.13610237762331961,
      "num_tokens": 2231131.0,
      "step": 995
    },
    {
      "entropy": 6.300810956954956,
      "epoch": 0.11231538159150896,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995,
      "loss": 6.2892,
      "mean_token_accuracy": 0.1371823087334633,
      "num_tokens": 2242309.0,
      "step": 1000
    },
    {
      "entropy": 6.2305169105529785,
      "epoch": 0.1128769584994665,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499999997707492,
      "loss": 6.1273,
      "mean_token_accuracy": 0.1387334682047367,
      "num_tokens": 2252236.0,
      "step": 1005
    },
    {
      "entropy": 6.3840419292449955,
      "epoch": 0.11343853540742405,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999999883941787,
      "loss": 6.4504,
      "mean_token_accuracy": 0.1293959580361843,
      "num_tokens": 2263529.0,
      "step": 1010
    },
    {
      "entropy": 6.223633098602295,
      "epoch": 0.1140001123153816,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999999719167785,
      "loss": 6.2809,
      "mean_token_accuracy": 0.13737061843276024,
      "num_tokens": 2275318.0,
      "step": 1015
    },
    {
      "entropy": 6.3728053092956545,
      "epoch": 0.11456168922333913,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999999482752917,
      "loss": 6.3265,
      "mean_token_accuracy": 0.13787831887602806,
      "num_tokens": 2287434.0,
      "step": 1020
    },
    {
      "entropy": 6.254129695892334,
      "epoch": 0.11512326613129668,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999999174697195,
      "loss": 6.2315,
      "mean_token_accuracy": 0.13839245587587357,
      "num_tokens": 2299145.0,
      "step": 1025
    },
    {
      "entropy": 6.261599349975586,
      "epoch": 0.11568484303925422,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999998795000627,
      "loss": 6.3257,
      "mean_token_accuracy": 0.13926222771406174,
      "num_tokens": 2309841.0,
      "step": 1030
    },
    {
      "entropy": 6.315525674819947,
      "epoch": 0.11624641994721177,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999998343663223,
      "loss": 6.3193,
      "mean_token_accuracy": 0.1403449870646,
      "num_tokens": 2322143.0,
      "step": 1035
    },
    {
      "entropy": 6.261302518844604,
      "epoch": 0.11680799685516932,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999997820685,
      "loss": 6.2402,
      "mean_token_accuracy": 0.13240669071674346,
      "num_tokens": 2333207.0,
      "step": 1040
    },
    {
      "entropy": 6.1786919116973875,
      "epoch": 0.11736957376312686,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999997226065974,
      "loss": 6.3105,
      "mean_token_accuracy": 0.13664863780140876,
      "num_tokens": 2345542.0,
      "step": 1045
    },
    {
      "entropy": 6.371749353408814,
      "epoch": 0.11793115067108441,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999996559806163,
      "loss": 6.3079,
      "mean_token_accuracy": 0.14061758145689965,
      "num_tokens": 2356612.0,
      "step": 1050
    },
    {
      "entropy": 6.225208377838134,
      "epoch": 0.11849272757904195,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999995821905589,
      "loss": 6.2306,
      "mean_token_accuracy": 0.13814162313938141,
      "num_tokens": 2368204.0,
      "step": 1055
    },
    {
      "entropy": 6.262781095504761,
      "epoch": 0.1190543044869995,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999995012364276,
      "loss": 6.264,
      "mean_token_accuracy": 0.14789239913225175,
      "num_tokens": 2379114.0,
      "step": 1060
    },
    {
      "entropy": 6.258517122268676,
      "epoch": 0.11961588139495703,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999994131182248,
      "loss": 6.3024,
      "mean_token_accuracy": 0.1361288011074066,
      "num_tokens": 2390562.0,
      "step": 1065
    },
    {
      "entropy": 6.30267333984375,
      "epoch": 0.12017745830291458,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999993178359535,
      "loss": 6.3561,
      "mean_token_accuracy": 0.13576339259743692,
      "num_tokens": 2402352.0,
      "step": 1070
    },
    {
      "entropy": 6.422991371154785,
      "epoch": 0.12073903521087213,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999992153896166,
      "loss": 6.3906,
      "mean_token_accuracy": 0.1379263438284397,
      "num_tokens": 2415140.0,
      "step": 1075
    },
    {
      "entropy": 6.226438283920288,
      "epoch": 0.12130061211882967,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999991057792175,
      "loss": 6.2828,
      "mean_token_accuracy": 0.13986248895525932,
      "num_tokens": 2426679.0,
      "step": 1080
    },
    {
      "entropy": 6.307008361816406,
      "epoch": 0.12186218902678722,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999989890047596,
      "loss": 6.3213,
      "mean_token_accuracy": 0.14080586433410644,
      "num_tokens": 2437735.0,
      "step": 1085
    },
    {
      "entropy": 6.244773292541504,
      "epoch": 0.12242376593474476,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999988650662467,
      "loss": 6.2433,
      "mean_token_accuracy": 0.13999149277806283,
      "num_tokens": 2448611.0,
      "step": 1090
    },
    {
      "entropy": 6.176812791824341,
      "epoch": 0.12298534284270231,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999987339636825,
      "loss": 6.1729,
      "mean_token_accuracy": 0.14464567825198174,
      "num_tokens": 2461101.0,
      "step": 1095
    },
    {
      "entropy": 6.265973329544067,
      "epoch": 0.12354691975065986,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999985956970713,
      "loss": 6.3176,
      "mean_token_accuracy": 0.13885853216052055,
      "num_tokens": 2473569.0,
      "step": 1100
    },
    {
      "entropy": 6.171710395812989,
      "epoch": 0.1241084966586174,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999984502664178,
      "loss": 6.2382,
      "mean_token_accuracy": 0.13876865357160567,
      "num_tokens": 2483970.0,
      "step": 1105
    },
    {
      "entropy": 6.363559436798096,
      "epoch": 0.12467007356657495,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999982976717261,
      "loss": 6.2895,
      "mean_token_accuracy": 0.14146780967712402,
      "num_tokens": 2494184.0,
      "step": 1110
    },
    {
      "entropy": 6.217218828201294,
      "epoch": 0.1252316504745325,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999981379130016,
      "loss": 6.3041,
      "mean_token_accuracy": 0.14307402372360228,
      "num_tokens": 2505715.0,
      "step": 1115
    },
    {
      "entropy": 6.386610078811645,
      "epoch": 0.12579322738249002,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499997970990249,
      "loss": 6.3412,
      "mean_token_accuracy": 0.14221135526895523,
      "num_tokens": 2517168.0,
      "step": 1120
    },
    {
      "entropy": 6.303906965255737,
      "epoch": 0.12635480429044757,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999977969034737,
      "loss": 6.3674,
      "mean_token_accuracy": 0.1398570343852043,
      "num_tokens": 2528296.0,
      "step": 1125
    },
    {
      "entropy": 6.249405193328857,
      "epoch": 0.12691638119840512,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999976156526814,
      "loss": 6.3056,
      "mean_token_accuracy": 0.14569928944110871,
      "num_tokens": 2540648.0,
      "step": 1130
    },
    {
      "entropy": 6.177938985824585,
      "epoch": 0.12747795810636267,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999974272378776,
      "loss": 6.1885,
      "mean_token_accuracy": 0.14313262403011323,
      "num_tokens": 2551840.0,
      "step": 1135
    },
    {
      "entropy": 6.324184131622315,
      "epoch": 0.12803953501432022,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999972316590685,
      "loss": 6.2524,
      "mean_token_accuracy": 0.1379991091787815,
      "num_tokens": 2561362.0,
      "step": 1140
    },
    {
      "entropy": 6.31357970237732,
      "epoch": 0.12860111192227774,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999970289162604,
      "loss": 6.298,
      "mean_token_accuracy": 0.14433052241802216,
      "num_tokens": 2573584.0,
      "step": 1145
    },
    {
      "entropy": 6.446721315383911,
      "epoch": 0.1291626888302353,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999968190094595,
      "loss": 6.4204,
      "mean_token_accuracy": 0.1336354859173298,
      "num_tokens": 2585255.0,
      "step": 1150
    },
    {
      "entropy": 6.054596328735352,
      "epoch": 0.12972426573819285,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999966019386727,
      "loss": 6.226,
      "mean_token_accuracy": 0.14078792110085486,
      "num_tokens": 2596849.0,
      "step": 1155
    },
    {
      "entropy": 6.258234024047852,
      "epoch": 0.1302858426461504,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999963777039067,
      "loss": 6.2622,
      "mean_token_accuracy": 0.13961924761533737,
      "num_tokens": 2606918.0,
      "step": 1160
    },
    {
      "entropy": 6.1849829196929935,
      "epoch": 0.13084741955410795,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499996146305169,
      "loss": 6.1952,
      "mean_token_accuracy": 0.14556236416101456,
      "num_tokens": 2617779.0,
      "step": 1165
    },
    {
      "entropy": 6.273143911361695,
      "epoch": 0.13140899646206547,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999959077424665,
      "loss": 6.2513,
      "mean_token_accuracy": 0.13997638300061227,
      "num_tokens": 2630294.0,
      "step": 1170
    },
    {
      "entropy": 6.12783145904541,
      "epoch": 0.13197057337002302,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999956620158072,
      "loss": 6.1548,
      "mean_token_accuracy": 0.14402848556637765,
      "num_tokens": 2641002.0,
      "step": 1175
    },
    {
      "entropy": 6.1911145687103275,
      "epoch": 0.13253215027798057,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999954091251987,
      "loss": 6.1448,
      "mean_token_accuracy": 0.14548660516738893,
      "num_tokens": 2652188.0,
      "step": 1180
    },
    {
      "entropy": 6.125283575057983,
      "epoch": 0.13309372718593812,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999951490706491,
      "loss": 6.1865,
      "mean_token_accuracy": 0.14291539043188095,
      "num_tokens": 2662542.0,
      "step": 1185
    },
    {
      "entropy": 6.296076965332031,
      "epoch": 0.13365530409389567,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999948818521667,
      "loss": 6.2038,
      "mean_token_accuracy": 0.1425152152776718,
      "num_tokens": 2673292.0,
      "step": 1190
    },
    {
      "entropy": 6.177773904800415,
      "epoch": 0.1342168810018532,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00049999460746976,
      "loss": 6.2022,
      "mean_token_accuracy": 0.1450867235660553,
      "num_tokens": 2684042.0,
      "step": 1195
    },
    {
      "entropy": 6.199269771575928,
      "epoch": 0.13477845790981074,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999943259234377,
      "loss": 6.1304,
      "mean_token_accuracy": 0.1481417030096054,
      "num_tokens": 2694716.0,
      "step": 1200
    },
    {
      "entropy": 6.261506462097168,
      "epoch": 0.1353400348177683,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999940372132089,
      "loss": 6.2723,
      "mean_token_accuracy": 0.1427570842206478,
      "num_tokens": 2706200.0,
      "step": 1205
    },
    {
      "entropy": 6.097541093826294,
      "epoch": 0.13590161172572585,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999937413390826,
      "loss": 6.0978,
      "mean_token_accuracy": 0.14956343919038773,
      "num_tokens": 2716814.0,
      "step": 1210
    },
    {
      "entropy": 6.26879563331604,
      "epoch": 0.1364631886336834,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999934383010683,
      "loss": 6.2646,
      "mean_token_accuracy": 0.13633433505892753,
      "num_tokens": 2728042.0,
      "step": 1215
    },
    {
      "entropy": 6.117309093475342,
      "epoch": 0.13702476554164092,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999931280991757,
      "loss": 6.2276,
      "mean_token_accuracy": 0.13767654895782472,
      "num_tokens": 2739093.0,
      "step": 1220
    },
    {
      "entropy": 6.323478746414184,
      "epoch": 0.13758634244959847,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999928107334146,
      "loss": 6.2653,
      "mean_token_accuracy": 0.14359876736998559,
      "num_tokens": 2750307.0,
      "step": 1225
    },
    {
      "entropy": 6.146844863891602,
      "epoch": 0.13814791935755602,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999924862037952,
      "loss": 6.1785,
      "mean_token_accuracy": 0.142026449739933,
      "num_tokens": 2760534.0,
      "step": 1230
    },
    {
      "entropy": 6.146856021881104,
      "epoch": 0.13870949626551357,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999921545103278,
      "loss": 6.2017,
      "mean_token_accuracy": 0.1402510516345501,
      "num_tokens": 2772459.0,
      "step": 1235
    },
    {
      "entropy": 6.118765497207642,
      "epoch": 0.1392710731734711,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999918156530229,
      "loss": 6.1162,
      "mean_token_accuracy": 0.15379080772399903,
      "num_tokens": 2784038.0,
      "step": 1240
    },
    {
      "entropy": 6.237395238876343,
      "epoch": 0.13983265008142864,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999914696318913,
      "loss": 6.24,
      "mean_token_accuracy": 0.14800693467259407,
      "num_tokens": 2795756.0,
      "step": 1245
    },
    {
      "entropy": 6.129542446136474,
      "epoch": 0.1403942269893862,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999911164469441,
      "loss": 6.1721,
      "mean_token_accuracy": 0.14674090445041657,
      "num_tokens": 2806242.0,
      "step": 1250
    },
    {
      "entropy": 6.290287971496582,
      "epoch": 0.14095580389734375,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999907560981925,
      "loss": 6.2556,
      "mean_token_accuracy": 0.14589913040399552,
      "num_tokens": 2817017.0,
      "step": 1255
    },
    {
      "entropy": 6.189040660858154,
      "epoch": 0.1415173808053013,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999903885856479,
      "loss": 6.1629,
      "mean_token_accuracy": 0.15110919922590255,
      "num_tokens": 2827415.0,
      "step": 1260
    },
    {
      "entropy": 6.141000080108642,
      "epoch": 0.14207895771325882,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499990013909322,
      "loss": 6.1726,
      "mean_token_accuracy": 0.14932918325066566,
      "num_tokens": 2837595.0,
      "step": 1265
    },
    {
      "entropy": 6.088430595397949,
      "epoch": 0.14264053462121637,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499989632069227,
      "loss": 6.1339,
      "mean_token_accuracy": 0.14398894980549812,
      "num_tokens": 2847906.0,
      "step": 1270
    },
    {
      "entropy": 6.23743634223938,
      "epoch": 0.14320211152917392,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999892430653747,
      "loss": 6.2346,
      "mean_token_accuracy": 0.13501434922218322,
      "num_tokens": 2859145.0,
      "step": 1275
    },
    {
      "entropy": 6.056405162811279,
      "epoch": 0.14376368843713147,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999888468977776,
      "loss": 6.1648,
      "mean_token_accuracy": 0.14927329123020172,
      "num_tokens": 2870704.0,
      "step": 1280
    },
    {
      "entropy": 6.249700832366943,
      "epoch": 0.14432526534508902,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999884435664485,
      "loss": 6.2647,
      "mean_token_accuracy": 0.14412776455283166,
      "num_tokens": 2881861.0,
      "step": 1285
    },
    {
      "entropy": 6.240557241439819,
      "epoch": 0.14488684225304654,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004999880330714,
      "loss": 6.1861,
      "mean_token_accuracy": 0.14692625999450684,
      "num_tokens": 2893247.0,
      "step": 1290
    },
    {
      "entropy": 6.075284337997436,
      "epoch": 0.1454484191610041,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999876154126453,
      "loss": 6.0185,
      "mean_token_accuracy": 0.1487583711743355,
      "num_tokens": 2905351.0,
      "step": 1295
    },
    {
      "entropy": 6.0624950408935545,
      "epoch": 0.14600999606896164,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999871905901977,
      "loss": 6.1821,
      "mean_token_accuracy": 0.14871341213583947,
      "num_tokens": 2915153.0,
      "step": 1300
    },
    {
      "entropy": 6.169287776947021,
      "epoch": 0.1465715729769192,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999867586040706,
      "loss": 6.1416,
      "mean_token_accuracy": 0.13760501518845558,
      "num_tokens": 2926234.0,
      "step": 1305
    },
    {
      "entropy": 6.069522142410278,
      "epoch": 0.14713314988487675,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999863194542779,
      "loss": 6.0943,
      "mean_token_accuracy": 0.14693765938282013,
      "num_tokens": 2937033.0,
      "step": 1310
    },
    {
      "entropy": 6.068335819244385,
      "epoch": 0.14769472679283427,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999858731408336,
      "loss": 6.0518,
      "mean_token_accuracy": 0.14859263896942138,
      "num_tokens": 2945904.0,
      "step": 1315
    },
    {
      "entropy": 6.208248567581177,
      "epoch": 0.14825630370079182,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999854196637518,
      "loss": 6.253,
      "mean_token_accuracy": 0.14146910086274148,
      "num_tokens": 2956298.0,
      "step": 1320
    },
    {
      "entropy": 6.151482725143433,
      "epoch": 0.14881788060874937,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999849590230469,
      "loss": 6.1522,
      "mean_token_accuracy": 0.14781234338879584,
      "num_tokens": 2967660.0,
      "step": 1325
    },
    {
      "entropy": 6.198898839950561,
      "epoch": 0.14937945751670692,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999844912187336,
      "loss": 6.2205,
      "mean_token_accuracy": 0.1416272446513176,
      "num_tokens": 2978616.0,
      "step": 1330
    },
    {
      "entropy": 6.0869788646698,
      "epoch": 0.14994103442466447,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499984016250827,
      "loss": 6.0738,
      "mean_token_accuracy": 0.15207655876874923,
      "num_tokens": 2989731.0,
      "step": 1335
    },
    {
      "entropy": 6.171562576293946,
      "epoch": 0.150502611332622,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499983534119342,
      "loss": 6.177,
      "mean_token_accuracy": 0.14310620874166488,
      "num_tokens": 3002110.0,
      "step": 1340
    },
    {
      "entropy": 5.928503322601318,
      "epoch": 0.15106418824057954,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999830448242941,
      "loss": 6.0135,
      "mean_token_accuracy": 0.15312863737344742,
      "num_tokens": 3012709.0,
      "step": 1345
    },
    {
      "entropy": 6.257901430130005,
      "epoch": 0.1516257651485371,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999825483656986,
      "loss": 6.3139,
      "mean_token_accuracy": 0.13134505301713945,
      "num_tokens": 3025738.0,
      "step": 1350
    },
    {
      "entropy": 6.151828098297119,
      "epoch": 0.15218734205649465,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999820447435716,
      "loss": 6.1416,
      "mean_token_accuracy": 0.14338411912322044,
      "num_tokens": 3037711.0,
      "step": 1355
    },
    {
      "entropy": 6.239610385894776,
      "epoch": 0.15274891896445217,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999815339579291,
      "loss": 6.3286,
      "mean_token_accuracy": 0.13714671283960342,
      "num_tokens": 3048929.0,
      "step": 1360
    },
    {
      "entropy": 6.117274618148803,
      "epoch": 0.15331049587240972,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999810160087872,
      "loss": 6.0739,
      "mean_token_accuracy": 0.15920419096946717,
      "num_tokens": 3059717.0,
      "step": 1365
    },
    {
      "entropy": 6.133962488174438,
      "epoch": 0.15387207278036727,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999804908961626,
      "loss": 6.0778,
      "mean_token_accuracy": 0.15217963755130767,
      "num_tokens": 3069530.0,
      "step": 1370
    },
    {
      "entropy": 6.133673429489136,
      "epoch": 0.15443364968832482,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999799586200719,
      "loss": 6.2715,
      "mean_token_accuracy": 0.14009033963084222,
      "num_tokens": 3080705.0,
      "step": 1375
    },
    {
      "entropy": 6.1995704650878904,
      "epoch": 0.15499522659628237,
      "grad_norm": 1.0,
      "learning_rate": 0.000499979419180532,
      "loss": 6.1198,
      "mean_token_accuracy": 0.1425781100988388,
      "num_tokens": 3092553.0,
      "step": 1380
    },
    {
      "entropy": 6.193589210510254,
      "epoch": 0.1555568035042399,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999788725775601,
      "loss": 6.2416,
      "mean_token_accuracy": 0.1423923633992672,
      "num_tokens": 3102947.0,
      "step": 1385
    },
    {
      "entropy": 6.08302493095398,
      "epoch": 0.15611838041219744,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999783188111737,
      "loss": 6.0833,
      "mean_token_accuracy": 0.15142102390527726,
      "num_tokens": 3114157.0,
      "step": 1390
    },
    {
      "entropy": 6.019034147262573,
      "epoch": 0.156679957320155,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999777578813904,
      "loss": 5.9512,
      "mean_token_accuracy": 0.14669259935617446,
      "num_tokens": 3125739.0,
      "step": 1395
    },
    {
      "entropy": 6.063040637969971,
      "epoch": 0.15724153422811254,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499977189788228,
      "loss": 6.1128,
      "mean_token_accuracy": 0.15026577562093735,
      "num_tokens": 3136970.0,
      "step": 1400
    },
    {
      "entropy": 6.108362627029419,
      "epoch": 0.1578031111360701,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999766145317046,
      "loss": 6.0792,
      "mean_token_accuracy": 0.14896431863307952,
      "num_tokens": 3148512.0,
      "step": 1405
    },
    {
      "entropy": 6.111115980148315,
      "epoch": 0.15836468804402762,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999760321118385,
      "loss": 6.1135,
      "mean_token_accuracy": 0.15590201914310456,
      "num_tokens": 3159477.0,
      "step": 1410
    },
    {
      "entropy": 6.05187873840332,
      "epoch": 0.15892626495198517,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999754425286483,
      "loss": 6.128,
      "mean_token_accuracy": 0.15035361796617508,
      "num_tokens": 3171361.0,
      "step": 1415
    },
    {
      "entropy": 6.220038843154907,
      "epoch": 0.15948784185994272,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999748457821529,
      "loss": 6.2881,
      "mean_token_accuracy": 0.13868645653128625,
      "num_tokens": 3184652.0,
      "step": 1420
    },
    {
      "entropy": 6.1836763381958,
      "epoch": 0.16004941876790027,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499974241872371,
      "loss": 6.1675,
      "mean_token_accuracy": 0.14523480385541915,
      "num_tokens": 3195557.0,
      "step": 1425
    },
    {
      "entropy": 6.168182754516602,
      "epoch": 0.16061099567585782,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999736307993221,
      "loss": 6.1987,
      "mean_token_accuracy": 0.1331624984741211,
      "num_tokens": 3207139.0,
      "step": 1430
    },
    {
      "entropy": 6.287556171417236,
      "epoch": 0.16117257258381534,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999730125630255,
      "loss": 6.2968,
      "mean_token_accuracy": 0.13671731874346732,
      "num_tokens": 3218967.0,
      "step": 1435
    },
    {
      "entropy": 6.173974084854126,
      "epoch": 0.1617341494917729,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999723871635008,
      "loss": 6.1019,
      "mean_token_accuracy": 0.1492471754550934,
      "num_tokens": 3230168.0,
      "step": 1440
    },
    {
      "entropy": 6.022477769851685,
      "epoch": 0.16229572639973044,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999717546007681,
      "loss": 6.1451,
      "mean_token_accuracy": 0.1437738373875618,
      "num_tokens": 3241829.0,
      "step": 1445
    },
    {
      "entropy": 6.239676189422608,
      "epoch": 0.162857303307688,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999711148748476,
      "loss": 6.1597,
      "mean_token_accuracy": 0.14430854842066765,
      "num_tokens": 3253195.0,
      "step": 1450
    },
    {
      "entropy": 6.084852933883667,
      "epoch": 0.16341888021564555,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999704679857594,
      "loss": 6.2113,
      "mean_token_accuracy": 0.14120337069034578,
      "num_tokens": 3264951.0,
      "step": 1455
    },
    {
      "entropy": 6.022989225387573,
      "epoch": 0.16398045712360307,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999698139335243,
      "loss": 6.0438,
      "mean_token_accuracy": 0.14825797379016875,
      "num_tokens": 3276054.0,
      "step": 1460
    },
    {
      "entropy": 6.0944970607757565,
      "epoch": 0.16454203403156062,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999691527181632,
      "loss": 6.0089,
      "mean_token_accuracy": 0.1612110674381256,
      "num_tokens": 3285540.0,
      "step": 1465
    },
    {
      "entropy": 6.161299562454223,
      "epoch": 0.16510361093951817,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999684843396968,
      "loss": 6.1427,
      "mean_token_accuracy": 0.14661948680877684,
      "num_tokens": 3296869.0,
      "step": 1470
    },
    {
      "entropy": 6.089122295379639,
      "epoch": 0.16566518784747572,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999678087981467,
      "loss": 6.0813,
      "mean_token_accuracy": 0.1550757810473442,
      "num_tokens": 3307554.0,
      "step": 1475
    },
    {
      "entropy": 5.974129676818848,
      "epoch": 0.16622676475543324,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999671260935343,
      "loss": 6.0369,
      "mean_token_accuracy": 0.14905909150838853,
      "num_tokens": 3319685.0,
      "step": 1480
    },
    {
      "entropy": 6.121063327789306,
      "epoch": 0.1667883416633908,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999664362258814,
      "loss": 6.1516,
      "mean_token_accuracy": 0.14900618344545363,
      "num_tokens": 3330742.0,
      "step": 1485
    },
    {
      "entropy": 6.037702798843384,
      "epoch": 0.16734991857134834,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999657391952099,
      "loss": 6.1774,
      "mean_token_accuracy": 0.14220552295446395,
      "num_tokens": 3343613.0,
      "step": 1490
    },
    {
      "entropy": 6.289458322525024,
      "epoch": 0.1679114954793059,
      "grad_norm": 0.875,
      "learning_rate": 0.000499965035001542,
      "loss": 6.2146,
      "mean_token_accuracy": 0.1457619272172451,
      "num_tokens": 3355162.0,
      "step": 1495
    },
    {
      "entropy": 6.107675218582154,
      "epoch": 0.16847307238726345,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999643236449001,
      "loss": 6.1513,
      "mean_token_accuracy": 0.15243639126420022,
      "num_tokens": 3366324.0,
      "step": 1500
    },
    {
      "entropy": 6.264118766784668,
      "epoch": 0.16903464929522097,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999636051253069,
      "loss": 6.292,
      "mean_token_accuracy": 0.13710838407278061,
      "num_tokens": 3378571.0,
      "step": 1505
    },
    {
      "entropy": 6.134381484985352,
      "epoch": 0.16959622620317852,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999628794427854,
      "loss": 6.1696,
      "mean_token_accuracy": 0.13963365852832793,
      "num_tokens": 3390605.0,
      "step": 1510
    },
    {
      "entropy": 6.0607460021972654,
      "epoch": 0.17015780311113607,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999621465973583,
      "loss": 6.0866,
      "mean_token_accuracy": 0.14871912077069283,
      "num_tokens": 3402735.0,
      "step": 1515
    },
    {
      "entropy": 6.100008773803711,
      "epoch": 0.17071938001909362,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999614065890495,
      "loss": 6.0991,
      "mean_token_accuracy": 0.15326170325279237,
      "num_tokens": 3413079.0,
      "step": 1520
    },
    {
      "entropy": 6.077723407745362,
      "epoch": 0.17128095692705117,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999606594178821,
      "loss": 6.0422,
      "mean_token_accuracy": 0.14912576824426652,
      "num_tokens": 3423098.0,
      "step": 1525
    },
    {
      "entropy": 6.100780582427978,
      "epoch": 0.1718425338350087,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00049995990508388,
      "loss": 6.0769,
      "mean_token_accuracy": 0.14803334027528764,
      "num_tokens": 3435794.0,
      "step": 1530
    },
    {
      "entropy": 6.115665483474731,
      "epoch": 0.17240411074296624,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999591435870674,
      "loss": 6.2069,
      "mean_token_accuracy": 0.13976619020104408,
      "num_tokens": 3448125.0,
      "step": 1535
    },
    {
      "entropy": 6.065301561355591,
      "epoch": 0.1729656876509238,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999583749274684,
      "loss": 6.0397,
      "mean_token_accuracy": 0.14901299476623536,
      "num_tokens": 3458217.0,
      "step": 1540
    },
    {
      "entropy": 6.002480125427246,
      "epoch": 0.17352726455888134,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999575991051074,
      "loss": 6.0901,
      "mean_token_accuracy": 0.1483938679099083,
      "num_tokens": 3469553.0,
      "step": 1545
    },
    {
      "entropy": 6.146899223327637,
      "epoch": 0.1740888414668389,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999568161200093,
      "loss": 6.0427,
      "mean_token_accuracy": 0.15437329113483428,
      "num_tokens": 3480928.0,
      "step": 1550
    },
    {
      "entropy": 6.065676021575928,
      "epoch": 0.17465041837479642,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999560259721989,
      "loss": 6.1347,
      "mean_token_accuracy": 0.1498277299106121,
      "num_tokens": 3492074.0,
      "step": 1555
    },
    {
      "entropy": 6.13408784866333,
      "epoch": 0.17521199528275397,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999552286617014,
      "loss": 6.1425,
      "mean_token_accuracy": 0.14677510857582093,
      "num_tokens": 3503320.0,
      "step": 1560
    },
    {
      "entropy": 6.015400695800781,
      "epoch": 0.17577357219071152,
      "grad_norm": 1.375,
      "learning_rate": 0.0004999544241885421,
      "loss": 6.136,
      "mean_token_accuracy": 0.15524707436561586,
      "num_tokens": 3515466.0,
      "step": 1565
    },
    {
      "entropy": 6.1630960464477536,
      "epoch": 0.17633514909866907,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999536125527467,
      "loss": 6.0079,
      "mean_token_accuracy": 0.15108589679002762,
      "num_tokens": 3525584.0,
      "step": 1570
    },
    {
      "entropy": 5.98162202835083,
      "epoch": 0.17689672600662662,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499952793754341,
      "loss": 6.0963,
      "mean_token_accuracy": 0.1469439223408699,
      "num_tokens": 3536973.0,
      "step": 1575
    },
    {
      "entropy": 6.10207724571228,
      "epoch": 0.17745830291458414,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999519677933512,
      "loss": 6.1286,
      "mean_token_accuracy": 0.14693209826946257,
      "num_tokens": 3547799.0,
      "step": 1580
    },
    {
      "entropy": 6.159058141708374,
      "epoch": 0.1780198798225417,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999511346698035,
      "loss": 6.1786,
      "mean_token_accuracy": 0.15070472061634063,
      "num_tokens": 3558436.0,
      "step": 1585
    },
    {
      "entropy": 6.149427270889282,
      "epoch": 0.17858145673049924,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999502943837244,
      "loss": 6.1659,
      "mean_token_accuracy": 0.15004989951848985,
      "num_tokens": 3569851.0,
      "step": 1590
    },
    {
      "entropy": 6.135163640975952,
      "epoch": 0.1791430336384568,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999494469351406,
      "loss": 6.0449,
      "mean_token_accuracy": 0.14947315603494643,
      "num_tokens": 3580797.0,
      "step": 1595
    },
    {
      "entropy": 6.05308346748352,
      "epoch": 0.17970461054641435,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999485923240793,
      "loss": 6.0467,
      "mean_token_accuracy": 0.1534357637166977,
      "num_tokens": 3592552.0,
      "step": 1600
    },
    {
      "entropy": 5.9738929748535154,
      "epoch": 0.18026618745437187,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999477305505675,
      "loss": 6.0287,
      "mean_token_accuracy": 0.15556652545928956,
      "num_tokens": 3603932.0,
      "step": 1605
    },
    {
      "entropy": 6.095688581466675,
      "epoch": 0.18082776436232942,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999468616146328,
      "loss": 6.052,
      "mean_token_accuracy": 0.1578151524066925,
      "num_tokens": 3614865.0,
      "step": 1610
    },
    {
      "entropy": 5.861964559555053,
      "epoch": 0.18138934127028697,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999459855163026,
      "loss": 5.9784,
      "mean_token_accuracy": 0.1593295931816101,
      "num_tokens": 3626275.0,
      "step": 1615
    },
    {
      "entropy": 6.140000581741333,
      "epoch": 0.18195091817824452,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499945102255605,
      "loss": 6.1078,
      "mean_token_accuracy": 0.1455901011824608,
      "num_tokens": 3638416.0,
      "step": 1620
    },
    {
      "entropy": 5.946600866317749,
      "epoch": 0.18251249508620204,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999442118325682,
      "loss": 5.9553,
      "mean_token_accuracy": 0.1524277485907078,
      "num_tokens": 3649651.0,
      "step": 1625
    },
    {
      "entropy": 6.132726192474365,
      "epoch": 0.1830740719941596,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999433142472205,
      "loss": 6.192,
      "mean_token_accuracy": 0.15279360264539718,
      "num_tokens": 3661172.0,
      "step": 1630
    },
    {
      "entropy": 6.108269739151001,
      "epoch": 0.18363564890211714,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999424094995902,
      "loss": 6.0268,
      "mean_token_accuracy": 0.15440748035907745,
      "num_tokens": 3672086.0,
      "step": 1635
    },
    {
      "entropy": 6.029655122756958,
      "epoch": 0.1841972258100747,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999414975897066,
      "loss": 6.0021,
      "mean_token_accuracy": 0.14995888769626617,
      "num_tokens": 3682579.0,
      "step": 1640
    },
    {
      "entropy": 6.10533299446106,
      "epoch": 0.18475880271803224,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999405785175983,
      "loss": 6.1253,
      "mean_token_accuracy": 0.14912039935588836,
      "num_tokens": 3694616.0,
      "step": 1645
    },
    {
      "entropy": 5.989877414703369,
      "epoch": 0.18532037962598977,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999396522832949,
      "loss": 5.909,
      "mean_token_accuracy": 0.1535977005958557,
      "num_tokens": 3706821.0,
      "step": 1650
    },
    {
      "entropy": 5.970361423492432,
      "epoch": 0.18588195653394732,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999387188868256,
      "loss": 5.9652,
      "mean_token_accuracy": 0.1572886511683464,
      "num_tokens": 3717552.0,
      "step": 1655
    },
    {
      "entropy": 5.978477573394775,
      "epoch": 0.18644353344190487,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999377783282203,
      "loss": 6.043,
      "mean_token_accuracy": 0.15294748544692993,
      "num_tokens": 3728300.0,
      "step": 1660
    },
    {
      "entropy": 6.069652843475342,
      "epoch": 0.18700511034986242,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999368306075089,
      "loss": 6.0383,
      "mean_token_accuracy": 0.15796971321105957,
      "num_tokens": 3738545.0,
      "step": 1665
    },
    {
      "entropy": 6.053223991394043,
      "epoch": 0.18756668725781997,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999358757247215,
      "loss": 6.031,
      "mean_token_accuracy": 0.14949157685041428,
      "num_tokens": 3749190.0,
      "step": 1670
    },
    {
      "entropy": 5.973500061035156,
      "epoch": 0.1881282641657775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999349136798886,
      "loss": 6.0387,
      "mean_token_accuracy": 0.15051646381616593,
      "num_tokens": 3760978.0,
      "step": 1675
    },
    {
      "entropy": 5.94163761138916,
      "epoch": 0.18868984107373504,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999339444730409,
      "loss": 5.9304,
      "mean_token_accuracy": 0.1564078763127327,
      "num_tokens": 3771339.0,
      "step": 1680
    },
    {
      "entropy": 6.000215005874634,
      "epoch": 0.1892514179816926,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999329681042092,
      "loss": 5.9158,
      "mean_token_accuracy": 0.15469563752412796,
      "num_tokens": 3781550.0,
      "step": 1685
    },
    {
      "entropy": 5.960664081573486,
      "epoch": 0.18981299488965014,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999319845734243,
      "loss": 6.1339,
      "mean_token_accuracy": 0.14286689832806587,
      "num_tokens": 3793700.0,
      "step": 1690
    },
    {
      "entropy": 6.237479496002197,
      "epoch": 0.1903745717976077,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999309938807181,
      "loss": 6.1323,
      "mean_token_accuracy": 0.14619217813014984,
      "num_tokens": 3806114.0,
      "step": 1695
    },
    {
      "entropy": 5.988754749298096,
      "epoch": 0.19093614870556522,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999299960261216,
      "loss": 6.0421,
      "mean_token_accuracy": 0.15423230230808258,
      "num_tokens": 3817251.0,
      "step": 1700
    },
    {
      "entropy": 6.091027498245239,
      "epoch": 0.19149772561352277,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999289910096669,
      "loss": 6.0464,
      "mean_token_accuracy": 0.15106936246156694,
      "num_tokens": 3828544.0,
      "step": 1705
    },
    {
      "entropy": 5.994654273986816,
      "epoch": 0.19205930252148032,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999279788313858,
      "loss": 6.1189,
      "mean_token_accuracy": 0.14720231145620347,
      "num_tokens": 3840423.0,
      "step": 1710
    },
    {
      "entropy": 6.1074625015258786,
      "epoch": 0.19262087942943787,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999269594913107,
      "loss": 6.0199,
      "mean_token_accuracy": 0.15324137806892396,
      "num_tokens": 3852751.0,
      "step": 1715
    },
    {
      "entropy": 5.966987991333008,
      "epoch": 0.19318245633739542,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999259329894739,
      "loss": 6.0043,
      "mean_token_accuracy": 0.15583723783493042,
      "num_tokens": 3863382.0,
      "step": 1720
    },
    {
      "entropy": 5.978365659713745,
      "epoch": 0.19374403324535294,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999248993259081,
      "loss": 6.0004,
      "mean_token_accuracy": 0.1522401787340641,
      "num_tokens": 3874294.0,
      "step": 1725
    },
    {
      "entropy": 6.057207489013672,
      "epoch": 0.1943056101533105,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999238585006463,
      "loss": 6.0364,
      "mean_token_accuracy": 0.14614153355360032,
      "num_tokens": 3886693.0,
      "step": 1730
    },
    {
      "entropy": 5.985201740264893,
      "epoch": 0.19486718706126804,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999228105137218,
      "loss": 5.9488,
      "mean_token_accuracy": 0.1587449014186859,
      "num_tokens": 3897755.0,
      "step": 1735
    },
    {
      "entropy": 5.956290817260742,
      "epoch": 0.1954287639692256,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999217553651676,
      "loss": 5.9546,
      "mean_token_accuracy": 0.15806302726268767,
      "num_tokens": 3909418.0,
      "step": 1740
    },
    {
      "entropy": 6.051894617080689,
      "epoch": 0.19599034087718312,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999206930550175,
      "loss": 6.1081,
      "mean_token_accuracy": 0.14818969666957854,
      "num_tokens": 3919992.0,
      "step": 1745
    },
    {
      "entropy": 6.063493871688843,
      "epoch": 0.19655191778514067,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999196235833054,
      "loss": 5.9663,
      "mean_token_accuracy": 0.16147716641426085,
      "num_tokens": 3930648.0,
      "step": 1750
    },
    {
      "entropy": 5.885579061508179,
      "epoch": 0.19711349469309822,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999185469500651,
      "loss": 5.9529,
      "mean_token_accuracy": 0.15516203343868257,
      "num_tokens": 3941883.0,
      "step": 1755
    },
    {
      "entropy": 6.019442367553711,
      "epoch": 0.19767507160105577,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499917463155331,
      "loss": 5.9987,
      "mean_token_accuracy": 0.15911805033683776,
      "num_tokens": 3952005.0,
      "step": 1760
    },
    {
      "entropy": 5.935513496398926,
      "epoch": 0.19823664850901332,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999163721991378,
      "loss": 5.9213,
      "mean_token_accuracy": 0.15570384711027146,
      "num_tokens": 3963720.0,
      "step": 1765
    },
    {
      "entropy": 5.93169755935669,
      "epoch": 0.19879822541697084,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049991527408152,
      "loss": 6.0081,
      "mean_token_accuracy": 0.15529269427061082,
      "num_tokens": 3974998.0,
      "step": 1770
    },
    {
      "entropy": 6.000465297698975,
      "epoch": 0.1993598023249284,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999141688025127,
      "loss": 6.048,
      "mean_token_accuracy": 0.14723881632089614,
      "num_tokens": 3986469.0,
      "step": 1775
    },
    {
      "entropy": 6.037735271453857,
      "epoch": 0.19992137923288594,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499913056362151,
      "loss": 5.8733,
      "mean_token_accuracy": 0.1559750333428383,
      "num_tokens": 3997553.0,
      "step": 1780
    },
    {
      "entropy": 5.848225831985474,
      "epoch": 0.2004829561408435,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999119367604703,
      "loss": 5.958,
      "mean_token_accuracy": 0.15345809012651443,
      "num_tokens": 4008613.0,
      "step": 1785
    },
    {
      "entropy": 6.148313570022583,
      "epoch": 0.20104453304880104,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999108099975064,
      "loss": 6.1178,
      "mean_token_accuracy": 0.1494819074869156,
      "num_tokens": 4019860.0,
      "step": 1790
    },
    {
      "entropy": 5.971745538711548,
      "epoch": 0.20160610995675857,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499909676073295,
      "loss": 6.0125,
      "mean_token_accuracy": 0.15912038534879686,
      "num_tokens": 4030602.0,
      "step": 1795
    },
    {
      "entropy": 5.967540597915649,
      "epoch": 0.20216768686471612,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999085349878723,
      "loss": 5.8881,
      "mean_token_accuracy": 0.1551676794886589,
      "num_tokens": 4041878.0,
      "step": 1800
    },
    {
      "entropy": 5.902796602249145,
      "epoch": 0.20272926377267367,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999073867412746,
      "loss": 5.985,
      "mean_token_accuracy": 0.15298098027706147,
      "num_tokens": 4052799.0,
      "step": 1805
    },
    {
      "entropy": 6.052335357666015,
      "epoch": 0.20329084068063122,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999062313335385,
      "loss": 5.977,
      "mean_token_accuracy": 0.1513774089515209,
      "num_tokens": 4064339.0,
      "step": 1810
    },
    {
      "entropy": 5.881051683425904,
      "epoch": 0.20385241758858877,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999050687647008,
      "loss": 5.9321,
      "mean_token_accuracy": 0.15355904400348663,
      "num_tokens": 4076246.0,
      "step": 1815
    },
    {
      "entropy": 5.9952904224395756,
      "epoch": 0.2044139944965463,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999038990347984,
      "loss": 6.0102,
      "mean_token_accuracy": 0.15132154226303102,
      "num_tokens": 4088006.0,
      "step": 1820
    },
    {
      "entropy": 5.986813735961914,
      "epoch": 0.20497557140450384,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999027221438688,
      "loss": 6.0135,
      "mean_token_accuracy": 0.1533118411898613,
      "num_tokens": 4098122.0,
      "step": 1825
    },
    {
      "entropy": 6.179859924316406,
      "epoch": 0.2055371483124614,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499901538091949,
      "loss": 6.2072,
      "mean_token_accuracy": 0.14116512686014177,
      "num_tokens": 4109958.0,
      "step": 1830
    },
    {
      "entropy": 6.044462394714356,
      "epoch": 0.20609872522041894,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999003468790773,
      "loss": 6.0967,
      "mean_token_accuracy": 0.14568886607885362,
      "num_tokens": 4121262.0,
      "step": 1835
    },
    {
      "entropy": 6.016641092300415,
      "epoch": 0.2066603021283765,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998991485052911,
      "loss": 5.87,
      "mean_token_accuracy": 0.16044027954339982,
      "num_tokens": 4130958.0,
      "step": 1840
    },
    {
      "entropy": 5.966278457641602,
      "epoch": 0.20722187903633402,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499897942970629,
      "loss": 6.0778,
      "mean_token_accuracy": 0.1423511266708374,
      "num_tokens": 4143779.0,
      "step": 1845
    },
    {
      "entropy": 6.1042670726776125,
      "epoch": 0.20778345594429157,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499896730275129,
      "loss": 6.1007,
      "mean_token_accuracy": 0.14468785226345063,
      "num_tokens": 4155367.0,
      "step": 1850
    },
    {
      "entropy": 5.985637760162353,
      "epoch": 0.20834503285224912,
      "grad_norm": 1.015625,
      "learning_rate": 0.00049989551041883,
      "loss": 5.9152,
      "mean_token_accuracy": 0.15388040691614152,
      "num_tokens": 4166657.0,
      "step": 1855
    },
    {
      "entropy": 5.961990070343018,
      "epoch": 0.20890660976020667,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998942834017707,
      "loss": 6.0298,
      "mean_token_accuracy": 0.15152619555592536,
      "num_tokens": 4177517.0,
      "step": 1860
    },
    {
      "entropy": 5.9690758228302006,
      "epoch": 0.2094681866681642,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998930492239901,
      "loss": 5.8913,
      "mean_token_accuracy": 0.15565723478794097,
      "num_tokens": 4188182.0,
      "step": 1865
    },
    {
      "entropy": 6.060051155090332,
      "epoch": 0.21002976357612174,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998918078855277,
      "loss": 6.0221,
      "mean_token_accuracy": 0.1551324889063835,
      "num_tokens": 4199415.0,
      "step": 1870
    },
    {
      "entropy": 5.884836721420288,
      "epoch": 0.2105913404840793,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998905593864227,
      "loss": 5.967,
      "mean_token_accuracy": 0.15338498651981353,
      "num_tokens": 4210025.0,
      "step": 1875
    },
    {
      "entropy": 5.985698986053467,
      "epoch": 0.21115291739203684,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998893037267153,
      "loss": 5.8966,
      "mean_token_accuracy": 0.15888059735298157,
      "num_tokens": 4219920.0,
      "step": 1880
    },
    {
      "entropy": 5.979580545425415,
      "epoch": 0.2117144942999944,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998880409064452,
      "loss": 6.0678,
      "mean_token_accuracy": 0.1490791544318199,
      "num_tokens": 4232647.0,
      "step": 1885
    },
    {
      "entropy": 6.082404136657715,
      "epoch": 0.21227607120795192,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998867709256527,
      "loss": 5.9527,
      "mean_token_accuracy": 0.15514377057552337,
      "num_tokens": 4242637.0,
      "step": 1890
    },
    {
      "entropy": 5.876793336868286,
      "epoch": 0.21283764811590947,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499885493784378,
      "loss": 5.9242,
      "mean_token_accuracy": 0.15735216736793517,
      "num_tokens": 4253633.0,
      "step": 1895
    },
    {
      "entropy": 5.9101646900177,
      "epoch": 0.21339922502386702,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998842094826621,
      "loss": 5.8893,
      "mean_token_accuracy": 0.15901758521795273,
      "num_tokens": 4264472.0,
      "step": 1900
    },
    {
      "entropy": 6.013765573501587,
      "epoch": 0.21396080193182457,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998829180205458,
      "loss": 6.0058,
      "mean_token_accuracy": 0.15594724267721177,
      "num_tokens": 4275233.0,
      "step": 1905
    },
    {
      "entropy": 6.0052587509155275,
      "epoch": 0.21452237883978212,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998816193980701,
      "loss": 5.9606,
      "mean_token_accuracy": 0.15338825732469558,
      "num_tokens": 4287063.0,
      "step": 1910
    },
    {
      "entropy": 6.001710557937622,
      "epoch": 0.21508395574773964,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998803136152764,
      "loss": 5.9866,
      "mean_token_accuracy": 0.15025851130485535,
      "num_tokens": 4298761.0,
      "step": 1915
    },
    {
      "entropy": 5.735988998413086,
      "epoch": 0.2156455326556972,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998790006722063,
      "loss": 5.8469,
      "mean_token_accuracy": 0.16698734015226363,
      "num_tokens": 4309650.0,
      "step": 1920
    },
    {
      "entropy": 6.018524694442749,
      "epoch": 0.21620710956365474,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998776805689016,
      "loss": 5.9063,
      "mean_token_accuracy": 0.16533491015434265,
      "num_tokens": 4320403.0,
      "step": 1925
    },
    {
      "entropy": 5.84126296043396,
      "epoch": 0.2167686864716123,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998763533054045,
      "loss": 5.932,
      "mean_token_accuracy": 0.1564387321472168,
      "num_tokens": 4332049.0,
      "step": 1930
    },
    {
      "entropy": 5.947682619094849,
      "epoch": 0.21733026337956984,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998750188817568,
      "loss": 5.8358,
      "mean_token_accuracy": 0.16291145533323287,
      "num_tokens": 4342551.0,
      "step": 1935
    },
    {
      "entropy": 5.899789810180664,
      "epoch": 0.21789184028752737,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998736772980015,
      "loss": 5.9692,
      "mean_token_accuracy": 0.1594847857952118,
      "num_tokens": 4355156.0,
      "step": 1940
    },
    {
      "entropy": 5.908867359161377,
      "epoch": 0.21845341719548492,
      "grad_norm": 1.0,
      "learning_rate": 0.000499872328554181,
      "loss": 6.0218,
      "mean_token_accuracy": 0.15519974678754805,
      "num_tokens": 4367402.0,
      "step": 1945
    },
    {
      "entropy": 6.017113590240479,
      "epoch": 0.21901499410344247,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998709726503383,
      "loss": 5.8973,
      "mean_token_accuracy": 0.16066040098667145,
      "num_tokens": 4378687.0,
      "step": 1950
    },
    {
      "entropy": 5.973746299743652,
      "epoch": 0.21957657101140002,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998696095865168,
      "loss": 6.0371,
      "mean_token_accuracy": 0.1629161447286606,
      "num_tokens": 4388871.0,
      "step": 1955
    },
    {
      "entropy": 5.922299194335937,
      "epoch": 0.22013814791935757,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998682393627595,
      "loss": 5.8576,
      "mean_token_accuracy": 0.16151300221681594,
      "num_tokens": 4399440.0,
      "step": 1960
    },
    {
      "entropy": 5.825509119033813,
      "epoch": 0.2206997248273151,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998668619791104,
      "loss": 5.8555,
      "mean_token_accuracy": 0.16079302430152892,
      "num_tokens": 4409989.0,
      "step": 1965
    },
    {
      "entropy": 6.02225980758667,
      "epoch": 0.22126130173527264,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998654774356131,
      "loss": 6.0152,
      "mean_token_accuracy": 0.15131131708621978,
      "num_tokens": 4422149.0,
      "step": 1970
    },
    {
      "entropy": 6.051222705841065,
      "epoch": 0.2218228786432302,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998640857323118,
      "loss": 6.0486,
      "mean_token_accuracy": 0.15000825822353364,
      "num_tokens": 4433157.0,
      "step": 1975
    },
    {
      "entropy": 6.00563907623291,
      "epoch": 0.22238445555118774,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998626868692509,
      "loss": 5.973,
      "mean_token_accuracy": 0.15266723185777664,
      "num_tokens": 4444595.0,
      "step": 1980
    },
    {
      "entropy": 6.015877962112427,
      "epoch": 0.2229460324591453,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998612808464748,
      "loss": 6.1061,
      "mean_token_accuracy": 0.14999096170067788,
      "num_tokens": 4456592.0,
      "step": 1985
    },
    {
      "entropy": 6.005351543426514,
      "epoch": 0.22350760936710282,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998598676640283,
      "loss": 5.8623,
      "mean_token_accuracy": 0.16657372564077377,
      "num_tokens": 4467624.0,
      "step": 1990
    },
    {
      "entropy": 5.870333099365235,
      "epoch": 0.22406918627506037,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998584473219563,
      "loss": 6.0578,
      "mean_token_accuracy": 0.1522249922156334,
      "num_tokens": 4479766.0,
      "step": 1995
    },
    {
      "entropy": 6.041551637649536,
      "epoch": 0.22463076318301792,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998570198203043,
      "loss": 5.927,
      "mean_token_accuracy": 0.1541971117258072,
      "num_tokens": 4490023.0,
      "step": 2000
    },
    {
      "entropy": 5.939703559875488,
      "epoch": 0.22519234009097547,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998555851591175,
      "loss": 5.9578,
      "mean_token_accuracy": 0.15466774851083756,
      "num_tokens": 4500651.0,
      "step": 2005
    },
    {
      "entropy": 6.019627428054809,
      "epoch": 0.225753916998933,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998541433384416,
      "loss": 6.0605,
      "mean_token_accuracy": 0.1519089847803116,
      "num_tokens": 4512934.0,
      "step": 2010
    },
    {
      "entropy": 5.9258240222930905,
      "epoch": 0.22631549390689054,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998526943583226,
      "loss": 5.9379,
      "mean_token_accuracy": 0.16064678132534027,
      "num_tokens": 4524251.0,
      "step": 2015
    },
    {
      "entropy": 5.996880960464478,
      "epoch": 0.2268770708148481,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998512382188066,
      "loss": 6.022,
      "mean_token_accuracy": 0.14892027527093887,
      "num_tokens": 4536025.0,
      "step": 2020
    },
    {
      "entropy": 6.004436492919922,
      "epoch": 0.22743864772280564,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998497749199401,
      "loss": 5.9701,
      "mean_token_accuracy": 0.158987794816494,
      "num_tokens": 4546717.0,
      "step": 2025
    },
    {
      "entropy": 5.910756921768188,
      "epoch": 0.2280002246307632,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998483044617694,
      "loss": 5.8651,
      "mean_token_accuracy": 0.1554678872227669,
      "num_tokens": 4557206.0,
      "step": 2030
    },
    {
      "entropy": 5.822503995895386,
      "epoch": 0.22856180153872072,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998468268443416,
      "loss": 5.8574,
      "mean_token_accuracy": 0.16205435246229172,
      "num_tokens": 4568123.0,
      "step": 2035
    },
    {
      "entropy": 5.961733341217041,
      "epoch": 0.22912337844667827,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998453420677034,
      "loss": 5.9819,
      "mean_token_accuracy": 0.15619362890720367,
      "num_tokens": 4579385.0,
      "step": 2040
    },
    {
      "entropy": 5.9808855056762695,
      "epoch": 0.22968495535463582,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998438501319025,
      "loss": 6.0248,
      "mean_token_accuracy": 0.15128423795104026,
      "num_tokens": 4590514.0,
      "step": 2045
    },
    {
      "entropy": 6.025481367111206,
      "epoch": 0.23024653226259337,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998423510369862,
      "loss": 5.9706,
      "mean_token_accuracy": 0.15017401427030563,
      "num_tokens": 4601624.0,
      "step": 2050
    },
    {
      "entropy": 5.990740871429443,
      "epoch": 0.23080810917055092,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998408447830022,
      "loss": 5.9518,
      "mean_token_accuracy": 0.16115492284297944,
      "num_tokens": 4612822.0,
      "step": 2055
    },
    {
      "entropy": 5.939420604705811,
      "epoch": 0.23136968607850844,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998393313699985,
      "loss": 5.9466,
      "mean_token_accuracy": 0.16056140810251235,
      "num_tokens": 4623443.0,
      "step": 2060
    },
    {
      "entropy": 5.930271577835083,
      "epoch": 0.231931262986466,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998378107980233,
      "loss": 5.8579,
      "mean_token_accuracy": 0.16032032370567323,
      "num_tokens": 4633600.0,
      "step": 2065
    },
    {
      "entropy": 5.932555675506592,
      "epoch": 0.23249283989442354,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998362830671249,
      "loss": 6.0102,
      "mean_token_accuracy": 0.15435325503349304,
      "num_tokens": 4644217.0,
      "step": 2070
    },
    {
      "entropy": 5.97379846572876,
      "epoch": 0.2330544168023811,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998347481773522,
      "loss": 6.0249,
      "mean_token_accuracy": 0.1527538165450096,
      "num_tokens": 4654671.0,
      "step": 2075
    },
    {
      "entropy": 5.9626545906066895,
      "epoch": 0.23361599371033864,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998332061287538,
      "loss": 6.0209,
      "mean_token_accuracy": 0.1507769837975502,
      "num_tokens": 4665920.0,
      "step": 2080
    },
    {
      "entropy": 5.939901208877563,
      "epoch": 0.23417757061829617,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499831656921379,
      "loss": 5.8997,
      "mean_token_accuracy": 0.1583072543144226,
      "num_tokens": 4675406.0,
      "step": 2085
    },
    {
      "entropy": 6.018610954284668,
      "epoch": 0.23473914752625372,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499830100555277,
      "loss": 5.9357,
      "mean_token_accuracy": 0.15790688395500183,
      "num_tokens": 4686870.0,
      "step": 2090
    },
    {
      "entropy": 5.953982305526734,
      "epoch": 0.23530072443421127,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998285370304974,
      "loss": 6.0124,
      "mean_token_accuracy": 0.15949206352233886,
      "num_tokens": 4698926.0,
      "step": 2095
    },
    {
      "entropy": 6.0622725009918215,
      "epoch": 0.23586230134216882,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00049982696634709,
      "loss": 6.0882,
      "mean_token_accuracy": 0.14462199285626412,
      "num_tokens": 4709318.0,
      "step": 2100
    },
    {
      "entropy": 5.849839591979981,
      "epoch": 0.23642387825012637,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998253885051048,
      "loss": 5.7796,
      "mean_token_accuracy": 0.16810209602117537,
      "num_tokens": 4718819.0,
      "step": 2105
    },
    {
      "entropy": 5.884680938720703,
      "epoch": 0.2369854551580839,
      "grad_norm": 1.0,
      "learning_rate": 0.000499823803504592,
      "loss": 5.8314,
      "mean_token_accuracy": 0.16093335300683975,
      "num_tokens": 4729513.0,
      "step": 2110
    },
    {
      "entropy": 5.859940433502198,
      "epoch": 0.23754703206604144,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998222113456022,
      "loss": 5.9287,
      "mean_token_accuracy": 0.15834257006645203,
      "num_tokens": 4739875.0,
      "step": 2115
    },
    {
      "entropy": 6.060333490371704,
      "epoch": 0.238108608973999,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998206120281858,
      "loss": 6.0609,
      "mean_token_accuracy": 0.14767249599099158,
      "num_tokens": 4752933.0,
      "step": 2120
    },
    {
      "entropy": 5.938970279693604,
      "epoch": 0.23867018588195654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499819005552394,
      "loss": 5.8526,
      "mean_token_accuracy": 0.16048028618097304,
      "num_tokens": 4762969.0,
      "step": 2125
    },
    {
      "entropy": 5.958646631240844,
      "epoch": 0.23923176278991407,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998173919182779,
      "loss": 5.9439,
      "mean_token_accuracy": 0.15564859807491302,
      "num_tokens": 4773495.0,
      "step": 2130
    },
    {
      "entropy": 5.948372316360474,
      "epoch": 0.23979333969787162,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499815771125889,
      "loss": 5.9773,
      "mean_token_accuracy": 0.16150738894939423,
      "num_tokens": 4785335.0,
      "step": 2135
    },
    {
      "entropy": 6.081469774246216,
      "epoch": 0.24035491660582917,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998141431752785,
      "loss": 6.1175,
      "mean_token_accuracy": 0.14878264293074608,
      "num_tokens": 4798068.0,
      "step": 2140
    },
    {
      "entropy": 5.92440824508667,
      "epoch": 0.24091649351378672,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998125080664987,
      "loss": 5.9373,
      "mean_token_accuracy": 0.1596187397837639,
      "num_tokens": 4809338.0,
      "step": 2145
    },
    {
      "entropy": 5.900418853759765,
      "epoch": 0.24147807042174427,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998108657996012,
      "loss": 5.9278,
      "mean_token_accuracy": 0.1604372188448906,
      "num_tokens": 4819636.0,
      "step": 2150
    },
    {
      "entropy": 5.9336934089660645,
      "epoch": 0.2420396473297018,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998092163746387,
      "loss": 5.9066,
      "mean_token_accuracy": 0.16119810789823533,
      "num_tokens": 4830975.0,
      "step": 2155
    },
    {
      "entropy": 5.893611574172974,
      "epoch": 0.24260122423765934,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998075597916636,
      "loss": 5.9428,
      "mean_token_accuracy": 0.15798580050468444,
      "num_tokens": 4841951.0,
      "step": 2160
    },
    {
      "entropy": 6.011291074752807,
      "epoch": 0.2431628011456169,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998058960507286,
      "loss": 5.9185,
      "mean_token_accuracy": 0.15732098147273063,
      "num_tokens": 4852919.0,
      "step": 2165
    },
    {
      "entropy": 5.955287456512451,
      "epoch": 0.24372437805357444,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998042251518866,
      "loss": 5.9264,
      "mean_token_accuracy": 0.1588020294904709,
      "num_tokens": 4863433.0,
      "step": 2170
    },
    {
      "entropy": 6.019167184829712,
      "epoch": 0.244285954961532,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998025470951908,
      "loss": 6.0132,
      "mean_token_accuracy": 0.15414848923683167,
      "num_tokens": 4875685.0,
      "step": 2175
    },
    {
      "entropy": 5.848620128631592,
      "epoch": 0.24484753186948952,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998008618806949,
      "loss": 5.9081,
      "mean_token_accuracy": 0.15716440230607986,
      "num_tokens": 4886347.0,
      "step": 2180
    },
    {
      "entropy": 5.924430418014526,
      "epoch": 0.24540910877744707,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997991695084523,
      "loss": 5.8998,
      "mean_token_accuracy": 0.16334613561630248,
      "num_tokens": 4896898.0,
      "step": 2185
    },
    {
      "entropy": 5.999474620819091,
      "epoch": 0.24597068568540462,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997974699785169,
      "loss": 6.0128,
      "mean_token_accuracy": 0.15362882167100905,
      "num_tokens": 4908892.0,
      "step": 2190
    },
    {
      "entropy": 5.907583093643188,
      "epoch": 0.24653226259336217,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997957632909429,
      "loss": 5.9471,
      "mean_token_accuracy": 0.15109152495861053,
      "num_tokens": 4920855.0,
      "step": 2195
    },
    {
      "entropy": 6.013603639602661,
      "epoch": 0.24709383950131972,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997940494457846,
      "loss": 5.9325,
      "mean_token_accuracy": 0.16000126004219056,
      "num_tokens": 4932857.0,
      "step": 2200
    },
    {
      "entropy": 5.8864202976226805,
      "epoch": 0.24765541640927724,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997923284430967,
      "loss": 5.9747,
      "mean_token_accuracy": 0.15626695305109023,
      "num_tokens": 4944063.0,
      "step": 2205
    },
    {
      "entropy": 5.871590852737427,
      "epoch": 0.2482169933172348,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997906002829338,
      "loss": 5.8432,
      "mean_token_accuracy": 0.16545321345329284,
      "num_tokens": 4955126.0,
      "step": 2210
    },
    {
      "entropy": 6.042226839065552,
      "epoch": 0.24877857022519234,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499788864965351,
      "loss": 6.0353,
      "mean_token_accuracy": 0.15824319273233414,
      "num_tokens": 4965738.0,
      "step": 2215
    },
    {
      "entropy": 6.032025718688965,
      "epoch": 0.2493401471331499,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997871224904036,
      "loss": 6.0553,
      "mean_token_accuracy": 0.15088909566402436,
      "num_tokens": 4976963.0,
      "step": 2220
    },
    {
      "entropy": 5.968750953674316,
      "epoch": 0.24990172404110744,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997853728581468,
      "loss": 5.9356,
      "mean_token_accuracy": 0.15454284846782684,
      "num_tokens": 4987577.0,
      "step": 2225
    },
    {
      "entropy": 5.988882637023925,
      "epoch": 0.250463300949065,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997836160686368,
      "loss": 6.0657,
      "mean_token_accuracy": 0.15284020453691483,
      "num_tokens": 5000234.0,
      "step": 2230
    },
    {
      "entropy": 5.972144842147827,
      "epoch": 0.25102487785702254,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997818521219291,
      "loss": 5.8895,
      "mean_token_accuracy": 0.15354024320840837,
      "num_tokens": 5011652.0,
      "step": 2235
    },
    {
      "entropy": 5.973802328109741,
      "epoch": 0.25158645476498004,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997800810180803,
      "loss": 5.9329,
      "mean_token_accuracy": 0.1593118280172348,
      "num_tokens": 5023603.0,
      "step": 2240
    },
    {
      "entropy": 5.892467927932739,
      "epoch": 0.2521480316729376,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997783027571464,
      "loss": 5.9091,
      "mean_token_accuracy": 0.15956079661846162,
      "num_tokens": 5034427.0,
      "step": 2245
    },
    {
      "entropy": 5.859362268447876,
      "epoch": 0.25270960858089514,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997765173391841,
      "loss": 5.8988,
      "mean_token_accuracy": 0.15956594049930573,
      "num_tokens": 5045738.0,
      "step": 2250
    },
    {
      "entropy": 5.869083404541016,
      "epoch": 0.2532711854888527,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997747247642505,
      "loss": 5.8353,
      "mean_token_accuracy": 0.15759642720222472,
      "num_tokens": 5057021.0,
      "step": 2255
    },
    {
      "entropy": 5.92234172821045,
      "epoch": 0.25383276239681024,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997729250324023,
      "loss": 5.9244,
      "mean_token_accuracy": 0.15921089351177214,
      "num_tokens": 5067676.0,
      "step": 2260
    },
    {
      "entropy": 5.886710166931152,
      "epoch": 0.2543943393047678,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997711181436971,
      "loss": 5.8615,
      "mean_token_accuracy": 0.15837088972330093,
      "num_tokens": 5079422.0,
      "step": 2265
    },
    {
      "entropy": 5.822842216491699,
      "epoch": 0.25495591621272534,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997693040981922,
      "loss": 5.8851,
      "mean_token_accuracy": 0.16085467636585235,
      "num_tokens": 5091242.0,
      "step": 2270
    },
    {
      "entropy": 6.006762552261352,
      "epoch": 0.2555174931206829,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997674828959456,
      "loss": 5.8909,
      "mean_token_accuracy": 0.15583527684211732,
      "num_tokens": 5102251.0,
      "step": 2275
    },
    {
      "entropy": 5.857793092727661,
      "epoch": 0.25607907002864044,
      "grad_norm": 1.5,
      "learning_rate": 0.0004997656545370151,
      "loss": 5.9104,
      "mean_token_accuracy": 0.1610156461596489,
      "num_tokens": 5113276.0,
      "step": 2280
    },
    {
      "entropy": 5.934245014190674,
      "epoch": 0.256640646936598,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499763819021459,
      "loss": 5.9191,
      "mean_token_accuracy": 0.15731906294822692,
      "num_tokens": 5124472.0,
      "step": 2285
    },
    {
      "entropy": 5.863279104232788,
      "epoch": 0.2572022238445555,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997619763493357,
      "loss": 5.8698,
      "mean_token_accuracy": 0.1613835647702217,
      "num_tokens": 5134896.0,
      "step": 2290
    },
    {
      "entropy": 5.962464809417725,
      "epoch": 0.25776380075251304,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499760126520704,
      "loss": 5.95,
      "mean_token_accuracy": 0.1634097322821617,
      "num_tokens": 5145115.0,
      "step": 2295
    },
    {
      "entropy": 5.927261829376221,
      "epoch": 0.2583253776604706,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997582695356225,
      "loss": 5.9289,
      "mean_token_accuracy": 0.15928202718496323,
      "num_tokens": 5155620.0,
      "step": 2300
    },
    {
      "entropy": 6.070884418487549,
      "epoch": 0.25888695456842814,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997564053941507,
      "loss": 6.0126,
      "mean_token_accuracy": 0.15451653599739074,
      "num_tokens": 5166260.0,
      "step": 2305
    },
    {
      "entropy": 5.954886245727539,
      "epoch": 0.2594485314763857,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997545340963478,
      "loss": 5.8733,
      "mean_token_accuracy": 0.15934662520885468,
      "num_tokens": 5177143.0,
      "step": 2310
    },
    {
      "entropy": 5.921825218200683,
      "epoch": 0.26001010838434324,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997526556422733,
      "loss": 6.0233,
      "mean_token_accuracy": 0.15293444991111754,
      "num_tokens": 5187795.0,
      "step": 2315
    },
    {
      "entropy": 5.922110557556152,
      "epoch": 0.2605716852923008,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997507700319872,
      "loss": 5.8811,
      "mean_token_accuracy": 0.16001545637845993,
      "num_tokens": 5199365.0,
      "step": 2320
    },
    {
      "entropy": 5.971580982208252,
      "epoch": 0.26113326220025834,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004997488772655492,
      "loss": 6.0976,
      "mean_token_accuracy": 0.15617025047540664,
      "num_tokens": 5212264.0,
      "step": 2325
    },
    {
      "entropy": 6.042904090881348,
      "epoch": 0.2616948391082159,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997469773430199,
      "loss": 5.9253,
      "mean_token_accuracy": 0.15450814068317414,
      "num_tokens": 5223544.0,
      "step": 2330
    },
    {
      "entropy": 5.868034029006958,
      "epoch": 0.2622564160161734,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997450702644596,
      "loss": 5.875,
      "mean_token_accuracy": 0.1604365214705467,
      "num_tokens": 5234713.0,
      "step": 2335
    },
    {
      "entropy": 5.9823534965515135,
      "epoch": 0.26281799292413094,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997431560299292,
      "loss": 5.9591,
      "mean_token_accuracy": 0.15298095047473909,
      "num_tokens": 5246103.0,
      "step": 2340
    },
    {
      "entropy": 5.814671277999878,
      "epoch": 0.2633795698320885,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997412346394894,
      "loss": 5.7917,
      "mean_token_accuracy": 0.16624176353216172,
      "num_tokens": 5257372.0,
      "step": 2345
    },
    {
      "entropy": 5.792899942398071,
      "epoch": 0.26394114674004604,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997393060932017,
      "loss": 5.8609,
      "mean_token_accuracy": 0.16722605973482133,
      "num_tokens": 5268677.0,
      "step": 2350
    },
    {
      "entropy": 5.997194576263428,
      "epoch": 0.2645027236480036,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997373703911273,
      "loss": 5.9247,
      "mean_token_accuracy": 0.16152105778455733,
      "num_tokens": 5279056.0,
      "step": 2355
    },
    {
      "entropy": 5.786316013336181,
      "epoch": 0.26506430055596114,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997354275333277,
      "loss": 5.8069,
      "mean_token_accuracy": 0.16789699494838714,
      "num_tokens": 5289310.0,
      "step": 2360
    },
    {
      "entropy": 5.953582143783569,
      "epoch": 0.2656258774639187,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000499733477519865,
      "loss": 5.8956,
      "mean_token_accuracy": 0.16771894991397857,
      "num_tokens": 5300619.0,
      "step": 2365
    },
    {
      "entropy": 6.053738594055176,
      "epoch": 0.26618745437187624,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997315203508013,
      "loss": 6.0608,
      "mean_token_accuracy": 0.1562006339430809,
      "num_tokens": 5311554.0,
      "step": 2370
    },
    {
      "entropy": 5.966435718536377,
      "epoch": 0.2667490312798338,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997295560261988,
      "loss": 5.883,
      "mean_token_accuracy": 0.1612299293279648,
      "num_tokens": 5323592.0,
      "step": 2375
    },
    {
      "entropy": 5.942288255691528,
      "epoch": 0.26731060818779134,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049972758454612,
      "loss": 5.9767,
      "mean_token_accuracy": 0.15383181422948838,
      "num_tokens": 5334978.0,
      "step": 2380
    },
    {
      "entropy": 5.898320913314819,
      "epoch": 0.26787218509574884,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997256059106279,
      "loss": 5.8667,
      "mean_token_accuracy": 0.15778945684432982,
      "num_tokens": 5345724.0,
      "step": 2385
    },
    {
      "entropy": 5.94132432937622,
      "epoch": 0.2684337620037064,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997236201197853,
      "loss": 5.9989,
      "mean_token_accuracy": 0.15381949245929719,
      "num_tokens": 5357141.0,
      "step": 2390
    },
    {
      "entropy": 5.9436756610870365,
      "epoch": 0.26899533891166394,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997216271736554,
      "loss": 5.995,
      "mean_token_accuracy": 0.15819256901741027,
      "num_tokens": 5367889.0,
      "step": 2395
    },
    {
      "entropy": 5.786690282821655,
      "epoch": 0.2695569158196215,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997196270723018,
      "loss": 5.7261,
      "mean_token_accuracy": 0.16399870365858077,
      "num_tokens": 5379329.0,
      "step": 2400
    },
    {
      "entropy": 5.895531129837036,
      "epoch": 0.27011849272757904,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997176198157881,
      "loss": 5.8852,
      "mean_token_accuracy": 0.1617034539580345,
      "num_tokens": 5389525.0,
      "step": 2405
    },
    {
      "entropy": 5.895331811904907,
      "epoch": 0.2706800696355366,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997156054041781,
      "loss": 5.7958,
      "mean_token_accuracy": 0.16883354038000106,
      "num_tokens": 5401061.0,
      "step": 2410
    },
    {
      "entropy": 5.873462343215943,
      "epoch": 0.27124164654349414,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997135838375362,
      "loss": 5.8949,
      "mean_token_accuracy": 0.1616640195250511,
      "num_tokens": 5411791.0,
      "step": 2415
    },
    {
      "entropy": 5.908325338363648,
      "epoch": 0.2718032234514517,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997115551159267,
      "loss": 5.8467,
      "mean_token_accuracy": 0.16689947098493577,
      "num_tokens": 5422790.0,
      "step": 2420
    },
    {
      "entropy": 5.738713026046753,
      "epoch": 0.27236480035940924,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499709519239414,
      "loss": 5.7951,
      "mean_token_accuracy": 0.1616228774189949,
      "num_tokens": 5434170.0,
      "step": 2425
    },
    {
      "entropy": 5.972346496582031,
      "epoch": 0.2729263772673668,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004997074762080632,
      "loss": 5.922,
      "mean_token_accuracy": 0.15746588557958602,
      "num_tokens": 5445948.0,
      "step": 2430
    },
    {
      "entropy": 5.907391738891602,
      "epoch": 0.2734879541753243,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499705426021939,
      "loss": 5.8828,
      "mean_token_accuracy": 0.15904785543680192,
      "num_tokens": 5457786.0,
      "step": 2435
    },
    {
      "entropy": 5.784610605239868,
      "epoch": 0.27404953108328184,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499703368681107,
      "loss": 5.8589,
      "mean_token_accuracy": 0.16070218682289122,
      "num_tokens": 5469434.0,
      "step": 2440
    },
    {
      "entropy": 5.8818888664245605,
      "epoch": 0.2746111079912394,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997013041856326,
      "loss": 5.8549,
      "mean_token_accuracy": 0.1531570464372635,
      "num_tokens": 5480393.0,
      "step": 2445
    },
    {
      "entropy": 5.844133996963501,
      "epoch": 0.27517268489919694,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996992325355815,
      "loss": 5.8264,
      "mean_token_accuracy": 0.16376967430114747,
      "num_tokens": 5490452.0,
      "step": 2450
    },
    {
      "entropy": 5.843231868743897,
      "epoch": 0.2757342618071545,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996971537310198,
      "loss": 5.7996,
      "mean_token_accuracy": 0.16604796648025513,
      "num_tokens": 5501404.0,
      "step": 2455
    },
    {
      "entropy": 5.97615795135498,
      "epoch": 0.27629583871511204,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996950677720133,
      "loss": 5.9411,
      "mean_token_accuracy": 0.15309768170118332,
      "num_tokens": 5512938.0,
      "step": 2460
    },
    {
      "entropy": 5.879659843444824,
      "epoch": 0.2768574156230696,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499692974658629,
      "loss": 5.8641,
      "mean_token_accuracy": 0.15448336452245712,
      "num_tokens": 5524009.0,
      "step": 2465
    },
    {
      "entropy": 5.961763477325439,
      "epoch": 0.27741899253102714,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996908743909331,
      "loss": 5.9491,
      "mean_token_accuracy": 0.15658549070358277,
      "num_tokens": 5535930.0,
      "step": 2470
    },
    {
      "entropy": 5.943891477584839,
      "epoch": 0.2779805694389847,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996887669689926,
      "loss": 5.9378,
      "mean_token_accuracy": 0.1560988336801529,
      "num_tokens": 5547295.0,
      "step": 2475
    },
    {
      "entropy": 5.934115314483643,
      "epoch": 0.2785421463469422,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996866523928747,
      "loss": 5.8393,
      "mean_token_accuracy": 0.1640307277441025,
      "num_tokens": 5558447.0,
      "step": 2480
    },
    {
      "entropy": 5.91972165107727,
      "epoch": 0.27910372325489974,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996845306626467,
      "loss": 5.9369,
      "mean_token_accuracy": 0.16300042271614074,
      "num_tokens": 5569822.0,
      "step": 2485
    },
    {
      "entropy": 5.9508904933929445,
      "epoch": 0.2796653001628573,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996824017783759,
      "loss": 5.9902,
      "mean_token_accuracy": 0.15508754998445512,
      "num_tokens": 5582284.0,
      "step": 2490
    },
    {
      "entropy": 5.9683637619018555,
      "epoch": 0.28022687707081484,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996802657401304,
      "loss": 5.879,
      "mean_token_accuracy": 0.1611318975687027,
      "num_tokens": 5593765.0,
      "step": 2495
    },
    {
      "entropy": 5.801318073272705,
      "epoch": 0.2807884539787724,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996781225479781,
      "loss": 5.8487,
      "mean_token_accuracy": 0.15998225808143615,
      "num_tokens": 5605080.0,
      "step": 2500
    },
    {
      "entropy": 6.015531349182129,
      "epoch": 0.28135003088672994,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996759722019872,
      "loss": 6.011,
      "mean_token_accuracy": 0.15439892411231995,
      "num_tokens": 5616739.0,
      "step": 2505
    },
    {
      "entropy": 5.904989242553711,
      "epoch": 0.2819116077946875,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996738147022262,
      "loss": 5.8735,
      "mean_token_accuracy": 0.1576665908098221,
      "num_tokens": 5628779.0,
      "step": 2510
    },
    {
      "entropy": 5.889083909988403,
      "epoch": 0.28247318470264504,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996716500487638,
      "loss": 5.909,
      "mean_token_accuracy": 0.15748686641454696,
      "num_tokens": 5639168.0,
      "step": 2515
    },
    {
      "entropy": 5.893275928497315,
      "epoch": 0.2830347616106026,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499669478241669,
      "loss": 5.885,
      "mean_token_accuracy": 0.15599167197942734,
      "num_tokens": 5650739.0,
      "step": 2520
    },
    {
      "entropy": 5.933381366729736,
      "epoch": 0.28359633851856014,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996672992810108,
      "loss": 5.9054,
      "mean_token_accuracy": 0.15816623866558074,
      "num_tokens": 5662630.0,
      "step": 2525
    },
    {
      "entropy": 5.921469497680664,
      "epoch": 0.28415791542651764,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996651131668587,
      "loss": 5.9083,
      "mean_token_accuracy": 0.1671989306807518,
      "num_tokens": 5673473.0,
      "step": 2530
    },
    {
      "entropy": 5.915191793441773,
      "epoch": 0.2847194923344752,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996629198992823,
      "loss": 6.0544,
      "mean_token_accuracy": 0.15136269852519035,
      "num_tokens": 5686577.0,
      "step": 2535
    },
    {
      "entropy": 5.963024854660034,
      "epoch": 0.28528106924243274,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996607194783513,
      "loss": 5.8035,
      "mean_token_accuracy": 0.15977411419153215,
      "num_tokens": 5697570.0,
      "step": 2540
    },
    {
      "entropy": 5.849263906478882,
      "epoch": 0.2858426461503903,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996585119041359,
      "loss": 5.8717,
      "mean_token_accuracy": 0.15102790147066117,
      "num_tokens": 5708601.0,
      "step": 2545
    },
    {
      "entropy": 5.817156410217285,
      "epoch": 0.28640422305834784,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996562971767063,
      "loss": 5.8333,
      "mean_token_accuracy": 0.1580057695508003,
      "num_tokens": 5719619.0,
      "step": 2550
    },
    {
      "entropy": 5.881134510040283,
      "epoch": 0.2869657999663054,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499654075296133,
      "loss": 5.854,
      "mean_token_accuracy": 0.16304896473884584,
      "num_tokens": 5730061.0,
      "step": 2555
    },
    {
      "entropy": 5.887245082855225,
      "epoch": 0.28752737687426294,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499651846262487,
      "loss": 5.8781,
      "mean_token_accuracy": 0.16307636797428132,
      "num_tokens": 5741579.0,
      "step": 2560
    },
    {
      "entropy": 5.929854440689087,
      "epoch": 0.2880889537822205,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996496100758389,
      "loss": 5.8899,
      "mean_token_accuracy": 0.16015600562095642,
      "num_tokens": 5754001.0,
      "step": 2565
    },
    {
      "entropy": 5.956954383850098,
      "epoch": 0.28865053069017804,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996473667362604,
      "loss": 5.9762,
      "mean_token_accuracy": 0.15367593243718147,
      "num_tokens": 5766224.0,
      "step": 2570
    },
    {
      "entropy": 5.9173064708709715,
      "epoch": 0.28921210759813554,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996451162438224,
      "loss": 5.9514,
      "mean_token_accuracy": 0.1529170662164688,
      "num_tokens": 5777475.0,
      "step": 2575
    },
    {
      "entropy": 5.9922263622283936,
      "epoch": 0.2897736845060931,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996428585985967,
      "loss": 5.9243,
      "mean_token_accuracy": 0.15300409942865373,
      "num_tokens": 5788359.0,
      "step": 2580
    },
    {
      "entropy": 5.833764743804932,
      "epoch": 0.29033526141405064,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996405938006553,
      "loss": 5.88,
      "mean_token_accuracy": 0.1589822694659233,
      "num_tokens": 5799116.0,
      "step": 2585
    },
    {
      "entropy": 5.849058294296265,
      "epoch": 0.2908968383220082,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996383218500703,
      "loss": 5.8163,
      "mean_token_accuracy": 0.16485902518033982,
      "num_tokens": 5811221.0,
      "step": 2590
    },
    {
      "entropy": 5.903125619888305,
      "epoch": 0.29145841522996574,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499636042746914,
      "loss": 5.8986,
      "mean_token_accuracy": 0.1597510278224945,
      "num_tokens": 5822099.0,
      "step": 2595
    },
    {
      "entropy": 5.908592653274536,
      "epoch": 0.2920199921379233,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499633756491259,
      "loss": 5.8603,
      "mean_token_accuracy": 0.154727041721344,
      "num_tokens": 5832836.0,
      "step": 2600
    },
    {
      "entropy": 5.813444709777832,
      "epoch": 0.29258156904588084,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996314630831781,
      "loss": 5.8066,
      "mean_token_accuracy": 0.15686748921871185,
      "num_tokens": 5842892.0,
      "step": 2605
    },
    {
      "entropy": 5.8215662956237795,
      "epoch": 0.2931431459538384,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996291625227443,
      "loss": 5.7707,
      "mean_token_accuracy": 0.17027000188827515,
      "num_tokens": 5852909.0,
      "step": 2610
    },
    {
      "entropy": 5.975628280639649,
      "epoch": 0.29370472286179594,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996268548100307,
      "loss": 5.9509,
      "mean_token_accuracy": 0.15499749705195426,
      "num_tokens": 5864676.0,
      "step": 2615
    },
    {
      "entropy": 5.67653169631958,
      "epoch": 0.2942662997697535,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499624539945111,
      "loss": 5.6832,
      "mean_token_accuracy": 0.17367247641086578,
      "num_tokens": 5875258.0,
      "step": 2620
    },
    {
      "entropy": 5.918656921386718,
      "epoch": 0.294827876677711,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996222179280589,
      "loss": 5.8505,
      "mean_token_accuracy": 0.16116076856851577,
      "num_tokens": 5885611.0,
      "step": 2625
    },
    {
      "entropy": 5.846941518783569,
      "epoch": 0.29538945358566854,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996198887589481,
      "loss": 5.8257,
      "mean_token_accuracy": 0.15930989608168603,
      "num_tokens": 5896726.0,
      "step": 2630
    },
    {
      "entropy": 5.917367076873779,
      "epoch": 0.2959510304936261,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996175524378531,
      "loss": 5.851,
      "mean_token_accuracy": 0.1528210312128067,
      "num_tokens": 5907860.0,
      "step": 2635
    },
    {
      "entropy": 5.806666278839112,
      "epoch": 0.29651260740158364,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499615208964848,
      "loss": 5.798,
      "mean_token_accuracy": 0.1638117402791977,
      "num_tokens": 5918676.0,
      "step": 2640
    },
    {
      "entropy": 5.9251220703125,
      "epoch": 0.2970741843095412,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996128583400077,
      "loss": 5.8618,
      "mean_token_accuracy": 0.16403974294662477,
      "num_tokens": 5928646.0,
      "step": 2645
    },
    {
      "entropy": 5.816641330718994,
      "epoch": 0.29763576121749874,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996105005634066,
      "loss": 5.8735,
      "mean_token_accuracy": 0.1569816768169403,
      "num_tokens": 5938990.0,
      "step": 2650
    },
    {
      "entropy": 5.8617603302001955,
      "epoch": 0.2981973381254563,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996081356351202,
      "loss": 5.8194,
      "mean_token_accuracy": 0.16183129400014878,
      "num_tokens": 5948811.0,
      "step": 2655
    },
    {
      "entropy": 5.812265920639038,
      "epoch": 0.29875891503341384,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996057635552237,
      "loss": 5.8406,
      "mean_token_accuracy": 0.16514092236757277,
      "num_tokens": 5959227.0,
      "step": 2660
    },
    {
      "entropy": 5.8553258895874025,
      "epoch": 0.2993204919413714,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996033843237925,
      "loss": 5.7535,
      "mean_token_accuracy": 0.16563905775547028,
      "num_tokens": 5970546.0,
      "step": 2665
    },
    {
      "entropy": 5.982068204879761,
      "epoch": 0.29988206884932894,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004996009979409025,
      "loss": 6.0502,
      "mean_token_accuracy": 0.147099506855011,
      "num_tokens": 5983936.0,
      "step": 2670
    },
    {
      "entropy": 5.802636957168579,
      "epoch": 0.30044364575728644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995986044066295,
      "loss": 5.8494,
      "mean_token_accuracy": 0.15568828135728835,
      "num_tokens": 5994963.0,
      "step": 2675
    },
    {
      "entropy": 5.762459421157837,
      "epoch": 0.301005222665244,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00049959620372105,
      "loss": 5.8079,
      "mean_token_accuracy": 0.16270437240600585,
      "num_tokens": 6005836.0,
      "step": 2680
    },
    {
      "entropy": 5.895258378982544,
      "epoch": 0.30156679957320154,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995937958842401,
      "loss": 5.8057,
      "mean_token_accuracy": 0.16682981103658676,
      "num_tokens": 6017408.0,
      "step": 2685
    },
    {
      "entropy": 5.811723136901856,
      "epoch": 0.3021283764811591,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995913808962767,
      "loss": 5.8192,
      "mean_token_accuracy": 0.15878787338733674,
      "num_tokens": 6029281.0,
      "step": 2690
    },
    {
      "entropy": 5.850067806243897,
      "epoch": 0.30268995338911664,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995889587572366,
      "loss": 5.8088,
      "mean_token_accuracy": 0.16636761128902436,
      "num_tokens": 6039354.0,
      "step": 2695
    },
    {
      "entropy": 5.8515503883361815,
      "epoch": 0.3032515302970742,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499586529467197,
      "loss": 5.785,
      "mean_token_accuracy": 0.15962323993444444,
      "num_tokens": 6050053.0,
      "step": 2700
    },
    {
      "entropy": 5.90614161491394,
      "epoch": 0.30381310720503174,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995840930262351,
      "loss": 5.8451,
      "mean_token_accuracy": 0.15831548869609832,
      "num_tokens": 6060221.0,
      "step": 2705
    },
    {
      "entropy": 5.824668931961059,
      "epoch": 0.3043746841129893,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995816494344287,
      "loss": 5.859,
      "mean_token_accuracy": 0.1601285994052887,
      "num_tokens": 6070963.0,
      "step": 2710
    },
    {
      "entropy": 5.989436864852905,
      "epoch": 0.30493626102094684,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995791986918554,
      "loss": 5.9386,
      "mean_token_accuracy": 0.15705745071172714,
      "num_tokens": 6082028.0,
      "step": 2715
    },
    {
      "entropy": 5.949339580535889,
      "epoch": 0.30549783792890434,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995767407985933,
      "loss": 5.8736,
      "mean_token_accuracy": 0.16265355795621872,
      "num_tokens": 6092778.0,
      "step": 2720
    },
    {
      "entropy": 5.677318954467774,
      "epoch": 0.3060594148368619,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995742757547207,
      "loss": 5.6421,
      "mean_token_accuracy": 0.16859253942966462,
      "num_tokens": 6103093.0,
      "step": 2725
    },
    {
      "entropy": 5.772259378433228,
      "epoch": 0.30662099174481944,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499571803560316,
      "loss": 5.7347,
      "mean_token_accuracy": 0.16355275958776475,
      "num_tokens": 6113516.0,
      "step": 2730
    },
    {
      "entropy": 5.810375738143921,
      "epoch": 0.307182568652777,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995693242154581,
      "loss": 5.8036,
      "mean_token_accuracy": 0.15890752524137497,
      "num_tokens": 6125723.0,
      "step": 2735
    },
    {
      "entropy": 5.847309589385986,
      "epoch": 0.30774414556073454,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004995668377202258,
      "loss": 5.8136,
      "mean_token_accuracy": 0.16112632900476456,
      "num_tokens": 6137091.0,
      "step": 2740
    },
    {
      "entropy": 5.77405891418457,
      "epoch": 0.3083057224686921,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995643440746982,
      "loss": 5.7528,
      "mean_token_accuracy": 0.16945552676916123,
      "num_tokens": 6147941.0,
      "step": 2745
    },
    {
      "entropy": 5.732852792739868,
      "epoch": 0.30886729937664964,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995618432789548,
      "loss": 5.7549,
      "mean_token_accuracy": 0.16648126393556595,
      "num_tokens": 6158493.0,
      "step": 2750
    },
    {
      "entropy": 5.8287242412567135,
      "epoch": 0.3094288762846072,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995593353330753,
      "loss": 5.8915,
      "mean_token_accuracy": 0.15722588896751405,
      "num_tokens": 6170759.0,
      "step": 2755
    },
    {
      "entropy": 5.898643875122071,
      "epoch": 0.30999045319256474,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995568202371393,
      "loss": 5.724,
      "mean_token_accuracy": 0.16514806002378463,
      "num_tokens": 6182488.0,
      "step": 2760
    },
    {
      "entropy": 5.841040992736817,
      "epoch": 0.3105520301005223,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995542979912272,
      "loss": 5.9165,
      "mean_token_accuracy": 0.15282527059316636,
      "num_tokens": 6193849.0,
      "step": 2765
    },
    {
      "entropy": 5.800279855728149,
      "epoch": 0.3111136070084798,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995517685954192,
      "loss": 5.7598,
      "mean_token_accuracy": 0.16608615219593048,
      "num_tokens": 6204132.0,
      "step": 2770
    },
    {
      "entropy": 5.946976375579834,
      "epoch": 0.31167518391643734,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995492320497958,
      "loss": 5.9018,
      "mean_token_accuracy": 0.1554347574710846,
      "num_tokens": 6215547.0,
      "step": 2775
    },
    {
      "entropy": 5.905535459518433,
      "epoch": 0.3122367608243949,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995466883544376,
      "loss": 5.9293,
      "mean_token_accuracy": 0.15042147785425186,
      "num_tokens": 6227150.0,
      "step": 2780
    },
    {
      "entropy": 5.8855243682861325,
      "epoch": 0.31279833773235244,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995441375094258,
      "loss": 5.9248,
      "mean_token_accuracy": 0.15840522199869156,
      "num_tokens": 6237387.0,
      "step": 2785
    },
    {
      "entropy": 5.872175312042236,
      "epoch": 0.31335991464031,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995415795148417,
      "loss": 5.812,
      "mean_token_accuracy": 0.16368252038955688,
      "num_tokens": 6248623.0,
      "step": 2790
    },
    {
      "entropy": 5.76340684890747,
      "epoch": 0.31392149154826754,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995390143707665,
      "loss": 5.7152,
      "mean_token_accuracy": 0.17597836554050444,
      "num_tokens": 6258596.0,
      "step": 2795
    },
    {
      "entropy": 5.822847366333008,
      "epoch": 0.3144830684562251,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995364420772821,
      "loss": 5.9642,
      "mean_token_accuracy": 0.15163593217730523,
      "num_tokens": 6270529.0,
      "step": 2800
    },
    {
      "entropy": 6.0933667659759525,
      "epoch": 0.31504464536418264,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995338626344702,
      "loss": 5.8579,
      "mean_token_accuracy": 0.16046109050512314,
      "num_tokens": 6280891.0,
      "step": 2805
    },
    {
      "entropy": 5.7229210376739506,
      "epoch": 0.3156062222721402,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499531276042413,
      "loss": 5.7988,
      "mean_token_accuracy": 0.15810492336750032,
      "num_tokens": 6292247.0,
      "step": 2810
    },
    {
      "entropy": 5.79867639541626,
      "epoch": 0.31616779918009774,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995286823011929,
      "loss": 5.7847,
      "mean_token_accuracy": 0.16373041421175002,
      "num_tokens": 6302403.0,
      "step": 2815
    },
    {
      "entropy": 5.7128393173217775,
      "epoch": 0.31672937608805524,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995260814108926,
      "loss": 5.6696,
      "mean_token_accuracy": 0.1683839187026024,
      "num_tokens": 6313901.0,
      "step": 2820
    },
    {
      "entropy": 5.921040821075439,
      "epoch": 0.3172909529960128,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995234733715947,
      "loss": 5.8928,
      "mean_token_accuracy": 0.15783516615629195,
      "num_tokens": 6323711.0,
      "step": 2825
    },
    {
      "entropy": 5.817658710479736,
      "epoch": 0.31785252990397034,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995208581833824,
      "loss": 5.7877,
      "mean_token_accuracy": 0.1626381605863571,
      "num_tokens": 6334937.0,
      "step": 2830
    },
    {
      "entropy": 5.928051614761353,
      "epoch": 0.3184141068119279,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995182358463388,
      "loss": 5.8838,
      "mean_token_accuracy": 0.15879348814487457,
      "num_tokens": 6345723.0,
      "step": 2835
    },
    {
      "entropy": 5.844352579116821,
      "epoch": 0.31897568371988544,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995156063605475,
      "loss": 5.8219,
      "mean_token_accuracy": 0.15776338502764703,
      "num_tokens": 6357573.0,
      "step": 2840
    },
    {
      "entropy": 5.8675610542297365,
      "epoch": 0.319537260627843,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995129697260922,
      "loss": 5.8322,
      "mean_token_accuracy": 0.16539549678564072,
      "num_tokens": 6367417.0,
      "step": 2845
    },
    {
      "entropy": 5.872849082946777,
      "epoch": 0.32009883753580054,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995103259430569,
      "loss": 5.841,
      "mean_token_accuracy": 0.16827336698770523,
      "num_tokens": 6377753.0,
      "step": 2850
    },
    {
      "entropy": 5.839739036560059,
      "epoch": 0.3206604144437581,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995076750115258,
      "loss": 5.8514,
      "mean_token_accuracy": 0.15937740802764894,
      "num_tokens": 6388613.0,
      "step": 2855
    },
    {
      "entropy": 5.815576601028442,
      "epoch": 0.32122199135171564,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995050169315832,
      "loss": 5.796,
      "mean_token_accuracy": 0.1593346953392029,
      "num_tokens": 6399415.0,
      "step": 2860
    },
    {
      "entropy": 5.867472791671753,
      "epoch": 0.32178356825967314,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995023517033138,
      "loss": 5.8285,
      "mean_token_accuracy": 0.15754698514938353,
      "num_tokens": 6411435.0,
      "step": 2865
    },
    {
      "entropy": 5.926137781143188,
      "epoch": 0.3223451451676307,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994996793268024,
      "loss": 5.976,
      "mean_token_accuracy": 0.15850219279527664,
      "num_tokens": 6423030.0,
      "step": 2870
    },
    {
      "entropy": 5.888271427154541,
      "epoch": 0.32290672207558824,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994969998021342,
      "loss": 5.7765,
      "mean_token_accuracy": 0.16209373772144317,
      "num_tokens": 6434330.0,
      "step": 2875
    },
    {
      "entropy": 5.859491014480591,
      "epoch": 0.3234682989835458,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994943131293944,
      "loss": 5.8646,
      "mean_token_accuracy": 0.16070864647626876,
      "num_tokens": 6446452.0,
      "step": 2880
    },
    {
      "entropy": 5.8355710983276365,
      "epoch": 0.32402987589150334,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994916193086687,
      "loss": 5.8139,
      "mean_token_accuracy": 0.1690973237156868,
      "num_tokens": 6456723.0,
      "step": 2885
    },
    {
      "entropy": 5.899684286117553,
      "epoch": 0.3245914527994609,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994889183400428,
      "loss": 5.9365,
      "mean_token_accuracy": 0.15811522603034972,
      "num_tokens": 6468117.0,
      "step": 2890
    },
    {
      "entropy": 5.826454973220825,
      "epoch": 0.32515302970741844,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994862102236025,
      "loss": 5.7204,
      "mean_token_accuracy": 0.16832807213068007,
      "num_tokens": 6479465.0,
      "step": 2895
    },
    {
      "entropy": 5.89757194519043,
      "epoch": 0.325714606615376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994834949594342,
      "loss": 5.9526,
      "mean_token_accuracy": 0.1629941999912262,
      "num_tokens": 6490868.0,
      "step": 2900
    },
    {
      "entropy": 5.972553682327271,
      "epoch": 0.32627618352333354,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994807725476246,
      "loss": 5.8245,
      "mean_token_accuracy": 0.16008032113313675,
      "num_tokens": 6502126.0,
      "step": 2905
    },
    {
      "entropy": 5.72076644897461,
      "epoch": 0.3268377604312911,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994780429882599,
      "loss": 5.9091,
      "mean_token_accuracy": 0.15670095086097718,
      "num_tokens": 6514382.0,
      "step": 2910
    },
    {
      "entropy": 5.882012605667114,
      "epoch": 0.3273993373392486,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004994753062814274,
      "loss": 5.7914,
      "mean_token_accuracy": 0.15964423269033431,
      "num_tokens": 6526272.0,
      "step": 2915
    },
    {
      "entropy": 5.866258335113526,
      "epoch": 0.32796091424720614,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994725624272139,
      "loss": 5.8134,
      "mean_token_accuracy": 0.16369341611862182,
      "num_tokens": 6537511.0,
      "step": 2920
    },
    {
      "entropy": 5.805434799194336,
      "epoch": 0.3285224911551637,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994698114257071,
      "loss": 5.7523,
      "mean_token_accuracy": 0.16356185525655748,
      "num_tokens": 6549737.0,
      "step": 2925
    },
    {
      "entropy": 5.735032749176026,
      "epoch": 0.32908406806312124,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994670532769944,
      "loss": 5.842,
      "mean_token_accuracy": 0.16296304613351822,
      "num_tokens": 6560318.0,
      "step": 2930
    },
    {
      "entropy": 5.818574094772339,
      "epoch": 0.3296456449710788,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994642879811637,
      "loss": 5.7497,
      "mean_token_accuracy": 0.16560823023319243,
      "num_tokens": 6570575.0,
      "step": 2935
    },
    {
      "entropy": 5.807329988479614,
      "epoch": 0.33020722187903634,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499461515538303,
      "loss": 5.8172,
      "mean_token_accuracy": 0.1601566642522812,
      "num_tokens": 6582083.0,
      "step": 2940
    },
    {
      "entropy": 6.00074577331543,
      "epoch": 0.3307687987869939,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994587359485006,
      "loss": 5.9414,
      "mean_token_accuracy": 0.16097475886344909,
      "num_tokens": 6592240.0,
      "step": 2945
    },
    {
      "entropy": 5.806456136703491,
      "epoch": 0.33133037569495144,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499455949211845,
      "loss": 5.7185,
      "mean_token_accuracy": 0.1677255555987358,
      "num_tokens": 6602680.0,
      "step": 2950
    },
    {
      "entropy": 5.796722078323365,
      "epoch": 0.331891952602909,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994531553284248,
      "loss": 5.8381,
      "mean_token_accuracy": 0.16280671954154968,
      "num_tokens": 6613399.0,
      "step": 2955
    },
    {
      "entropy": 5.918023681640625,
      "epoch": 0.3324535295108665,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994503542983293,
      "loss": 5.91,
      "mean_token_accuracy": 0.15692875981330873,
      "num_tokens": 6624410.0,
      "step": 2960
    },
    {
      "entropy": 5.893260526657104,
      "epoch": 0.33301510641882404,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994475461216472,
      "loss": 5.7623,
      "mean_token_accuracy": 0.1689491778612137,
      "num_tokens": 6635523.0,
      "step": 2965
    },
    {
      "entropy": 5.8270707607269285,
      "epoch": 0.3335766833267816,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994447307984684,
      "loss": 5.9249,
      "mean_token_accuracy": 0.1531895712018013,
      "num_tokens": 6648416.0,
      "step": 2970
    },
    {
      "entropy": 5.764549970626831,
      "epoch": 0.33413826023473914,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994419083288823,
      "loss": 5.7112,
      "mean_token_accuracy": 0.16065409630537034,
      "num_tokens": 6658907.0,
      "step": 2975
    },
    {
      "entropy": 5.840141725540161,
      "epoch": 0.3346998371426967,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994390787129787,
      "loss": 5.8754,
      "mean_token_accuracy": 0.1610553666949272,
      "num_tokens": 6669682.0,
      "step": 2980
    },
    {
      "entropy": 5.9260087490081785,
      "epoch": 0.33526141405065424,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994362419508478,
      "loss": 5.9334,
      "mean_token_accuracy": 0.1567075252532959,
      "num_tokens": 6681919.0,
      "step": 2985
    },
    {
      "entropy": 5.9462815284729,
      "epoch": 0.3358229909586118,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00049943339804258,
      "loss": 5.8707,
      "mean_token_accuracy": 0.1613776594400406,
      "num_tokens": 6693988.0,
      "step": 2990
    },
    {
      "entropy": 5.7616931915283205,
      "epoch": 0.33638456786656934,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004994305469882657,
      "loss": 5.8059,
      "mean_token_accuracy": 0.15830975472927095,
      "num_tokens": 6705413.0,
      "step": 2995
    },
    {
      "entropy": 5.859702157974243,
      "epoch": 0.3369461447745269,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994276887879957,
      "loss": 5.763,
      "mean_token_accuracy": 0.16223098933696747,
      "num_tokens": 6716217.0,
      "step": 3000
    },
    {
      "epoch": 0.3369461447745269,
      "eval_entropy": 5.720634406333059,
      "eval_loss": 5.844816207885742,
      "eval_mean_token_accuracy": 0.1682441681402459,
      "eval_num_tokens": 6716217.0,
      "eval_runtime": 23.7412,
      "eval_samples_per_second": 1306.296,
      "eval_steps_per_second": 163.303,
      "step": 3000
    },
    {
      "entropy": 5.799144506454468,
      "epoch": 0.33750772168248444,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994248234418611,
      "loss": 5.8366,
      "mean_token_accuracy": 0.162978595495224,
      "num_tokens": 6728059.0,
      "step": 3005
    },
    {
      "entropy": 5.893440103530883,
      "epoch": 0.33806929859044194,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499421950949953,
      "loss": 5.7807,
      "mean_token_accuracy": 0.16390309780836104,
      "num_tokens": 6740650.0,
      "step": 3010
    },
    {
      "entropy": 5.698744678497315,
      "epoch": 0.3386308754983995,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499419071312363,
      "loss": 5.6836,
      "mean_token_accuracy": 0.16469298750162126,
      "num_tokens": 6751467.0,
      "step": 3015
    },
    {
      "entropy": 5.7961132526397705,
      "epoch": 0.33919245240635704,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994161845291828,
      "loss": 5.7861,
      "mean_token_accuracy": 0.16712161004543305,
      "num_tokens": 6761377.0,
      "step": 3020
    },
    {
      "entropy": 5.79654335975647,
      "epoch": 0.3397540293143146,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994132906005041,
      "loss": 5.7393,
      "mean_token_accuracy": 0.16078124195337296,
      "num_tokens": 6773296.0,
      "step": 3025
    },
    {
      "entropy": 5.845974779129028,
      "epoch": 0.34031560622227214,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994103895264193,
      "loss": 5.7954,
      "mean_token_accuracy": 0.15900143682956697,
      "num_tokens": 6784122.0,
      "step": 3030
    },
    {
      "entropy": 5.704025363922119,
      "epoch": 0.3408771831302297,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994074813070207,
      "loss": 5.6383,
      "mean_token_accuracy": 0.16705414056777954,
      "num_tokens": 6795604.0,
      "step": 3035
    },
    {
      "entropy": 5.769487285614014,
      "epoch": 0.34143876003818724,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994045659424007,
      "loss": 5.7981,
      "mean_token_accuracy": 0.16261281818151474,
      "num_tokens": 6806671.0,
      "step": 3040
    },
    {
      "entropy": 5.9057597637176515,
      "epoch": 0.3420003369461448,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994016434326524,
      "loss": 5.8216,
      "mean_token_accuracy": 0.16332198679447174,
      "num_tokens": 6818271.0,
      "step": 3045
    },
    {
      "entropy": 5.7448649406433105,
      "epoch": 0.34256191385410234,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993987137778687,
      "loss": 5.8169,
      "mean_token_accuracy": 0.1635512739419937,
      "num_tokens": 6830049.0,
      "step": 3050
    },
    {
      "entropy": 5.942811059951782,
      "epoch": 0.3431234907620599,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499395776978143,
      "loss": 5.9403,
      "mean_token_accuracy": 0.15486131906509398,
      "num_tokens": 6842275.0,
      "step": 3055
    },
    {
      "entropy": 5.956660890579224,
      "epoch": 0.3436850676700174,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993928330335686,
      "loss": 5.8766,
      "mean_token_accuracy": 0.1637545719742775,
      "num_tokens": 6854910.0,
      "step": 3060
    },
    {
      "entropy": 5.791312170028687,
      "epoch": 0.34424664457797494,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993898819442395,
      "loss": 5.8284,
      "mean_token_accuracy": 0.16965375244617462,
      "num_tokens": 6866031.0,
      "step": 3065
    },
    {
      "entropy": 5.93616361618042,
      "epoch": 0.3448082214859325,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993869237102494,
      "loss": 5.8471,
      "mean_token_accuracy": 0.15883387625217438,
      "num_tokens": 6877672.0,
      "step": 3070
    },
    {
      "entropy": 5.765334987640381,
      "epoch": 0.34536979839389004,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993839583316927,
      "loss": 5.7543,
      "mean_token_accuracy": 0.16836509853601456,
      "num_tokens": 6889089.0,
      "step": 3075
    },
    {
      "entropy": 5.673844814300537,
      "epoch": 0.3459313753018476,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993809858086638,
      "loss": 5.6994,
      "mean_token_accuracy": 0.1676324874162674,
      "num_tokens": 6899877.0,
      "step": 3080
    },
    {
      "entropy": 5.793374395370483,
      "epoch": 0.34649295220980514,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993780061412571,
      "loss": 5.6829,
      "mean_token_accuracy": 0.1715866893529892,
      "num_tokens": 6910686.0,
      "step": 3085
    },
    {
      "entropy": 5.734429454803466,
      "epoch": 0.3470545291177627,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993750193295678,
      "loss": 5.7143,
      "mean_token_accuracy": 0.17135517448186874,
      "num_tokens": 6920867.0,
      "step": 3090
    },
    {
      "entropy": 5.899741554260254,
      "epoch": 0.34761610602572024,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993720253736909,
      "loss": 5.9044,
      "mean_token_accuracy": 0.16466446667909623,
      "num_tokens": 6932779.0,
      "step": 3095
    },
    {
      "entropy": 5.86834020614624,
      "epoch": 0.3481776829336778,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004993690242737216,
      "loss": 5.8055,
      "mean_token_accuracy": 0.16673210114240647,
      "num_tokens": 6943378.0,
      "step": 3100
    },
    {
      "entropy": 5.760788822174073,
      "epoch": 0.3487392598416353,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993660160297555,
      "loss": 5.7867,
      "mean_token_accuracy": 0.16799272298812867,
      "num_tokens": 6955469.0,
      "step": 3105
    },
    {
      "entropy": 5.971051931381226,
      "epoch": 0.34930083674959284,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004993630006418885,
      "loss": 5.8788,
      "mean_token_accuracy": 0.16197884529829026,
      "num_tokens": 6966750.0,
      "step": 3110
    },
    {
      "entropy": 5.811239528656006,
      "epoch": 0.3498624136575504,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993599781102166,
      "loss": 5.7912,
      "mean_token_accuracy": 0.16256844401359558,
      "num_tokens": 6979010.0,
      "step": 3115
    },
    {
      "entropy": 5.775843524932862,
      "epoch": 0.35042399056550794,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993569484348359,
      "loss": 5.7764,
      "mean_token_accuracy": 0.16187913715839386,
      "num_tokens": 6990566.0,
      "step": 3120
    },
    {
      "entropy": 5.8379576206207275,
      "epoch": 0.3509855674734655,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004993539116158429,
      "loss": 5.7852,
      "mean_token_accuracy": 0.1654660627245903,
      "num_tokens": 7000288.0,
      "step": 3125
    },
    {
      "entropy": 5.689918518066406,
      "epoch": 0.35154714438142304,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993508676533346,
      "loss": 5.7248,
      "mean_token_accuracy": 0.16818051636219025,
      "num_tokens": 7011143.0,
      "step": 3130
    },
    {
      "entropy": 5.967591333389282,
      "epoch": 0.3521087212893806,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993478165474074,
      "loss": 5.8802,
      "mean_token_accuracy": 0.15500763505697251,
      "num_tokens": 7022797.0,
      "step": 3135
    },
    {
      "entropy": 5.829250907897949,
      "epoch": 0.35267029819733814,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993447582981588,
      "loss": 5.8523,
      "mean_token_accuracy": 0.15978631526231765,
      "num_tokens": 7033477.0,
      "step": 3140
    },
    {
      "entropy": 5.825192642211914,
      "epoch": 0.3532318751052957,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499341692905686,
      "loss": 5.8713,
      "mean_token_accuracy": 0.16687492281198502,
      "num_tokens": 7045249.0,
      "step": 3145
    },
    {
      "entropy": 5.924324893951416,
      "epoch": 0.35379345201325324,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993386203700868,
      "loss": 5.8924,
      "mean_token_accuracy": 0.16145936846733094,
      "num_tokens": 7056890.0,
      "step": 3150
    },
    {
      "entropy": 5.861443424224854,
      "epoch": 0.35435502892121074,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993355406914589,
      "loss": 5.7678,
      "mean_token_accuracy": 0.1675198808312416,
      "num_tokens": 7067181.0,
      "step": 3155
    },
    {
      "entropy": 5.785042476654053,
      "epoch": 0.3549166058291683,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993324538699003,
      "loss": 5.7324,
      "mean_token_accuracy": 0.16697331219911576,
      "num_tokens": 7078220.0,
      "step": 3160
    },
    {
      "entropy": 5.833939266204834,
      "epoch": 0.35547818273712584,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993293599055094,
      "loss": 5.8066,
      "mean_token_accuracy": 0.16511004269123078,
      "num_tokens": 7089536.0,
      "step": 3165
    },
    {
      "entropy": 5.8496153354644775,
      "epoch": 0.3560397596450834,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993262587983848,
      "loss": 5.8848,
      "mean_token_accuracy": 0.16400568187236786,
      "num_tokens": 7101535.0,
      "step": 3170
    },
    {
      "entropy": 5.862979078292847,
      "epoch": 0.35660133655304094,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499323150548625,
      "loss": 5.7782,
      "mean_token_accuracy": 0.1608920007944107,
      "num_tokens": 7112468.0,
      "step": 3175
    },
    {
      "entropy": 5.801843690872192,
      "epoch": 0.3571629134609985,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004993200351563291,
      "loss": 5.7695,
      "mean_token_accuracy": 0.16856954991817474,
      "num_tokens": 7123192.0,
      "step": 3180
    },
    {
      "entropy": 5.776076078414917,
      "epoch": 0.35772449036895604,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993169126215962,
      "loss": 5.6971,
      "mean_token_accuracy": 0.17152681350708007,
      "num_tokens": 7133063.0,
      "step": 3185
    },
    {
      "entropy": 5.832470464706421,
      "epoch": 0.3582860672769136,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993137829445259,
      "loss": 5.7899,
      "mean_token_accuracy": 0.16362055987119675,
      "num_tokens": 7144190.0,
      "step": 3190
    },
    {
      "entropy": 5.7901458740234375,
      "epoch": 0.35884764418487114,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993106461252177,
      "loss": 5.779,
      "mean_token_accuracy": 0.1660066932439804,
      "num_tokens": 7155433.0,
      "step": 3195
    },
    {
      "entropy": 5.8804432392120365,
      "epoch": 0.3594092210928287,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993075021637716,
      "loss": 5.8442,
      "mean_token_accuracy": 0.16579061150550842,
      "num_tokens": 7166084.0,
      "step": 3200
    },
    {
      "entropy": 5.849256277084351,
      "epoch": 0.3599707980007862,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993043510602875,
      "loss": 5.8714,
      "mean_token_accuracy": 0.1630292773246765,
      "num_tokens": 7176948.0,
      "step": 3205
    },
    {
      "entropy": 5.846883821487427,
      "epoch": 0.36053237490874374,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499301192814866,
      "loss": 5.7289,
      "mean_token_accuracy": 0.16961728185415267,
      "num_tokens": 7188601.0,
      "step": 3210
    },
    {
      "entropy": 5.7951243877410885,
      "epoch": 0.3610939518167013,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004992980274276074,
      "loss": 5.7635,
      "mean_token_accuracy": 0.16846298575401306,
      "num_tokens": 7198495.0,
      "step": 3215
    },
    {
      "entropy": 5.826298189163208,
      "epoch": 0.36165552872465884,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004992948548986128,
      "loss": 5.753,
      "mean_token_accuracy": 0.16387602537870408,
      "num_tokens": 7209848.0,
      "step": 3220
    },
    {
      "entropy": 5.814381456375122,
      "epoch": 0.3622171056326164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004992916752279829,
      "loss": 5.8787,
      "mean_token_accuracy": 0.16065715700387956,
      "num_tokens": 7220366.0,
      "step": 3225
    },
    {
      "entropy": 5.806219387054443,
      "epoch": 0.36277868254057394,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499288488415819,
      "loss": 5.7833,
      "mean_token_accuracy": 0.16056074649095536,
      "num_tokens": 7231442.0,
      "step": 3230
    },
    {
      "entropy": 5.850152111053466,
      "epoch": 0.3633402594485315,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004992852944622229,
      "loss": 5.825,
      "mean_token_accuracy": 0.1608233243227005,
      "num_tokens": 7242391.0,
      "step": 3235
    },
    {
      "entropy": 5.733070659637451,
      "epoch": 0.36390183635648904,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004992820933672958,
      "loss": 5.7124,
      "mean_token_accuracy": 0.16292389035224913,
      "num_tokens": 7252061.0,
      "step": 3240
    },
    {
      "entropy": 5.832701826095581,
      "epoch": 0.3644634132644466,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992788851311399,
      "loss": 5.7926,
      "mean_token_accuracy": 0.1660913810133934,
      "num_tokens": 7264035.0,
      "step": 3245
    },
    {
      "entropy": 5.919405937194824,
      "epoch": 0.3650249901724041,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992756697538574,
      "loss": 5.8652,
      "mean_token_accuracy": 0.16157617121934892,
      "num_tokens": 7275988.0,
      "step": 3250
    },
    {
      "entropy": 5.77119779586792,
      "epoch": 0.36558656708036164,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992724472355505,
      "loss": 5.7206,
      "mean_token_accuracy": 0.16446294635534286,
      "num_tokens": 7286509.0,
      "step": 3255
    },
    {
      "entropy": 5.865907192230225,
      "epoch": 0.3661481439883192,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004992692175763219,
      "loss": 5.8134,
      "mean_token_accuracy": 0.16690248399972915,
      "num_tokens": 7296916.0,
      "step": 3260
    },
    {
      "entropy": 5.820225429534912,
      "epoch": 0.36670972089627674,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992659807762745,
      "loss": 5.8982,
      "mean_token_accuracy": 0.15828224942088126,
      "num_tokens": 7309563.0,
      "step": 3265
    },
    {
      "entropy": 5.88502197265625,
      "epoch": 0.3672712978042343,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992627368355112,
      "loss": 5.7505,
      "mean_token_accuracy": 0.16168039441108703,
      "num_tokens": 7320264.0,
      "step": 3270
    },
    {
      "entropy": 5.719604015350342,
      "epoch": 0.36783287471219184,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004992594857541355,
      "loss": 5.7186,
      "mean_token_accuracy": 0.16655558794736863,
      "num_tokens": 7330442.0,
      "step": 3275
    },
    {
      "entropy": 5.853730297088623,
      "epoch": 0.3683944516201494,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004992562275322506,
      "loss": 5.9159,
      "mean_token_accuracy": 0.1572217419743538,
      "num_tokens": 7342292.0,
      "step": 3280
    },
    {
      "entropy": 5.909031820297241,
      "epoch": 0.36895602852810694,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004992529621699606,
      "loss": 5.8141,
      "mean_token_accuracy": 0.16075354367494582,
      "num_tokens": 7353857.0,
      "step": 3285
    },
    {
      "entropy": 5.804742479324341,
      "epoch": 0.3695176054360645,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992496896673692,
      "loss": 5.7377,
      "mean_token_accuracy": 0.16423878073692322,
      "num_tokens": 7365653.0,
      "step": 3290
    },
    {
      "entropy": 5.790262269973755,
      "epoch": 0.37007918234402204,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992464100245809,
      "loss": 5.7455,
      "mean_token_accuracy": 0.15999912321567536,
      "num_tokens": 7376522.0,
      "step": 3295
    },
    {
      "entropy": 5.796634531021118,
      "epoch": 0.37064075925197953,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992431232416997,
      "loss": 5.7407,
      "mean_token_accuracy": 0.1694473460316658,
      "num_tokens": 7388562.0,
      "step": 3300
    },
    {
      "entropy": 5.8152648448944095,
      "epoch": 0.3712023361599371,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992398293188307,
      "loss": 5.794,
      "mean_token_accuracy": 0.16164605543017388,
      "num_tokens": 7398597.0,
      "step": 3305
    },
    {
      "entropy": 5.75906810760498,
      "epoch": 0.37176391306789464,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992365282560784,
      "loss": 5.7432,
      "mean_token_accuracy": 0.16698759496212007,
      "num_tokens": 7409656.0,
      "step": 3310
    },
    {
      "entropy": 5.769649410247803,
      "epoch": 0.3723254899758522,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004992332200535483,
      "loss": 5.7906,
      "mean_token_accuracy": 0.1635170191526413,
      "num_tokens": 7420621.0,
      "step": 3315
    },
    {
      "entropy": 5.914590549468994,
      "epoch": 0.37288706688380974,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992299047113452,
      "loss": 5.8258,
      "mean_token_accuracy": 0.1644177705049515,
      "num_tokens": 7432951.0,
      "step": 3320
    },
    {
      "entropy": 5.746472978591919,
      "epoch": 0.3734486437917673,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992265822295752,
      "loss": 5.8086,
      "mean_token_accuracy": 0.1657330572605133,
      "num_tokens": 7443630.0,
      "step": 3325
    },
    {
      "entropy": 5.878232765197754,
      "epoch": 0.37401022069972484,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992232526083438,
      "loss": 5.7904,
      "mean_token_accuracy": 0.1630190759897232,
      "num_tokens": 7454535.0,
      "step": 3330
    },
    {
      "entropy": 5.851468801498413,
      "epoch": 0.3745717976076824,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499219915847757,
      "loss": 5.8617,
      "mean_token_accuracy": 0.16609936952590942,
      "num_tokens": 7467751.0,
      "step": 3335
    },
    {
      "entropy": 5.7327392578125,
      "epoch": 0.37513337451563994,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004992165719479211,
      "loss": 5.7229,
      "mean_token_accuracy": 0.1723914325237274,
      "num_tokens": 7479347.0,
      "step": 3340
    },
    {
      "entropy": 5.792557239532471,
      "epoch": 0.37569495142359743,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004992132209089427,
      "loss": 5.7559,
      "mean_token_accuracy": 0.16422603875398636,
      "num_tokens": 7490916.0,
      "step": 3345
    },
    {
      "entropy": 5.839074325561524,
      "epoch": 0.376256528331555,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004992098627309283,
      "loss": 5.7285,
      "mean_token_accuracy": 0.16606951653957366,
      "num_tokens": 7500824.0,
      "step": 3350
    },
    {
      "entropy": 5.8109547138214115,
      "epoch": 0.37681810523951254,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499206497413985,
      "loss": 5.8818,
      "mean_token_accuracy": 0.15842843651771546,
      "num_tokens": 7511778.0,
      "step": 3355
    },
    {
      "entropy": 5.980834197998047,
      "epoch": 0.3773796821474701,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992031249582197,
      "loss": 5.9424,
      "mean_token_accuracy": 0.15256146043539048,
      "num_tokens": 7522616.0,
      "step": 3360
    },
    {
      "entropy": 5.881027460098267,
      "epoch": 0.37794125905542764,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991997453637401,
      "loss": 5.8116,
      "mean_token_accuracy": 0.1650218188762665,
      "num_tokens": 7533019.0,
      "step": 3365
    },
    {
      "entropy": 5.749409818649292,
      "epoch": 0.3785028359633852,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991963586306536,
      "loss": 5.7544,
      "mean_token_accuracy": 0.16154202967882156,
      "num_tokens": 7544139.0,
      "step": 3370
    },
    {
      "entropy": 5.80066704750061,
      "epoch": 0.37906441287134274,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991929647590682,
      "loss": 5.7377,
      "mean_token_accuracy": 0.16260901242494583,
      "num_tokens": 7554638.0,
      "step": 3375
    },
    {
      "entropy": 5.953235530853272,
      "epoch": 0.3796259897793003,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991895637490917,
      "loss": 5.8049,
      "mean_token_accuracy": 0.16714229732751845,
      "num_tokens": 7565158.0,
      "step": 3380
    },
    {
      "entropy": 5.809656095504761,
      "epoch": 0.38018756668725784,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991861556008326,
      "loss": 5.811,
      "mean_token_accuracy": 0.16301512867212295,
      "num_tokens": 7576381.0,
      "step": 3385
    },
    {
      "entropy": 5.718403387069702,
      "epoch": 0.3807491435952154,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991827403143992,
      "loss": 5.6948,
      "mean_token_accuracy": 0.1689286082983017,
      "num_tokens": 7586631.0,
      "step": 3390
    },
    {
      "entropy": 5.807317209243775,
      "epoch": 0.3813107205031729,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991793178899006,
      "loss": 5.7382,
      "mean_token_accuracy": 0.165502168238163,
      "num_tokens": 7598321.0,
      "step": 3395
    },
    {
      "entropy": 5.771745109558106,
      "epoch": 0.38187229741113043,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991758883274455,
      "loss": 5.7698,
      "mean_token_accuracy": 0.1654894694685936,
      "num_tokens": 7609336.0,
      "step": 3400
    },
    {
      "entropy": 5.721436452865601,
      "epoch": 0.382433874319088,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004991724516271432,
      "loss": 5.6716,
      "mean_token_accuracy": 0.16680669486522676,
      "num_tokens": 7619116.0,
      "step": 3405
    },
    {
      "entropy": 5.82664680480957,
      "epoch": 0.38299545122704554,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991690077891032,
      "loss": 5.8092,
      "mean_token_accuracy": 0.15846319049596785,
      "num_tokens": 7630567.0,
      "step": 3410
    },
    {
      "entropy": 5.8768939018249515,
      "epoch": 0.3835570281350031,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991655568134348,
      "loss": 5.8718,
      "mean_token_accuracy": 0.16050247251987457,
      "num_tokens": 7642564.0,
      "step": 3415
    },
    {
      "entropy": 5.909596014022827,
      "epoch": 0.38411860504296064,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991620987002483,
      "loss": 5.8983,
      "mean_token_accuracy": 0.15932083278894424,
      "num_tokens": 7654588.0,
      "step": 3420
    },
    {
      "entropy": 5.780777931213379,
      "epoch": 0.3846801819509182,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004991586334496536,
      "loss": 5.6903,
      "mean_token_accuracy": 0.1646028310060501,
      "num_tokens": 7666120.0,
      "step": 3425
    },
    {
      "entropy": 5.796514987945557,
      "epoch": 0.38524175885887574,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991551610617611,
      "loss": 5.8408,
      "mean_token_accuracy": 0.16049385517835618,
      "num_tokens": 7677132.0,
      "step": 3430
    },
    {
      "entropy": 5.968781232833862,
      "epoch": 0.3858033357668333,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991516815366812,
      "loss": 5.9041,
      "mean_token_accuracy": 0.1601366326212883,
      "num_tokens": 7687515.0,
      "step": 3435
    },
    {
      "entropy": 5.767029762268066,
      "epoch": 0.38636491267479084,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499148194874525,
      "loss": 5.7118,
      "mean_token_accuracy": 0.16504708528518677,
      "num_tokens": 7698454.0,
      "step": 3440
    },
    {
      "entropy": 5.698157548904419,
      "epoch": 0.38692648958274833,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991447010754032,
      "loss": 5.6899,
      "mean_token_accuracy": 0.1678033024072647,
      "num_tokens": 7708940.0,
      "step": 3445
    },
    {
      "entropy": 5.8262519359588625,
      "epoch": 0.3874880664907059,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004991412001394273,
      "loss": 5.7239,
      "mean_token_accuracy": 0.16594443321228028,
      "num_tokens": 7719705.0,
      "step": 3450
    },
    {
      "entropy": 5.863242149353027,
      "epoch": 0.38804964339866344,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991376920667086,
      "loss": 5.8811,
      "mean_token_accuracy": 0.15880899280309677,
      "num_tokens": 7730649.0,
      "step": 3455
    },
    {
      "entropy": 5.779720497131348,
      "epoch": 0.388611220306621,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991341768573589,
      "loss": 5.8953,
      "mean_token_accuracy": 0.1590091198682785,
      "num_tokens": 7742014.0,
      "step": 3460
    },
    {
      "entropy": 5.901068115234375,
      "epoch": 0.38917279721457854,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049913065451149,
      "loss": 5.799,
      "mean_token_accuracy": 0.1558951511979103,
      "num_tokens": 7752892.0,
      "step": 3465
    },
    {
      "entropy": 5.8446044921875,
      "epoch": 0.3897343741225361,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991271250292141,
      "loss": 5.81,
      "mean_token_accuracy": 0.157416270673275,
      "num_tokens": 7762959.0,
      "step": 3470
    },
    {
      "entropy": 5.81669659614563,
      "epoch": 0.39029595103049364,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991235884106437,
      "loss": 5.7144,
      "mean_token_accuracy": 0.16509674340486527,
      "num_tokens": 7773566.0,
      "step": 3475
    },
    {
      "entropy": 5.7814857959747314,
      "epoch": 0.3908575279384512,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991200446558913,
      "loss": 5.748,
      "mean_token_accuracy": 0.15518904775381087,
      "num_tokens": 7785782.0,
      "step": 3480
    },
    {
      "entropy": 5.841710948944092,
      "epoch": 0.39141910484640874,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991164937650698,
      "loss": 5.9032,
      "mean_token_accuracy": 0.16198572665452957,
      "num_tokens": 7797551.0,
      "step": 3485
    },
    {
      "entropy": 5.751931428909302,
      "epoch": 0.39198068175436623,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991129357382922,
      "loss": 5.6772,
      "mean_token_accuracy": 0.17679855972528458,
      "num_tokens": 7808819.0,
      "step": 3490
    },
    {
      "entropy": 5.647474765777588,
      "epoch": 0.3925422586623238,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991093705756718,
      "loss": 5.6283,
      "mean_token_accuracy": 0.16952918767929076,
      "num_tokens": 7818940.0,
      "step": 3495
    },
    {
      "entropy": 5.671689987182617,
      "epoch": 0.39310383557028133,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991057982773221,
      "loss": 5.6507,
      "mean_token_accuracy": 0.17054159790277482,
      "num_tokens": 7829733.0,
      "step": 3500
    },
    {
      "entropy": 5.847164916992187,
      "epoch": 0.3936654124782389,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991022188433569,
      "loss": 5.7133,
      "mean_token_accuracy": 0.17065165787935258,
      "num_tokens": 7840427.0,
      "step": 3505
    },
    {
      "entropy": 5.859947633743286,
      "epoch": 0.39422698938619644,
      "grad_norm": 1.125,
      "learning_rate": 0.00049909863227389,
      "loss": 5.775,
      "mean_token_accuracy": 0.16124107241630553,
      "num_tokens": 7852834.0,
      "step": 3510
    },
    {
      "entropy": 5.7385765552520756,
      "epoch": 0.394788566294154,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004990950385690358,
      "loss": 5.8178,
      "mean_token_accuracy": 0.1656463325023651,
      "num_tokens": 7864993.0,
      "step": 3515
    },
    {
      "entropy": 5.750460052490235,
      "epoch": 0.39535014320211154,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990914377289088,
      "loss": 5.793,
      "mean_token_accuracy": 0.16076247096061708,
      "num_tokens": 7876578.0,
      "step": 3520
    },
    {
      "entropy": 5.729760694503784,
      "epoch": 0.3959117201100691,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990878297536233,
      "loss": 5.6272,
      "mean_token_accuracy": 0.17054133117198944,
      "num_tokens": 7886770.0,
      "step": 3525
    },
    {
      "entropy": 5.800265884399414,
      "epoch": 0.39647329701802664,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004990842146432946,
      "loss": 5.7904,
      "mean_token_accuracy": 0.1667746052145958,
      "num_tokens": 7897106.0,
      "step": 3530
    },
    {
      "entropy": 5.749977684020996,
      "epoch": 0.3970348739259842,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990805923980375,
      "loss": 5.6444,
      "mean_token_accuracy": 0.17652151733636856,
      "num_tokens": 7907321.0,
      "step": 3535
    },
    {
      "entropy": 5.651272344589233,
      "epoch": 0.3975964508339417,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990769630179673,
      "loss": 5.6569,
      "mean_token_accuracy": 0.170053867995739,
      "num_tokens": 7918628.0,
      "step": 3540
    },
    {
      "entropy": 5.872209167480468,
      "epoch": 0.39815802774189923,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004990733265032,
      "loss": 5.8387,
      "mean_token_accuracy": 0.15291183888912202,
      "num_tokens": 7929387.0,
      "step": 3545
    },
    {
      "entropy": 5.92167649269104,
      "epoch": 0.3987196046498568,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990696828538508,
      "loss": 5.8595,
      "mean_token_accuracy": 0.16960857212543487,
      "num_tokens": 7940840.0,
      "step": 3550
    },
    {
      "entropy": 5.664200639724731,
      "epoch": 0.39928118155781434,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499066032070036,
      "loss": 5.7269,
      "mean_token_accuracy": 0.1674627184867859,
      "num_tokens": 7952243.0,
      "step": 3555
    },
    {
      "entropy": 5.863605737686157,
      "epoch": 0.3998427584657719,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004990623741518719,
      "loss": 5.7724,
      "mean_token_accuracy": 0.16179074645042418,
      "num_tokens": 7963150.0,
      "step": 3560
    },
    {
      "entropy": 5.818461227416992,
      "epoch": 0.40040433537372944,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499058709099475,
      "loss": 5.8741,
      "mean_token_accuracy": 0.15937056988477707,
      "num_tokens": 7974862.0,
      "step": 3565
    },
    {
      "entropy": 5.877570915222168,
      "epoch": 0.400965912281687,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990550369129618,
      "loss": 5.8819,
      "mean_token_accuracy": 0.16194065660238266,
      "num_tokens": 7986438.0,
      "step": 3570
    },
    {
      "entropy": 5.8217613697052,
      "epoch": 0.40152748918964454,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004990513575924494,
      "loss": 5.7173,
      "mean_token_accuracy": 0.16897953897714615,
      "num_tokens": 7996958.0,
      "step": 3575
    },
    {
      "entropy": 5.835204267501831,
      "epoch": 0.4020890660976021,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990476711380548,
      "loss": 5.8001,
      "mean_token_accuracy": 0.1581471100449562,
      "num_tokens": 8008862.0,
      "step": 3580
    },
    {
      "entropy": 5.774871110916138,
      "epoch": 0.40265064300555964,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990439775498954,
      "loss": 5.7798,
      "mean_token_accuracy": 0.1676494598388672,
      "num_tokens": 8018940.0,
      "step": 3585
    },
    {
      "entropy": 5.914262056350708,
      "epoch": 0.40321221991351713,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499040276828089,
      "loss": 5.7946,
      "mean_token_accuracy": 0.15799015909433364,
      "num_tokens": 8029148.0,
      "step": 3590
    },
    {
      "entropy": 5.742766189575195,
      "epoch": 0.4037737968214747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990365689727533,
      "loss": 5.7104,
      "mean_token_accuracy": 0.1686326339840889,
      "num_tokens": 8039618.0,
      "step": 3595
    },
    {
      "entropy": 5.744286060333252,
      "epoch": 0.40433537372943223,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990328539840062,
      "loss": 5.6934,
      "mean_token_accuracy": 0.16474545300006865,
      "num_tokens": 8051145.0,
      "step": 3600
    },
    {
      "entropy": 5.7836991310119625,
      "epoch": 0.4048969506373898,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004990291318619663,
      "loss": 5.7609,
      "mean_token_accuracy": 0.1606362998485565,
      "num_tokens": 8061520.0,
      "step": 3605
    },
    {
      "entropy": 5.726670598983764,
      "epoch": 0.40545852754534734,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990254026067518,
      "loss": 5.7099,
      "mean_token_accuracy": 0.16285099536180497,
      "num_tokens": 8072876.0,
      "step": 3610
    },
    {
      "entropy": 5.934379577636719,
      "epoch": 0.4060201044533049,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990216662184817,
      "loss": 5.8154,
      "mean_token_accuracy": 0.16516112834215163,
      "num_tokens": 8084240.0,
      "step": 3615
    },
    {
      "entropy": 5.753713512420655,
      "epoch": 0.40658168136126244,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990179226972749,
      "loss": 5.6932,
      "mean_token_accuracy": 0.16864500045776368,
      "num_tokens": 8095773.0,
      "step": 3620
    },
    {
      "entropy": 5.629334354400635,
      "epoch": 0.40714325826922,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990141720432504,
      "loss": 5.6819,
      "mean_token_accuracy": 0.167998369038105,
      "num_tokens": 8107003.0,
      "step": 3625
    },
    {
      "entropy": 5.72817096710205,
      "epoch": 0.40770483517717754,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990104142565279,
      "loss": 5.6439,
      "mean_token_accuracy": 0.17401827424764632,
      "num_tokens": 8117864.0,
      "step": 3630
    },
    {
      "entropy": 5.835469007492065,
      "epoch": 0.40826641208513503,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499006649337227,
      "loss": 5.86,
      "mean_token_accuracy": 0.16062401831150055,
      "num_tokens": 8129363.0,
      "step": 3635
    },
    {
      "entropy": 5.805969905853272,
      "epoch": 0.4088279889930926,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990028772854674,
      "loss": 5.7791,
      "mean_token_accuracy": 0.1627799391746521,
      "num_tokens": 8141319.0,
      "step": 3640
    },
    {
      "entropy": 5.784128999710083,
      "epoch": 0.40938956590105013,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004989990981013693,
      "loss": 5.6802,
      "mean_token_accuracy": 0.1702434554696083,
      "num_tokens": 8153344.0,
      "step": 3645
    },
    {
      "entropy": 5.774145793914795,
      "epoch": 0.4099511428090077,
      "grad_norm": 1.15625,
      "learning_rate": 0.000498995311785053,
      "loss": 5.8402,
      "mean_token_accuracy": 0.162425434589386,
      "num_tokens": 8164196.0,
      "step": 3650
    },
    {
      "entropy": 5.816594219207763,
      "epoch": 0.41051271971696524,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004989915183366392,
      "loss": 5.7159,
      "mean_token_accuracy": 0.16790103316307067,
      "num_tokens": 8174663.0,
      "step": 3655
    },
    {
      "entropy": 5.713854694366455,
      "epoch": 0.4110742966249228,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989877177562485,
      "loss": 5.683,
      "mean_token_accuracy": 0.1685110315680504,
      "num_tokens": 8185215.0,
      "step": 3660
    },
    {
      "entropy": 5.739247465133667,
      "epoch": 0.41163587353288034,
      "grad_norm": 1.234375,
      "learning_rate": 0.000498983910044002,
      "loss": 5.7765,
      "mean_token_accuracy": 0.17141249477863313,
      "num_tokens": 8195694.0,
      "step": 3665
    },
    {
      "entropy": 5.681199550628662,
      "epoch": 0.4121974504408379,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989800952000209,
      "loss": 5.6616,
      "mean_token_accuracy": 0.17553078830242158,
      "num_tokens": 8206112.0,
      "step": 3670
    },
    {
      "entropy": 5.731632661819458,
      "epoch": 0.41275902734879544,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004989762732244268,
      "loss": 5.7515,
      "mean_token_accuracy": 0.1661844179034233,
      "num_tokens": 8216548.0,
      "step": 3675
    },
    {
      "entropy": 5.843146276473999,
      "epoch": 0.413320604256753,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004989724441173413,
      "loss": 5.7867,
      "mean_token_accuracy": 0.16781337559223175,
      "num_tokens": 8226643.0,
      "step": 3680
    },
    {
      "entropy": 5.857947587966919,
      "epoch": 0.4138821811647105,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989686078788862,
      "loss": 5.8183,
      "mean_token_accuracy": 0.16166319102048873,
      "num_tokens": 8236755.0,
      "step": 3685
    },
    {
      "entropy": 5.782751274108887,
      "epoch": 0.41444375807266803,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004989647645091839,
      "loss": 5.6963,
      "mean_token_accuracy": 0.16801404505968093,
      "num_tokens": 8247526.0,
      "step": 3690
    },
    {
      "entropy": 5.734130477905273,
      "epoch": 0.4150053349806256,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004989609140083565,
      "loss": 5.794,
      "mean_token_accuracy": 0.15583045184612274,
      "num_tokens": 8258649.0,
      "step": 3695
    },
    {
      "entropy": 5.765203428268433,
      "epoch": 0.41556691188858313,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004989570563765269,
      "loss": 5.676,
      "mean_token_accuracy": 0.1731864795088768,
      "num_tokens": 8268260.0,
      "step": 3700
    },
    {
      "entropy": 5.9308096408844,
      "epoch": 0.4161284887965407,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004989531916138176,
      "loss": 5.8785,
      "mean_token_accuracy": 0.1627708837389946,
      "num_tokens": 8280090.0,
      "step": 3705
    },
    {
      "entropy": 5.841350793838501,
      "epoch": 0.41669006570449824,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989493197203519,
      "loss": 5.7489,
      "mean_token_accuracy": 0.1697134330868721,
      "num_tokens": 8291217.0,
      "step": 3710
    },
    {
      "entropy": 5.754220199584961,
      "epoch": 0.4172516426124558,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000498945440696253,
      "loss": 5.6683,
      "mean_token_accuracy": 0.17081218361854553,
      "num_tokens": 8303541.0,
      "step": 3715
    },
    {
      "entropy": 5.719450426101685,
      "epoch": 0.41781321952041334,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989415545416444,
      "loss": 5.7355,
      "mean_token_accuracy": 0.1609584018588066,
      "num_tokens": 8316104.0,
      "step": 3720
    },
    {
      "entropy": 5.714957666397095,
      "epoch": 0.4183747964283709,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989376612566498,
      "loss": 5.6832,
      "mean_token_accuracy": 0.16480520740151405,
      "num_tokens": 8327158.0,
      "step": 3725
    },
    {
      "entropy": 5.846664381027222,
      "epoch": 0.4189363733363284,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989337608413933,
      "loss": 5.8338,
      "mean_token_accuracy": 0.16076735854148866,
      "num_tokens": 8337685.0,
      "step": 3730
    },
    {
      "entropy": 5.870333480834961,
      "epoch": 0.41949795024428593,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989298532959989,
      "loss": 5.8683,
      "mean_token_accuracy": 0.16113072335720063,
      "num_tokens": 8349778.0,
      "step": 3735
    },
    {
      "entropy": 5.7798456192016605,
      "epoch": 0.4200595271522435,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004989259386205912,
      "loss": 5.7038,
      "mean_token_accuracy": 0.176594515144825,
      "num_tokens": 8361041.0,
      "step": 3740
    },
    {
      "entropy": 5.8628821849823,
      "epoch": 0.42062110406020103,
      "grad_norm": 1.25,
      "learning_rate": 0.0004989220168152948,
      "loss": 5.8965,
      "mean_token_accuracy": 0.15907129645347595,
      "num_tokens": 8373227.0,
      "step": 3745
    },
    {
      "entropy": 5.7868818759918215,
      "epoch": 0.4211826809681586,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004989180878802346,
      "loss": 5.7209,
      "mean_token_accuracy": 0.16800411641597748,
      "num_tokens": 8384223.0,
      "step": 3750
    },
    {
      "entropy": 5.731780624389648,
      "epoch": 0.42174425787611614,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004989141518155354,
      "loss": 5.7506,
      "mean_token_accuracy": 0.1691988691687584,
      "num_tokens": 8395014.0,
      "step": 3755
    },
    {
      "entropy": 5.823819208145141,
      "epoch": 0.4223058347840737,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004989102086213229,
      "loss": 5.7283,
      "mean_token_accuracy": 0.16270644217729568,
      "num_tokens": 8406445.0,
      "step": 3760
    },
    {
      "entropy": 5.693364000320434,
      "epoch": 0.42286741169203124,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004989062582977225,
      "loss": 5.665,
      "mean_token_accuracy": 0.1706314980983734,
      "num_tokens": 8417488.0,
      "step": 3765
    },
    {
      "entropy": 5.793571662902832,
      "epoch": 0.4234289885999888,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989023008448599,
      "loss": 5.8689,
      "mean_token_accuracy": 0.15837376266717912,
      "num_tokens": 8430582.0,
      "step": 3770
    },
    {
      "entropy": 5.831238794326782,
      "epoch": 0.42399056550794634,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004988983362628614,
      "loss": 5.6906,
      "mean_token_accuracy": 0.164572674036026,
      "num_tokens": 8442033.0,
      "step": 3775
    },
    {
      "entropy": 5.690927267074585,
      "epoch": 0.42455214241590383,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004988943645518528,
      "loss": 5.6636,
      "mean_token_accuracy": 0.16775935739278794,
      "num_tokens": 8451722.0,
      "step": 3780
    },
    {
      "entropy": 5.668238401412964,
      "epoch": 0.4251137193238614,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004988903857119611,
      "loss": 5.75,
      "mean_token_accuracy": 0.1658486858010292,
      "num_tokens": 8462753.0,
      "step": 3785
    },
    {
      "entropy": 5.861290073394775,
      "epoch": 0.42567529623181893,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004988863997433125,
      "loss": 5.7587,
      "mean_token_accuracy": 0.16550534069538117,
      "num_tokens": 8472755.0,
      "step": 3790
    },
    {
      "entropy": 5.766337490081787,
      "epoch": 0.4262368731397765,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004988824066460341,
      "loss": 5.7399,
      "mean_token_accuracy": 0.16958845853805543,
      "num_tokens": 8483296.0,
      "step": 3795
    },
    {
      "entropy": 5.7786054611206055,
      "epoch": 0.42679845004773403,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000498878406420253,
      "loss": 5.6058,
      "mean_token_accuracy": 0.17350622415542602,
      "num_tokens": 8493739.0,
      "step": 3800
    },
    {
      "entropy": 5.682381916046142,
      "epoch": 0.4273600269556916,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988743990660967,
      "loss": 5.7767,
      "mean_token_accuracy": 0.162274731695652,
      "num_tokens": 8504788.0,
      "step": 3805
    },
    {
      "entropy": 5.797802543640136,
      "epoch": 0.42792160386364914,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004988703845836927,
      "loss": 5.7306,
      "mean_token_accuracy": 0.1645368903875351,
      "num_tokens": 8514708.0,
      "step": 3810
    },
    {
      "entropy": 5.790168046951294,
      "epoch": 0.4284831807716067,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004988663629731687,
      "loss": 5.6135,
      "mean_token_accuracy": 0.17496834099292755,
      "num_tokens": 8525227.0,
      "step": 3815
    },
    {
      "entropy": 5.622102165222168,
      "epoch": 0.42904475767956424,
      "grad_norm": 1.171875,
      "learning_rate": 0.000498862334234653,
      "loss": 5.6619,
      "mean_token_accuracy": 0.17011212110519408,
      "num_tokens": 8536502.0,
      "step": 3820
    },
    {
      "entropy": 5.7700787544250485,
      "epoch": 0.4296063345875218,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004988582983682737,
      "loss": 5.8033,
      "mean_token_accuracy": 0.16546136736869813,
      "num_tokens": 8548087.0,
      "step": 3825
    },
    {
      "entropy": 5.85570330619812,
      "epoch": 0.4301679114954793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988542553741594,
      "loss": 5.809,
      "mean_token_accuracy": 0.1667978622019291,
      "num_tokens": 8560856.0,
      "step": 3830
    },
    {
      "entropy": 5.931719970703125,
      "epoch": 0.43072948840343683,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004988502052524388,
      "loss": 5.9145,
      "mean_token_accuracy": 0.16501590609550476,
      "num_tokens": 8572521.0,
      "step": 3835
    },
    {
      "entropy": 5.84990963935852,
      "epoch": 0.4312910653113944,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004988461480032408,
      "loss": 5.7442,
      "mean_token_accuracy": 0.1674473449587822,
      "num_tokens": 8583977.0,
      "step": 3840
    },
    {
      "entropy": 5.704129409790039,
      "epoch": 0.43185264221935193,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988420836266946,
      "loss": 5.6924,
      "mean_token_accuracy": 0.16754663586616517,
      "num_tokens": 8594441.0,
      "step": 3845
    },
    {
      "entropy": 5.84066071510315,
      "epoch": 0.4324142191273095,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004988380121229298,
      "loss": 5.7392,
      "mean_token_accuracy": 0.16255563497543335,
      "num_tokens": 8604843.0,
      "step": 3850
    },
    {
      "entropy": 5.746822309494019,
      "epoch": 0.43297579603526704,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988339334920757,
      "loss": 5.7298,
      "mean_token_accuracy": 0.16818083822727203,
      "num_tokens": 8616059.0,
      "step": 3855
    },
    {
      "entropy": 5.74391360282898,
      "epoch": 0.4335373729432246,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004988298477342625,
      "loss": 5.6845,
      "mean_token_accuracy": 0.17156679034233094,
      "num_tokens": 8626203.0,
      "step": 3860
    },
    {
      "entropy": 5.739895057678223,
      "epoch": 0.43409894985118214,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00049882575484962,
      "loss": 5.684,
      "mean_token_accuracy": 0.16509181261062622,
      "num_tokens": 8636921.0,
      "step": 3865
    },
    {
      "entropy": 5.6815019130706785,
      "epoch": 0.4346605267591397,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004988216548382787,
      "loss": 5.6579,
      "mean_token_accuracy": 0.17124742418527603,
      "num_tokens": 8648526.0,
      "step": 3870
    },
    {
      "entropy": 5.796351671218872,
      "epoch": 0.4352221036670972,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004988175477003691,
      "loss": 5.8049,
      "mean_token_accuracy": 0.15377746522426605,
      "num_tokens": 8660339.0,
      "step": 3875
    },
    {
      "entropy": 5.9060732364654545,
      "epoch": 0.43578368057505473,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000498813433436022,
      "loss": 5.8632,
      "mean_token_accuracy": 0.1561919257044792,
      "num_tokens": 8671835.0,
      "step": 3880
    },
    {
      "entropy": 5.678639030456543,
      "epoch": 0.4363452574830123,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004988093120453684,
      "loss": 5.5652,
      "mean_token_accuracy": 0.17134900540113449,
      "num_tokens": 8681146.0,
      "step": 3885
    },
    {
      "entropy": 5.650815439224243,
      "epoch": 0.43690683439096983,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004988051835285394,
      "loss": 5.6231,
      "mean_token_accuracy": 0.17194372713565825,
      "num_tokens": 8691623.0,
      "step": 3890
    },
    {
      "entropy": 5.765362548828125,
      "epoch": 0.4374684112989274,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004988010478856667,
      "loss": 5.7243,
      "mean_token_accuracy": 0.16024973541498183,
      "num_tokens": 8703702.0,
      "step": 3895
    },
    {
      "entropy": 5.937811851501465,
      "epoch": 0.43802998820688493,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004987969051168816,
      "loss": 5.8895,
      "mean_token_accuracy": 0.16733449399471284,
      "num_tokens": 8715434.0,
      "step": 3900
    },
    {
      "entropy": 5.7049939155578615,
      "epoch": 0.4385915651148425,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987927552223165,
      "loss": 5.6636,
      "mean_token_accuracy": 0.17958703339099885,
      "num_tokens": 8726135.0,
      "step": 3905
    },
    {
      "entropy": 5.635489463806152,
      "epoch": 0.43915314202280004,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004987885982021032,
      "loss": 5.761,
      "mean_token_accuracy": 0.16973788887262345,
      "num_tokens": 8736538.0,
      "step": 3910
    },
    {
      "entropy": 5.879006290435791,
      "epoch": 0.4397147189307576,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004987844340563741,
      "loss": 5.6806,
      "mean_token_accuracy": 0.17044662386178971,
      "num_tokens": 8746456.0,
      "step": 3915
    },
    {
      "entropy": 5.7940206050872805,
      "epoch": 0.44027629583871514,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004987802627852619,
      "loss": 5.7909,
      "mean_token_accuracy": 0.16806718856096267,
      "num_tokens": 8757143.0,
      "step": 3920
    },
    {
      "entropy": 5.737888336181641,
      "epoch": 0.44083787274667263,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004987760843888992,
      "loss": 5.7794,
      "mean_token_accuracy": 0.17163707613945006,
      "num_tokens": 8769084.0,
      "step": 3925
    },
    {
      "entropy": 5.792534828186035,
      "epoch": 0.4413994496546302,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004987718988674193,
      "loss": 5.7753,
      "mean_token_accuracy": 0.16346744000911712,
      "num_tokens": 8780707.0,
      "step": 3930
    },
    {
      "entropy": 5.78266863822937,
      "epoch": 0.44196102656258773,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004987677062209552,
      "loss": 5.7305,
      "mean_token_accuracy": 0.16145867258310317,
      "num_tokens": 8792157.0,
      "step": 3935
    },
    {
      "entropy": 5.824311637878418,
      "epoch": 0.4425226034705453,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004987635064496407,
      "loss": 5.7576,
      "mean_token_accuracy": 0.16702643185853958,
      "num_tokens": 8802691.0,
      "step": 3940
    },
    {
      "entropy": 5.8741062641143795,
      "epoch": 0.44308418037850283,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004987592995536093,
      "loss": 5.7672,
      "mean_token_accuracy": 0.1658184953033924,
      "num_tokens": 8814869.0,
      "step": 3945
    },
    {
      "entropy": 5.73998441696167,
      "epoch": 0.4436457572864604,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004987550855329951,
      "loss": 5.7099,
      "mean_token_accuracy": 0.17231065332889556,
      "num_tokens": 8825264.0,
      "step": 3950
    },
    {
      "entropy": 5.8049551486969,
      "epoch": 0.44420733419441794,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004987508643879322,
      "loss": 5.787,
      "mean_token_accuracy": 0.16581701785326003,
      "num_tokens": 8837128.0,
      "step": 3955
    },
    {
      "entropy": 5.801353025436401,
      "epoch": 0.4447689111023755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004987466361185549,
      "loss": 5.7861,
      "mean_token_accuracy": 0.15883654057979585,
      "num_tokens": 8848587.0,
      "step": 3960
    },
    {
      "entropy": 5.848768043518066,
      "epoch": 0.44533048801033304,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004987424007249979,
      "loss": 5.8546,
      "mean_token_accuracy": 0.16280081868171692,
      "num_tokens": 8859219.0,
      "step": 3965
    },
    {
      "entropy": 5.788423871994018,
      "epoch": 0.4458920649182906,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004987381582073962,
      "loss": 5.7406,
      "mean_token_accuracy": 0.17112360298633575,
      "num_tokens": 8868780.0,
      "step": 3970
    },
    {
      "entropy": 5.834964323043823,
      "epoch": 0.4464536418262481,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004987339085658848,
      "loss": 5.7172,
      "mean_token_accuracy": 0.1637463465332985,
      "num_tokens": 8880903.0,
      "step": 3975
    },
    {
      "entropy": 5.660829162597656,
      "epoch": 0.44701521873420563,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004987296518005988,
      "loss": 5.5606,
      "mean_token_accuracy": 0.17245692163705825,
      "num_tokens": 8890335.0,
      "step": 3980
    },
    {
      "entropy": 5.665368318557739,
      "epoch": 0.4475767956421632,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498725387911674,
      "loss": 5.7033,
      "mean_token_accuracy": 0.16330161690711975,
      "num_tokens": 8904104.0,
      "step": 3985
    },
    {
      "entropy": 5.9002443790435795,
      "epoch": 0.44813837255012073,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004987211168992461,
      "loss": 5.7955,
      "mean_token_accuracy": 0.1588960111141205,
      "num_tokens": 8915118.0,
      "step": 3990
    },
    {
      "entropy": 5.748874807357788,
      "epoch": 0.4486999494580783,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000498716838763451,
      "loss": 5.7518,
      "mean_token_accuracy": 0.16183157116174698,
      "num_tokens": 8927450.0,
      "step": 3995
    },
    {
      "entropy": 5.855266857147217,
      "epoch": 0.44926152636603583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000498712553504425,
      "loss": 5.8327,
      "mean_token_accuracy": 0.16530437469482423,
      "num_tokens": 8939902.0,
      "step": 4000
    },
    {
      "entropy": 5.803925609588623,
      "epoch": 0.4498231032739934,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004987082611223046,
      "loss": 5.6966,
      "mean_token_accuracy": 0.16004012376070023,
      "num_tokens": 8951736.0,
      "step": 4005
    },
    {
      "entropy": 5.824748182296753,
      "epoch": 0.45038468018195094,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004987039616172265,
      "loss": 5.7918,
      "mean_token_accuracy": 0.16543588787317276,
      "num_tokens": 8963283.0,
      "step": 4010
    },
    {
      "entropy": 5.746975517272949,
      "epoch": 0.4509462570899085,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004986996549893273,
      "loss": 5.6875,
      "mean_token_accuracy": 0.1728332132101059,
      "num_tokens": 8974323.0,
      "step": 4015
    },
    {
      "entropy": 5.742187404632569,
      "epoch": 0.451507833997866,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004986953412387445,
      "loss": 5.7779,
      "mean_token_accuracy": 0.1621204733848572,
      "num_tokens": 8986687.0,
      "step": 4020
    },
    {
      "entropy": 5.705374097824096,
      "epoch": 0.45206941090582353,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004986910203656151,
      "loss": 5.6662,
      "mean_token_accuracy": 0.17241707146167756,
      "num_tokens": 8997734.0,
      "step": 4025
    },
    {
      "entropy": 5.743109798431396,
      "epoch": 0.4526309878137811,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004986866923700769,
      "loss": 5.79,
      "mean_token_accuracy": 0.15861670374870301,
      "num_tokens": 9008743.0,
      "step": 4030
    },
    {
      "entropy": 5.864490699768067,
      "epoch": 0.45319256472173863,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004986823572522677,
      "loss": 5.649,
      "mean_token_accuracy": 0.167484150826931,
      "num_tokens": 9019495.0,
      "step": 4035
    },
    {
      "entropy": 5.750733947753906,
      "epoch": 0.4537541416296962,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004986780150123254,
      "loss": 5.7626,
      "mean_token_accuracy": 0.16347892284393312,
      "num_tokens": 9029546.0,
      "step": 4040
    },
    {
      "entropy": 5.62654070854187,
      "epoch": 0.45431571853765373,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986736656503885,
      "loss": 5.5203,
      "mean_token_accuracy": 0.17364439517259597,
      "num_tokens": 9040393.0,
      "step": 4045
    },
    {
      "entropy": 5.732595634460449,
      "epoch": 0.4548772954456113,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004986693091665952,
      "loss": 5.7232,
      "mean_token_accuracy": 0.16511399149894715,
      "num_tokens": 9050840.0,
      "step": 4050
    },
    {
      "entropy": 5.76080846786499,
      "epoch": 0.45543887235356884,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004986649455610845,
      "loss": 5.6551,
      "mean_token_accuracy": 0.16657507717609404,
      "num_tokens": 9061615.0,
      "step": 4055
    },
    {
      "entropy": 5.688541269302368,
      "epoch": 0.4560004492615264,
      "grad_norm": 1.328125,
      "learning_rate": 0.000498660574833995,
      "loss": 5.6307,
      "mean_token_accuracy": 0.17303157597780228,
      "num_tokens": 9071902.0,
      "step": 4060
    },
    {
      "entropy": 5.747698211669922,
      "epoch": 0.45656202616948394,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004986561969854661,
      "loss": 5.7872,
      "mean_token_accuracy": 0.16532022655010223,
      "num_tokens": 9083613.0,
      "step": 4065
    },
    {
      "entropy": 5.754197359085083,
      "epoch": 0.45712360307744143,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004986518120156373,
      "loss": 5.6888,
      "mean_token_accuracy": 0.16673017293214798,
      "num_tokens": 9095678.0,
      "step": 4070
    },
    {
      "entropy": 5.761954212188721,
      "epoch": 0.457685179985399,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000498647419924648,
      "loss": 5.696,
      "mean_token_accuracy": 0.17291476875543593,
      "num_tokens": 9106637.0,
      "step": 4075
    },
    {
      "entropy": 5.667988586425781,
      "epoch": 0.45824675689335653,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000498643020712638,
      "loss": 5.6597,
      "mean_token_accuracy": 0.17288853377103805,
      "num_tokens": 9117343.0,
      "step": 4080
    },
    {
      "entropy": 5.798785257339477,
      "epoch": 0.4588083338013141,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986386143797475,
      "loss": 5.7887,
      "mean_token_accuracy": 0.15990065932273864,
      "num_tokens": 9129315.0,
      "step": 4085
    },
    {
      "entropy": 5.7697852611541744,
      "epoch": 0.45936991070927163,
      "grad_norm": 1.25,
      "learning_rate": 0.0004986342009261169,
      "loss": 5.6784,
      "mean_token_accuracy": 0.16411612257361413,
      "num_tokens": 9139701.0,
      "step": 4090
    },
    {
      "entropy": 5.79812912940979,
      "epoch": 0.4599314876172292,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004986297803518865,
      "loss": 5.8133,
      "mean_token_accuracy": 0.16509659737348556,
      "num_tokens": 9152334.0,
      "step": 4095
    },
    {
      "entropy": 5.762947416305542,
      "epoch": 0.46049306452518673,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004986253526571971,
      "loss": 5.6195,
      "mean_token_accuracy": 0.17192814201116563,
      "num_tokens": 9163277.0,
      "step": 4100
    },
    {
      "entropy": 5.772461366653443,
      "epoch": 0.4610546414331443,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004986209178421899,
      "loss": 5.802,
      "mean_token_accuracy": 0.1619513064622879,
      "num_tokens": 9174909.0,
      "step": 4105
    },
    {
      "entropy": 5.647283840179443,
      "epoch": 0.46161621834110184,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000498616475907006,
      "loss": 5.5928,
      "mean_token_accuracy": 0.16549195647239684,
      "num_tokens": 9186058.0,
      "step": 4110
    },
    {
      "entropy": 5.73241286277771,
      "epoch": 0.46217779524905933,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004986120268517865,
      "loss": 5.7547,
      "mean_token_accuracy": 0.16822669506072999,
      "num_tokens": 9196919.0,
      "step": 4115
    },
    {
      "entropy": 5.772916507720947,
      "epoch": 0.4627393721570169,
      "grad_norm": 1.25,
      "learning_rate": 0.0004986075706766736,
      "loss": 5.7294,
      "mean_token_accuracy": 0.16655745655298232,
      "num_tokens": 9208452.0,
      "step": 4120
    },
    {
      "entropy": 5.783675956726074,
      "epoch": 0.46330094906497443,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004986031073818088,
      "loss": 5.7842,
      "mean_token_accuracy": 0.1649884581565857,
      "num_tokens": 9219605.0,
      "step": 4125
    },
    {
      "entropy": 5.74765625,
      "epoch": 0.463862525972932,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004985986369673343,
      "loss": 5.6159,
      "mean_token_accuracy": 0.17641097754240037,
      "num_tokens": 9230491.0,
      "step": 4130
    },
    {
      "entropy": 5.717433166503906,
      "epoch": 0.46442410288088953,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004985941594333926,
      "loss": 5.7567,
      "mean_token_accuracy": 0.16423618644475937,
      "num_tokens": 9241768.0,
      "step": 4135
    },
    {
      "entropy": 5.800970554351807,
      "epoch": 0.4649856797888471,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498589674780126,
      "loss": 5.7934,
      "mean_token_accuracy": 0.16235399544239043,
      "num_tokens": 9253779.0,
      "step": 4140
    },
    {
      "entropy": 5.94872088432312,
      "epoch": 0.46554725669680463,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004985851830076775,
      "loss": 5.6916,
      "mean_token_accuracy": 0.17534759044647216,
      "num_tokens": 9264567.0,
      "step": 4145
    },
    {
      "entropy": 5.591338491439819,
      "epoch": 0.4661088336047622,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004985806841161901,
      "loss": 5.6529,
      "mean_token_accuracy": 0.17427073419094086,
      "num_tokens": 9275960.0,
      "step": 4150
    },
    {
      "entropy": 5.708569431304932,
      "epoch": 0.46667041051271974,
      "grad_norm": 1.390625,
      "learning_rate": 0.000498576178105807,
      "loss": 5.7891,
      "mean_token_accuracy": 0.17374880164861678,
      "num_tokens": 9288305.0,
      "step": 4155
    },
    {
      "entropy": 5.802067136764526,
      "epoch": 0.4672319874206773,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004985716649766717,
      "loss": 5.6165,
      "mean_token_accuracy": 0.17416664063930512,
      "num_tokens": 9299579.0,
      "step": 4160
    },
    {
      "entropy": 5.667328643798828,
      "epoch": 0.4677935643286348,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004985671447289277,
      "loss": 5.6468,
      "mean_token_accuracy": 0.16909621357917787,
      "num_tokens": 9311221.0,
      "step": 4165
    },
    {
      "entropy": 5.732243394851684,
      "epoch": 0.46835514123659233,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004985626173627194,
      "loss": 5.7524,
      "mean_token_accuracy": 0.16208277642726898,
      "num_tokens": 9322488.0,
      "step": 4170
    },
    {
      "entropy": 5.74122986793518,
      "epoch": 0.4689167181445499,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004985580828781904,
      "loss": 5.7099,
      "mean_token_accuracy": 0.16819628775119783,
      "num_tokens": 9333672.0,
      "step": 4175
    },
    {
      "entropy": 5.8608152866363525,
      "epoch": 0.46947829505250743,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004985535412754855,
      "loss": 5.7418,
      "mean_token_accuracy": 0.16545120030641555,
      "num_tokens": 9344449.0,
      "step": 4180
    },
    {
      "entropy": 5.826900577545166,
      "epoch": 0.470039871960465,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000498548992554749,
      "loss": 5.8074,
      "mean_token_accuracy": 0.1609753578901291,
      "num_tokens": 9355279.0,
      "step": 4185
    },
    {
      "entropy": 5.751828336715699,
      "epoch": 0.47060144886842253,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985444367161259,
      "loss": 5.7103,
      "mean_token_accuracy": 0.16859401911497116,
      "num_tokens": 9366942.0,
      "step": 4190
    },
    {
      "entropy": 5.685658740997314,
      "epoch": 0.4711630257763801,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004985398737597614,
      "loss": 5.6355,
      "mean_token_accuracy": 0.1736157312989235,
      "num_tokens": 9378129.0,
      "step": 4195
    },
    {
      "entropy": 5.801950359344483,
      "epoch": 0.47172460268433763,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004985353036858004,
      "loss": 5.7287,
      "mean_token_accuracy": 0.16660770624876023,
      "num_tokens": 9388757.0,
      "step": 4200
    },
    {
      "entropy": 5.744151067733765,
      "epoch": 0.4722861795922952,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004985307264943887,
      "loss": 5.6745,
      "mean_token_accuracy": 0.16941459774971007,
      "num_tokens": 9399112.0,
      "step": 4205
    },
    {
      "entropy": 5.7844664573669435,
      "epoch": 0.47284775650025274,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004985261421856721,
      "loss": 5.7691,
      "mean_token_accuracy": 0.16700249612331391,
      "num_tokens": 9410156.0,
      "step": 4210
    },
    {
      "entropy": 5.833684635162354,
      "epoch": 0.47340933340821023,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004985215507597963,
      "loss": 5.7515,
      "mean_token_accuracy": 0.17010293602943422,
      "num_tokens": 9420402.0,
      "step": 4215
    },
    {
      "entropy": 5.729236125946045,
      "epoch": 0.4739709103161678,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004985169522169076,
      "loss": 5.7532,
      "mean_token_accuracy": 0.1657623678445816,
      "num_tokens": 9431159.0,
      "step": 4220
    },
    {
      "entropy": 5.701308298110962,
      "epoch": 0.47453248722412533,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004985123465571526,
      "loss": 5.6592,
      "mean_token_accuracy": 0.17336241155862808,
      "num_tokens": 9441737.0,
      "step": 4225
    },
    {
      "entropy": 5.6865349292755125,
      "epoch": 0.4750940641320829,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985077337806777,
      "loss": 5.6196,
      "mean_token_accuracy": 0.1779865562915802,
      "num_tokens": 9452756.0,
      "step": 4230
    },
    {
      "entropy": 5.739285612106324,
      "epoch": 0.47565564104004043,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004985031138876299,
      "loss": 5.6878,
      "mean_token_accuracy": 0.1680801346898079,
      "num_tokens": 9464231.0,
      "step": 4235
    },
    {
      "entropy": 5.764004802703857,
      "epoch": 0.476217217947998,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984984868781563,
      "loss": 5.7067,
      "mean_token_accuracy": 0.1760090947151184,
      "num_tokens": 9475648.0,
      "step": 4240
    },
    {
      "entropy": 5.744698429107666,
      "epoch": 0.47677879485595553,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984938527524041,
      "loss": 5.751,
      "mean_token_accuracy": 0.16820831894874572,
      "num_tokens": 9486594.0,
      "step": 4245
    },
    {
      "entropy": 5.672558355331421,
      "epoch": 0.4773403717639131,
      "grad_norm": 1.1875,
      "learning_rate": 0.000498489211510521,
      "loss": 5.6621,
      "mean_token_accuracy": 0.17208858728408813,
      "num_tokens": 9497085.0,
      "step": 4250
    },
    {
      "entropy": 5.784761142730713,
      "epoch": 0.47790194867187064,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984845631526547,
      "loss": 5.7477,
      "mean_token_accuracy": 0.16722307056188584,
      "num_tokens": 9509765.0,
      "step": 4255
    },
    {
      "entropy": 5.725100564956665,
      "epoch": 0.47846352557982813,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004984799076789534,
      "loss": 5.5731,
      "mean_token_accuracy": 0.16857376992702483,
      "num_tokens": 9521341.0,
      "step": 4260
    },
    {
      "entropy": 5.630601024627685,
      "epoch": 0.4790251024877857,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004984752450895649,
      "loss": 5.6929,
      "mean_token_accuracy": 0.1659396380186081,
      "num_tokens": 9533196.0,
      "step": 4265
    },
    {
      "entropy": 5.8102864742279055,
      "epoch": 0.47958667939574323,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000498470575384638,
      "loss": 5.8286,
      "mean_token_accuracy": 0.15910447090864183,
      "num_tokens": 9543687.0,
      "step": 4270
    },
    {
      "entropy": 5.903976631164551,
      "epoch": 0.4801482563037008,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004984658985643214,
      "loss": 5.7156,
      "mean_token_accuracy": 0.16543746441602708,
      "num_tokens": 9555320.0,
      "step": 4275
    },
    {
      "entropy": 5.636949110031128,
      "epoch": 0.48070983321165833,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004984612146287638,
      "loss": 5.6659,
      "mean_token_accuracy": 0.16941267549991607,
      "num_tokens": 9566347.0,
      "step": 4280
    },
    {
      "entropy": 5.636159801483155,
      "epoch": 0.4812714101196159,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984565235781147,
      "loss": 5.6669,
      "mean_token_accuracy": 0.16980277746915817,
      "num_tokens": 9577111.0,
      "step": 4285
    },
    {
      "entropy": 5.761268234252929,
      "epoch": 0.48183298702757343,
      "grad_norm": 1.5,
      "learning_rate": 0.000498451825412523,
      "loss": 5.671,
      "mean_token_accuracy": 0.16593389809131623,
      "num_tokens": 9586612.0,
      "step": 4290
    },
    {
      "entropy": 5.774903297424316,
      "epoch": 0.482394563935531,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984471201321386,
      "loss": 5.6786,
      "mean_token_accuracy": 0.17062007635831833,
      "num_tokens": 9597347.0,
      "step": 4295
    },
    {
      "entropy": 5.674345445632935,
      "epoch": 0.48295614084348854,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004984424077371112,
      "loss": 5.6524,
      "mean_token_accuracy": 0.17706192433834075,
      "num_tokens": 9608507.0,
      "step": 4300
    },
    {
      "entropy": 5.708226156234741,
      "epoch": 0.4835177177514461,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004984376882275909,
      "loss": 5.7082,
      "mean_token_accuracy": 0.17436117678880692,
      "num_tokens": 9619449.0,
      "step": 4305
    },
    {
      "entropy": 5.680789613723755,
      "epoch": 0.4840792946594036,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004984329616037279,
      "loss": 5.6346,
      "mean_token_accuracy": 0.16903165727853775,
      "num_tokens": 9630613.0,
      "step": 4310
    },
    {
      "entropy": 5.761481714248657,
      "epoch": 0.48464087156736113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004984282278656727,
      "loss": 5.7086,
      "mean_token_accuracy": 0.16750625967979432,
      "num_tokens": 9642149.0,
      "step": 4315
    },
    {
      "entropy": 5.755221033096314,
      "epoch": 0.4852024484753187,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984234870135762,
      "loss": 5.6496,
      "mean_token_accuracy": 0.17484654635190963,
      "num_tokens": 9652960.0,
      "step": 4320
    },
    {
      "entropy": 5.68124532699585,
      "epoch": 0.48576402538327623,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004984187390475891,
      "loss": 5.6417,
      "mean_token_accuracy": 0.17295996993780136,
      "num_tokens": 9662676.0,
      "step": 4325
    },
    {
      "entropy": 5.838757276535034,
      "epoch": 0.4863256022912338,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004984139839678628,
      "loss": 5.7493,
      "mean_token_accuracy": 0.16618652939796447,
      "num_tokens": 9674138.0,
      "step": 4330
    },
    {
      "entropy": 5.714073657989502,
      "epoch": 0.48688717919919133,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004984092217745486,
      "loss": 5.7113,
      "mean_token_accuracy": 0.1709674835205078,
      "num_tokens": 9685397.0,
      "step": 4335
    },
    {
      "entropy": 5.790579319000244,
      "epoch": 0.4874487561071489,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984044524677981,
      "loss": 5.8231,
      "mean_token_accuracy": 0.1605924040079117,
      "num_tokens": 9697266.0,
      "step": 4340
    },
    {
      "entropy": 5.828333997726441,
      "epoch": 0.48801033301510643,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004983996760477632,
      "loss": 5.6552,
      "mean_token_accuracy": 0.16677314341068267,
      "num_tokens": 9707967.0,
      "step": 4345
    },
    {
      "entropy": 5.62044587135315,
      "epoch": 0.488571909923064,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000498394892514596,
      "loss": 5.6341,
      "mean_token_accuracy": 0.17084338515996933,
      "num_tokens": 9718162.0,
      "step": 4350
    },
    {
      "entropy": 5.717535734176636,
      "epoch": 0.48913348683102154,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004983901018684486,
      "loss": 5.6886,
      "mean_token_accuracy": 0.16895658969879152,
      "num_tokens": 9729439.0,
      "step": 4355
    },
    {
      "entropy": 5.775072956085205,
      "epoch": 0.48969506373897903,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004983853041094739,
      "loss": 5.7112,
      "mean_token_accuracy": 0.16845377534627914,
      "num_tokens": 9741383.0,
      "step": 4360
    },
    {
      "entropy": 5.712487268447876,
      "epoch": 0.4902566406469366,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004983804992378243,
      "loss": 5.7325,
      "mean_token_accuracy": 0.17263987958431243,
      "num_tokens": 9753050.0,
      "step": 4365
    },
    {
      "entropy": 5.681834554672241,
      "epoch": 0.49081821755489413,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004983756872536532,
      "loss": 5.6141,
      "mean_token_accuracy": 0.1744478538632393,
      "num_tokens": 9763715.0,
      "step": 4370
    },
    {
      "entropy": 5.699193477630615,
      "epoch": 0.4913797944628517,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004983708681571134,
      "loss": 5.5872,
      "mean_token_accuracy": 0.1801524966955185,
      "num_tokens": 9774145.0,
      "step": 4375
    },
    {
      "entropy": 5.711805200576782,
      "epoch": 0.49194137137080923,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983660419483586,
      "loss": 5.6484,
      "mean_token_accuracy": 0.1740181788802147,
      "num_tokens": 9784703.0,
      "step": 4380
    },
    {
      "entropy": 5.7284644603729244,
      "epoch": 0.4925029482787668,
      "grad_norm": 1.125,
      "learning_rate": 0.0004983612086275424,
      "loss": 5.6657,
      "mean_token_accuracy": 0.16666370034217834,
      "num_tokens": 9795426.0,
      "step": 4385
    },
    {
      "entropy": 5.744356107711792,
      "epoch": 0.49306452518672433,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004983563681948187,
      "loss": 5.7488,
      "mean_token_accuracy": 0.16563860327005386,
      "num_tokens": 9806539.0,
      "step": 4390
    },
    {
      "entropy": 5.838592100143432,
      "epoch": 0.4936261020946819,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004983515206503416,
      "loss": 5.7354,
      "mean_token_accuracy": 0.16389266550540924,
      "num_tokens": 9816806.0,
      "step": 4395
    },
    {
      "entropy": 5.741393136978149,
      "epoch": 0.49418767900263944,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004983466659942655,
      "loss": 5.6239,
      "mean_token_accuracy": 0.1763622909784317,
      "num_tokens": 9828001.0,
      "step": 4400
    },
    {
      "entropy": 5.568348503112793,
      "epoch": 0.49474925591059693,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983418042267449,
      "loss": 5.6675,
      "mean_token_accuracy": 0.17209520637989045,
      "num_tokens": 9838773.0,
      "step": 4405
    },
    {
      "entropy": 5.723768329620361,
      "epoch": 0.4953108328185545,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983369353479345,
      "loss": 5.664,
      "mean_token_accuracy": 0.17434099316596985,
      "num_tokens": 9848680.0,
      "step": 4410
    },
    {
      "entropy": 5.700279903411865,
      "epoch": 0.49587240972651203,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004983320593579895,
      "loss": 5.6349,
      "mean_token_accuracy": 0.16858525723218917,
      "num_tokens": 9859168.0,
      "step": 4415
    },
    {
      "entropy": 5.66675705909729,
      "epoch": 0.4964339866344696,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983271762570653,
      "loss": 5.671,
      "mean_token_accuracy": 0.17040987610816954,
      "num_tokens": 9869071.0,
      "step": 4420
    },
    {
      "entropy": 5.750021266937256,
      "epoch": 0.49699556354242713,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498322286045317,
      "loss": 5.6767,
      "mean_token_accuracy": 0.17082203477621077,
      "num_tokens": 9880208.0,
      "step": 4425
    },
    {
      "entropy": 5.7331212043762205,
      "epoch": 0.4975571404503847,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004983173887229006,
      "loss": 5.7172,
      "mean_token_accuracy": 0.17034917324781418,
      "num_tokens": 9891192.0,
      "step": 4430
    },
    {
      "entropy": 5.790711212158203,
      "epoch": 0.49811871735834223,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004983124842899719,
      "loss": 5.7379,
      "mean_token_accuracy": 0.16513569355010987,
      "num_tokens": 9903038.0,
      "step": 4435
    },
    {
      "entropy": 5.830787420272827,
      "epoch": 0.4986802942662998,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004983075727466871,
      "loss": 5.7921,
      "mean_token_accuracy": 0.1677357569336891,
      "num_tokens": 9916104.0,
      "step": 4440
    },
    {
      "entropy": 5.6895956039428714,
      "epoch": 0.49924187117425733,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004983026540932026,
      "loss": 5.6103,
      "mean_token_accuracy": 0.171831975877285,
      "num_tokens": 9927094.0,
      "step": 4445
    },
    {
      "entropy": 5.784496068954468,
      "epoch": 0.4998034480822149,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004982977283296749,
      "loss": 5.8228,
      "mean_token_accuracy": 0.1600668027997017,
      "num_tokens": 9937964.0,
      "step": 4450
    },
    {
      "entropy": 5.8589826107025145,
      "epoch": 0.5003650249901724,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000498292795456261,
      "loss": 5.7841,
      "mean_token_accuracy": 0.16068200916051864,
      "num_tokens": 9949238.0,
      "step": 4455
    },
    {
      "entropy": 5.740886306762695,
      "epoch": 0.50092660189813,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004982878554731178,
      "loss": 5.6144,
      "mean_token_accuracy": 0.17186491191387177,
      "num_tokens": 9960125.0,
      "step": 4460
    },
    {
      "entropy": 5.579822969436646,
      "epoch": 0.5014881788060875,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982829083804028,
      "loss": 5.6379,
      "mean_token_accuracy": 0.1658797100186348,
      "num_tokens": 9971130.0,
      "step": 4465
    },
    {
      "entropy": 5.683795070648193,
      "epoch": 0.5020497557140451,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004982779541782733,
      "loss": 5.5301,
      "mean_token_accuracy": 0.17982134222984314,
      "num_tokens": 9981516.0,
      "step": 4470
    },
    {
      "entropy": 5.691908359527588,
      "epoch": 0.5026113326220026,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004982729928668872,
      "loss": 5.765,
      "mean_token_accuracy": 0.16426052749156952,
      "num_tokens": 9992360.0,
      "step": 4475
    },
    {
      "entropy": 5.844257497787476,
      "epoch": 0.5031729095299601,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004982680244464024,
      "loss": 5.865,
      "mean_token_accuracy": 0.16424636989831926,
      "num_tokens": 10003410.0,
      "step": 4480
    },
    {
      "entropy": 5.831128740310669,
      "epoch": 0.5037344864379176,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004982630489169771,
      "loss": 5.6665,
      "mean_token_accuracy": 0.1697389677166939,
      "num_tokens": 10014974.0,
      "step": 4485
    },
    {
      "entropy": 5.730645370483399,
      "epoch": 0.5042960633458752,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004982580662787697,
      "loss": 5.7355,
      "mean_token_accuracy": 0.1644341304898262,
      "num_tokens": 10026702.0,
      "step": 4490
    },
    {
      "entropy": 5.7113964557647705,
      "epoch": 0.5048576402538327,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004982530765319389,
      "loss": 5.7192,
      "mean_token_accuracy": 0.1672709807753563,
      "num_tokens": 10039318.0,
      "step": 4495
    },
    {
      "entropy": 5.741805553436279,
      "epoch": 0.5054192171617903,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004982480796766436,
      "loss": 5.6632,
      "mean_token_accuracy": 0.1635326474905014,
      "num_tokens": 10049726.0,
      "step": 4500
    },
    {
      "entropy": 5.691263341903687,
      "epoch": 0.5059807940697478,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004982430757130429,
      "loss": 5.5585,
      "mean_token_accuracy": 0.1718662425875664,
      "num_tokens": 10059663.0,
      "step": 4505
    },
    {
      "entropy": 5.583151340484619,
      "epoch": 0.5065423709777054,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004982380646412959,
      "loss": 5.6013,
      "mean_token_accuracy": 0.16774161756038666,
      "num_tokens": 10071195.0,
      "step": 4510
    },
    {
      "entropy": 5.827329444885254,
      "epoch": 0.5071039478856629,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004982330464615625,
      "loss": 5.8071,
      "mean_token_accuracy": 0.16533977538347244,
      "num_tokens": 10082014.0,
      "step": 4515
    },
    {
      "entropy": 5.811189365386963,
      "epoch": 0.5076655247936205,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004982280211740023,
      "loss": 5.6143,
      "mean_token_accuracy": 0.17575001269578933,
      "num_tokens": 10093156.0,
      "step": 4520
    },
    {
      "entropy": 5.631749963760376,
      "epoch": 0.508227101701578,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004982229887787753,
      "loss": 5.7599,
      "mean_token_accuracy": 0.1561910793185234,
      "num_tokens": 10104993.0,
      "step": 4525
    },
    {
      "entropy": 5.7125547409057615,
      "epoch": 0.5087886786095356,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004982179492760418,
      "loss": 5.5668,
      "mean_token_accuracy": 0.18113714456558228,
      "num_tokens": 10114590.0,
      "step": 4530
    },
    {
      "entropy": 5.67285680770874,
      "epoch": 0.5093502555174931,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982129026659621,
      "loss": 5.5738,
      "mean_token_accuracy": 0.17499979734420776,
      "num_tokens": 10125849.0,
      "step": 4535
    },
    {
      "entropy": 5.626177024841309,
      "epoch": 0.5099118324254507,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004982078489486972,
      "loss": 5.5941,
      "mean_token_accuracy": 0.17779029458761214,
      "num_tokens": 10135493.0,
      "step": 4540
    },
    {
      "entropy": 5.7630730152130125,
      "epoch": 0.5104734093334082,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004982027881244076,
      "loss": 5.7775,
      "mean_token_accuracy": 0.1641564130783081,
      "num_tokens": 10146987.0,
      "step": 4545
    },
    {
      "entropy": 5.765534067153931,
      "epoch": 0.5110349862413658,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004981977201932549,
      "loss": 5.7086,
      "mean_token_accuracy": 0.17529913783073425,
      "num_tokens": 10157823.0,
      "step": 4550
    },
    {
      "entropy": 5.639018440246582,
      "epoch": 0.5115965631493233,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004981926451554001,
      "loss": 5.6758,
      "mean_token_accuracy": 0.17088306099176406,
      "num_tokens": 10169305.0,
      "step": 4555
    },
    {
      "entropy": 5.824262237548828,
      "epoch": 0.5121581400572809,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004981875630110048,
      "loss": 5.7553,
      "mean_token_accuracy": 0.1605099081993103,
      "num_tokens": 10180346.0,
      "step": 4560
    },
    {
      "entropy": 5.753881549835205,
      "epoch": 0.5127197169652384,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004981824737602311,
      "loss": 5.6425,
      "mean_token_accuracy": 0.17663276344537734,
      "num_tokens": 10190870.0,
      "step": 4565
    },
    {
      "entropy": 5.745822525024414,
      "epoch": 0.513281293873196,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004981773774032409,
      "loss": 5.7113,
      "mean_token_accuracy": 0.15780248045921325,
      "num_tokens": 10202408.0,
      "step": 4570
    },
    {
      "entropy": 5.637575244903564,
      "epoch": 0.5138428707811534,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004981722739401962,
      "loss": 5.5405,
      "mean_token_accuracy": 0.17590135782957078,
      "num_tokens": 10213367.0,
      "step": 4575
    },
    {
      "entropy": 5.701434135437012,
      "epoch": 0.514404447689111,
      "grad_norm": 1.3125,
      "learning_rate": 0.00049816716337126,
      "loss": 5.6765,
      "mean_token_accuracy": 0.16563864946365356,
      "num_tokens": 10223865.0,
      "step": 4580
    },
    {
      "entropy": 5.599912309646607,
      "epoch": 0.5149660245970685,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004981620456965946,
      "loss": 5.4468,
      "mean_token_accuracy": 0.17510706782341004,
      "num_tokens": 10235268.0,
      "step": 4585
    },
    {
      "entropy": 5.6726030826568605,
      "epoch": 0.5155276015050261,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004981569209163633,
      "loss": 5.7124,
      "mean_token_accuracy": 0.16576266586780547,
      "num_tokens": 10247700.0,
      "step": 4590
    },
    {
      "entropy": 5.761053800582886,
      "epoch": 0.5160891784129836,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004981517890307288,
      "loss": 5.76,
      "mean_token_accuracy": 0.16685858815908433,
      "num_tokens": 10258553.0,
      "step": 4595
    },
    {
      "entropy": 5.758320093154907,
      "epoch": 0.5166507553209412,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000498146650039855,
      "loss": 5.6481,
      "mean_token_accuracy": 0.17269993275403978,
      "num_tokens": 10268496.0,
      "step": 4600
    },
    {
      "entropy": 5.774955701828003,
      "epoch": 0.5172123322288987,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004981415039439054,
      "loss": 5.7253,
      "mean_token_accuracy": 0.16718546748161317,
      "num_tokens": 10278803.0,
      "step": 4605
    },
    {
      "entropy": 5.711667108535766,
      "epoch": 0.5177739091368563,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004981363507430436,
      "loss": 5.5835,
      "mean_token_accuracy": 0.1724689558148384,
      "num_tokens": 10289805.0,
      "step": 4610
    },
    {
      "entropy": 5.688625192642212,
      "epoch": 0.5183354860448138,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004981311904374339,
      "loss": 5.7046,
      "mean_token_accuracy": 0.174521566927433,
      "num_tokens": 10300295.0,
      "step": 4615
    },
    {
      "entropy": 5.841267871856689,
      "epoch": 0.5188970629527714,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004981260230272405,
      "loss": 5.7889,
      "mean_token_accuracy": 0.16799403876066207,
      "num_tokens": 10312175.0,
      "step": 4620
    },
    {
      "entropy": 5.63864483833313,
      "epoch": 0.5194586398607289,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004981208485126281,
      "loss": 5.6779,
      "mean_token_accuracy": 0.1660699799656868,
      "num_tokens": 10322888.0,
      "step": 4625
    },
    {
      "entropy": 5.693124485015869,
      "epoch": 0.5200202167686865,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004981156668937613,
      "loss": 5.6516,
      "mean_token_accuracy": 0.17714009433984756,
      "num_tokens": 10333305.0,
      "step": 4630
    },
    {
      "entropy": 5.757165384292603,
      "epoch": 0.520581793676644,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004981104781708051,
      "loss": 5.6988,
      "mean_token_accuracy": 0.16862452477216722,
      "num_tokens": 10344190.0,
      "step": 4635
    },
    {
      "entropy": 5.797582530975342,
      "epoch": 0.5211433705846016,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004981052823439247,
      "loss": 5.7742,
      "mean_token_accuracy": 0.16717695221304893,
      "num_tokens": 10356271.0,
      "step": 4640
    },
    {
      "entropy": 5.836018705368042,
      "epoch": 0.5217049474925591,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004981000794132858,
      "loss": 5.8366,
      "mean_token_accuracy": 0.15832895487546922,
      "num_tokens": 10368756.0,
      "step": 4645
    },
    {
      "entropy": 5.814976835250855,
      "epoch": 0.5222665244005167,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004980948693790536,
      "loss": 5.6962,
      "mean_token_accuracy": 0.1687784492969513,
      "num_tokens": 10380029.0,
      "step": 4650
    },
    {
      "entropy": 5.691782474517822,
      "epoch": 0.5228281013084742,
      "grad_norm": 1.25,
      "learning_rate": 0.0004980896522413943,
      "loss": 5.6803,
      "mean_token_accuracy": 0.16841885596513748,
      "num_tokens": 10390358.0,
      "step": 4655
    },
    {
      "entropy": 5.686275625228882,
      "epoch": 0.5233896782164318,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004980844280004739,
      "loss": 5.6551,
      "mean_token_accuracy": 0.1711376890540123,
      "num_tokens": 10401331.0,
      "step": 4660
    },
    {
      "entropy": 5.753196573257446,
      "epoch": 0.5239512551243893,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004980791966564589,
      "loss": 5.6783,
      "mean_token_accuracy": 0.16949976235628128,
      "num_tokens": 10412755.0,
      "step": 4665
    },
    {
      "entropy": 5.692148733139038,
      "epoch": 0.5245128320323468,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004980739582095158,
      "loss": 5.6194,
      "mean_token_accuracy": 0.17993575930595399,
      "num_tokens": 10422698.0,
      "step": 4670
    },
    {
      "entropy": 5.739432430267334,
      "epoch": 0.5250744089403043,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004980687126598111,
      "loss": 5.753,
      "mean_token_accuracy": 0.16443512141704558,
      "num_tokens": 10434467.0,
      "step": 4675
    },
    {
      "entropy": 5.691123342514038,
      "epoch": 0.5256359858482619,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004980634600075123,
      "loss": 5.6647,
      "mean_token_accuracy": 0.17190593779087066,
      "num_tokens": 10446007.0,
      "step": 4680
    },
    {
      "entropy": 5.758034658432007,
      "epoch": 0.5261975627562194,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004980582002527863,
      "loss": 5.6857,
      "mean_token_accuracy": 0.17210673540830612,
      "num_tokens": 10457538.0,
      "step": 4685
    },
    {
      "entropy": 5.737921142578125,
      "epoch": 0.526759139664177,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004980529333958008,
      "loss": 5.6438,
      "mean_token_accuracy": 0.17155391573905945,
      "num_tokens": 10468389.0,
      "step": 4690
    },
    {
      "entropy": 5.64755654335022,
      "epoch": 0.5273207165721345,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004980476594367233,
      "loss": 5.6167,
      "mean_token_accuracy": 0.166074176132679,
      "num_tokens": 10479032.0,
      "step": 4695
    },
    {
      "entropy": 5.7380961894989015,
      "epoch": 0.5278822934800921,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004980423783757219,
      "loss": 5.6203,
      "mean_token_accuracy": 0.1761274218559265,
      "num_tokens": 10490505.0,
      "step": 4700
    },
    {
      "entropy": 5.777146339416504,
      "epoch": 0.5284438703880496,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004980370902129646,
      "loss": 5.6841,
      "mean_token_accuracy": 0.17220447063446045,
      "num_tokens": 10501382.0,
      "step": 4705
    },
    {
      "entropy": 5.7597027778625485,
      "epoch": 0.5290054472960072,
      "grad_norm": 1.171875,
      "learning_rate": 0.00049803179494862,
      "loss": 5.6519,
      "mean_token_accuracy": 0.17078416049480438,
      "num_tokens": 10512727.0,
      "step": 4710
    },
    {
      "entropy": 5.627700519561768,
      "epoch": 0.5295670242039647,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004980264925828563,
      "loss": 5.6379,
      "mean_token_accuracy": 0.17503894567489625,
      "num_tokens": 10523383.0,
      "step": 4715
    },
    {
      "entropy": 5.723991012573242,
      "epoch": 0.5301286011119223,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004980211831158427,
      "loss": 5.6443,
      "mean_token_accuracy": 0.17036737948656083,
      "num_tokens": 10534416.0,
      "step": 4720
    },
    {
      "entropy": 5.713537502288818,
      "epoch": 0.5306901780198798,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004980158665477481,
      "loss": 5.6192,
      "mean_token_accuracy": 0.1723712608218193,
      "num_tokens": 10544291.0,
      "step": 4725
    },
    {
      "entropy": 5.703956174850464,
      "epoch": 0.5312517549278374,
      "grad_norm": 1.25,
      "learning_rate": 0.0004980105428787419,
      "loss": 5.585,
      "mean_token_accuracy": 0.1744757741689682,
      "num_tokens": 10555291.0,
      "step": 4730
    },
    {
      "entropy": 5.739058351516723,
      "epoch": 0.5318133318357949,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004980052121089934,
      "loss": 5.6839,
      "mean_token_accuracy": 0.1688266322016716,
      "num_tokens": 10565812.0,
      "step": 4735
    },
    {
      "entropy": 5.644025564193726,
      "epoch": 0.5323749087437525,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004979998742386725,
      "loss": 5.6093,
      "mean_token_accuracy": 0.17441638559103012,
      "num_tokens": 10577351.0,
      "step": 4740
    },
    {
      "entropy": 5.761340379714966,
      "epoch": 0.53293648565171,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004979945292679492,
      "loss": 5.6919,
      "mean_token_accuracy": 0.1682047963142395,
      "num_tokens": 10589236.0,
      "step": 4745
    },
    {
      "entropy": 5.650798368453979,
      "epoch": 0.5334980625596676,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004979891771969935,
      "loss": 5.6363,
      "mean_token_accuracy": 0.17235926836729049,
      "num_tokens": 10599926.0,
      "step": 4750
    },
    {
      "entropy": 5.7528852939605715,
      "epoch": 0.5340596394676251,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000497983818025976,
      "loss": 5.7496,
      "mean_token_accuracy": 0.16808155328035354,
      "num_tokens": 10611495.0,
      "step": 4755
    },
    {
      "entropy": 5.749897480010986,
      "epoch": 0.5346212163755827,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004979784517550671,
      "loss": 5.7608,
      "mean_token_accuracy": 0.1683972179889679,
      "num_tokens": 10622455.0,
      "step": 4760
    },
    {
      "entropy": 5.70879487991333,
      "epoch": 0.5351827932835401,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004979730783844379,
      "loss": 5.6031,
      "mean_token_accuracy": 0.17482960671186448,
      "num_tokens": 10634255.0,
      "step": 4765
    },
    {
      "entropy": 5.737195301055908,
      "epoch": 0.5357443701914977,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004979676979142595,
      "loss": 5.7169,
      "mean_token_accuracy": 0.1722305104136467,
      "num_tokens": 10644816.0,
      "step": 4770
    },
    {
      "entropy": 5.732215309143067,
      "epoch": 0.5363059470994552,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000497962310344703,
      "loss": 5.6459,
      "mean_token_accuracy": 0.17166779786348343,
      "num_tokens": 10655449.0,
      "step": 4775
    },
    {
      "entropy": 5.718864822387696,
      "epoch": 0.5368675240074128,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00049795691567594,
      "loss": 5.6553,
      "mean_token_accuracy": 0.16806258261203766,
      "num_tokens": 10666343.0,
      "step": 4780
    },
    {
      "entropy": 5.72328372001648,
      "epoch": 0.5374291009153703,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004979515139081426,
      "loss": 5.7064,
      "mean_token_accuracy": 0.16548898369073867,
      "num_tokens": 10678511.0,
      "step": 4785
    },
    {
      "entropy": 5.665549087524414,
      "epoch": 0.5379906778233279,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004979461050414822,
      "loss": 5.6909,
      "mean_token_accuracy": 0.17560625523328782,
      "num_tokens": 10690005.0,
      "step": 4790
    },
    {
      "entropy": 5.679067897796631,
      "epoch": 0.5385522547312854,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004979406890761315,
      "loss": 5.681,
      "mean_token_accuracy": 0.1686423435807228,
      "num_tokens": 10701062.0,
      "step": 4795
    },
    {
      "entropy": 5.821629285812378,
      "epoch": 0.539113831639243,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004979352660122629,
      "loss": 5.5993,
      "mean_token_accuracy": 0.1772200345993042,
      "num_tokens": 10711986.0,
      "step": 4800
    },
    {
      "entropy": 5.744456338882446,
      "epoch": 0.5396754085472005,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004979298358500488,
      "loss": 5.7093,
      "mean_token_accuracy": 0.16843186989426612,
      "num_tokens": 10722509.0,
      "step": 4805
    },
    {
      "entropy": 5.742955589294434,
      "epoch": 0.5402369854551581,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004979243985896624,
      "loss": 5.6577,
      "mean_token_accuracy": 0.16745564192533494,
      "num_tokens": 10733755.0,
      "step": 4810
    },
    {
      "entropy": 5.69589319229126,
      "epoch": 0.5407985623631156,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004979189542312765,
      "loss": 5.688,
      "mean_token_accuracy": 0.16879853308200837,
      "num_tokens": 10744356.0,
      "step": 4815
    },
    {
      "entropy": 5.710256290435791,
      "epoch": 0.5413601392710732,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004979135027750649,
      "loss": 5.5928,
      "mean_token_accuracy": 0.17196511775255202,
      "num_tokens": 10756474.0,
      "step": 4820
    },
    {
      "entropy": 5.702636575698852,
      "epoch": 0.5419217161790307,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004979080442212007,
      "loss": 5.6718,
      "mean_token_accuracy": 0.17200553864240647,
      "num_tokens": 10767551.0,
      "step": 4825
    },
    {
      "entropy": 5.7792284965515135,
      "epoch": 0.5424832930869883,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004979025785698581,
      "loss": 5.7624,
      "mean_token_accuracy": 0.16749392002820968,
      "num_tokens": 10778524.0,
      "step": 4830
    },
    {
      "entropy": 5.679055833816529,
      "epoch": 0.5430448699949458,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004978971058212108,
      "loss": 5.6711,
      "mean_token_accuracy": 0.17368635684251785,
      "num_tokens": 10789798.0,
      "step": 4835
    },
    {
      "entropy": 5.730114889144898,
      "epoch": 0.5436064469029034,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004978916259754331,
      "loss": 5.6202,
      "mean_token_accuracy": 0.1732513874769211,
      "num_tokens": 10800500.0,
      "step": 4840
    },
    {
      "entropy": 5.761911201477051,
      "epoch": 0.5441680238108609,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004978861390326998,
      "loss": 5.6492,
      "mean_token_accuracy": 0.17786261439323425,
      "num_tokens": 10812047.0,
      "step": 4845
    },
    {
      "entropy": 5.655855131149292,
      "epoch": 0.5447296007188185,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004978806449931854,
      "loss": 5.6656,
      "mean_token_accuracy": 0.1673392355442047,
      "num_tokens": 10823272.0,
      "step": 4850
    },
    {
      "entropy": 5.585034036636353,
      "epoch": 0.545291177626776,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004978751438570647,
      "loss": 5.5368,
      "mean_token_accuracy": 0.17088105529546738,
      "num_tokens": 10834485.0,
      "step": 4855
    },
    {
      "entropy": 5.683445978164673,
      "epoch": 0.5458527545347336,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004978696356245129,
      "loss": 5.614,
      "mean_token_accuracy": 0.16837672442197799,
      "num_tokens": 10843547.0,
      "step": 4860
    },
    {
      "entropy": 5.719631719589233,
      "epoch": 0.546414331442691,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004978641202957057,
      "loss": 5.7497,
      "mean_token_accuracy": 0.16445043236017226,
      "num_tokens": 10855890.0,
      "step": 4865
    },
    {
      "entropy": 5.91353702545166,
      "epoch": 0.5469759083506486,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004978585978708183,
      "loss": 5.7949,
      "mean_token_accuracy": 0.16377429962158202,
      "num_tokens": 10868060.0,
      "step": 4870
    },
    {
      "entropy": 5.783869218826294,
      "epoch": 0.5475374852586061,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004978530683500267,
      "loss": 5.7163,
      "mean_token_accuracy": 0.1670852780342102,
      "num_tokens": 10879323.0,
      "step": 4875
    },
    {
      "entropy": 5.661490631103516,
      "epoch": 0.5480990621665637,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004978475317335071,
      "loss": 5.5598,
      "mean_token_accuracy": 0.1758361741900444,
      "num_tokens": 10890545.0,
      "step": 4880
    },
    {
      "entropy": 5.658501005172729,
      "epoch": 0.5486606390745212,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004978419880214356,
      "loss": 5.6612,
      "mean_token_accuracy": 0.16851151436567308,
      "num_tokens": 10900761.0,
      "step": 4885
    },
    {
      "entropy": 5.717454147338867,
      "epoch": 0.5492222159824788,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004978364372139887,
      "loss": 5.6471,
      "mean_token_accuracy": 0.17221338003873826,
      "num_tokens": 10911874.0,
      "step": 4890
    },
    {
      "entropy": 5.68519868850708,
      "epoch": 0.5497837928904363,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004978308793113434,
      "loss": 5.5666,
      "mean_token_accuracy": 0.1797461986541748,
      "num_tokens": 10922949.0,
      "step": 4895
    },
    {
      "entropy": 5.734178066253662,
      "epoch": 0.5503453697983939,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004978253143136762,
      "loss": 5.7637,
      "mean_token_accuracy": 0.16003435403108596,
      "num_tokens": 10935415.0,
      "step": 4900
    },
    {
      "entropy": 5.712212610244751,
      "epoch": 0.5509069467063514,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004978197422211647,
      "loss": 5.5989,
      "mean_token_accuracy": 0.1794376477599144,
      "num_tokens": 10946983.0,
      "step": 4905
    },
    {
      "entropy": 5.645990180969238,
      "epoch": 0.551468523614309,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004978141630339863,
      "loss": 5.6755,
      "mean_token_accuracy": 0.16025410294532777,
      "num_tokens": 10957884.0,
      "step": 4910
    },
    {
      "entropy": 5.732716608047485,
      "epoch": 0.5520301005222665,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004978085767523184,
      "loss": 5.6752,
      "mean_token_accuracy": 0.17449066042900085,
      "num_tokens": 10968518.0,
      "step": 4915
    },
    {
      "entropy": 5.708969449996948,
      "epoch": 0.5525916774302241,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000497802983376339,
      "loss": 5.6864,
      "mean_token_accuracy": 0.16695939302444457,
      "num_tokens": 10979621.0,
      "step": 4920
    },
    {
      "entropy": 5.764673709869385,
      "epoch": 0.5531532543381816,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004977973829062263,
      "loss": 5.619,
      "mean_token_accuracy": 0.1710394114255905,
      "num_tokens": 10990158.0,
      "step": 4925
    },
    {
      "entropy": 5.673544979095459,
      "epoch": 0.5537148312461392,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004977917753421585,
      "loss": 5.7305,
      "mean_token_accuracy": 0.16447483003139496,
      "num_tokens": 11001732.0,
      "step": 4930
    },
    {
      "entropy": 5.843268728256225,
      "epoch": 0.5542764081540967,
      "grad_norm": 1.390625,
      "learning_rate": 0.000497786160684314,
      "loss": 5.7428,
      "mean_token_accuracy": 0.1634986564517021,
      "num_tokens": 11013483.0,
      "step": 4935
    },
    {
      "entropy": 5.796664619445801,
      "epoch": 0.5548379850620543,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004977805389328719,
      "loss": 5.7391,
      "mean_token_accuracy": 0.16766392290592194,
      "num_tokens": 11025423.0,
      "step": 4940
    },
    {
      "entropy": 5.703360080718994,
      "epoch": 0.5553995619700118,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000497774910088011,
      "loss": 5.5716,
      "mean_token_accuracy": 0.17025255560874938,
      "num_tokens": 11037041.0,
      "step": 4945
    },
    {
      "entropy": 5.723888826370239,
      "epoch": 0.5559611388779694,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004977692741499106,
      "loss": 5.7208,
      "mean_token_accuracy": 0.16566304564476014,
      "num_tokens": 11049206.0,
      "step": 4950
    },
    {
      "entropy": 5.74290828704834,
      "epoch": 0.5565227157859269,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00049776363111875,
      "loss": 5.6884,
      "mean_token_accuracy": 0.17315239310264588,
      "num_tokens": 11060407.0,
      "step": 4955
    },
    {
      "entropy": 5.688162040710449,
      "epoch": 0.5570842926938844,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004977579809947091,
      "loss": 5.6215,
      "mean_token_accuracy": 0.1681760311126709,
      "num_tokens": 11071269.0,
      "step": 4960
    },
    {
      "entropy": 5.701694965362549,
      "epoch": 0.5576458696018419,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004977523237779676,
      "loss": 5.6215,
      "mean_token_accuracy": 0.17521134316921233,
      "num_tokens": 11082078.0,
      "step": 4965
    },
    {
      "entropy": 5.906933164596557,
      "epoch": 0.5582074465097995,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004977466594687057,
      "loss": 5.8526,
      "mean_token_accuracy": 0.1614489182829857,
      "num_tokens": 11094596.0,
      "step": 4970
    },
    {
      "entropy": 5.737157964706421,
      "epoch": 0.558769023417757,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004977409880671039,
      "loss": 5.7163,
      "mean_token_accuracy": 0.16220240145921708,
      "num_tokens": 11107705.0,
      "step": 4975
    },
    {
      "entropy": 5.703377056121826,
      "epoch": 0.5593306003257146,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004977353095733427,
      "loss": 5.622,
      "mean_token_accuracy": 0.17367908358573914,
      "num_tokens": 11118152.0,
      "step": 4980
    },
    {
      "entropy": 5.711778259277343,
      "epoch": 0.5598921772336721,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004977296239876026,
      "loss": 5.6611,
      "mean_token_accuracy": 0.17005926817655564,
      "num_tokens": 11129261.0,
      "step": 4985
    },
    {
      "entropy": 5.815555191040039,
      "epoch": 0.5604537541416297,
      "grad_norm": 1.203125,
      "learning_rate": 0.000497723931310065,
      "loss": 5.7661,
      "mean_token_accuracy": 0.1613462135195732,
      "num_tokens": 11142227.0,
      "step": 4990
    },
    {
      "entropy": 5.781078052520752,
      "epoch": 0.5610153310495872,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004977182315409111,
      "loss": 5.6717,
      "mean_token_accuracy": 0.1686568960547447,
      "num_tokens": 11154277.0,
      "step": 4995
    },
    {
      "entropy": 5.681065464019776,
      "epoch": 0.5615769079575448,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004977125246803224,
      "loss": 5.6813,
      "mean_token_accuracy": 0.1690026491880417,
      "num_tokens": 11164902.0,
      "step": 5000
    },
    {
      "entropy": 5.634833431243896,
      "epoch": 0.5621384848655023,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004977068107284803,
      "loss": 5.5775,
      "mean_token_accuracy": 0.16621919125318527,
      "num_tokens": 11176010.0,
      "step": 5005
    },
    {
      "entropy": 5.798631000518799,
      "epoch": 0.5627000617734599,
      "grad_norm": 1.375,
      "learning_rate": 0.0004977010896855672,
      "loss": 5.6064,
      "mean_token_accuracy": 0.17368722558021546,
      "num_tokens": 11187773.0,
      "step": 5010
    },
    {
      "entropy": 5.680552387237549,
      "epoch": 0.5632616386814174,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004976953615517649,
      "loss": 5.6802,
      "mean_token_accuracy": 0.17477964162826537,
      "num_tokens": 11198420.0,
      "step": 5015
    },
    {
      "entropy": 5.5892163753509525,
      "epoch": 0.563823215589375,
      "grad_norm": 2.109375,
      "learning_rate": 0.000497689626327256,
      "loss": 5.4906,
      "mean_token_accuracy": 0.17867352068424225,
      "num_tokens": 11208553.0,
      "step": 5020
    },
    {
      "entropy": 5.70281949043274,
      "epoch": 0.5643847924973325,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000497683884012223,
      "loss": 5.6524,
      "mean_token_accuracy": 0.1714022785425186,
      "num_tokens": 11219737.0,
      "step": 5025
    },
    {
      "entropy": 5.650163793563843,
      "epoch": 0.5649463694052901,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004976781346068487,
      "loss": 5.6602,
      "mean_token_accuracy": 0.16621274948120118,
      "num_tokens": 11230997.0,
      "step": 5030
    },
    {
      "entropy": 5.775289058685303,
      "epoch": 0.5655079463132476,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004976723781113163,
      "loss": 5.658,
      "mean_token_accuracy": 0.17257270216941833,
      "num_tokens": 11242338.0,
      "step": 5035
    },
    {
      "entropy": 5.775376224517823,
      "epoch": 0.5660695232212052,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004976666145258089,
      "loss": 5.6957,
      "mean_token_accuracy": 0.17020781189203263,
      "num_tokens": 11252671.0,
      "step": 5040
    },
    {
      "entropy": 5.555762767791748,
      "epoch": 0.5666311001291627,
      "grad_norm": 1.25,
      "learning_rate": 0.0004976608438505103,
      "loss": 5.5207,
      "mean_token_accuracy": 0.17949869185686113,
      "num_tokens": 11262721.0,
      "step": 5045
    },
    {
      "entropy": 5.700231218338013,
      "epoch": 0.5671926770371203,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004976550660856041,
      "loss": 5.7102,
      "mean_token_accuracy": 0.16806024610996245,
      "num_tokens": 11274074.0,
      "step": 5050
    },
    {
      "entropy": 5.789868927001953,
      "epoch": 0.5677542539450777,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000497649281231274,
      "loss": 5.6467,
      "mean_token_accuracy": 0.1738978683948517,
      "num_tokens": 11286051.0,
      "step": 5055
    },
    {
      "entropy": 5.801263046264649,
      "epoch": 0.5683158308530353,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004976434892877046,
      "loss": 5.7707,
      "mean_token_accuracy": 0.16562236547470094,
      "num_tokens": 11298841.0,
      "step": 5060
    },
    {
      "entropy": 5.760428190231323,
      "epoch": 0.5688774077609928,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004976376902550802,
      "loss": 5.7601,
      "mean_token_accuracy": 0.1714622288942337,
      "num_tokens": 11309898.0,
      "step": 5065
    },
    {
      "entropy": 5.770657253265381,
      "epoch": 0.5694389846689504,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004976318841335854,
      "loss": 5.7862,
      "mean_token_accuracy": 0.16856923401355745,
      "num_tokens": 11322163.0,
      "step": 5070
    },
    {
      "entropy": 5.779578781127929,
      "epoch": 0.5700005615769079,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004976260709234049,
      "loss": 5.7287,
      "mean_token_accuracy": 0.165048947930336,
      "num_tokens": 11333405.0,
      "step": 5075
    },
    {
      "entropy": 5.838383293151855,
      "epoch": 0.5705621384848655,
      "grad_norm": 1.1875,
      "learning_rate": 0.000497620250624724,
      "loss": 5.7896,
      "mean_token_accuracy": 0.16699133217334747,
      "num_tokens": 11345037.0,
      "step": 5080
    },
    {
      "entropy": 5.788431072235108,
      "epoch": 0.571123715392823,
      "grad_norm": 1.40625,
      "learning_rate": 0.000497614423237728,
      "loss": 5.7095,
      "mean_token_accuracy": 0.16881436556577684,
      "num_tokens": 11356403.0,
      "step": 5085
    },
    {
      "entropy": 5.6463860988616945,
      "epoch": 0.5716852923007806,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004976085887626026,
      "loss": 5.6361,
      "mean_token_accuracy": 0.1750362992286682,
      "num_tokens": 11367913.0,
      "step": 5090
    },
    {
      "entropy": 5.6775815963745115,
      "epoch": 0.5722468692087381,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004976027471995331,
      "loss": 5.6461,
      "mean_token_accuracy": 0.17520785629749297,
      "num_tokens": 11378519.0,
      "step": 5095
    },
    {
      "entropy": 5.759826040267944,
      "epoch": 0.5728084461166957,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000497596898548706,
      "loss": 5.6073,
      "mean_token_accuracy": 0.17434988915920258,
      "num_tokens": 11389530.0,
      "step": 5100
    },
    {
      "entropy": 5.59257664680481,
      "epoch": 0.5733700230246532,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004975910428103072,
      "loss": 5.5348,
      "mean_token_accuracy": 0.17843396216630936,
      "num_tokens": 11400468.0,
      "step": 5105
    },
    {
      "entropy": 5.500880479812622,
      "epoch": 0.5739315999326108,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004975851799845233,
      "loss": 5.544,
      "mean_token_accuracy": 0.1740495353937149,
      "num_tokens": 11411242.0,
      "step": 5110
    },
    {
      "entropy": 5.757795429229736,
      "epoch": 0.5744931768405683,
      "grad_norm": 1.375,
      "learning_rate": 0.000497579310071541,
      "loss": 5.6211,
      "mean_token_accuracy": 0.17112922221422194,
      "num_tokens": 11420909.0,
      "step": 5115
    },
    {
      "entropy": 5.599933767318726,
      "epoch": 0.5750547537485259,
      "grad_norm": 1.375,
      "learning_rate": 0.0004975734330715471,
      "loss": 5.5078,
      "mean_token_accuracy": 0.17252909243106843,
      "num_tokens": 11431791.0,
      "step": 5120
    },
    {
      "entropy": 5.802357625961304,
      "epoch": 0.5756163306564834,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004975675489847287,
      "loss": 5.7397,
      "mean_token_accuracy": 0.16458898186683654,
      "num_tokens": 11443056.0,
      "step": 5125
    },
    {
      "entropy": 5.745362329483032,
      "epoch": 0.576177907564441,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004975616578112734,
      "loss": 5.7048,
      "mean_token_accuracy": 0.17184705585241317,
      "num_tokens": 11453413.0,
      "step": 5130
    },
    {
      "entropy": 5.848228311538696,
      "epoch": 0.5767394844723985,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004975557595513684,
      "loss": 5.9266,
      "mean_token_accuracy": 0.15690554976463317,
      "num_tokens": 11465457.0,
      "step": 5135
    },
    {
      "entropy": 5.914571619033813,
      "epoch": 0.5773010613803561,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004975498542052018,
      "loss": 5.7023,
      "mean_token_accuracy": 0.1729892149567604,
      "num_tokens": 11476899.0,
      "step": 5140
    },
    {
      "entropy": 5.682943296432495,
      "epoch": 0.5778626382883136,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004975439417729615,
      "loss": 5.6611,
      "mean_token_accuracy": 0.17298050820827485,
      "num_tokens": 11488652.0,
      "step": 5145
    },
    {
      "entropy": 5.711069393157959,
      "epoch": 0.5784242151962711,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004975380222548358,
      "loss": 5.7013,
      "mean_token_accuracy": 0.16747490912675858,
      "num_tokens": 11499540.0,
      "step": 5150
    },
    {
      "entropy": 5.864771842956543,
      "epoch": 0.5789857921042286,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004975320956510131,
      "loss": 5.7277,
      "mean_token_accuracy": 0.17383481860160827,
      "num_tokens": 11511156.0,
      "step": 5155
    },
    {
      "entropy": 5.793862915039062,
      "epoch": 0.5795473690121862,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004975261619616822,
      "loss": 5.7969,
      "mean_token_accuracy": 0.16044799983501434,
      "num_tokens": 11523489.0,
      "step": 5160
    },
    {
      "entropy": 5.701024532318115,
      "epoch": 0.5801089459201437,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004975202211870321,
      "loss": 5.62,
      "mean_token_accuracy": 0.16963277608156205,
      "num_tokens": 11534497.0,
      "step": 5165
    },
    {
      "entropy": 5.849947166442871,
      "epoch": 0.5806705228281013,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004975142733272518,
      "loss": 5.7043,
      "mean_token_accuracy": 0.16341470927000046,
      "num_tokens": 11545585.0,
      "step": 5170
    },
    {
      "entropy": 5.594762372970581,
      "epoch": 0.5812320997360588,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004975083183825306,
      "loss": 5.553,
      "mean_token_accuracy": 0.17195095419883727,
      "num_tokens": 11556278.0,
      "step": 5175
    },
    {
      "entropy": 5.634889745712281,
      "epoch": 0.5817936766440164,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004975023563530584,
      "loss": 5.6343,
      "mean_token_accuracy": 0.17483613938093184,
      "num_tokens": 11568358.0,
      "step": 5180
    },
    {
      "entropy": 5.6792529106140135,
      "epoch": 0.5823552535519739,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004974963872390249,
      "loss": 5.6092,
      "mean_token_accuracy": 0.17366641908884048,
      "num_tokens": 11579777.0,
      "step": 5185
    },
    {
      "entropy": 5.678589296340943,
      "epoch": 0.5829168304599315,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00049749041104062,
      "loss": 5.6641,
      "mean_token_accuracy": 0.1732245460152626,
      "num_tokens": 11591479.0,
      "step": 5190
    },
    {
      "entropy": 5.741658306121826,
      "epoch": 0.583478407367889,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004974844277580343,
      "loss": 5.6151,
      "mean_token_accuracy": 0.1689417690038681,
      "num_tokens": 11601452.0,
      "step": 5195
    },
    {
      "entropy": 5.639203977584839,
      "epoch": 0.5840399842758466,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004974784373914579,
      "loss": 5.479,
      "mean_token_accuracy": 0.1778860330581665,
      "num_tokens": 11611661.0,
      "step": 5200
    },
    {
      "entropy": 5.637208557128906,
      "epoch": 0.5846015611838041,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004974724399410818,
      "loss": 5.662,
      "mean_token_accuracy": 0.16758041232824325,
      "num_tokens": 11622940.0,
      "step": 5205
    },
    {
      "entropy": 5.700867700576782,
      "epoch": 0.5851631380917617,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004974664354070971,
      "loss": 5.6744,
      "mean_token_accuracy": 0.17014639526605607,
      "num_tokens": 11634290.0,
      "step": 5210
    },
    {
      "entropy": 5.802171611785889,
      "epoch": 0.5857247149997192,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004974604237896947,
      "loss": 5.7173,
      "mean_token_accuracy": 0.16413109749555588,
      "num_tokens": 11645812.0,
      "step": 5215
    },
    {
      "entropy": 5.730967092514038,
      "epoch": 0.5862862919076768,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004974544050890662,
      "loss": 5.711,
      "mean_token_accuracy": 0.16343657672405243,
      "num_tokens": 11657520.0,
      "step": 5220
    },
    {
      "entropy": 5.6769603252410885,
      "epoch": 0.5868478688156343,
      "grad_norm": 1.328125,
      "learning_rate": 0.000497448379305403,
      "loss": 5.567,
      "mean_token_accuracy": 0.17653037905693053,
      "num_tokens": 11666887.0,
      "step": 5225
    },
    {
      "entropy": 5.796042203903198,
      "epoch": 0.5874094457235919,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004974423464388973,
      "loss": 5.7423,
      "mean_token_accuracy": 0.16953345388174057,
      "num_tokens": 11677337.0,
      "step": 5230
    },
    {
      "entropy": 5.756515645980835,
      "epoch": 0.5879710226315494,
      "grad_norm": 1.453125,
      "learning_rate": 0.000497436306489741,
      "loss": 5.6623,
      "mean_token_accuracy": 0.17309503853321076,
      "num_tokens": 11688203.0,
      "step": 5235
    },
    {
      "entropy": 5.702898836135864,
      "epoch": 0.588532599539507,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004974302594581264,
      "loss": 5.702,
      "mean_token_accuracy": 0.16658136248588562,
      "num_tokens": 11699727.0,
      "step": 5240
    },
    {
      "entropy": 5.732393646240235,
      "epoch": 0.5890941764474645,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004974242053442461,
      "loss": 5.7085,
      "mean_token_accuracy": 0.16990188509225845,
      "num_tokens": 11710956.0,
      "step": 5245
    },
    {
      "entropy": 5.743574380874634,
      "epoch": 0.589655753355422,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004974181441482927,
      "loss": 5.5437,
      "mean_token_accuracy": 0.17698764950037002,
      "num_tokens": 11721681.0,
      "step": 5250
    },
    {
      "entropy": 5.713166904449463,
      "epoch": 0.5902173302633795,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004974120758704595,
      "loss": 5.6719,
      "mean_token_accuracy": 0.16816408634185792,
      "num_tokens": 11731679.0,
      "step": 5255
    },
    {
      "entropy": 5.8048162937164305,
      "epoch": 0.5907789071713371,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004974060005109394,
      "loss": 5.8048,
      "mean_token_accuracy": 0.1657121554017067,
      "num_tokens": 11743953.0,
      "step": 5260
    },
    {
      "entropy": 5.6638165473937985,
      "epoch": 0.5913404840792946,
      "grad_norm": 1.3125,
      "learning_rate": 0.000497399918069926,
      "loss": 5.5795,
      "mean_token_accuracy": 0.17744518518447877,
      "num_tokens": 11754091.0,
      "step": 5265
    },
    {
      "entropy": 5.639192199707031,
      "epoch": 0.5919020609872522,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004973938285476131,
      "loss": 5.6066,
      "mean_token_accuracy": 0.1745060995221138,
      "num_tokens": 11764771.0,
      "step": 5270
    },
    {
      "entropy": 5.734188079833984,
      "epoch": 0.5924636378952097,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004973877319441943,
      "loss": 5.6757,
      "mean_token_accuracy": 0.16847294121980666,
      "num_tokens": 11776287.0,
      "step": 5275
    },
    {
      "entropy": 5.747639417648315,
      "epoch": 0.5930252148031673,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004973816282598639,
      "loss": 5.6861,
      "mean_token_accuracy": 0.16214133352041243,
      "num_tokens": 11787142.0,
      "step": 5280
    },
    {
      "entropy": 5.72074031829834,
      "epoch": 0.5935867917111248,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004973755174948162,
      "loss": 5.6593,
      "mean_token_accuracy": 0.16755084693431854,
      "num_tokens": 11798966.0,
      "step": 5285
    },
    {
      "entropy": 5.771004104614258,
      "epoch": 0.5941483686190824,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004973693996492458,
      "loss": 5.7396,
      "mean_token_accuracy": 0.16607423722743989,
      "num_tokens": 11810529.0,
      "step": 5290
    },
    {
      "entropy": 5.65753321647644,
      "epoch": 0.5947099455270399,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004973632747233475,
      "loss": 5.584,
      "mean_token_accuracy": 0.1776696190237999,
      "num_tokens": 11821965.0,
      "step": 5295
    },
    {
      "entropy": 5.7572873592376705,
      "epoch": 0.5952715224349975,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004973571427173163,
      "loss": 5.7412,
      "mean_token_accuracy": 0.1662243828177452,
      "num_tokens": 11833936.0,
      "step": 5300
    },
    {
      "entropy": 5.793422603607178,
      "epoch": 0.595833099342955,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004973510036313473,
      "loss": 5.6572,
      "mean_token_accuracy": 0.17151601612567902,
      "num_tokens": 11845245.0,
      "step": 5305
    },
    {
      "entropy": 5.701361179351807,
      "epoch": 0.5963946762509126,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004973448574656362,
      "loss": 5.6248,
      "mean_token_accuracy": 0.17118371576070784,
      "num_tokens": 11856010.0,
      "step": 5310
    },
    {
      "entropy": 5.748444557189941,
      "epoch": 0.5969562531588701,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004973387042203786,
      "loss": 5.7801,
      "mean_token_accuracy": 0.16333963572978974,
      "num_tokens": 11866194.0,
      "step": 5315
    },
    {
      "entropy": 5.785281944274902,
      "epoch": 0.5975178300668277,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004973325438957704,
      "loss": 5.6244,
      "mean_token_accuracy": 0.170792356133461,
      "num_tokens": 11877224.0,
      "step": 5320
    },
    {
      "entropy": 5.662514305114746,
      "epoch": 0.5980794069747852,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004973263764920079,
      "loss": 5.5478,
      "mean_token_accuracy": 0.17639639377593994,
      "num_tokens": 11887410.0,
      "step": 5325
    },
    {
      "entropy": 5.627243566513061,
      "epoch": 0.5986409838827428,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004973202020092871,
      "loss": 5.7129,
      "mean_token_accuracy": 0.1726754367351532,
      "num_tokens": 11898816.0,
      "step": 5330
    },
    {
      "entropy": 5.750458002090454,
      "epoch": 0.5992025607907003,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000497314020447805,
      "loss": 5.6623,
      "mean_token_accuracy": 0.16915512531995774,
      "num_tokens": 11909220.0,
      "step": 5335
    },
    {
      "entropy": 5.7871287822723385,
      "epoch": 0.5997641376986579,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004973078318077583,
      "loss": 5.6694,
      "mean_token_accuracy": 0.16608534008264542,
      "num_tokens": 11919584.0,
      "step": 5340
    },
    {
      "entropy": 5.664968776702881,
      "epoch": 0.6003257146066153,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000497301636089344,
      "loss": 5.6782,
      "mean_token_accuracy": 0.16855976581573487,
      "num_tokens": 11929298.0,
      "step": 5345
    },
    {
      "entropy": 5.701796770095825,
      "epoch": 0.6008872915145729,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004972954332927592,
      "loss": 5.6465,
      "mean_token_accuracy": 0.17208643406629562,
      "num_tokens": 11939809.0,
      "step": 5350
    },
    {
      "entropy": 5.727794981002807,
      "epoch": 0.6014488684225304,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004972892234182018,
      "loss": 5.541,
      "mean_token_accuracy": 0.1775716796517372,
      "num_tokens": 11950292.0,
      "step": 5355
    },
    {
      "entropy": 5.610636901855469,
      "epoch": 0.602010445330488,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004972830064658692,
      "loss": 5.568,
      "mean_token_accuracy": 0.17599292993545532,
      "num_tokens": 11961774.0,
      "step": 5360
    },
    {
      "entropy": 5.764124965667724,
      "epoch": 0.6025720222384455,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004972767824359596,
      "loss": 5.6824,
      "mean_token_accuracy": 0.16461727768182755,
      "num_tokens": 11973351.0,
      "step": 5365
    },
    {
      "entropy": 5.664096021652222,
      "epoch": 0.6031335991464031,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004972705513286709,
      "loss": 5.5476,
      "mean_token_accuracy": 0.178704696893692,
      "num_tokens": 11984472.0,
      "step": 5370
    },
    {
      "entropy": 5.647040939331054,
      "epoch": 0.6036951760543606,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004972643131442017,
      "loss": 5.6518,
      "mean_token_accuracy": 0.17527108937501906,
      "num_tokens": 11994271.0,
      "step": 5375
    },
    {
      "entropy": 5.860633420944214,
      "epoch": 0.6042567529623182,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004972580678827506,
      "loss": 5.7576,
      "mean_token_accuracy": 0.16238637864589692,
      "num_tokens": 12007647.0,
      "step": 5380
    },
    {
      "entropy": 5.786860418319702,
      "epoch": 0.6048183298702757,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004972518155445163,
      "loss": 5.7092,
      "mean_token_accuracy": 0.16535102427005768,
      "num_tokens": 12020331.0,
      "step": 5385
    },
    {
      "entropy": 5.657439231872559,
      "epoch": 0.6053799067782333,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004972455561296982,
      "loss": 5.7088,
      "mean_token_accuracy": 0.16050323843955994,
      "num_tokens": 12031616.0,
      "step": 5390
    },
    {
      "entropy": 5.811110305786133,
      "epoch": 0.6059414836861908,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004972392896384952,
      "loss": 5.6699,
      "mean_token_accuracy": 0.1691645324230194,
      "num_tokens": 12044107.0,
      "step": 5395
    },
    {
      "entropy": 5.748564100265503,
      "epoch": 0.6065030605941484,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004972330160711071,
      "loss": 5.6608,
      "mean_token_accuracy": 0.17236095368862153,
      "num_tokens": 12055747.0,
      "step": 5400
    },
    {
      "entropy": 5.757844591140747,
      "epoch": 0.6070646375021059,
      "grad_norm": 1.25,
      "learning_rate": 0.0004972267354277336,
      "loss": 5.6581,
      "mean_token_accuracy": 0.17029646039009094,
      "num_tokens": 12067849.0,
      "step": 5405
    },
    {
      "entropy": 5.722942781448364,
      "epoch": 0.6076262144100635,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004972204477085746,
      "loss": 5.6759,
      "mean_token_accuracy": 0.16894526779651642,
      "num_tokens": 12078534.0,
      "step": 5410
    },
    {
      "entropy": 5.657567405700684,
      "epoch": 0.608187791318021,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004972141529138303,
      "loss": 5.6041,
      "mean_token_accuracy": 0.17073164135217667,
      "num_tokens": 12089534.0,
      "step": 5415
    },
    {
      "entropy": 5.70810079574585,
      "epoch": 0.6087493682259786,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004972078510437013,
      "loss": 5.5066,
      "mean_token_accuracy": 0.18743069767951964,
      "num_tokens": 12099513.0,
      "step": 5420
    },
    {
      "entropy": 5.70379114151001,
      "epoch": 0.6093109451339361,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004972015420983882,
      "loss": 5.712,
      "mean_token_accuracy": 0.17635903805494307,
      "num_tokens": 12110106.0,
      "step": 5425
    },
    {
      "entropy": 5.705994033813477,
      "epoch": 0.6098725220418937,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004971952260780915,
      "loss": 5.604,
      "mean_token_accuracy": 0.1767367586493492,
      "num_tokens": 12121302.0,
      "step": 5430
    },
    {
      "entropy": 5.692618274688721,
      "epoch": 0.6104340989498512,
      "grad_norm": 1.328125,
      "learning_rate": 0.000497188902983013,
      "loss": 5.6502,
      "mean_token_accuracy": 0.17055889666080476,
      "num_tokens": 12132879.0,
      "step": 5435
    },
    {
      "entropy": 5.642355585098267,
      "epoch": 0.6109956758578087,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004971825728133535,
      "loss": 5.5734,
      "mean_token_accuracy": 0.17844846993684768,
      "num_tokens": 12143597.0,
      "step": 5440
    },
    {
      "entropy": 5.729561996459961,
      "epoch": 0.6115572527657662,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004971762355693146,
      "loss": 5.6705,
      "mean_token_accuracy": 0.17112083733081818,
      "num_tokens": 12156036.0,
      "step": 5445
    },
    {
      "entropy": 5.774557399749756,
      "epoch": 0.6121188296737238,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004971698912510983,
      "loss": 5.6849,
      "mean_token_accuracy": 0.1741700991988182,
      "num_tokens": 12167305.0,
      "step": 5450
    },
    {
      "entropy": 5.718290376663208,
      "epoch": 0.6126804065816813,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004971635398589066,
      "loss": 5.6458,
      "mean_token_accuracy": 0.17421892285346985,
      "num_tokens": 12176772.0,
      "step": 5455
    },
    {
      "entropy": 5.706922245025635,
      "epoch": 0.6132419834896389,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004971571813929415,
      "loss": 5.6452,
      "mean_token_accuracy": 0.17209067046642304,
      "num_tokens": 12187852.0,
      "step": 5460
    },
    {
      "entropy": 5.790262699127197,
      "epoch": 0.6138035603975964,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004971508158534056,
      "loss": 5.7057,
      "mean_token_accuracy": 0.17010461091995238,
      "num_tokens": 12198574.0,
      "step": 5465
    },
    {
      "entropy": 5.622075748443604,
      "epoch": 0.614365137305554,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004971444432405015,
      "loss": 5.4607,
      "mean_token_accuracy": 0.18198293149471284,
      "num_tokens": 12209463.0,
      "step": 5470
    },
    {
      "entropy": 5.731533527374268,
      "epoch": 0.6149267142135115,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004971380635544323,
      "loss": 5.7366,
      "mean_token_accuracy": 0.16922779083251954,
      "num_tokens": 12220993.0,
      "step": 5475
    },
    {
      "entropy": 5.730971240997315,
      "epoch": 0.6154882911214691,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004971316767954009,
      "loss": 5.6102,
      "mean_token_accuracy": 0.17507601529359818,
      "num_tokens": 12232648.0,
      "step": 5480
    },
    {
      "entropy": 5.736530876159668,
      "epoch": 0.6160498680294266,
      "grad_norm": 1.625,
      "learning_rate": 0.0004971252829636108,
      "loss": 5.65,
      "mean_token_accuracy": 0.17610020488500594,
      "num_tokens": 12243931.0,
      "step": 5485
    },
    {
      "entropy": 5.787777423858643,
      "epoch": 0.6166114449373842,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004971188820592656,
      "loss": 5.6509,
      "mean_token_accuracy": 0.17270122319459916,
      "num_tokens": 12254821.0,
      "step": 5490
    },
    {
      "entropy": 5.583421754837036,
      "epoch": 0.6171730218453417,
      "grad_norm": 1.203125,
      "learning_rate": 0.000497112474082569,
      "loss": 5.6572,
      "mean_token_accuracy": 0.17220793664455414,
      "num_tokens": 12265440.0,
      "step": 5495
    },
    {
      "entropy": 5.727280187606811,
      "epoch": 0.6177345987532993,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000497106059033725,
      "loss": 5.6329,
      "mean_token_accuracy": 0.17365415692329406,
      "num_tokens": 12276636.0,
      "step": 5500
    },
    {
      "entropy": 5.766627502441406,
      "epoch": 0.6182961756612568,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004970996369129381,
      "loss": 5.5133,
      "mean_token_accuracy": 0.17947188764810562,
      "num_tokens": 12287404.0,
      "step": 5505
    },
    {
      "entropy": 5.672186088562012,
      "epoch": 0.6188577525692144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004970932077204125,
      "loss": 5.7397,
      "mean_token_accuracy": 0.1642993837594986,
      "num_tokens": 12299283.0,
      "step": 5510
    },
    {
      "entropy": 5.666852855682373,
      "epoch": 0.6194193294771719,
      "grad_norm": 1.296875,
      "learning_rate": 0.000497086771456353,
      "loss": 5.6404,
      "mean_token_accuracy": 0.16676075011491776,
      "num_tokens": 12310281.0,
      "step": 5515
    },
    {
      "entropy": 5.804997062683105,
      "epoch": 0.6199809063851295,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004970803281209646,
      "loss": 5.6661,
      "mean_token_accuracy": 0.17195305079221726,
      "num_tokens": 12320839.0,
      "step": 5520
    },
    {
      "entropy": 5.676700162887573,
      "epoch": 0.620542483293087,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004970738777144524,
      "loss": 5.6211,
      "mean_token_accuracy": 0.17245084047317505,
      "num_tokens": 12332845.0,
      "step": 5525
    },
    {
      "entropy": 5.554922676086425,
      "epoch": 0.6211040602010446,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004970674202370218,
      "loss": 5.5204,
      "mean_token_accuracy": 0.1806936517357826,
      "num_tokens": 12343327.0,
      "step": 5530
    },
    {
      "entropy": 5.72095046043396,
      "epoch": 0.621665637109002,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004970609556888785,
      "loss": 5.6445,
      "mean_token_accuracy": 0.17047364860773087,
      "num_tokens": 12353224.0,
      "step": 5535
    },
    {
      "entropy": 5.682199335098266,
      "epoch": 0.6222272140169596,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004970544840702282,
      "loss": 5.5634,
      "mean_token_accuracy": 0.17255714684724807,
      "num_tokens": 12363981.0,
      "step": 5540
    },
    {
      "entropy": 5.584892797470093,
      "epoch": 0.6227887909249171,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004970480053812771,
      "loss": 5.5853,
      "mean_token_accuracy": 0.1800747200846672,
      "num_tokens": 12374769.0,
      "step": 5545
    },
    {
      "entropy": 5.814059019088745,
      "epoch": 0.6233503678328747,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004970415196222312,
      "loss": 5.8038,
      "mean_token_accuracy": 0.16921334117650985,
      "num_tokens": 12387247.0,
      "step": 5550
    },
    {
      "entropy": 5.793169641494751,
      "epoch": 0.6239119447408322,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004970350267932973,
      "loss": 5.6566,
      "mean_token_accuracy": 0.17780971080064772,
      "num_tokens": 12398347.0,
      "step": 5555
    },
    {
      "entropy": 5.531301593780517,
      "epoch": 0.6244735216487898,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000497028526894682,
      "loss": 5.5502,
      "mean_token_accuracy": 0.17427618205547332,
      "num_tokens": 12409763.0,
      "step": 5560
    },
    {
      "entropy": 5.740422582626342,
      "epoch": 0.6250350985567473,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004970220199265923,
      "loss": 5.6429,
      "mean_token_accuracy": 0.17239355593919753,
      "num_tokens": 12420658.0,
      "step": 5565
    },
    {
      "entropy": 5.830804109573364,
      "epoch": 0.6255966754647049,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004970155058892354,
      "loss": 5.7334,
      "mean_token_accuracy": 0.16390733271837235,
      "num_tokens": 12433959.0,
      "step": 5570
    },
    {
      "entropy": 5.738416051864624,
      "epoch": 0.6261582523726624,
      "grad_norm": 1.5,
      "learning_rate": 0.0004970089847828187,
      "loss": 5.7218,
      "mean_token_accuracy": 0.16629742383956908,
      "num_tokens": 12444756.0,
      "step": 5575
    },
    {
      "entropy": 5.73380479812622,
      "epoch": 0.62671982928062,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004970024566075497,
      "loss": 5.6352,
      "mean_token_accuracy": 0.17590337246656418,
      "num_tokens": 12456114.0,
      "step": 5580
    },
    {
      "entropy": 5.736455249786377,
      "epoch": 0.6272814061885775,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004969959213636365,
      "loss": 5.6718,
      "mean_token_accuracy": 0.17344282269477845,
      "num_tokens": 12466927.0,
      "step": 5585
    },
    {
      "entropy": 5.701116847991943,
      "epoch": 0.6278429830965351,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004969893790512871,
      "loss": 5.723,
      "mean_token_accuracy": 0.16070540398359298,
      "num_tokens": 12477834.0,
      "step": 5590
    },
    {
      "entropy": 5.750099849700928,
      "epoch": 0.6284045600044926,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004969828296707097,
      "loss": 5.6247,
      "mean_token_accuracy": 0.1745440736413002,
      "num_tokens": 12488219.0,
      "step": 5595
    },
    {
      "entropy": 5.8359832763671875,
      "epoch": 0.6289661369124502,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004969762732221129,
      "loss": 5.7385,
      "mean_token_accuracy": 0.17294516116380693,
      "num_tokens": 12498947.0,
      "step": 5600
    },
    {
      "entropy": 5.66269965171814,
      "epoch": 0.6295277138204077,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004969697097057054,
      "loss": 5.6343,
      "mean_token_accuracy": 0.1720503583550453,
      "num_tokens": 12510142.0,
      "step": 5605
    },
    {
      "entropy": 5.79200439453125,
      "epoch": 0.6300892907283653,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004969631391216964,
      "loss": 5.6472,
      "mean_token_accuracy": 0.1704936742782593,
      "num_tokens": 12521350.0,
      "step": 5610
    },
    {
      "entropy": 5.521368408203125,
      "epoch": 0.6306508676363228,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004969565614702948,
      "loss": 5.4855,
      "mean_token_accuracy": 0.1774906650185585,
      "num_tokens": 12532579.0,
      "step": 5615
    },
    {
      "entropy": 5.734045362472534,
      "epoch": 0.6312124445442804,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004969499767517104,
      "loss": 5.6044,
      "mean_token_accuracy": 0.17279993891716003,
      "num_tokens": 12542885.0,
      "step": 5620
    },
    {
      "entropy": 5.80007586479187,
      "epoch": 0.6317740214522379,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004969433849661525,
      "loss": 5.7554,
      "mean_token_accuracy": 0.1622144266963005,
      "num_tokens": 12554881.0,
      "step": 5625
    },
    {
      "entropy": 5.737228202819824,
      "epoch": 0.6323355983601955,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004969367861138313,
      "loss": 5.6993,
      "mean_token_accuracy": 0.16859257966279984,
      "num_tokens": 12564626.0,
      "step": 5630
    },
    {
      "entropy": 5.745629692077637,
      "epoch": 0.6328971752681529,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004969301801949565,
      "loss": 5.5502,
      "mean_token_accuracy": 0.17777176648378373,
      "num_tokens": 12576180.0,
      "step": 5635
    },
    {
      "entropy": 5.658581924438477,
      "epoch": 0.6334587521761105,
      "grad_norm": 1.234375,
      "learning_rate": 0.000496923567209739,
      "loss": 5.6427,
      "mean_token_accuracy": 0.16927312016487123,
      "num_tokens": 12587487.0,
      "step": 5640
    },
    {
      "entropy": 5.788752126693725,
      "epoch": 0.634020329084068,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004969169471583889,
      "loss": 5.7908,
      "mean_token_accuracy": 0.16504471004009247,
      "num_tokens": 12598268.0,
      "step": 5645
    },
    {
      "entropy": 5.807561635971069,
      "epoch": 0.6345819059920256,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004969103200411171,
      "loss": 5.6686,
      "mean_token_accuracy": 0.1691132053732872,
      "num_tokens": 12609287.0,
      "step": 5650
    },
    {
      "entropy": 5.669478559494019,
      "epoch": 0.6351434828999831,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004969036858581346,
      "loss": 5.5726,
      "mean_token_accuracy": 0.17225826978683473,
      "num_tokens": 12620317.0,
      "step": 5655
    },
    {
      "entropy": 5.70239109992981,
      "epoch": 0.6357050598079407,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004968970446096527,
      "loss": 5.6395,
      "mean_token_accuracy": 0.17853050231933593,
      "num_tokens": 12631420.0,
      "step": 5660
    },
    {
      "entropy": 5.694774198532104,
      "epoch": 0.6362666367158982,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000496890396295883,
      "loss": 5.5798,
      "mean_token_accuracy": 0.1807517871260643,
      "num_tokens": 12641319.0,
      "step": 5665
    },
    {
      "entropy": 5.760483884811402,
      "epoch": 0.6368282136238558,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004968837409170368,
      "loss": 5.6805,
      "mean_token_accuracy": 0.1660526767373085,
      "num_tokens": 12651662.0,
      "step": 5670
    },
    {
      "entropy": 5.768562316894531,
      "epoch": 0.6373897905318133,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004968770784733263,
      "loss": 5.7284,
      "mean_token_accuracy": 0.1657192662358284,
      "num_tokens": 12663488.0,
      "step": 5675
    },
    {
      "entropy": 5.687350702285767,
      "epoch": 0.6379513674397709,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004968704089649636,
      "loss": 5.6362,
      "mean_token_accuracy": 0.1720639333128929,
      "num_tokens": 12675848.0,
      "step": 5680
    },
    {
      "entropy": 5.735997581481934,
      "epoch": 0.6385129443477284,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004968637323921611,
      "loss": 5.6178,
      "mean_token_accuracy": 0.1734489381313324,
      "num_tokens": 12686541.0,
      "step": 5685
    },
    {
      "entropy": 5.710645627975464,
      "epoch": 0.639074521255686,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004968570487551312,
      "loss": 5.6499,
      "mean_token_accuracy": 0.17180876433849335,
      "num_tokens": 12698001.0,
      "step": 5690
    },
    {
      "entropy": 5.720677328109741,
      "epoch": 0.6396360981636435,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004968503580540869,
      "loss": 5.6477,
      "mean_token_accuracy": 0.1650005429983139,
      "num_tokens": 12709069.0,
      "step": 5695
    },
    {
      "entropy": 5.759660863876343,
      "epoch": 0.6401976750716011,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004968436602892411,
      "loss": 5.6475,
      "mean_token_accuracy": 0.16808155924081802,
      "num_tokens": 12719916.0,
      "step": 5700
    },
    {
      "entropy": 5.743867063522339,
      "epoch": 0.6407592519795586,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004968369554608071,
      "loss": 5.7584,
      "mean_token_accuracy": 0.16313536167144777,
      "num_tokens": 12732079.0,
      "step": 5705
    },
    {
      "entropy": 5.852132272720337,
      "epoch": 0.6413208288875162,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004968302435689985,
      "loss": 5.7338,
      "mean_token_accuracy": 0.16560664474964143,
      "num_tokens": 12742708.0,
      "step": 5710
    },
    {
      "entropy": 5.792535924911499,
      "epoch": 0.6418824057954737,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004968235246140288,
      "loss": 5.6781,
      "mean_token_accuracy": 0.1690981075167656,
      "num_tokens": 12753952.0,
      "step": 5715
    },
    {
      "entropy": 5.677698373794556,
      "epoch": 0.6424439827034313,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004968167985961121,
      "loss": 5.6078,
      "mean_token_accuracy": 0.17256635576486587,
      "num_tokens": 12765226.0,
      "step": 5720
    },
    {
      "entropy": 5.602059602737427,
      "epoch": 0.6430055596113888,
      "grad_norm": 1.5,
      "learning_rate": 0.0004968100655154625,
      "loss": 5.5366,
      "mean_token_accuracy": 0.17826073467731476,
      "num_tokens": 12776455.0,
      "step": 5725
    },
    {
      "entropy": 5.672917366027832,
      "epoch": 0.6435671365193463,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004968033253722943,
      "loss": 5.5564,
      "mean_token_accuracy": 0.17404053062200547,
      "num_tokens": 12787054.0,
      "step": 5730
    },
    {
      "entropy": 5.719582462310791,
      "epoch": 0.6441287134273038,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004967965781668223,
      "loss": 5.6866,
      "mean_token_accuracy": 0.170250004529953,
      "num_tokens": 12798223.0,
      "step": 5735
    },
    {
      "entropy": 5.715972423553467,
      "epoch": 0.6446902903352614,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004967898238992613,
      "loss": 5.5732,
      "mean_token_accuracy": 0.17579154819250106,
      "num_tokens": 12809388.0,
      "step": 5740
    },
    {
      "entropy": 5.686679077148438,
      "epoch": 0.6452518672432189,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004967830625698262,
      "loss": 5.6769,
      "mean_token_accuracy": 0.17444596737623214,
      "num_tokens": 12820116.0,
      "step": 5745
    },
    {
      "entropy": 5.6671665668487545,
      "epoch": 0.6458134441511765,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004967762941787323,
      "loss": 5.5967,
      "mean_token_accuracy": 0.17919077277183532,
      "num_tokens": 12831590.0,
      "step": 5750
    },
    {
      "entropy": 5.747342586517334,
      "epoch": 0.646375021059134,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004967695187261952,
      "loss": 5.7564,
      "mean_token_accuracy": 0.16459435522556304,
      "num_tokens": 12843510.0,
      "step": 5755
    },
    {
      "entropy": 5.7117125511169435,
      "epoch": 0.6469365979670916,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004967627362124306,
      "loss": 5.6307,
      "mean_token_accuracy": 0.1705598846077919,
      "num_tokens": 12854907.0,
      "step": 5760
    },
    {
      "entropy": 5.658813190460205,
      "epoch": 0.6474981748750491,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004967559466376546,
      "loss": 5.5959,
      "mean_token_accuracy": 0.1710912510752678,
      "num_tokens": 12865753.0,
      "step": 5765
    },
    {
      "entropy": 5.6811473846435545,
      "epoch": 0.6480597517830067,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004967491500020831,
      "loss": 5.5035,
      "mean_token_accuracy": 0.17676436007022858,
      "num_tokens": 12877116.0,
      "step": 5770
    },
    {
      "entropy": 5.5508228778839115,
      "epoch": 0.6486213286909642,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004967423463059328,
      "loss": 5.4959,
      "mean_token_accuracy": 0.17579397261142732,
      "num_tokens": 12887685.0,
      "step": 5775
    },
    {
      "entropy": 5.713850355148315,
      "epoch": 0.6491829055989218,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004967355355494201,
      "loss": 5.6584,
      "mean_token_accuracy": 0.1692407101392746,
      "num_tokens": 12898519.0,
      "step": 5780
    },
    {
      "entropy": 5.742898797988891,
      "epoch": 0.6497444825068793,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000496728717732762,
      "loss": 5.586,
      "mean_token_accuracy": 0.17314111143350602,
      "num_tokens": 12908400.0,
      "step": 5785
    },
    {
      "entropy": 5.602768039703369,
      "epoch": 0.6503060594148369,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004967218928561755,
      "loss": 5.56,
      "mean_token_accuracy": 0.17106268554925919,
      "num_tokens": 12919159.0,
      "step": 5790
    },
    {
      "entropy": 5.631039237976074,
      "epoch": 0.6508676363227944,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004967150609198779,
      "loss": 5.523,
      "mean_token_accuracy": 0.1793055310845375,
      "num_tokens": 12929838.0,
      "step": 5795
    },
    {
      "entropy": 5.732568550109863,
      "epoch": 0.651429213230752,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004967082219240869,
      "loss": 5.6285,
      "mean_token_accuracy": 0.1695741206407547,
      "num_tokens": 12939976.0,
      "step": 5800
    },
    {
      "entropy": 5.758488702774048,
      "epoch": 0.6519907901387095,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004967013758690199,
      "loss": 5.6404,
      "mean_token_accuracy": 0.1749062106013298,
      "num_tokens": 12950493.0,
      "step": 5805
    },
    {
      "entropy": 5.654942989349365,
      "epoch": 0.6525523670466671,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004966945227548953,
      "loss": 5.6052,
      "mean_token_accuracy": 0.17201246023178102,
      "num_tokens": 12961267.0,
      "step": 5810
    },
    {
      "entropy": 5.66741909980774,
      "epoch": 0.6531139439546246,
      "grad_norm": 1.25,
      "learning_rate": 0.0004966876625819309,
      "loss": 5.6493,
      "mean_token_accuracy": 0.16587674468755723,
      "num_tokens": 12972720.0,
      "step": 5815
    },
    {
      "entropy": 5.8147331237792965,
      "epoch": 0.6536755208625822,
      "grad_norm": 1.375,
      "learning_rate": 0.0004966807953503456,
      "loss": 5.7294,
      "mean_token_accuracy": 0.16530007719993592,
      "num_tokens": 12984025.0,
      "step": 5820
    },
    {
      "entropy": 5.751352405548095,
      "epoch": 0.6542370977705396,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004966739210603578,
      "loss": 5.6064,
      "mean_token_accuracy": 0.17820507735013963,
      "num_tokens": 12994422.0,
      "step": 5825
    },
    {
      "entropy": 5.753329849243164,
      "epoch": 0.6547986746784972,
      "grad_norm": 1.5,
      "learning_rate": 0.0004966670397121862,
      "loss": 5.7651,
      "mean_token_accuracy": 0.1632770538330078,
      "num_tokens": 13005040.0,
      "step": 5830
    },
    {
      "entropy": 5.65441665649414,
      "epoch": 0.6553602515864547,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004966601513060502,
      "loss": 5.5259,
      "mean_token_accuracy": 0.17237963378429413,
      "num_tokens": 13015720.0,
      "step": 5835
    },
    {
      "entropy": 5.689071893692017,
      "epoch": 0.6559218284944123,
      "grad_norm": 1.25,
      "learning_rate": 0.000496653255842169,
      "loss": 5.4925,
      "mean_token_accuracy": 0.17613681703805922,
      "num_tokens": 13026269.0,
      "step": 5840
    },
    {
      "entropy": 5.638681697845459,
      "epoch": 0.6564834054023698,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004966463533207621,
      "loss": 5.6659,
      "mean_token_accuracy": 0.16914643421769143,
      "num_tokens": 13037102.0,
      "step": 5845
    },
    {
      "entropy": 5.634537220001221,
      "epoch": 0.6570449823103274,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004966394437420493,
      "loss": 5.5301,
      "mean_token_accuracy": 0.1786402016878128,
      "num_tokens": 13047149.0,
      "step": 5850
    },
    {
      "entropy": 5.6855401515960695,
      "epoch": 0.6576065592182849,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004966325271062508,
      "loss": 5.5626,
      "mean_token_accuracy": 0.17522660344839097,
      "num_tokens": 13058196.0,
      "step": 5855
    },
    {
      "entropy": 5.617813682556152,
      "epoch": 0.6581681361262425,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004966256034135866,
      "loss": 5.575,
      "mean_token_accuracy": 0.17152782380580903,
      "num_tokens": 13067897.0,
      "step": 5860
    },
    {
      "entropy": 5.704078626632691,
      "epoch": 0.6587297130342,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004966186726642772,
      "loss": 5.6755,
      "mean_token_accuracy": 0.1717715382575989,
      "num_tokens": 13079720.0,
      "step": 5865
    },
    {
      "entropy": 5.634116268157959,
      "epoch": 0.6592912899421576,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004966117348585434,
      "loss": 5.5258,
      "mean_token_accuracy": 0.167197148501873,
      "num_tokens": 13090995.0,
      "step": 5870
    },
    {
      "entropy": 5.605711269378662,
      "epoch": 0.6598528668501151,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004966047899966059,
      "loss": 5.572,
      "mean_token_accuracy": 0.17753074020147325,
      "num_tokens": 13101945.0,
      "step": 5875
    },
    {
      "entropy": 5.679044914245606,
      "epoch": 0.6604144437580727,
      "grad_norm": 1.3125,
      "learning_rate": 0.000496597838078686,
      "loss": 5.5155,
      "mean_token_accuracy": 0.1752203419804573,
      "num_tokens": 13112869.0,
      "step": 5880
    },
    {
      "entropy": 5.710168552398682,
      "epoch": 0.6609760206660302,
      "grad_norm": 1.390625,
      "learning_rate": 0.000496590879105005,
      "loss": 5.6074,
      "mean_token_accuracy": 0.16586030423641204,
      "num_tokens": 13123242.0,
      "step": 5885
    },
    {
      "entropy": 5.776198816299439,
      "epoch": 0.6615375975739878,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004965839130757845,
      "loss": 5.7182,
      "mean_token_accuracy": 0.1584482118487358,
      "num_tokens": 13133981.0,
      "step": 5890
    },
    {
      "entropy": 5.67058835029602,
      "epoch": 0.6620991744819453,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004965769399912462,
      "loss": 5.6804,
      "mean_token_accuracy": 0.16925962790846824,
      "num_tokens": 13144696.0,
      "step": 5895
    },
    {
      "entropy": 5.770441007614136,
      "epoch": 0.6626607513899029,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004965699598516122,
      "loss": 5.7219,
      "mean_token_accuracy": 0.17378196567296983,
      "num_tokens": 13156674.0,
      "step": 5900
    },
    {
      "entropy": 5.704008340835571,
      "epoch": 0.6632223282978604,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004965629726571047,
      "loss": 5.5632,
      "mean_token_accuracy": 0.17381847351789476,
      "num_tokens": 13167757.0,
      "step": 5905
    },
    {
      "entropy": 5.748448419570923,
      "epoch": 0.663783905205818,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004965559784079463,
      "loss": 5.6488,
      "mean_token_accuracy": 0.16911055743694306,
      "num_tokens": 13177701.0,
      "step": 5910
    },
    {
      "entropy": 5.785614109039306,
      "epoch": 0.6643454821137755,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004965489771043596,
      "loss": 5.6658,
      "mean_token_accuracy": 0.16642796993255615,
      "num_tokens": 13189840.0,
      "step": 5915
    },
    {
      "entropy": 5.5934004306793215,
      "epoch": 0.664907059021733,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004965419687465676,
      "loss": 5.4739,
      "mean_token_accuracy": 0.17768412381410598,
      "num_tokens": 13200206.0,
      "step": 5920
    },
    {
      "entropy": 5.586026859283447,
      "epoch": 0.6654686359296905,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004965349533347934,
      "loss": 5.588,
      "mean_token_accuracy": 0.17564436197280883,
      "num_tokens": 13211474.0,
      "step": 5925
    },
    {
      "entropy": 5.733428430557251,
      "epoch": 0.6660302128376481,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004965279308692602,
      "loss": 5.6484,
      "mean_token_accuracy": 0.1843220919370651,
      "num_tokens": 13223337.0,
      "step": 5930
    },
    {
      "entropy": 5.801119422912597,
      "epoch": 0.6665917897456056,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004965209013501919,
      "loss": 5.7285,
      "mean_token_accuracy": 0.17110735177993774,
      "num_tokens": 13234977.0,
      "step": 5935
    },
    {
      "entropy": 5.70616512298584,
      "epoch": 0.6671533666535632,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004965138647778122,
      "loss": 5.6768,
      "mean_token_accuracy": 0.17690228521823884,
      "num_tokens": 13245315.0,
      "step": 5940
    },
    {
      "entropy": 5.685105466842652,
      "epoch": 0.6677149435615207,
      "grad_norm": 1.453125,
      "learning_rate": 0.000496506821152345,
      "loss": 5.5582,
      "mean_token_accuracy": 0.1749824121594429,
      "num_tokens": 13255829.0,
      "step": 5945
    },
    {
      "entropy": 5.66689453125,
      "epoch": 0.6682765204694783,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004964997704740147,
      "loss": 5.6256,
      "mean_token_accuracy": 0.170874160528183,
      "num_tokens": 13266964.0,
      "step": 5950
    },
    {
      "entropy": 5.796186685562134,
      "epoch": 0.6688380973774358,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000496492712743046,
      "loss": 5.6614,
      "mean_token_accuracy": 0.1689567595720291,
      "num_tokens": 13279416.0,
      "step": 5955
    },
    {
      "entropy": 5.723680639266968,
      "epoch": 0.6693996742853934,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004964856479596632,
      "loss": 5.6919,
      "mean_token_accuracy": 0.1676722764968872,
      "num_tokens": 13290633.0,
      "step": 5960
    },
    {
      "entropy": 5.7067337989807125,
      "epoch": 0.6699612511933509,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004964785761240915,
      "loss": 5.5734,
      "mean_token_accuracy": 0.17108363509178162,
      "num_tokens": 13302236.0,
      "step": 5965
    },
    {
      "entropy": 5.7176580905914305,
      "epoch": 0.6705228281013085,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004964714972365562,
      "loss": 5.6724,
      "mean_token_accuracy": 0.17194082885980605,
      "num_tokens": 13315356.0,
      "step": 5970
    },
    {
      "entropy": 5.516757583618164,
      "epoch": 0.671084405009266,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004964644112972824,
      "loss": 5.5327,
      "mean_token_accuracy": 0.1755928486585617,
      "num_tokens": 13326037.0,
      "step": 5975
    },
    {
      "entropy": 5.632549905776978,
      "epoch": 0.6716459819172236,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004964573183064959,
      "loss": 5.5914,
      "mean_token_accuracy": 0.17360655069351197,
      "num_tokens": 13336278.0,
      "step": 5980
    },
    {
      "entropy": 5.736466360092163,
      "epoch": 0.6722075588251811,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004964502182644225,
      "loss": 5.7065,
      "mean_token_accuracy": 0.17162298858165742,
      "num_tokens": 13348147.0,
      "step": 5985
    },
    {
      "entropy": 5.71574764251709,
      "epoch": 0.6727691357331387,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004964431111712883,
      "loss": 5.6413,
      "mean_token_accuracy": 0.17189154624938965,
      "num_tokens": 13360181.0,
      "step": 5990
    },
    {
      "entropy": 5.774587678909302,
      "epoch": 0.6733307126410962,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004964359970273195,
      "loss": 5.7266,
      "mean_token_accuracy": 0.16633295267820358,
      "num_tokens": 13372258.0,
      "step": 5995
    },
    {
      "entropy": 5.71021146774292,
      "epoch": 0.6738922895490538,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004964288758327426,
      "loss": 5.6378,
      "mean_token_accuracy": 0.16402822285890578,
      "num_tokens": 13384332.0,
      "step": 6000
    },
    {
      "epoch": 0.6738922895490538,
      "eval_entropy": 5.5617586172269675,
      "eval_loss": 5.65424108505249,
      "eval_mean_token_accuracy": 0.17736698931749925,
      "eval_num_tokens": 13384332.0,
      "eval_runtime": 23.6897,
      "eval_samples_per_second": 1309.134,
      "eval_steps_per_second": 163.658,
      "step": 6000
    },
    {
      "entropy": 5.7171698093414305,
      "epoch": 0.6744538664570113,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004964217475877846,
      "loss": 5.6314,
      "mean_token_accuracy": 0.16942399442195893,
      "num_tokens": 13396345.0,
      "step": 6005
    },
    {
      "entropy": 5.591657829284668,
      "epoch": 0.6750154433649689,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000496414612292672,
      "loss": 5.4644,
      "mean_token_accuracy": 0.17689456790685654,
      "num_tokens": 13407325.0,
      "step": 6010
    },
    {
      "entropy": 5.689947080612183,
      "epoch": 0.6755770202729264,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004964074699476325,
      "loss": 5.6021,
      "mean_token_accuracy": 0.17259112149477004,
      "num_tokens": 13417476.0,
      "step": 6015
    },
    {
      "entropy": 5.656712484359741,
      "epoch": 0.6761385971808839,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004964003205528933,
      "loss": 5.6096,
      "mean_token_accuracy": 0.1726845309138298,
      "num_tokens": 13428792.0,
      "step": 6020
    },
    {
      "entropy": 5.700129842758178,
      "epoch": 0.6767001740888414,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004963931641086819,
      "loss": 5.6682,
      "mean_token_accuracy": 0.1713205248117447,
      "num_tokens": 13440043.0,
      "step": 6025
    },
    {
      "entropy": 5.7401378631591795,
      "epoch": 0.677261750996799,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004963860006152262,
      "loss": 5.6353,
      "mean_token_accuracy": 0.17090512961149215,
      "num_tokens": 13451637.0,
      "step": 6030
    },
    {
      "entropy": 5.824400568008423,
      "epoch": 0.6778233279047565,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004963788300727546,
      "loss": 5.6046,
      "mean_token_accuracy": 0.17313029766082763,
      "num_tokens": 13463205.0,
      "step": 6035
    },
    {
      "entropy": 5.660120248794556,
      "epoch": 0.6783849048127141,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004963716524814951,
      "loss": 5.6569,
      "mean_token_accuracy": 0.17166511565446854,
      "num_tokens": 13474201.0,
      "step": 6040
    },
    {
      "entropy": 5.62141170501709,
      "epoch": 0.6789464817206716,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004963644678416763,
      "loss": 5.5434,
      "mean_token_accuracy": 0.17378048449754716,
      "num_tokens": 13485439.0,
      "step": 6045
    },
    {
      "entropy": 5.730491065979004,
      "epoch": 0.6795080586286292,
      "grad_norm": 1.1875,
      "learning_rate": 0.000496357276153527,
      "loss": 5.7492,
      "mean_token_accuracy": 0.16852689236402513,
      "num_tokens": 13497257.0,
      "step": 6050
    },
    {
      "entropy": 5.730785655975342,
      "epoch": 0.6800696355365867,
      "grad_norm": 1.125,
      "learning_rate": 0.0004963500774172763,
      "loss": 5.5958,
      "mean_token_accuracy": 0.17269592732191086,
      "num_tokens": 13508689.0,
      "step": 6055
    },
    {
      "entropy": 5.698699808120727,
      "epoch": 0.6806312124445443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000496342871633153,
      "loss": 5.5497,
      "mean_token_accuracy": 0.18048133552074433,
      "num_tokens": 13519392.0,
      "step": 6060
    },
    {
      "entropy": 5.62058744430542,
      "epoch": 0.6811927893525018,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004963356588013871,
      "loss": 5.605,
      "mean_token_accuracy": 0.1779961258172989,
      "num_tokens": 13530503.0,
      "step": 6065
    },
    {
      "entropy": 5.707237100601196,
      "epoch": 0.6817543662604594,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004963284389222078,
      "loss": 5.674,
      "mean_token_accuracy": 0.1740332290530205,
      "num_tokens": 13541668.0,
      "step": 6070
    },
    {
      "entropy": 5.699452447891235,
      "epoch": 0.6823159431684169,
      "grad_norm": 1.25,
      "learning_rate": 0.0004963212119958453,
      "loss": 5.5067,
      "mean_token_accuracy": 0.1787132978439331,
      "num_tokens": 13552692.0,
      "step": 6075
    },
    {
      "entropy": 5.577724456787109,
      "epoch": 0.6828775200763745,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004963139780225295,
      "loss": 5.6074,
      "mean_token_accuracy": 0.1706698417663574,
      "num_tokens": 13563813.0,
      "step": 6080
    },
    {
      "entropy": 5.6306346416473385,
      "epoch": 0.683439096984332,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004963067370024909,
      "loss": 5.5537,
      "mean_token_accuracy": 0.17718004286289216,
      "num_tokens": 13573818.0,
      "step": 6085
    },
    {
      "entropy": 5.760622024536133,
      "epoch": 0.6840006738922896,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004962994889359598,
      "loss": 5.6771,
      "mean_token_accuracy": 0.16787791550159453,
      "num_tokens": 13585416.0,
      "step": 6090
    },
    {
      "entropy": 5.631905651092529,
      "epoch": 0.6845622508002471,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004962922338231673,
      "loss": 5.5727,
      "mean_token_accuracy": 0.17555365860462188,
      "num_tokens": 13596645.0,
      "step": 6095
    },
    {
      "entropy": 5.68451681137085,
      "epoch": 0.6851238277082047,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004962849716643441,
      "loss": 5.5218,
      "mean_token_accuracy": 0.177066071331501,
      "num_tokens": 13607206.0,
      "step": 6100
    },
    {
      "entropy": 5.7633359909057615,
      "epoch": 0.6856854046161622,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004962777024597216,
      "loss": 5.7316,
      "mean_token_accuracy": 0.16855940222740173,
      "num_tokens": 13619418.0,
      "step": 6105
    },
    {
      "entropy": 5.643983459472656,
      "epoch": 0.6862469815241198,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004962704262095314,
      "loss": 5.5224,
      "mean_token_accuracy": 0.17984748929738997,
      "num_tokens": 13630098.0,
      "step": 6110
    },
    {
      "entropy": 5.625010013580322,
      "epoch": 0.6868085584320772,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004962631429140048,
      "loss": 5.6491,
      "mean_token_accuracy": 0.17171284705400466,
      "num_tokens": 13641401.0,
      "step": 6115
    },
    {
      "entropy": 5.739342164993286,
      "epoch": 0.6873701353400348,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000496255852573374,
      "loss": 5.6075,
      "mean_token_accuracy": 0.1670444905757904,
      "num_tokens": 13653099.0,
      "step": 6120
    },
    {
      "entropy": 5.795357084274292,
      "epoch": 0.6879317122479923,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004962485551878711,
      "loss": 5.622,
      "mean_token_accuracy": 0.16640905737876893,
      "num_tokens": 13665381.0,
      "step": 6125
    },
    {
      "entropy": 5.661467742919922,
      "epoch": 0.6884932891559499,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004962412507577283,
      "loss": 5.6346,
      "mean_token_accuracy": 0.17221850603818895,
      "num_tokens": 13676767.0,
      "step": 6130
    },
    {
      "entropy": 5.627943468093872,
      "epoch": 0.6890548660639074,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004962339392831783,
      "loss": 5.6228,
      "mean_token_accuracy": 0.17215524911880492,
      "num_tokens": 13688552.0,
      "step": 6135
    },
    {
      "entropy": 5.70760407447815,
      "epoch": 0.689616442971865,
      "grad_norm": 1.46875,
      "learning_rate": 0.000496226620764454,
      "loss": 5.5953,
      "mean_token_accuracy": 0.17065925300121307,
      "num_tokens": 13700160.0,
      "step": 6140
    },
    {
      "entropy": 5.671698188781738,
      "epoch": 0.6901780198798225,
      "grad_norm": 2.015625,
      "learning_rate": 0.000496219295201788,
      "loss": 5.6011,
      "mean_token_accuracy": 0.1736286088824272,
      "num_tokens": 13710828.0,
      "step": 6145
    },
    {
      "entropy": 5.744783687591553,
      "epoch": 0.6907395967877801,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004962119625954141,
      "loss": 5.7163,
      "mean_token_accuracy": 0.16494496762752534,
      "num_tokens": 13721912.0,
      "step": 6150
    },
    {
      "entropy": 5.773995733261108,
      "epoch": 0.6913011736957376,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004962046229455653,
      "loss": 5.7312,
      "mean_token_accuracy": 0.1703624963760376,
      "num_tokens": 13733662.0,
      "step": 6155
    },
    {
      "entropy": 5.625052499771118,
      "epoch": 0.6918627506036952,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004961972762524757,
      "loss": 5.549,
      "mean_token_accuracy": 0.1760735273361206,
      "num_tokens": 13743699.0,
      "step": 6160
    },
    {
      "entropy": 5.629493951797485,
      "epoch": 0.6924243275116527,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000496189922516379,
      "loss": 5.5554,
      "mean_token_accuracy": 0.17835567742586136,
      "num_tokens": 13754918.0,
      "step": 6165
    },
    {
      "entropy": 5.802048397064209,
      "epoch": 0.6929859044196103,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004961825617375093,
      "loss": 5.589,
      "mean_token_accuracy": 0.1810780867934227,
      "num_tokens": 13764432.0,
      "step": 6170
    },
    {
      "entropy": 5.56612720489502,
      "epoch": 0.6935474813275678,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004961751939161011,
      "loss": 5.5101,
      "mean_token_accuracy": 0.1665957048535347,
      "num_tokens": 13777049.0,
      "step": 6175
    },
    {
      "entropy": 5.613159084320069,
      "epoch": 0.6941090582355254,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000496167819052389,
      "loss": 5.6285,
      "mean_token_accuracy": 0.17187547385692598,
      "num_tokens": 13788324.0,
      "step": 6180
    },
    {
      "entropy": 5.774720954895019,
      "epoch": 0.6946706351434829,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004961604371466077,
      "loss": 5.7071,
      "mean_token_accuracy": 0.1715664029121399,
      "num_tokens": 13800443.0,
      "step": 6185
    },
    {
      "entropy": 5.813463592529297,
      "epoch": 0.6952322120514405,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004961530481989923,
      "loss": 5.7471,
      "mean_token_accuracy": 0.1697026953101158,
      "num_tokens": 13811982.0,
      "step": 6190
    },
    {
      "entropy": 5.664430713653564,
      "epoch": 0.695793788959398,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004961456522097782,
      "loss": 5.5307,
      "mean_token_accuracy": 0.17545289546251297,
      "num_tokens": 13822005.0,
      "step": 6195
    },
    {
      "entropy": 5.580399084091186,
      "epoch": 0.6963553658673556,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004961382491792006,
      "loss": 5.5458,
      "mean_token_accuracy": 0.1745297133922577,
      "num_tokens": 13831943.0,
      "step": 6200
    },
    {
      "entropy": 5.800264644622803,
      "epoch": 0.6969169427753131,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004961308391074953,
      "loss": 5.7382,
      "mean_token_accuracy": 0.16787344366312026,
      "num_tokens": 13842517.0,
      "step": 6205
    },
    {
      "entropy": 5.726728725433349,
      "epoch": 0.6974785196832706,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004961234219948986,
      "loss": 5.6109,
      "mean_token_accuracy": 0.1686943292617798,
      "num_tokens": 13853636.0,
      "step": 6210
    },
    {
      "entropy": 5.655916929244995,
      "epoch": 0.6980400965912281,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004961159978416461,
      "loss": 5.5513,
      "mean_token_accuracy": 0.1721298947930336,
      "num_tokens": 13864378.0,
      "step": 6215
    },
    {
      "entropy": 5.713779544830322,
      "epoch": 0.6986016734991857,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004961085666479746,
      "loss": 5.5778,
      "mean_token_accuracy": 0.17656196355819703,
      "num_tokens": 13874964.0,
      "step": 6220
    },
    {
      "entropy": 5.689186382293701,
      "epoch": 0.6991632504071432,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004961011284141206,
      "loss": 5.5644,
      "mean_token_accuracy": 0.16974303126335144,
      "num_tokens": 13885852.0,
      "step": 6225
    },
    {
      "entropy": 5.630877304077148,
      "epoch": 0.6997248273151008,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004960936831403209,
      "loss": 5.6134,
      "mean_token_accuracy": 0.17150009721517562,
      "num_tokens": 13896101.0,
      "step": 6230
    },
    {
      "entropy": 5.648992586135864,
      "epoch": 0.7002864042230583,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004960862308268126,
      "loss": 5.5052,
      "mean_token_accuracy": 0.17886769622564316,
      "num_tokens": 13907129.0,
      "step": 6235
    },
    {
      "entropy": 5.720948791503906,
      "epoch": 0.7008479811310159,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004960787714738329,
      "loss": 5.6951,
      "mean_token_accuracy": 0.1737249031662941,
      "num_tokens": 13918662.0,
      "step": 6240
    },
    {
      "entropy": 5.685187721252442,
      "epoch": 0.7014095580389734,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004960713050816194,
      "loss": 5.6562,
      "mean_token_accuracy": 0.17299176901578903,
      "num_tokens": 13930717.0,
      "step": 6245
    },
    {
      "entropy": 5.633158874511719,
      "epoch": 0.701971134946931,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004960638316504098,
      "loss": 5.5279,
      "mean_token_accuracy": 0.17130255252122878,
      "num_tokens": 13941299.0,
      "step": 6250
    },
    {
      "entropy": 5.780350589752198,
      "epoch": 0.7025327118548885,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004960563511804421,
      "loss": 5.6119,
      "mean_token_accuracy": 0.17323518246412278,
      "num_tokens": 13951807.0,
      "step": 6255
    },
    {
      "entropy": 5.650942611694336,
      "epoch": 0.7030942887628461,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004960488636719545,
      "loss": 5.5209,
      "mean_token_accuracy": 0.17925544530153276,
      "num_tokens": 13962994.0,
      "step": 6260
    },
    {
      "entropy": 5.566241312026977,
      "epoch": 0.7036558656708036,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004960413691251852,
      "loss": 5.5587,
      "mean_token_accuracy": 0.17189284712076186,
      "num_tokens": 13973731.0,
      "step": 6265
    },
    {
      "entropy": 5.686014032363891,
      "epoch": 0.7042174425787612,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004960338675403731,
      "loss": 5.6194,
      "mean_token_accuracy": 0.16850770264863968,
      "num_tokens": 13983772.0,
      "step": 6270
    },
    {
      "entropy": 5.6831377983093265,
      "epoch": 0.7047790194867187,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004960263589177568,
      "loss": 5.6285,
      "mean_token_accuracy": 0.17523877322673798,
      "num_tokens": 13994935.0,
      "step": 6275
    },
    {
      "entropy": 5.693459796905517,
      "epoch": 0.7053405963946763,
      "grad_norm": 1.625,
      "learning_rate": 0.0004960188432575757,
      "loss": 5.5889,
      "mean_token_accuracy": 0.17878624945878982,
      "num_tokens": 14006225.0,
      "step": 6280
    },
    {
      "entropy": 5.69208722114563,
      "epoch": 0.7059021733026338,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004960113205600688,
      "loss": 5.6025,
      "mean_token_accuracy": 0.18202067613601686,
      "num_tokens": 14017693.0,
      "step": 6285
    },
    {
      "entropy": 5.67942214012146,
      "epoch": 0.7064637502105914,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004960037908254757,
      "loss": 5.5771,
      "mean_token_accuracy": 0.17393409013748168,
      "num_tokens": 14027264.0,
      "step": 6290
    },
    {
      "entropy": 5.7325080871582035,
      "epoch": 0.7070253271185489,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004959962540540361,
      "loss": 5.6781,
      "mean_token_accuracy": 0.17509326040744783,
      "num_tokens": 14039422.0,
      "step": 6295
    },
    {
      "entropy": 5.694810819625855,
      "epoch": 0.7075869040265065,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004959887102459903,
      "loss": 5.6026,
      "mean_token_accuracy": 0.17309763431549072,
      "num_tokens": 14051619.0,
      "step": 6300
    },
    {
      "entropy": 5.518226671218872,
      "epoch": 0.7081484809344639,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004959811594015781,
      "loss": 5.6008,
      "mean_token_accuracy": 0.17118653059005737,
      "num_tokens": 14061752.0,
      "step": 6305
    },
    {
      "entropy": 5.7187727928161625,
      "epoch": 0.7087100578424215,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004959736015210401,
      "loss": 5.5499,
      "mean_token_accuracy": 0.17870223075151442,
      "num_tokens": 14072931.0,
      "step": 6310
    },
    {
      "entropy": 5.670480346679687,
      "epoch": 0.709271634750379,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004959660366046169,
      "loss": 5.6354,
      "mean_token_accuracy": 0.1812950477004051,
      "num_tokens": 14084137.0,
      "step": 6315
    },
    {
      "entropy": 5.733542203903198,
      "epoch": 0.7098332116583366,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004959584646525494,
      "loss": 5.643,
      "mean_token_accuracy": 0.16594423800706865,
      "num_tokens": 14096251.0,
      "step": 6320
    },
    {
      "entropy": 5.834647226333618,
      "epoch": 0.7103947885662941,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004959508856650788,
      "loss": 5.747,
      "mean_token_accuracy": 0.15990046858787538,
      "num_tokens": 14107844.0,
      "step": 6325
    },
    {
      "entropy": 5.701244258880616,
      "epoch": 0.7109563654742517,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004959432996424463,
      "loss": 5.5159,
      "mean_token_accuracy": 0.1764766052365303,
      "num_tokens": 14118344.0,
      "step": 6330
    },
    {
      "entropy": 5.683355712890625,
      "epoch": 0.7115179423822092,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004959357065848932,
      "loss": 5.5731,
      "mean_token_accuracy": 0.17560436129570006,
      "num_tokens": 14128767.0,
      "step": 6335
    },
    {
      "entropy": 5.676946210861206,
      "epoch": 0.7120795192901668,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004959281064926618,
      "loss": 5.7624,
      "mean_token_accuracy": 0.1625056892633438,
      "num_tokens": 14141593.0,
      "step": 6340
    },
    {
      "entropy": 5.787940835952758,
      "epoch": 0.7126410961981243,
      "grad_norm": 1.375,
      "learning_rate": 0.0004959204993659936,
      "loss": 5.658,
      "mean_token_accuracy": 0.16435816437005996,
      "num_tokens": 14151488.0,
      "step": 6345
    },
    {
      "entropy": 5.709534454345703,
      "epoch": 0.7132026731060819,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004959128852051311,
      "loss": 5.6446,
      "mean_token_accuracy": 0.17000896334648133,
      "num_tokens": 14162399.0,
      "step": 6350
    },
    {
      "entropy": 5.681110572814942,
      "epoch": 0.7137642500140394,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004959052640103167,
      "loss": 5.6154,
      "mean_token_accuracy": 0.1655770495533943,
      "num_tokens": 14173528.0,
      "step": 6355
    },
    {
      "entropy": 5.737189197540284,
      "epoch": 0.714325826921997,
      "grad_norm": 1.578125,
      "learning_rate": 0.000495897635781793,
      "loss": 5.5977,
      "mean_token_accuracy": 0.1738772988319397,
      "num_tokens": 14184119.0,
      "step": 6360
    },
    {
      "entropy": 5.783509731292725,
      "epoch": 0.7148874038299545,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004958900005198029,
      "loss": 5.6966,
      "mean_token_accuracy": 0.16666410118341446,
      "num_tokens": 14196225.0,
      "step": 6365
    },
    {
      "entropy": 5.5992560386657715,
      "epoch": 0.7154489807379121,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004958823582245894,
      "loss": 5.5122,
      "mean_token_accuracy": 0.1770908787846565,
      "num_tokens": 14207541.0,
      "step": 6370
    },
    {
      "entropy": 5.713044595718384,
      "epoch": 0.7160105576458696,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004958747088963961,
      "loss": 5.62,
      "mean_token_accuracy": 0.17931632846593856,
      "num_tokens": 14218713.0,
      "step": 6375
    },
    {
      "entropy": 5.576667070388794,
      "epoch": 0.7165721345538272,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004958670525354663,
      "loss": 5.4876,
      "mean_token_accuracy": 0.17556754648685455,
      "num_tokens": 14229626.0,
      "step": 6380
    },
    {
      "entropy": 5.67002272605896,
      "epoch": 0.7171337114617847,
      "grad_norm": 1.421875,
      "learning_rate": 0.000495859389142044,
      "loss": 5.6633,
      "mean_token_accuracy": 0.16710093021392822,
      "num_tokens": 14240320.0,
      "step": 6385
    },
    {
      "entropy": 5.830556774139405,
      "epoch": 0.7176952883697423,
      "grad_norm": 1.390625,
      "learning_rate": 0.000495851718716373,
      "loss": 5.719,
      "mean_token_accuracy": 0.17310031801462172,
      "num_tokens": 14252233.0,
      "step": 6390
    },
    {
      "entropy": 5.79133095741272,
      "epoch": 0.7182568652776998,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004958440412586977,
      "loss": 5.7307,
      "mean_token_accuracy": 0.15904613435268403,
      "num_tokens": 14264421.0,
      "step": 6395
    },
    {
      "entropy": 5.731802177429199,
      "epoch": 0.7188184421856574,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004958363567692624,
      "loss": 5.6031,
      "mean_token_accuracy": 0.17360225617885588,
      "num_tokens": 14275320.0,
      "step": 6400
    },
    {
      "entropy": 5.638220739364624,
      "epoch": 0.7193800190936148,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004958286652483118,
      "loss": 5.5668,
      "mean_token_accuracy": 0.1780466377735138,
      "num_tokens": 14286099.0,
      "step": 6405
    },
    {
      "entropy": 5.66323037147522,
      "epoch": 0.7199415960015724,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004958209666960909,
      "loss": 5.6835,
      "mean_token_accuracy": 0.16914550662040712,
      "num_tokens": 14297795.0,
      "step": 6410
    },
    {
      "entropy": 5.638380813598633,
      "epoch": 0.7205031729095299,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004958132611128448,
      "loss": 5.5828,
      "mean_token_accuracy": 0.17390582710504532,
      "num_tokens": 14309424.0,
      "step": 6415
    },
    {
      "entropy": 5.784746742248535,
      "epoch": 0.7210647498174875,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004958055484988188,
      "loss": 5.6623,
      "mean_token_accuracy": 0.16943105459213256,
      "num_tokens": 14320763.0,
      "step": 6420
    },
    {
      "entropy": 5.7342438220977785,
      "epoch": 0.721626326725445,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004957978288542585,
      "loss": 5.6264,
      "mean_token_accuracy": 0.1723559558391571,
      "num_tokens": 14331704.0,
      "step": 6425
    },
    {
      "entropy": 5.756981945037841,
      "epoch": 0.7221879036334026,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004957901021794098,
      "loss": 5.675,
      "mean_token_accuracy": 0.16486780792474748,
      "num_tokens": 14343694.0,
      "step": 6430
    },
    {
      "entropy": 5.680773448944092,
      "epoch": 0.7227494805413601,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004957823684745184,
      "loss": 5.572,
      "mean_token_accuracy": 0.17167100310325623,
      "num_tokens": 14355408.0,
      "step": 6435
    },
    {
      "entropy": 5.703755664825439,
      "epoch": 0.7233110574493177,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004957746277398309,
      "loss": 5.7254,
      "mean_token_accuracy": 0.16364069730043412,
      "num_tokens": 14366778.0,
      "step": 6440
    },
    {
      "entropy": 5.712000751495362,
      "epoch": 0.7238726343572752,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004957668799755937,
      "loss": 5.5799,
      "mean_token_accuracy": 0.17299263626337053,
      "num_tokens": 14377947.0,
      "step": 6445
    },
    {
      "entropy": 5.705758762359619,
      "epoch": 0.7244342112652328,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004957591251820532,
      "loss": 5.6688,
      "mean_token_accuracy": 0.1714080259203911,
      "num_tokens": 14389571.0,
      "step": 6450
    },
    {
      "entropy": 5.8451415538787845,
      "epoch": 0.7249957881731903,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004957513633594567,
      "loss": 5.7455,
      "mean_token_accuracy": 0.1655985251069069,
      "num_tokens": 14401865.0,
      "step": 6455
    },
    {
      "entropy": 5.643240690231323,
      "epoch": 0.7255573650811479,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004957435945080512,
      "loss": 5.576,
      "mean_token_accuracy": 0.17257309556007386,
      "num_tokens": 14413676.0,
      "step": 6460
    },
    {
      "entropy": 5.6537106990814205,
      "epoch": 0.7261189419891054,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004957358186280839,
      "loss": 5.6016,
      "mean_token_accuracy": 0.1794086754322052,
      "num_tokens": 14424675.0,
      "step": 6465
    },
    {
      "entropy": 5.618784666061401,
      "epoch": 0.726680518897063,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004957280357198026,
      "loss": 5.5149,
      "mean_token_accuracy": 0.17647840976715087,
      "num_tokens": 14437238.0,
      "step": 6470
    },
    {
      "entropy": 5.549806547164917,
      "epoch": 0.7272420958050205,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004957202457834551,
      "loss": 5.4673,
      "mean_token_accuracy": 0.17938644886016847,
      "num_tokens": 14448003.0,
      "step": 6475
    },
    {
      "entropy": 5.528613138198852,
      "epoch": 0.7278036727129781,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004957124488192893,
      "loss": 5.4236,
      "mean_token_accuracy": 0.18780238777399064,
      "num_tokens": 14458440.0,
      "step": 6480
    },
    {
      "entropy": 5.688983345031739,
      "epoch": 0.7283652496209356,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004957046448275535,
      "loss": 5.6076,
      "mean_token_accuracy": 0.17096126079559326,
      "num_tokens": 14470127.0,
      "step": 6485
    },
    {
      "entropy": 5.652023792266846,
      "epoch": 0.7289268265288932,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004956968338084961,
      "loss": 5.5194,
      "mean_token_accuracy": 0.17799320966005325,
      "num_tokens": 14480993.0,
      "step": 6490
    },
    {
      "entropy": 5.667353630065918,
      "epoch": 0.7294884034368507,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004956890157623661,
      "loss": 5.6216,
      "mean_token_accuracy": 0.16608051508665084,
      "num_tokens": 14492789.0,
      "step": 6495
    },
    {
      "entropy": 5.735092353820801,
      "epoch": 0.7300499803448082,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004956811906894122,
      "loss": 5.7142,
      "mean_token_accuracy": 0.16898949593305587,
      "num_tokens": 14504528.0,
      "step": 6500
    },
    {
      "entropy": 5.777708959579468,
      "epoch": 0.7306115572527657,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004956733585898836,
      "loss": 5.6052,
      "mean_token_accuracy": 0.17461458444595337,
      "num_tokens": 14515739.0,
      "step": 6505
    },
    {
      "entropy": 5.7372722148895265,
      "epoch": 0.7311731341607233,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004956655194640297,
      "loss": 5.7253,
      "mean_token_accuracy": 0.165912227332592,
      "num_tokens": 14527120.0,
      "step": 6510
    },
    {
      "entropy": 5.734869432449341,
      "epoch": 0.7317347110686808,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004956576733121,
      "loss": 5.6732,
      "mean_token_accuracy": 0.17188447639346122,
      "num_tokens": 14539390.0,
      "step": 6515
    },
    {
      "entropy": 5.7428192615509035,
      "epoch": 0.7322962879766384,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004956498201343444,
      "loss": 5.6542,
      "mean_token_accuracy": 0.17126702666282653,
      "num_tokens": 14549905.0,
      "step": 6520
    },
    {
      "entropy": 5.7277122974395756,
      "epoch": 0.7328578648845959,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000495641959931013,
      "loss": 5.7013,
      "mean_token_accuracy": 0.17059755474328994,
      "num_tokens": 14561519.0,
      "step": 6525
    },
    {
      "entropy": 5.656032657623291,
      "epoch": 0.7334194417925535,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004956340927023561,
      "loss": 5.4721,
      "mean_token_accuracy": 0.17097557485103607,
      "num_tokens": 14572704.0,
      "step": 6530
    },
    {
      "entropy": 5.733268117904663,
      "epoch": 0.733981018700511,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000495626218448624,
      "loss": 5.6477,
      "mean_token_accuracy": 0.17309343367815017,
      "num_tokens": 14584055.0,
      "step": 6535
    },
    {
      "entropy": 5.528829097747803,
      "epoch": 0.7345425956084686,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004956183371700676,
      "loss": 5.4578,
      "mean_token_accuracy": 0.17761017978191376,
      "num_tokens": 14595271.0,
      "step": 6540
    },
    {
      "entropy": 5.687887334823609,
      "epoch": 0.7351041725164261,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004956104488669377,
      "loss": 5.6217,
      "mean_token_accuracy": 0.1771188572049141,
      "num_tokens": 14605748.0,
      "step": 6545
    },
    {
      "entropy": 5.658285093307495,
      "epoch": 0.7356657494243837,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004956025535394856,
      "loss": 5.5326,
      "mean_token_accuracy": 0.18142949491739274,
      "num_tokens": 14616295.0,
      "step": 6550
    },
    {
      "entropy": 5.633347415924073,
      "epoch": 0.7362273263323412,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004955946511879626,
      "loss": 5.625,
      "mean_token_accuracy": 0.16703985631465912,
      "num_tokens": 14626690.0,
      "step": 6555
    },
    {
      "entropy": 5.678591918945313,
      "epoch": 0.7367889032402988,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004955867418126205,
      "loss": 5.6293,
      "mean_token_accuracy": 0.17487271279096603,
      "num_tokens": 14637685.0,
      "step": 6560
    },
    {
      "entropy": 5.738638973236084,
      "epoch": 0.7373504801482563,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004955788254137108,
      "loss": 5.5306,
      "mean_token_accuracy": 0.17113898396492006,
      "num_tokens": 14648098.0,
      "step": 6565
    },
    {
      "entropy": 5.746050071716309,
      "epoch": 0.7379120570562139,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004955709019914857,
      "loss": 5.6727,
      "mean_token_accuracy": 0.16656773388385773,
      "num_tokens": 14661134.0,
      "step": 6570
    },
    {
      "entropy": 5.746538877487183,
      "epoch": 0.7384736339641714,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004955629715461977,
      "loss": 5.6445,
      "mean_token_accuracy": 0.17216980457305908,
      "num_tokens": 14672699.0,
      "step": 6575
    },
    {
      "entropy": 5.747246170043946,
      "epoch": 0.739035210872129,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004955550340780991,
      "loss": 5.6143,
      "mean_token_accuracy": 0.16882283240556717,
      "num_tokens": 14682994.0,
      "step": 6580
    },
    {
      "entropy": 5.672818374633789,
      "epoch": 0.7395967877800865,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004955470895874427,
      "loss": 5.6186,
      "mean_token_accuracy": 0.17457711100578308,
      "num_tokens": 14694828.0,
      "step": 6585
    },
    {
      "entropy": 5.648428583145142,
      "epoch": 0.7401583646880441,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004955391380744813,
      "loss": 5.6108,
      "mean_token_accuracy": 0.17359036207199097,
      "num_tokens": 14705975.0,
      "step": 6590
    },
    {
      "entropy": 5.726032495498657,
      "epoch": 0.7407199415960015,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004955311795394684,
      "loss": 5.5944,
      "mean_token_accuracy": 0.1734750136733055,
      "num_tokens": 14717717.0,
      "step": 6595
    },
    {
      "entropy": 5.711815071105957,
      "epoch": 0.7412815185039591,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000495523213982657,
      "loss": 5.5662,
      "mean_token_accuracy": 0.1741805464029312,
      "num_tokens": 14729254.0,
      "step": 6600
    },
    {
      "entropy": 5.6931883811950685,
      "epoch": 0.7418430954119166,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004955152414043011,
      "loss": 5.7119,
      "mean_token_accuracy": 0.1718486949801445,
      "num_tokens": 14739733.0,
      "step": 6605
    },
    {
      "entropy": 5.752079057693481,
      "epoch": 0.7424046723198742,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004955072618046543,
      "loss": 5.6666,
      "mean_token_accuracy": 0.17087932378053666,
      "num_tokens": 14750855.0,
      "step": 6610
    },
    {
      "entropy": 5.71418023109436,
      "epoch": 0.7429662492278317,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004954992751839708,
      "loss": 5.6614,
      "mean_token_accuracy": 0.1698993593454361,
      "num_tokens": 14762762.0,
      "step": 6615
    },
    {
      "entropy": 5.688072967529297,
      "epoch": 0.7435278261357893,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004954912815425048,
      "loss": 5.5299,
      "mean_token_accuracy": 0.1700862854719162,
      "num_tokens": 14773690.0,
      "step": 6620
    },
    {
      "entropy": 5.640633058547974,
      "epoch": 0.7440894030437468,
      "grad_norm": 1.390625,
      "learning_rate": 0.000495483280880511,
      "loss": 5.6447,
      "mean_token_accuracy": 0.17218244969844818,
      "num_tokens": 14785019.0,
      "step": 6625
    },
    {
      "entropy": 5.643097877502441,
      "epoch": 0.7446509799517044,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000495475273198244,
      "loss": 5.6206,
      "mean_token_accuracy": 0.16789701133966445,
      "num_tokens": 14795638.0,
      "step": 6630
    },
    {
      "entropy": 5.715163087844848,
      "epoch": 0.7452125568596619,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004954672584959588,
      "loss": 5.549,
      "mean_token_accuracy": 0.17813322842121124,
      "num_tokens": 14806546.0,
      "step": 6635
    },
    {
      "entropy": 5.596341991424561,
      "epoch": 0.7457741337676195,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004954592367739104,
      "loss": 5.5626,
      "mean_token_accuracy": 0.17497475147247316,
      "num_tokens": 14816918.0,
      "step": 6640
    },
    {
      "entropy": 5.6133809089660645,
      "epoch": 0.746335710675577,
      "grad_norm": 1.75,
      "learning_rate": 0.0004954512080323546,
      "loss": 5.5571,
      "mean_token_accuracy": 0.18172363042831421,
      "num_tokens": 14827420.0,
      "step": 6645
    },
    {
      "entropy": 5.7528026580810545,
      "epoch": 0.7468972875835346,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004954431722715468,
      "loss": 5.7731,
      "mean_token_accuracy": 0.17284041196107863,
      "num_tokens": 14839209.0,
      "step": 6650
    },
    {
      "entropy": 5.772123146057129,
      "epoch": 0.7474588644914921,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004954351294917428,
      "loss": 5.6546,
      "mean_token_accuracy": 0.16983917206525803,
      "num_tokens": 14849916.0,
      "step": 6655
    },
    {
      "entropy": 5.593883991241455,
      "epoch": 0.7480204413994497,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004954270796931989,
      "loss": 5.5467,
      "mean_token_accuracy": 0.17545504868030548,
      "num_tokens": 14860946.0,
      "step": 6660
    },
    {
      "entropy": 5.773710012435913,
      "epoch": 0.7485820183074072,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004954190228761712,
      "loss": 5.6383,
      "mean_token_accuracy": 0.16717524081468582,
      "num_tokens": 14872750.0,
      "step": 6665
    },
    {
      "entropy": 5.68846116065979,
      "epoch": 0.7491435952153648,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004954109590409164,
      "loss": 5.6227,
      "mean_token_accuracy": 0.17093631327152253,
      "num_tokens": 14883714.0,
      "step": 6670
    },
    {
      "entropy": 5.633935451507568,
      "epoch": 0.7497051721233223,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004954028881876911,
      "loss": 5.6325,
      "mean_token_accuracy": 0.17225925624370575,
      "num_tokens": 14895147.0,
      "step": 6675
    },
    {
      "entropy": 5.698961019515991,
      "epoch": 0.7502667490312799,
      "grad_norm": 1.375,
      "learning_rate": 0.0004953948103167524,
      "loss": 5.5536,
      "mean_token_accuracy": 0.1762816697359085,
      "num_tokens": 14905914.0,
      "step": 6680
    },
    {
      "entropy": 5.582836866378784,
      "epoch": 0.7508283259392374,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004953867254283575,
      "loss": 5.5428,
      "mean_token_accuracy": 0.17713676244020463,
      "num_tokens": 14916988.0,
      "step": 6685
    },
    {
      "entropy": 5.6897834777832035,
      "epoch": 0.7513899028471949,
      "grad_norm": 1.5,
      "learning_rate": 0.0004953786335227636,
      "loss": 5.6518,
      "mean_token_accuracy": 0.17040350288152695,
      "num_tokens": 14928675.0,
      "step": 6690
    },
    {
      "entropy": 5.795252656936645,
      "epoch": 0.7519514797551524,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004953705346002287,
      "loss": 5.6581,
      "mean_token_accuracy": 0.17117824405431747,
      "num_tokens": 14939422.0,
      "step": 6695
    },
    {
      "entropy": 5.6437677383422855,
      "epoch": 0.75251305666311,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004953624286610106,
      "loss": 5.6086,
      "mean_token_accuracy": 0.17704180479049683,
      "num_tokens": 14950156.0,
      "step": 6700
    },
    {
      "entropy": 5.65733003616333,
      "epoch": 0.7530746335710675,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004953543157053672,
      "loss": 5.5983,
      "mean_token_accuracy": 0.1764406979084015,
      "num_tokens": 14960535.0,
      "step": 6705
    },
    {
      "entropy": 5.767850303649903,
      "epoch": 0.7536362104790251,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004953461957335569,
      "loss": 5.6239,
      "mean_token_accuracy": 0.1750498056411743,
      "num_tokens": 14970975.0,
      "step": 6710
    },
    {
      "entropy": 5.632910871505738,
      "epoch": 0.7541977873869826,
      "grad_norm": 1.375,
      "learning_rate": 0.0004953380687458382,
      "loss": 5.4731,
      "mean_token_accuracy": 0.18237796127796174,
      "num_tokens": 14981204.0,
      "step": 6715
    },
    {
      "entropy": 5.653511953353882,
      "epoch": 0.7547593642949402,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004953299347424701,
      "loss": 5.6468,
      "mean_token_accuracy": 0.17152471095323563,
      "num_tokens": 14991631.0,
      "step": 6720
    },
    {
      "entropy": 5.657448101043701,
      "epoch": 0.7553209412028977,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004953217937237114,
      "loss": 5.5735,
      "mean_token_accuracy": 0.17640465199947358,
      "num_tokens": 15003873.0,
      "step": 6725
    },
    {
      "entropy": 5.69847002029419,
      "epoch": 0.7558825181108553,
      "grad_norm": 1.5,
      "learning_rate": 0.0004953136456898212,
      "loss": 5.6493,
      "mean_token_accuracy": 0.17125397473573684,
      "num_tokens": 15014890.0,
      "step": 6730
    },
    {
      "entropy": 5.604665327072143,
      "epoch": 0.7564440950188128,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004953054906410593,
      "loss": 5.5194,
      "mean_token_accuracy": 0.18065904527902604,
      "num_tokens": 15025037.0,
      "step": 6735
    },
    {
      "entropy": 5.690511035919189,
      "epoch": 0.7570056719267704,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004952973285776848,
      "loss": 5.6233,
      "mean_token_accuracy": 0.17375240176916124,
      "num_tokens": 15035722.0,
      "step": 6740
    },
    {
      "entropy": 5.696527576446533,
      "epoch": 0.7575672488347279,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004952891594999582,
      "loss": 5.6224,
      "mean_token_accuracy": 0.16907656490802764,
      "num_tokens": 15047138.0,
      "step": 6745
    },
    {
      "entropy": 5.616460132598877,
      "epoch": 0.7581288257426855,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004952809834081392,
      "loss": 5.5799,
      "mean_token_accuracy": 0.17677342891693115,
      "num_tokens": 15058559.0,
      "step": 6750
    },
    {
      "entropy": 5.669347715377808,
      "epoch": 0.758690402650643,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004952728003024883,
      "loss": 5.6034,
      "mean_token_accuracy": 0.17386410832405091,
      "num_tokens": 15070481.0,
      "step": 6755
    },
    {
      "entropy": 5.746075916290283,
      "epoch": 0.7592519795586006,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004952646101832659,
      "loss": 5.525,
      "mean_token_accuracy": 0.1773831456899643,
      "num_tokens": 15079844.0,
      "step": 6760
    },
    {
      "entropy": 5.575839519500732,
      "epoch": 0.7598135564665581,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004952564130507331,
      "loss": 5.5459,
      "mean_token_accuracy": 0.1829729363322258,
      "num_tokens": 15091197.0,
      "step": 6765
    },
    {
      "entropy": 5.63728141784668,
      "epoch": 0.7603751333745157,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004952482089051506,
      "loss": 5.6395,
      "mean_token_accuracy": 0.16757640540599822,
      "num_tokens": 15102130.0,
      "step": 6770
    },
    {
      "entropy": 5.80513825416565,
      "epoch": 0.7609367102824732,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004952399977467796,
      "loss": 5.7159,
      "mean_token_accuracy": 0.164860138297081,
      "num_tokens": 15113886.0,
      "step": 6775
    },
    {
      "entropy": 5.695002603530884,
      "epoch": 0.7614982871904308,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004952317795758817,
      "loss": 5.5309,
      "mean_token_accuracy": 0.1707255095243454,
      "num_tokens": 15125284.0,
      "step": 6780
    },
    {
      "entropy": 5.693516254425049,
      "epoch": 0.7620598640983883,
      "grad_norm": 1.375,
      "learning_rate": 0.0004952235543927185,
      "loss": 5.6013,
      "mean_token_accuracy": 0.17872631102800368,
      "num_tokens": 15136486.0,
      "step": 6785
    },
    {
      "entropy": 5.5827912330627445,
      "epoch": 0.7626214410063458,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000495215322197552,
      "loss": 5.4942,
      "mean_token_accuracy": 0.1766755223274231,
      "num_tokens": 15145937.0,
      "step": 6790
    },
    {
      "entropy": 5.606060981750488,
      "epoch": 0.7631830179143033,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004952070829906442,
      "loss": 5.5703,
      "mean_token_accuracy": 0.17622093558311464,
      "num_tokens": 15156923.0,
      "step": 6795
    },
    {
      "entropy": 5.764186716079712,
      "epoch": 0.7637445948222609,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004951988367722574,
      "loss": 5.701,
      "mean_token_accuracy": 0.17148951441049576,
      "num_tokens": 15169793.0,
      "step": 6800
    },
    {
      "entropy": 5.666046524047852,
      "epoch": 0.7643061717302184,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004951905835426544,
      "loss": 5.4701,
      "mean_token_accuracy": 0.18407936692237853,
      "num_tokens": 15180679.0,
      "step": 6805
    },
    {
      "entropy": 5.658257055282593,
      "epoch": 0.764867748638176,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004951823233020976,
      "loss": 5.6771,
      "mean_token_accuracy": 0.1721682906150818,
      "num_tokens": 15192298.0,
      "step": 6810
    },
    {
      "entropy": 5.657361793518066,
      "epoch": 0.7654293255461335,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004951740560508504,
      "loss": 5.7427,
      "mean_token_accuracy": 0.1654340222477913,
      "num_tokens": 15204759.0,
      "step": 6815
    },
    {
      "entropy": 5.795834302902222,
      "epoch": 0.7659909024540911,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004951657817891758,
      "loss": 5.6181,
      "mean_token_accuracy": 0.17172770649194719,
      "num_tokens": 15217166.0,
      "step": 6820
    },
    {
      "entropy": 5.596086502075195,
      "epoch": 0.7665524793620486,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004951575005173373,
      "loss": 5.5908,
      "mean_token_accuracy": 0.17772889286279678,
      "num_tokens": 15227022.0,
      "step": 6825
    },
    {
      "entropy": 5.7071445941925045,
      "epoch": 0.7671140562700062,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004951492122355986,
      "loss": 5.6438,
      "mean_token_accuracy": 0.16916121244430543,
      "num_tokens": 15239088.0,
      "step": 6830
    },
    {
      "entropy": 5.74067268371582,
      "epoch": 0.7676756331779637,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004951409169442236,
      "loss": 5.6636,
      "mean_token_accuracy": 0.17377155274152756,
      "num_tokens": 15250263.0,
      "step": 6835
    },
    {
      "entropy": 5.614690923690796,
      "epoch": 0.7682372100859213,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004951326146434765,
      "loss": 5.56,
      "mean_token_accuracy": 0.17508067339658737,
      "num_tokens": 15263014.0,
      "step": 6840
    },
    {
      "entropy": 5.698054409027099,
      "epoch": 0.7687987869938788,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004951243053336214,
      "loss": 5.5488,
      "mean_token_accuracy": 0.17616119384765624,
      "num_tokens": 15274261.0,
      "step": 6845
    },
    {
      "entropy": 5.617717838287353,
      "epoch": 0.7693603639018364,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004951159890149232,
      "loss": 5.4812,
      "mean_token_accuracy": 0.17868284434080123,
      "num_tokens": 15284929.0,
      "step": 6850
    },
    {
      "entropy": 5.601969814300537,
      "epoch": 0.7699219408097939,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004951076656876466,
      "loss": 5.5905,
      "mean_token_accuracy": 0.1784390926361084,
      "num_tokens": 15296622.0,
      "step": 6855
    },
    {
      "entropy": 5.671678590774536,
      "epoch": 0.7704835177177515,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004950993353520564,
      "loss": 5.6416,
      "mean_token_accuracy": 0.16959952116012572,
      "num_tokens": 15308373.0,
      "step": 6860
    },
    {
      "entropy": 5.633020448684692,
      "epoch": 0.771045094625709,
      "grad_norm": 1.515625,
      "learning_rate": 0.000495090998008418,
      "loss": 5.5145,
      "mean_token_accuracy": 0.18185887783765792,
      "num_tokens": 15319705.0,
      "step": 6865
    },
    {
      "entropy": 5.718202543258667,
      "epoch": 0.7716066715336666,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000495082653656997,
      "loss": 5.6745,
      "mean_token_accuracy": 0.16872720867395402,
      "num_tokens": 15330883.0,
      "step": 6870
    },
    {
      "entropy": 5.695891237258911,
      "epoch": 0.7721682484416241,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004950743022980587,
      "loss": 5.5826,
      "mean_token_accuracy": 0.17275900691747664,
      "num_tokens": 15342294.0,
      "step": 6875
    },
    {
      "entropy": 5.698135662078857,
      "epoch": 0.7727298253495817,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004950659439318696,
      "loss": 5.5901,
      "mean_token_accuracy": 0.1722102403640747,
      "num_tokens": 15352615.0,
      "step": 6880
    },
    {
      "entropy": 5.6508818626403805,
      "epoch": 0.7732914022575391,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004950575785586952,
      "loss": 5.6341,
      "mean_token_accuracy": 0.168660968542099,
      "num_tokens": 15364516.0,
      "step": 6885
    },
    {
      "entropy": 5.67797999382019,
      "epoch": 0.7738529791654967,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004950492061788023,
      "loss": 5.527,
      "mean_token_accuracy": 0.173749241232872,
      "num_tokens": 15374877.0,
      "step": 6890
    },
    {
      "entropy": 5.673302030563354,
      "epoch": 0.7744145560734542,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004950408267924573,
      "loss": 5.4688,
      "mean_token_accuracy": 0.18849669992923737,
      "num_tokens": 15385250.0,
      "step": 6895
    },
    {
      "entropy": 5.630706882476806,
      "epoch": 0.7749761329814118,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004950324403999271,
      "loss": 5.6217,
      "mean_token_accuracy": 0.178818379342556,
      "num_tokens": 15397074.0,
      "step": 6900
    },
    {
      "entropy": 5.6523487091064455,
      "epoch": 0.7755377098893693,
      "grad_norm": 1.5,
      "learning_rate": 0.0004950240470014785,
      "loss": 5.6633,
      "mean_token_accuracy": 0.16847054213285445,
      "num_tokens": 15408575.0,
      "step": 6905
    },
    {
      "entropy": 5.721060371398925,
      "epoch": 0.7760992867973269,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004950156465973791,
      "loss": 5.5786,
      "mean_token_accuracy": 0.17714389264583588,
      "num_tokens": 15419071.0,
      "step": 6910
    },
    {
      "entropy": 5.595917701721191,
      "epoch": 0.7766608637052844,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000495007239187896,
      "loss": 5.4627,
      "mean_token_accuracy": 0.1753948897123337,
      "num_tokens": 15430137.0,
      "step": 6915
    },
    {
      "entropy": 5.549996185302734,
      "epoch": 0.777222440613242,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004949988247732972,
      "loss": 5.536,
      "mean_token_accuracy": 0.18008177876472473,
      "num_tokens": 15440661.0,
      "step": 6920
    },
    {
      "entropy": 5.577071666717529,
      "epoch": 0.7777840175211995,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004949904033538504,
      "loss": 5.5443,
      "mean_token_accuracy": 0.18011982291936873,
      "num_tokens": 15451909.0,
      "step": 6925
    },
    {
      "entropy": 5.744392728805542,
      "epoch": 0.7783455944291571,
      "grad_norm": 1.390625,
      "learning_rate": 0.000494981974929824,
      "loss": 5.6133,
      "mean_token_accuracy": 0.17083452343940736,
      "num_tokens": 15462244.0,
      "step": 6930
    },
    {
      "entropy": 5.644606256484986,
      "epoch": 0.7789071713371146,
      "grad_norm": 1.375,
      "learning_rate": 0.000494973539501486,
      "loss": 5.6585,
      "mean_token_accuracy": 0.17281380891799927,
      "num_tokens": 15473284.0,
      "step": 6935
    },
    {
      "entropy": 5.587286949157715,
      "epoch": 0.7794687482450722,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004949650970691052,
      "loss": 5.5591,
      "mean_token_accuracy": 0.1845861554145813,
      "num_tokens": 15484964.0,
      "step": 6940
    },
    {
      "entropy": 5.646401405334473,
      "epoch": 0.7800303251530297,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004949566476329505,
      "loss": 5.5877,
      "mean_token_accuracy": 0.17619993090629577,
      "num_tokens": 15495922.0,
      "step": 6945
    },
    {
      "entropy": 5.780960130691528,
      "epoch": 0.7805919020609873,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004949481911932908,
      "loss": 5.6492,
      "mean_token_accuracy": 0.17256713807582855,
      "num_tokens": 15507173.0,
      "step": 6950
    },
    {
      "entropy": 5.70377368927002,
      "epoch": 0.7811534789689448,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004949397277503954,
      "loss": 5.6209,
      "mean_token_accuracy": 0.17474339306354522,
      "num_tokens": 15517720.0,
      "step": 6955
    },
    {
      "entropy": 5.71235990524292,
      "epoch": 0.7817150558769024,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004949312573045338,
      "loss": 5.6394,
      "mean_token_accuracy": 0.1703775092959404,
      "num_tokens": 15529482.0,
      "step": 6960
    },
    {
      "entropy": 5.76483187675476,
      "epoch": 0.7822766327848599,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004949227798559757,
      "loss": 5.6429,
      "mean_token_accuracy": 0.1658545657992363,
      "num_tokens": 15540522.0,
      "step": 6965
    },
    {
      "entropy": 5.700478982925415,
      "epoch": 0.7828382096928175,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004949142954049909,
      "loss": 5.6267,
      "mean_token_accuracy": 0.17000363171100616,
      "num_tokens": 15552443.0,
      "step": 6970
    },
    {
      "entropy": 5.766734600067139,
      "epoch": 0.783399786600775,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004949058039518496,
      "loss": 5.664,
      "mean_token_accuracy": 0.17442309558391572,
      "num_tokens": 15563418.0,
      "step": 6975
    },
    {
      "entropy": 5.721600914001465,
      "epoch": 0.7839613635087325,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004948973054968222,
      "loss": 5.6809,
      "mean_token_accuracy": 0.1688017174601555,
      "num_tokens": 15574301.0,
      "step": 6980
    },
    {
      "entropy": 5.780105543136597,
      "epoch": 0.78452294041669,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004948888000401794,
      "loss": 5.5844,
      "mean_token_accuracy": 0.1704667940735817,
      "num_tokens": 15585125.0,
      "step": 6985
    },
    {
      "entropy": 5.7562014102935795,
      "epoch": 0.7850845173246476,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004948802875821918,
      "loss": 5.6937,
      "mean_token_accuracy": 0.17066754549741744,
      "num_tokens": 15595822.0,
      "step": 6990
    },
    {
      "entropy": 5.5500236511230465,
      "epoch": 0.7856460942326051,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004948717681231308,
      "loss": 5.5279,
      "mean_token_accuracy": 0.17664736062288283,
      "num_tokens": 15606835.0,
      "step": 6995
    },
    {
      "entropy": 5.632187032699585,
      "epoch": 0.7862076711405627,
      "grad_norm": 3.125,
      "learning_rate": 0.0004948632416632672,
      "loss": 5.6124,
      "mean_token_accuracy": 0.17274107187986373,
      "num_tokens": 15617287.0,
      "step": 7000
    },
    {
      "entropy": 5.8710390567779545,
      "epoch": 0.7867692480485202,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004948547082028728,
      "loss": 5.6573,
      "mean_token_accuracy": 0.16424071788787842,
      "num_tokens": 15628440.0,
      "step": 7005
    },
    {
      "entropy": 5.7740159034729,
      "epoch": 0.7873308249564778,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004948461677422193,
      "loss": 5.7659,
      "mean_token_accuracy": 0.16086588203907012,
      "num_tokens": 15640677.0,
      "step": 7010
    },
    {
      "entropy": 5.702903985977173,
      "epoch": 0.7878924018644353,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004948376202815785,
      "loss": 5.5799,
      "mean_token_accuracy": 0.17863162457942963,
      "num_tokens": 15651941.0,
      "step": 7015
    },
    {
      "entropy": 5.643584585189819,
      "epoch": 0.7884539787723929,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004948290658212225,
      "loss": 5.5872,
      "mean_token_accuracy": 0.1744334802031517,
      "num_tokens": 15663513.0,
      "step": 7020
    },
    {
      "entropy": 5.612797451019287,
      "epoch": 0.7890155556803504,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004948205043614239,
      "loss": 5.5694,
      "mean_token_accuracy": 0.17362559586763382,
      "num_tokens": 15675548.0,
      "step": 7025
    },
    {
      "entropy": 5.7275231838226315,
      "epoch": 0.789577132588308,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004948119359024551,
      "loss": 5.6234,
      "mean_token_accuracy": 0.17916473597288132,
      "num_tokens": 15685365.0,
      "step": 7030
    },
    {
      "entropy": 5.686577701568604,
      "epoch": 0.7901387094962655,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004948033604445891,
      "loss": 5.6126,
      "mean_token_accuracy": 0.17018286287784576,
      "num_tokens": 15696112.0,
      "step": 7035
    },
    {
      "entropy": 5.621775579452515,
      "epoch": 0.7907002864042231,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004947947779880988,
      "loss": 5.5484,
      "mean_token_accuracy": 0.1771180063486099,
      "num_tokens": 15707414.0,
      "step": 7040
    },
    {
      "entropy": 5.734884214401245,
      "epoch": 0.7912618633121806,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004947861885332576,
      "loss": 5.6507,
      "mean_token_accuracy": 0.17650367319583893,
      "num_tokens": 15719077.0,
      "step": 7045
    },
    {
      "entropy": 5.712475728988648,
      "epoch": 0.7918234402201382,
      "grad_norm": 1.375,
      "learning_rate": 0.0004947775920803388,
      "loss": 5.6426,
      "mean_token_accuracy": 0.1715145543217659,
      "num_tokens": 15730497.0,
      "step": 7050
    },
    {
      "entropy": 5.666735076904297,
      "epoch": 0.7923850171280957,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004947689886296162,
      "loss": 5.5464,
      "mean_token_accuracy": 0.1741068556904793,
      "num_tokens": 15741724.0,
      "step": 7055
    },
    {
      "entropy": 5.691272640228272,
      "epoch": 0.7929465940360533,
      "grad_norm": 1.8125,
      "learning_rate": 0.000494760378181364,
      "loss": 5.627,
      "mean_token_accuracy": 0.1703273355960846,
      "num_tokens": 15751868.0,
      "step": 7060
    },
    {
      "entropy": 5.659696960449219,
      "epoch": 0.7935081709440108,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004947517607358559,
      "loss": 5.6401,
      "mean_token_accuracy": 0.16680581122636795,
      "num_tokens": 15762783.0,
      "step": 7065
    },
    {
      "entropy": 5.722996759414673,
      "epoch": 0.7940697478519684,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004947431362933666,
      "loss": 5.6208,
      "mean_token_accuracy": 0.17095878720283508,
      "num_tokens": 15774155.0,
      "step": 7070
    },
    {
      "entropy": 5.82449369430542,
      "epoch": 0.7946313247599258,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004947345048541706,
      "loss": 5.6404,
      "mean_token_accuracy": 0.1689218834042549,
      "num_tokens": 15785034.0,
      "step": 7075
    },
    {
      "entropy": 5.595941543579102,
      "epoch": 0.7951929016678834,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004947258664185427,
      "loss": 5.5844,
      "mean_token_accuracy": 0.17003095299005508,
      "num_tokens": 15795959.0,
      "step": 7080
    },
    {
      "entropy": 5.658422422409058,
      "epoch": 0.7957544785758409,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004947172209867581,
      "loss": 5.4638,
      "mean_token_accuracy": 0.18406398892402648,
      "num_tokens": 15805561.0,
      "step": 7085
    },
    {
      "entropy": 5.713173961639404,
      "epoch": 0.7963160554837985,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004947085685590919,
      "loss": 5.5692,
      "mean_token_accuracy": 0.17338566929101945,
      "num_tokens": 15816883.0,
      "step": 7090
    },
    {
      "entropy": 5.580157470703125,
      "epoch": 0.796877632391756,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004946999091358196,
      "loss": 5.5261,
      "mean_token_accuracy": 0.17477951496839522,
      "num_tokens": 15827753.0,
      "step": 7095
    },
    {
      "entropy": 5.665834951400757,
      "epoch": 0.7974392092997136,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000494691242717217,
      "loss": 5.6449,
      "mean_token_accuracy": 0.1671520546078682,
      "num_tokens": 15840346.0,
      "step": 7100
    },
    {
      "entropy": 5.718213987350464,
      "epoch": 0.7980007862076711,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004946825693035601,
      "loss": 5.5982,
      "mean_token_accuracy": 0.17325277775526046,
      "num_tokens": 15851036.0,
      "step": 7105
    },
    {
      "entropy": 5.6614522457122805,
      "epoch": 0.7985623631156287,
      "grad_norm": 1.25,
      "learning_rate": 0.000494673888895125,
      "loss": 5.5575,
      "mean_token_accuracy": 0.16837076246738433,
      "num_tokens": 15863154.0,
      "step": 7110
    },
    {
      "entropy": 5.720604181289673,
      "epoch": 0.7991239400235862,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004946652014921881,
      "loss": 5.6639,
      "mean_token_accuracy": 0.16528711915016175,
      "num_tokens": 15874543.0,
      "step": 7115
    },
    {
      "entropy": 5.706193590164185,
      "epoch": 0.7996855169315438,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004946565070950259,
      "loss": 5.5968,
      "mean_token_accuracy": 0.17797693461179734,
      "num_tokens": 15886221.0,
      "step": 7120
    },
    {
      "entropy": 5.709640121459961,
      "epoch": 0.8002470938395013,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004946478057039153,
      "loss": 5.6053,
      "mean_token_accuracy": 0.17097271531820296,
      "num_tokens": 15897602.0,
      "step": 7125
    },
    {
      "entropy": 5.709883308410644,
      "epoch": 0.8008086707474589,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004946390973191335,
      "loss": 5.6058,
      "mean_token_accuracy": 0.171327905356884,
      "num_tokens": 15908030.0,
      "step": 7130
    },
    {
      "entropy": 5.781009912490845,
      "epoch": 0.8013702476554164,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004946303819409575,
      "loss": 5.6467,
      "mean_token_accuracy": 0.17163679897785186,
      "num_tokens": 15918341.0,
      "step": 7135
    },
    {
      "entropy": 5.623637104034424,
      "epoch": 0.801931824563374,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004946216595696652,
      "loss": 5.6264,
      "mean_token_accuracy": 0.17764584273099898,
      "num_tokens": 15929144.0,
      "step": 7140
    },
    {
      "entropy": 5.7017920970916744,
      "epoch": 0.8024934014713315,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004946129302055338,
      "loss": 5.5534,
      "mean_token_accuracy": 0.1748026505112648,
      "num_tokens": 15940013.0,
      "step": 7145
    },
    {
      "entropy": 5.746622562408447,
      "epoch": 0.8030549783792891,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004946041938488417,
      "loss": 5.6151,
      "mean_token_accuracy": 0.17079914808273317,
      "num_tokens": 15950680.0,
      "step": 7150
    },
    {
      "entropy": 5.644058322906494,
      "epoch": 0.8036165552872466,
      "grad_norm": 1.625,
      "learning_rate": 0.0004945954504998668,
      "loss": 5.6012,
      "mean_token_accuracy": 0.17851518392562865,
      "num_tokens": 15961462.0,
      "step": 7155
    },
    {
      "entropy": 5.67428412437439,
      "epoch": 0.8041781321952042,
      "grad_norm": 1.375,
      "learning_rate": 0.0004945867001588877,
      "loss": 5.5232,
      "mean_token_accuracy": 0.17970660775899888,
      "num_tokens": 15972788.0,
      "step": 7160
    },
    {
      "entropy": 5.655921173095703,
      "epoch": 0.8047397091031617,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004945779428261827,
      "loss": 5.5807,
      "mean_token_accuracy": 0.17457723915576934,
      "num_tokens": 15983244.0,
      "step": 7165
    },
    {
      "entropy": 5.596137046813965,
      "epoch": 0.8053012860111193,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000494569178502031,
      "loss": 5.6844,
      "mean_token_accuracy": 0.16675399839878083,
      "num_tokens": 15994325.0,
      "step": 7170
    },
    {
      "entropy": 5.77509298324585,
      "epoch": 0.8058628629190767,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004945604071867114,
      "loss": 5.6754,
      "mean_token_accuracy": 0.1712609902024269,
      "num_tokens": 16005237.0,
      "step": 7175
    },
    {
      "entropy": 5.679333925247192,
      "epoch": 0.8064244398270343,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004945516288805035,
      "loss": 5.5028,
      "mean_token_accuracy": 0.180667944252491,
      "num_tokens": 16015943.0,
      "step": 7180
    },
    {
      "entropy": 5.576353931427002,
      "epoch": 0.8069860167349918,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004945428435836864,
      "loss": 5.5114,
      "mean_token_accuracy": 0.17413717806339263,
      "num_tokens": 16026727.0,
      "step": 7185
    },
    {
      "entropy": 5.622631072998047,
      "epoch": 0.8075475936429494,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004945340512965401,
      "loss": 5.5532,
      "mean_token_accuracy": 0.16718171387910843,
      "num_tokens": 16039021.0,
      "step": 7190
    },
    {
      "entropy": 5.626061344146729,
      "epoch": 0.8081091705509069,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004945252520193443,
      "loss": 5.5031,
      "mean_token_accuracy": 0.17845340073108673,
      "num_tokens": 16050312.0,
      "step": 7195
    },
    {
      "entropy": 5.668435192108154,
      "epoch": 0.8086707474588645,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004945164457523795,
      "loss": 5.4561,
      "mean_token_accuracy": 0.18104152232408524,
      "num_tokens": 16060009.0,
      "step": 7200
    },
    {
      "entropy": 5.589456796646118,
      "epoch": 0.809232324366822,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004945076324959258,
      "loss": 5.5798,
      "mean_token_accuracy": 0.17277259826660157,
      "num_tokens": 16072025.0,
      "step": 7205
    },
    {
      "entropy": 5.730244207382202,
      "epoch": 0.8097939012747796,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004944988122502641,
      "loss": 5.6001,
      "mean_token_accuracy": 0.17296409606933594,
      "num_tokens": 16083229.0,
      "step": 7210
    },
    {
      "entropy": 5.6034921169281,
      "epoch": 0.8103554781827371,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000494489985015675,
      "loss": 5.4848,
      "mean_token_accuracy": 0.17338486462831498,
      "num_tokens": 16094936.0,
      "step": 7215
    },
    {
      "entropy": 5.629249382019043,
      "epoch": 0.8109170550906947,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004944811507924396,
      "loss": 5.5966,
      "mean_token_accuracy": 0.17871529906988143,
      "num_tokens": 16106182.0,
      "step": 7220
    },
    {
      "entropy": 5.612615060806275,
      "epoch": 0.8114786319986522,
      "grad_norm": 1.25,
      "learning_rate": 0.0004944723095808393,
      "loss": 5.5095,
      "mean_token_accuracy": 0.18377710580825807,
      "num_tokens": 16117006.0,
      "step": 7225
    },
    {
      "entropy": 5.6077577590942385,
      "epoch": 0.8120402089066098,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004944634613811555,
      "loss": 5.5422,
      "mean_token_accuracy": 0.1714983433485031,
      "num_tokens": 16128892.0,
      "step": 7230
    },
    {
      "entropy": 5.727390718460083,
      "epoch": 0.8126017858145673,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00049445460619367,
      "loss": 5.6503,
      "mean_token_accuracy": 0.17741401940584184,
      "num_tokens": 16139953.0,
      "step": 7235
    },
    {
      "entropy": 5.771072483062744,
      "epoch": 0.8131633627225249,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004944457440186648,
      "loss": 5.6187,
      "mean_token_accuracy": 0.1663561910390854,
      "num_tokens": 16150103.0,
      "step": 7240
    },
    {
      "entropy": 5.621788215637207,
      "epoch": 0.8137249396304824,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004944368748564219,
      "loss": 5.7002,
      "mean_token_accuracy": 0.167274634540081,
      "num_tokens": 16161666.0,
      "step": 7245
    },
    {
      "entropy": 5.6615965366363525,
      "epoch": 0.81428651653844,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004944279987072238,
      "loss": 5.6388,
      "mean_token_accuracy": 0.16890483051538469,
      "num_tokens": 16172524.0,
      "step": 7250
    },
    {
      "entropy": 5.7637022018432615,
      "epoch": 0.8148480934463975,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004944191155713531,
      "loss": 5.5756,
      "mean_token_accuracy": 0.1765427216887474,
      "num_tokens": 16183023.0,
      "step": 7255
    },
    {
      "entropy": 5.686658239364624,
      "epoch": 0.8154096703543551,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004944102254490928,
      "loss": 5.5918,
      "mean_token_accuracy": 0.17796089202165605,
      "num_tokens": 16193504.0,
      "step": 7260
    },
    {
      "entropy": 5.630303049087525,
      "epoch": 0.8159712472623126,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004944013283407257,
      "loss": 5.5496,
      "mean_token_accuracy": 0.17709226608276368,
      "num_tokens": 16204730.0,
      "step": 7265
    },
    {
      "entropy": 5.6650426387786865,
      "epoch": 0.8165328241702701,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004943924242465352,
      "loss": 5.5613,
      "mean_token_accuracy": 0.1778118908405304,
      "num_tokens": 16214496.0,
      "step": 7270
    },
    {
      "entropy": 5.608745145797729,
      "epoch": 0.8170944010782276,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004943835131668049,
      "loss": 5.5023,
      "mean_token_accuracy": 0.18091920465230943,
      "num_tokens": 16225494.0,
      "step": 7275
    },
    {
      "entropy": 5.569278240203857,
      "epoch": 0.8176559779861852,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004943745951018184,
      "loss": 5.4483,
      "mean_token_accuracy": 0.17797216773033142,
      "num_tokens": 16236385.0,
      "step": 7280
    },
    {
      "entropy": 5.640453624725342,
      "epoch": 0.8182175548941427,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004943656700518597,
      "loss": 5.6211,
      "mean_token_accuracy": 0.1828884169459343,
      "num_tokens": 16247300.0,
      "step": 7285
    },
    {
      "entropy": 5.715027952194214,
      "epoch": 0.8187791318021003,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004943567380172129,
      "loss": 5.6553,
      "mean_token_accuracy": 0.17173643559217452,
      "num_tokens": 16259326.0,
      "step": 7290
    },
    {
      "entropy": 5.711592769622802,
      "epoch": 0.8193407087100578,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004943477989981627,
      "loss": 5.6291,
      "mean_token_accuracy": 0.17496313005685807,
      "num_tokens": 16270559.0,
      "step": 7295
    },
    {
      "entropy": 5.681230926513672,
      "epoch": 0.8199022856180154,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004943388529949933,
      "loss": 5.6011,
      "mean_token_accuracy": 0.16669342219829558,
      "num_tokens": 16283120.0,
      "step": 7300
    },
    {
      "entropy": 5.798020315170288,
      "epoch": 0.8204638625259729,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004943299000079899,
      "loss": 5.6213,
      "mean_token_accuracy": 0.17745826691389083,
      "num_tokens": 16293449.0,
      "step": 7305
    },
    {
      "entropy": 5.665607500076294,
      "epoch": 0.8210254394339305,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004943209400374373,
      "loss": 5.6342,
      "mean_token_accuracy": 0.16801470518112183,
      "num_tokens": 16305144.0,
      "step": 7310
    },
    {
      "entropy": 5.592559432983398,
      "epoch": 0.821587016341888,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000494311973083621,
      "loss": 5.4582,
      "mean_token_accuracy": 0.17840937823057174,
      "num_tokens": 16316728.0,
      "step": 7315
    },
    {
      "entropy": 5.7071305274963375,
      "epoch": 0.8221485932498456,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004943029991468264,
      "loss": 5.7375,
      "mean_token_accuracy": 0.16938112676143646,
      "num_tokens": 16329117.0,
      "step": 7320
    },
    {
      "entropy": 5.769921779632568,
      "epoch": 0.8227101701578031,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004942940182273393,
      "loss": 5.6999,
      "mean_token_accuracy": 0.1706177443265915,
      "num_tokens": 16341651.0,
      "step": 7325
    },
    {
      "entropy": 5.79766092300415,
      "epoch": 0.8232717470657607,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004942850303254455,
      "loss": 5.6691,
      "mean_token_accuracy": 0.17348309159278869,
      "num_tokens": 16353220.0,
      "step": 7330
    },
    {
      "entropy": 5.774486017227173,
      "epoch": 0.8238333239737182,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004942760354414313,
      "loss": 5.706,
      "mean_token_accuracy": 0.1695014789700508,
      "num_tokens": 16363748.0,
      "step": 7335
    },
    {
      "entropy": 5.521135091781616,
      "epoch": 0.8243949008816758,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004942670335755831,
      "loss": 5.3854,
      "mean_token_accuracy": 0.18615321964025497,
      "num_tokens": 16374086.0,
      "step": 7340
    },
    {
      "entropy": 5.6541694641113285,
      "epoch": 0.8249564777896333,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004942580247281876,
      "loss": 5.5564,
      "mean_token_accuracy": 0.17574853003025054,
      "num_tokens": 16384140.0,
      "step": 7345
    },
    {
      "entropy": 5.707195806503296,
      "epoch": 0.8255180546975909,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004942490088995315,
      "loss": 5.6636,
      "mean_token_accuracy": 0.1692671984434128,
      "num_tokens": 16395940.0,
      "step": 7350
    },
    {
      "entropy": 5.629947328567505,
      "epoch": 0.8260796316055484,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004942399860899018,
      "loss": 5.5087,
      "mean_token_accuracy": 0.18015119284391404,
      "num_tokens": 16406667.0,
      "step": 7355
    },
    {
      "entropy": 5.7489300727844235,
      "epoch": 0.826641208513506,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004942309562995862,
      "loss": 5.6606,
      "mean_token_accuracy": 0.165461665391922,
      "num_tokens": 16418621.0,
      "step": 7360
    },
    {
      "entropy": 5.661215448379517,
      "epoch": 0.8272027854214634,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004942219195288716,
      "loss": 5.5629,
      "mean_token_accuracy": 0.1782558485865593,
      "num_tokens": 16429576.0,
      "step": 7365
    },
    {
      "entropy": 5.675850677490234,
      "epoch": 0.827764362329421,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004942128757780463,
      "loss": 5.6791,
      "mean_token_accuracy": 0.16648157685995102,
      "num_tokens": 16441004.0,
      "step": 7370
    },
    {
      "entropy": 5.695897388458252,
      "epoch": 0.8283259392373785,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004942038250473979,
      "loss": 5.6208,
      "mean_token_accuracy": 0.16576655954122543,
      "num_tokens": 16452183.0,
      "step": 7375
    },
    {
      "entropy": 5.650100517272949,
      "epoch": 0.8288875161453361,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004941947673372147,
      "loss": 5.5744,
      "mean_token_accuracy": 0.17556263953447343,
      "num_tokens": 16463552.0,
      "step": 7380
    },
    {
      "entropy": 5.723178291320801,
      "epoch": 0.8294490930532936,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004941857026477851,
      "loss": 5.6099,
      "mean_token_accuracy": 0.1799635797739029,
      "num_tokens": 16474062.0,
      "step": 7385
    },
    {
      "entropy": 5.752410173416138,
      "epoch": 0.8300106699612512,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004941766309793978,
      "loss": 5.6976,
      "mean_token_accuracy": 0.17354531735181808,
      "num_tokens": 16487377.0,
      "step": 7390
    },
    {
      "entropy": 5.755760908126831,
      "epoch": 0.8305722468692087,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004941675523323414,
      "loss": 5.6332,
      "mean_token_accuracy": 0.18097470849752426,
      "num_tokens": 16499704.0,
      "step": 7395
    },
    {
      "entropy": 5.6458765983581545,
      "epoch": 0.8311338237771663,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004941584667069052,
      "loss": 5.6438,
      "mean_token_accuracy": 0.17571662366390228,
      "num_tokens": 16511954.0,
      "step": 7400
    },
    {
      "entropy": 5.5650115489959715,
      "epoch": 0.8316954006851238,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004941493741033784,
      "loss": 5.479,
      "mean_token_accuracy": 0.1725131541490555,
      "num_tokens": 16522602.0,
      "step": 7405
    },
    {
      "entropy": 5.740706253051758,
      "epoch": 0.8322569775930814,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004941402745220507,
      "loss": 5.6325,
      "mean_token_accuracy": 0.16962281316518785,
      "num_tokens": 16534233.0,
      "step": 7410
    },
    {
      "entropy": 5.7556068897247314,
      "epoch": 0.8328185545010389,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004941311679632115,
      "loss": 5.7121,
      "mean_token_accuracy": 0.17037609368562698,
      "num_tokens": 16546120.0,
      "step": 7415
    },
    {
      "entropy": 5.709545755386353,
      "epoch": 0.8333801314089965,
      "grad_norm": 1.421875,
      "learning_rate": 0.000494122054427151,
      "loss": 5.5935,
      "mean_token_accuracy": 0.1746015027165413,
      "num_tokens": 16556830.0,
      "step": 7420
    },
    {
      "entropy": 5.644628953933716,
      "epoch": 0.833941708316954,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004941129339141592,
      "loss": 5.5573,
      "mean_token_accuracy": 0.17804006040096282,
      "num_tokens": 16567436.0,
      "step": 7425
    },
    {
      "entropy": 5.660109615325927,
      "epoch": 0.8345032852249116,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004941038064245267,
      "loss": 5.5257,
      "mean_token_accuracy": 0.1747888594865799,
      "num_tokens": 16577807.0,
      "step": 7430
    },
    {
      "entropy": 5.707765102386475,
      "epoch": 0.8350648621328691,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000494094671958544,
      "loss": 5.6211,
      "mean_token_accuracy": 0.17427547425031661,
      "num_tokens": 16589745.0,
      "step": 7435
    },
    {
      "entropy": 5.779169940948487,
      "epoch": 0.8356264390408267,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004940855305165019,
      "loss": 5.6572,
      "mean_token_accuracy": 0.17683361768722533,
      "num_tokens": 16600727.0,
      "step": 7440
    },
    {
      "entropy": 5.661066389083862,
      "epoch": 0.8361880159487842,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004940763820986916,
      "loss": 5.5597,
      "mean_token_accuracy": 0.17430997490882874,
      "num_tokens": 16611091.0,
      "step": 7445
    },
    {
      "entropy": 5.582389688491821,
      "epoch": 0.8367495928567418,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004940672267054044,
      "loss": 5.5401,
      "mean_token_accuracy": 0.17366688549518586,
      "num_tokens": 16622550.0,
      "step": 7450
    },
    {
      "entropy": 5.662583303451538,
      "epoch": 0.8373111697646993,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004940580643369316,
      "loss": 5.5482,
      "mean_token_accuracy": 0.17264899015426635,
      "num_tokens": 16634243.0,
      "step": 7455
    },
    {
      "entropy": 5.637042140960693,
      "epoch": 0.8378727466726568,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004940488949935651,
      "loss": 5.5622,
      "mean_token_accuracy": 0.1711127832531929,
      "num_tokens": 16645266.0,
      "step": 7460
    },
    {
      "entropy": 5.686684608459473,
      "epoch": 0.8384343235806143,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004940397186755968,
      "loss": 5.5659,
      "mean_token_accuracy": 0.17493919879198075,
      "num_tokens": 16654365.0,
      "step": 7465
    },
    {
      "entropy": 5.683757019042969,
      "epoch": 0.8389959004885719,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004940305353833191,
      "loss": 5.5409,
      "mean_token_accuracy": 0.18146927505731583,
      "num_tokens": 16665251.0,
      "step": 7470
    },
    {
      "entropy": 5.676362562179565,
      "epoch": 0.8395574773965294,
      "grad_norm": 1.390625,
      "learning_rate": 0.000494021345117024,
      "loss": 5.5938,
      "mean_token_accuracy": 0.17529145181179046,
      "num_tokens": 16676209.0,
      "step": 7475
    },
    {
      "entropy": 5.618336915969849,
      "epoch": 0.840119054304487,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004940121478770044,
      "loss": 5.5403,
      "mean_token_accuracy": 0.17493845522403717,
      "num_tokens": 16687213.0,
      "step": 7480
    },
    {
      "entropy": 5.630944728851318,
      "epoch": 0.8406806312124445,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004940029436635531,
      "loss": 5.5802,
      "mean_token_accuracy": 0.18078657239675522,
      "num_tokens": 16697920.0,
      "step": 7485
    },
    {
      "entropy": 5.627004718780517,
      "epoch": 0.8412422081204021,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004939937324769631,
      "loss": 5.5623,
      "mean_token_accuracy": 0.17084627151489257,
      "num_tokens": 16707815.0,
      "step": 7490
    },
    {
      "entropy": 5.632286834716797,
      "epoch": 0.8418037850283596,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004939845143175276,
      "loss": 5.4714,
      "mean_token_accuracy": 0.1783707097172737,
      "num_tokens": 16719221.0,
      "step": 7495
    },
    {
      "entropy": 5.597582626342773,
      "epoch": 0.8423653619363172,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004939752891855404,
      "loss": 5.5105,
      "mean_token_accuracy": 0.17155885994434356,
      "num_tokens": 16731741.0,
      "step": 7500
    },
    {
      "entropy": 5.613684749603271,
      "epoch": 0.8429269388442747,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004939660570812951,
      "loss": 5.5279,
      "mean_token_accuracy": 0.18077764958143233,
      "num_tokens": 16742795.0,
      "step": 7505
    },
    {
      "entropy": 5.636168098449707,
      "epoch": 0.8434885157522323,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004939568180050855,
      "loss": 5.524,
      "mean_token_accuracy": 0.177456533908844,
      "num_tokens": 16753139.0,
      "step": 7510
    },
    {
      "entropy": 5.653849363327026,
      "epoch": 0.8440500926601898,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004939475719572059,
      "loss": 5.5368,
      "mean_token_accuracy": 0.17530351132154465,
      "num_tokens": 16764246.0,
      "step": 7515
    },
    {
      "entropy": 5.70451078414917,
      "epoch": 0.8446116695681474,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004939383189379507,
      "loss": 5.6509,
      "mean_token_accuracy": 0.1700239211320877,
      "num_tokens": 16776039.0,
      "step": 7520
    },
    {
      "entropy": 5.718337535858154,
      "epoch": 0.8451732464761049,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004939290589476145,
      "loss": 5.6344,
      "mean_token_accuracy": 0.17975671738386154,
      "num_tokens": 16786979.0,
      "step": 7525
    },
    {
      "entropy": 5.705946540832519,
      "epoch": 0.8457348233840625,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004939197919864922,
      "loss": 5.6571,
      "mean_token_accuracy": 0.17143665701150895,
      "num_tokens": 16798113.0,
      "step": 7530
    },
    {
      "entropy": 5.6824140548706055,
      "epoch": 0.84629640029202,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004939105180548788,
      "loss": 5.5921,
      "mean_token_accuracy": 0.17321973741054536,
      "num_tokens": 16809072.0,
      "step": 7535
    },
    {
      "entropy": 5.61365909576416,
      "epoch": 0.8468579771999776,
      "grad_norm": 1.5,
      "learning_rate": 0.0004939012371530696,
      "loss": 5.3964,
      "mean_token_accuracy": 0.17975719422101974,
      "num_tokens": 16819548.0,
      "step": 7540
    },
    {
      "entropy": 5.733366680145264,
      "epoch": 0.8474195541079351,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004938919492813601,
      "loss": 5.7082,
      "mean_token_accuracy": 0.16560092866420745,
      "num_tokens": 16831935.0,
      "step": 7545
    },
    {
      "entropy": 5.675737428665161,
      "epoch": 0.8479811310158927,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004938826544400461,
      "loss": 5.604,
      "mean_token_accuracy": 0.17931524962186812,
      "num_tokens": 16842961.0,
      "step": 7550
    },
    {
      "entropy": 5.6335454940795895,
      "epoch": 0.8485427079238502,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004938733526294234,
      "loss": 5.5092,
      "mean_token_accuracy": 0.17333014458417892,
      "num_tokens": 16854392.0,
      "step": 7555
    },
    {
      "entropy": 5.620349311828614,
      "epoch": 0.8491042848318077,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004938640438497883,
      "loss": 5.5988,
      "mean_token_accuracy": 0.16811076998710633,
      "num_tokens": 16865443.0,
      "step": 7560
    },
    {
      "entropy": 5.677863216400146,
      "epoch": 0.8496658617397652,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004938547281014372,
      "loss": 5.5517,
      "mean_token_accuracy": 0.17474559247493743,
      "num_tokens": 16876854.0,
      "step": 7565
    },
    {
      "entropy": 5.7703766345977785,
      "epoch": 0.8502274386477228,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004938454053846666,
      "loss": 5.6388,
      "mean_token_accuracy": 0.1716684192419052,
      "num_tokens": 16888823.0,
      "step": 7570
    },
    {
      "entropy": 5.61264853477478,
      "epoch": 0.8507890155556803,
      "grad_norm": 1.875,
      "learning_rate": 0.0004938360756997735,
      "loss": 5.565,
      "mean_token_accuracy": 0.1738715797662735,
      "num_tokens": 16898807.0,
      "step": 7575
    },
    {
      "entropy": 5.690545082092285,
      "epoch": 0.8513505924636379,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004938267390470547,
      "loss": 5.6812,
      "mean_token_accuracy": 0.16348067820072174,
      "num_tokens": 16911258.0,
      "step": 7580
    },
    {
      "entropy": 5.783576726913452,
      "epoch": 0.8519121693715954,
      "grad_norm": 1.625,
      "learning_rate": 0.0004938173954268079,
      "loss": 5.6417,
      "mean_token_accuracy": 0.17234276682138444,
      "num_tokens": 16922557.0,
      "step": 7585
    },
    {
      "entropy": 5.723300170898438,
      "epoch": 0.852473746279553,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004938080448393302,
      "loss": 5.66,
      "mean_token_accuracy": 0.16851891279220582,
      "num_tokens": 16935330.0,
      "step": 7590
    },
    {
      "entropy": 5.723967742919922,
      "epoch": 0.8530353231875105,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004937986872849195,
      "loss": 5.5626,
      "mean_token_accuracy": 0.1776669666171074,
      "num_tokens": 16945361.0,
      "step": 7595
    },
    {
      "entropy": 5.565654373168945,
      "epoch": 0.8535969000954681,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004937893227638738,
      "loss": 5.4383,
      "mean_token_accuracy": 0.18152663558721543,
      "num_tokens": 16955363.0,
      "step": 7600
    },
    {
      "entropy": 5.56234769821167,
      "epoch": 0.8541584770034256,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004937799512764912,
      "loss": 5.5148,
      "mean_token_accuracy": 0.17852590829133988,
      "num_tokens": 16965267.0,
      "step": 7605
    },
    {
      "entropy": 5.671128654479981,
      "epoch": 0.8547200539113832,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00049377057282307,
      "loss": 5.7321,
      "mean_token_accuracy": 0.17176787555217743,
      "num_tokens": 16977528.0,
      "step": 7610
    },
    {
      "entropy": 5.715606927871704,
      "epoch": 0.8552816308193407,
      "grad_norm": 1.796875,
      "learning_rate": 0.000493761187403909,
      "loss": 5.5643,
      "mean_token_accuracy": 0.17219118773937225,
      "num_tokens": 16987326.0,
      "step": 7615
    },
    {
      "entropy": 5.799017238616943,
      "epoch": 0.8558432077272983,
      "grad_norm": 2.078125,
      "learning_rate": 0.000493751795019307,
      "loss": 5.6114,
      "mean_token_accuracy": 0.1744072198867798,
      "num_tokens": 16999127.0,
      "step": 7620
    },
    {
      "entropy": 5.582177782058716,
      "epoch": 0.8564047846352558,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004937423956695629,
      "loss": 5.5232,
      "mean_token_accuracy": 0.18371396213769914,
      "num_tokens": 17009929.0,
      "step": 7625
    },
    {
      "entropy": 5.713424921035767,
      "epoch": 0.8569663615432134,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004937329893549762,
      "loss": 5.6514,
      "mean_token_accuracy": 0.17402349263429642,
      "num_tokens": 17020909.0,
      "step": 7630
    },
    {
      "entropy": 5.744926929473877,
      "epoch": 0.8575279384511709,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004937235760758462,
      "loss": 5.5807,
      "mean_token_accuracy": 0.17321668565273285,
      "num_tokens": 17031358.0,
      "step": 7635
    },
    {
      "entropy": 5.654985761642456,
      "epoch": 0.8580895153591285,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004937141558324726,
      "loss": 5.6242,
      "mean_token_accuracy": 0.17349520176649094,
      "num_tokens": 17042566.0,
      "step": 7640
    },
    {
      "entropy": 5.677562236785889,
      "epoch": 0.858651092267086,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004937047286251556,
      "loss": 5.6472,
      "mean_token_accuracy": 0.16312987059354783,
      "num_tokens": 17054998.0,
      "step": 7645
    },
    {
      "entropy": 5.6565553665161135,
      "epoch": 0.8592126691750436,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004936952944541952,
      "loss": 5.512,
      "mean_token_accuracy": 0.17477733939886092,
      "num_tokens": 17064876.0,
      "step": 7650
    },
    {
      "entropy": 5.647423267364502,
      "epoch": 0.859774246083001,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004936858533198917,
      "loss": 5.5621,
      "mean_token_accuracy": 0.17949966639280318,
      "num_tokens": 17075310.0,
      "step": 7655
    },
    {
      "entropy": 5.519606304168701,
      "epoch": 0.8603358229909586,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000493676405222546,
      "loss": 5.4676,
      "mean_token_accuracy": 0.17772774249315262,
      "num_tokens": 17086023.0,
      "step": 7660
    },
    {
      "entropy": 5.5962433338165285,
      "epoch": 0.8608973998989161,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004936669501624586,
      "loss": 5.5539,
      "mean_token_accuracy": 0.17585509717464448,
      "num_tokens": 17097554.0,
      "step": 7665
    },
    {
      "entropy": 5.8958173274993895,
      "epoch": 0.8614589768068737,
      "grad_norm": 6.15625,
      "learning_rate": 0.0004936574881399308,
      "loss": 5.6789,
      "mean_token_accuracy": 0.17030972838401795,
      "num_tokens": 17108890.0,
      "step": 7670
    },
    {
      "entropy": 5.694438219070435,
      "epoch": 0.8620205537148312,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004936480191552637,
      "loss": 5.5134,
      "mean_token_accuracy": 0.17637402266263963,
      "num_tokens": 17120651.0,
      "step": 7675
    },
    {
      "entropy": 5.666385555267334,
      "epoch": 0.8625821306227888,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004936385432087589,
      "loss": 5.651,
      "mean_token_accuracy": 0.17565001398324967,
      "num_tokens": 17131306.0,
      "step": 7680
    },
    {
      "entropy": 5.812044906616211,
      "epoch": 0.8631437075307463,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004936290603007181,
      "loss": 5.798,
      "mean_token_accuracy": 0.16602026224136351,
      "num_tokens": 17142779.0,
      "step": 7685
    },
    {
      "entropy": 5.860309505462647,
      "epoch": 0.8637052844387039,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004936195704314432,
      "loss": 5.6687,
      "mean_token_accuracy": 0.18110954910516738,
      "num_tokens": 17154992.0,
      "step": 7690
    },
    {
      "entropy": 5.662728500366211,
      "epoch": 0.8642668613466614,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004936100736012363,
      "loss": 5.6427,
      "mean_token_accuracy": 0.17859661132097243,
      "num_tokens": 17166089.0,
      "step": 7695
    },
    {
      "entropy": 5.610977554321289,
      "epoch": 0.864828438254619,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004936005698104,
      "loss": 5.5468,
      "mean_token_accuracy": 0.1707144245505333,
      "num_tokens": 17175691.0,
      "step": 7700
    },
    {
      "entropy": 5.6774708271026615,
      "epoch": 0.8653900151625765,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004935910590592367,
      "loss": 5.602,
      "mean_token_accuracy": 0.16899661272764205,
      "num_tokens": 17186466.0,
      "step": 7705
    },
    {
      "entropy": 5.582030820846557,
      "epoch": 0.8659515920705341,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004935815413480494,
      "loss": 5.3873,
      "mean_token_accuracy": 0.18827227801084517,
      "num_tokens": 17198520.0,
      "step": 7710
    },
    {
      "entropy": 5.670034170150757,
      "epoch": 0.8665131689784916,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000493572016677141,
      "loss": 5.6234,
      "mean_token_accuracy": 0.16890666335821153,
      "num_tokens": 17210496.0,
      "step": 7715
    },
    {
      "entropy": 5.68838415145874,
      "epoch": 0.8670747458864492,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004935624850468148,
      "loss": 5.6152,
      "mean_token_accuracy": 0.17695807665586472,
      "num_tokens": 17221631.0,
      "step": 7720
    },
    {
      "entropy": 5.682717370986938,
      "epoch": 0.8676363227944067,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004935529464573742,
      "loss": 5.65,
      "mean_token_accuracy": 0.16548071503639222,
      "num_tokens": 17233316.0,
      "step": 7725
    },
    {
      "entropy": 5.678200769424438,
      "epoch": 0.8681978997023643,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004935434009091232,
      "loss": 5.5573,
      "mean_token_accuracy": 0.1785666510462761,
      "num_tokens": 17245761.0,
      "step": 7730
    },
    {
      "entropy": 5.697708415985107,
      "epoch": 0.8687594766103218,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004935338484023655,
      "loss": 5.5596,
      "mean_token_accuracy": 0.1804763749241829,
      "num_tokens": 17255007.0,
      "step": 7735
    },
    {
      "entropy": 5.644683218002319,
      "epoch": 0.8693210535182794,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004935242889374052,
      "loss": 5.536,
      "mean_token_accuracy": 0.17835524082183837,
      "num_tokens": 17265378.0,
      "step": 7740
    },
    {
      "entropy": 5.724565315246582,
      "epoch": 0.8698826304262369,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004935147225145469,
      "loss": 5.6105,
      "mean_token_accuracy": 0.17023085206747054,
      "num_tokens": 17277202.0,
      "step": 7745
    },
    {
      "entropy": 5.6139226913452145,
      "epoch": 0.8704442073341944,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004935051491340951,
      "loss": 5.5543,
      "mean_token_accuracy": 0.1805954396724701,
      "num_tokens": 17287006.0,
      "step": 7750
    },
    {
      "entropy": 5.649082994461059,
      "epoch": 0.8710057842421519,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004934955687963546,
      "loss": 5.52,
      "mean_token_accuracy": 0.18563045263290406,
      "num_tokens": 17297387.0,
      "step": 7755
    },
    {
      "entropy": 5.797161531448364,
      "epoch": 0.8715673611501095,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004934859815016304,
      "loss": 5.7537,
      "mean_token_accuracy": 0.16585738137364386,
      "num_tokens": 17310364.0,
      "step": 7760
    },
    {
      "entropy": 5.6904988288879395,
      "epoch": 0.872128938058067,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004934763872502278,
      "loss": 5.5897,
      "mean_token_accuracy": 0.17995432764291763,
      "num_tokens": 17321282.0,
      "step": 7765
    },
    {
      "entropy": 5.684155511856079,
      "epoch": 0.8726905149660246,
      "grad_norm": 1.625,
      "learning_rate": 0.0004934667860424523,
      "loss": 5.632,
      "mean_token_accuracy": 0.1700143799185753,
      "num_tokens": 17331755.0,
      "step": 7770
    },
    {
      "entropy": 5.656625986099243,
      "epoch": 0.8732520918739821,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004934571778786096,
      "loss": 5.5745,
      "mean_token_accuracy": 0.17331324964761735,
      "num_tokens": 17342542.0,
      "step": 7775
    },
    {
      "entropy": 5.652045154571534,
      "epoch": 0.8738136687819397,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004934475627590057,
      "loss": 5.5215,
      "mean_token_accuracy": 0.17376778274774551,
      "num_tokens": 17353150.0,
      "step": 7780
    },
    {
      "entropy": 5.65459680557251,
      "epoch": 0.8743752456898972,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004934379406839466,
      "loss": 5.5644,
      "mean_token_accuracy": 0.17380731254816056,
      "num_tokens": 17363901.0,
      "step": 7785
    },
    {
      "entropy": 5.67426061630249,
      "epoch": 0.8749368225978548,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004934283116537387,
      "loss": 5.488,
      "mean_token_accuracy": 0.17831778526306152,
      "num_tokens": 17374429.0,
      "step": 7790
    },
    {
      "entropy": 5.673512268066406,
      "epoch": 0.8754983995058123,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004934186756686888,
      "loss": 5.6291,
      "mean_token_accuracy": 0.17127451598644255,
      "num_tokens": 17384952.0,
      "step": 7795
    },
    {
      "entropy": 5.670864820480347,
      "epoch": 0.8760599764137699,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004934090327291034,
      "loss": 5.5948,
      "mean_token_accuracy": 0.17161913961172104,
      "num_tokens": 17397771.0,
      "step": 7800
    },
    {
      "entropy": 5.670876312255859,
      "epoch": 0.8766215533217274,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004933993828352898,
      "loss": 5.5027,
      "mean_token_accuracy": 0.17967257499694825,
      "num_tokens": 17407748.0,
      "step": 7805
    },
    {
      "entropy": 5.611726522445679,
      "epoch": 0.877183130229685,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004933897259875551,
      "loss": 5.4843,
      "mean_token_accuracy": 0.175080007314682,
      "num_tokens": 17418572.0,
      "step": 7810
    },
    {
      "entropy": 5.52510027885437,
      "epoch": 0.8777447071376425,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004933800621862069,
      "loss": 5.5306,
      "mean_token_accuracy": 0.17571943253278732,
      "num_tokens": 17429909.0,
      "step": 7815
    },
    {
      "entropy": 5.638761234283447,
      "epoch": 0.8783062840456001,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004933703914315527,
      "loss": 5.5553,
      "mean_token_accuracy": 0.1714665785431862,
      "num_tokens": 17441143.0,
      "step": 7820
    },
    {
      "entropy": 5.736855459213257,
      "epoch": 0.8788678609535576,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004933607137239006,
      "loss": 5.6724,
      "mean_token_accuracy": 0.16394690573215484,
      "num_tokens": 17452156.0,
      "step": 7825
    },
    {
      "entropy": 5.618521738052368,
      "epoch": 0.8794294378615152,
      "grad_norm": 1.5,
      "learning_rate": 0.0004933510290635587,
      "loss": 5.574,
      "mean_token_accuracy": 0.17211691588163375,
      "num_tokens": 17463873.0,
      "step": 7830
    },
    {
      "entropy": 5.598127937316894,
      "epoch": 0.8799910147694727,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004933413374508353,
      "loss": 5.5071,
      "mean_token_accuracy": 0.17668081372976302,
      "num_tokens": 17474180.0,
      "step": 7835
    },
    {
      "entropy": 5.652832412719727,
      "epoch": 0.8805525916774303,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004933316388860391,
      "loss": 5.539,
      "mean_token_accuracy": 0.1772014617919922,
      "num_tokens": 17486903.0,
      "step": 7840
    },
    {
      "entropy": 5.573214387893676,
      "epoch": 0.8811141685853877,
      "grad_norm": 1.75,
      "learning_rate": 0.0004933219333694787,
      "loss": 5.567,
      "mean_token_accuracy": 0.17032884508371354,
      "num_tokens": 17498790.0,
      "step": 7845
    },
    {
      "entropy": 5.723428106307983,
      "epoch": 0.8816757454933453,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004933122209014634,
      "loss": 5.5983,
      "mean_token_accuracy": 0.16774654984474183,
      "num_tokens": 17510061.0,
      "step": 7850
    },
    {
      "entropy": 5.679811573028564,
      "epoch": 0.8822373224013028,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004933025014823022,
      "loss": 5.6114,
      "mean_token_accuracy": 0.17310027331113814,
      "num_tokens": 17522414.0,
      "step": 7855
    },
    {
      "entropy": 5.8511148452758786,
      "epoch": 0.8827988993092604,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004932927751123048,
      "loss": 5.7443,
      "mean_token_accuracy": 0.16237847954034806,
      "num_tokens": 17532745.0,
      "step": 7860
    },
    {
      "entropy": 5.631840419769287,
      "epoch": 0.8833604762172179,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004932830417917808,
      "loss": 5.5358,
      "mean_token_accuracy": 0.17639891058206558,
      "num_tokens": 17545329.0,
      "step": 7865
    },
    {
      "entropy": 5.729234743118286,
      "epoch": 0.8839220531251755,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00049327330152104,
      "loss": 5.6476,
      "mean_token_accuracy": 0.1724901929497719,
      "num_tokens": 17557789.0,
      "step": 7870
    },
    {
      "entropy": 5.672587299346924,
      "epoch": 0.884483630033133,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004932635543003927,
      "loss": 5.4769,
      "mean_token_accuracy": 0.18150333017110826,
      "num_tokens": 17568338.0,
      "step": 7875
    },
    {
      "entropy": 5.638689804077148,
      "epoch": 0.8850452069410906,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004932538001301491,
      "loss": 5.5781,
      "mean_token_accuracy": 0.1726062461733818,
      "num_tokens": 17581199.0,
      "step": 7880
    },
    {
      "entropy": 5.544777345657349,
      "epoch": 0.8856067838490481,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004932440390106199,
      "loss": 5.5334,
      "mean_token_accuracy": 0.1764208197593689,
      "num_tokens": 17592937.0,
      "step": 7885
    },
    {
      "entropy": 5.629625701904297,
      "epoch": 0.8861683607570057,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004932342709421159,
      "loss": 5.5926,
      "mean_token_accuracy": 0.17708465456962585,
      "num_tokens": 17604003.0,
      "step": 7890
    },
    {
      "entropy": 5.90011510848999,
      "epoch": 0.8867299376649632,
      "grad_norm": 1.3125,
      "learning_rate": 0.000493224495924948,
      "loss": 5.7302,
      "mean_token_accuracy": 0.16930215954780578,
      "num_tokens": 17617283.0,
      "step": 7895
    },
    {
      "entropy": 5.677170133590698,
      "epoch": 0.8872915145729208,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004932147139594276,
      "loss": 5.5707,
      "mean_token_accuracy": 0.1770807087421417,
      "num_tokens": 17628161.0,
      "step": 7900
    },
    {
      "entropy": 5.728165149688721,
      "epoch": 0.8878530914808783,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004932049250458659,
      "loss": 5.6503,
      "mean_token_accuracy": 0.16944217383861543,
      "num_tokens": 17639226.0,
      "step": 7905
    },
    {
      "entropy": 5.592314767837524,
      "epoch": 0.8884146683888359,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004931951291845749,
      "loss": 5.4893,
      "mean_token_accuracy": 0.18333953768014907,
      "num_tokens": 17652114.0,
      "step": 7910
    },
    {
      "entropy": 5.5825001239776615,
      "epoch": 0.8889762452967934,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004931853263758663,
      "loss": 5.529,
      "mean_token_accuracy": 0.18106401562690735,
      "num_tokens": 17663280.0,
      "step": 7915
    },
    {
      "entropy": 5.623987007141113,
      "epoch": 0.889537822204751,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004931755166200523,
      "loss": 5.6155,
      "mean_token_accuracy": 0.17368122637271882,
      "num_tokens": 17674445.0,
      "step": 7920
    },
    {
      "entropy": 5.691004753112793,
      "epoch": 0.8900993991127085,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004931656999174452,
      "loss": 5.5071,
      "mean_token_accuracy": 0.17749142944812774,
      "num_tokens": 17685046.0,
      "step": 7925
    },
    {
      "entropy": 5.591828203201294,
      "epoch": 0.8906609760206661,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004931558762683577,
      "loss": 5.507,
      "mean_token_accuracy": 0.17967617213726045,
      "num_tokens": 17695269.0,
      "step": 7930
    },
    {
      "entropy": 5.559813833236694,
      "epoch": 0.8912225529286236,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004931460456731024,
      "loss": 5.5466,
      "mean_token_accuracy": 0.17714540660381317,
      "num_tokens": 17706296.0,
      "step": 7935
    },
    {
      "entropy": 5.6044251918792725,
      "epoch": 0.8917841298365812,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004931362081319925,
      "loss": 5.4769,
      "mean_token_accuracy": 0.17915261387825013,
      "num_tokens": 17716262.0,
      "step": 7940
    },
    {
      "entropy": 5.66194977760315,
      "epoch": 0.8923457067445386,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004931263636453409,
      "loss": 5.5201,
      "mean_token_accuracy": 0.17992712408304215,
      "num_tokens": 17726818.0,
      "step": 7945
    },
    {
      "entropy": 5.589525651931763,
      "epoch": 0.8929072836524962,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004931165122134616,
      "loss": 5.5256,
      "mean_token_accuracy": 0.18117809742689134,
      "num_tokens": 17737112.0,
      "step": 7950
    },
    {
      "entropy": 5.660979413986206,
      "epoch": 0.8934688605604537,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004931066538366677,
      "loss": 5.5984,
      "mean_token_accuracy": 0.17052928656339644,
      "num_tokens": 17747558.0,
      "step": 7955
    },
    {
      "entropy": 5.712664318084717,
      "epoch": 0.8940304374684113,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004930967885152734,
      "loss": 5.536,
      "mean_token_accuracy": 0.17624609172344208,
      "num_tokens": 17757495.0,
      "step": 7960
    },
    {
      "entropy": 5.624363279342651,
      "epoch": 0.8945920143763688,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004930869162495929,
      "loss": 5.5294,
      "mean_token_accuracy": 0.1765795186161995,
      "num_tokens": 17767562.0,
      "step": 7965
    },
    {
      "entropy": 5.6667194843292235,
      "epoch": 0.8951535912843264,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004930770370399403,
      "loss": 5.6186,
      "mean_token_accuracy": 0.16974660009145737,
      "num_tokens": 17779800.0,
      "step": 7970
    },
    {
      "entropy": 5.610093069076538,
      "epoch": 0.8957151681922839,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004930671508866303,
      "loss": 5.4308,
      "mean_token_accuracy": 0.17751775681972504,
      "num_tokens": 17791183.0,
      "step": 7975
    },
    {
      "entropy": 5.681202840805054,
      "epoch": 0.8962767451002415,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004930572577899776,
      "loss": 5.5975,
      "mean_token_accuracy": 0.1753363460302353,
      "num_tokens": 17802547.0,
      "step": 7980
    },
    {
      "entropy": 5.511297702789307,
      "epoch": 0.896838322008199,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004930473577502972,
      "loss": 5.492,
      "mean_token_accuracy": 0.17547860294580458,
      "num_tokens": 17815406.0,
      "step": 7985
    },
    {
      "entropy": 5.631451606750488,
      "epoch": 0.8973998989161566,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004930374507679044,
      "loss": 5.521,
      "mean_token_accuracy": 0.17684534192085266,
      "num_tokens": 17826516.0,
      "step": 7990
    },
    {
      "entropy": 5.646472406387329,
      "epoch": 0.8979614758241141,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004930275368431146,
      "loss": 5.4978,
      "mean_token_accuracy": 0.1771871417760849,
      "num_tokens": 17838193.0,
      "step": 7995
    },
    {
      "entropy": 5.624480295181274,
      "epoch": 0.8985230527320717,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004930176159762435,
      "loss": 5.5205,
      "mean_token_accuracy": 0.1750570610165596,
      "num_tokens": 17849564.0,
      "step": 8000
    },
    {
      "entropy": 5.736857175827026,
      "epoch": 0.8990846296400292,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004930076881676069,
      "loss": 5.662,
      "mean_token_accuracy": 0.17136270999908448,
      "num_tokens": 17861874.0,
      "step": 8005
    },
    {
      "entropy": 5.575713205337524,
      "epoch": 0.8996462065479868,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004929977534175211,
      "loss": 5.5055,
      "mean_token_accuracy": 0.17983078062534333,
      "num_tokens": 17873139.0,
      "step": 8010
    },
    {
      "entropy": 5.736148262023926,
      "epoch": 0.9002077834559443,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004929878117263021,
      "loss": 5.572,
      "mean_token_accuracy": 0.17753154337406157,
      "num_tokens": 17883262.0,
      "step": 8015
    },
    {
      "entropy": 5.554032325744629,
      "epoch": 0.9007693603639019,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004929778630942668,
      "loss": 5.4604,
      "mean_token_accuracy": 0.18278993517160416,
      "num_tokens": 17895041.0,
      "step": 8020
    },
    {
      "entropy": 5.528722429275513,
      "epoch": 0.9013309372718594,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004929679075217317,
      "loss": 5.5034,
      "mean_token_accuracy": 0.17425708770751952,
      "num_tokens": 17905154.0,
      "step": 8025
    },
    {
      "entropy": 5.616612911224365,
      "epoch": 0.901892514179817,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004929579450090139,
      "loss": 5.5601,
      "mean_token_accuracy": 0.17341381311416626,
      "num_tokens": 17916287.0,
      "step": 8030
    },
    {
      "entropy": 5.7543384552001955,
      "epoch": 0.9024540910877745,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004929479755564305,
      "loss": 5.5673,
      "mean_token_accuracy": 0.17662589699029924,
      "num_tokens": 17927269.0,
      "step": 8035
    },
    {
      "entropy": 5.6863922595977785,
      "epoch": 0.903015667995732,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004929379991642991,
      "loss": 5.5818,
      "mean_token_accuracy": 0.1742201864719391,
      "num_tokens": 17938142.0,
      "step": 8040
    },
    {
      "entropy": 5.675807046890259,
      "epoch": 0.9035772449036895,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004929280158329373,
      "loss": 5.5645,
      "mean_token_accuracy": 0.1757965698838234,
      "num_tokens": 17950229.0,
      "step": 8045
    },
    {
      "entropy": 5.612723159790039,
      "epoch": 0.9041388218116471,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000492918025562663,
      "loss": 5.548,
      "mean_token_accuracy": 0.18230236321687698,
      "num_tokens": 17961757.0,
      "step": 8050
    },
    {
      "entropy": 5.685063695907592,
      "epoch": 0.9047003987196046,
      "grad_norm": 1.5,
      "learning_rate": 0.0004929080283537942,
      "loss": 5.6233,
      "mean_token_accuracy": 0.1663338214159012,
      "num_tokens": 17972564.0,
      "step": 8055
    },
    {
      "entropy": 5.735013771057129,
      "epoch": 0.9052619756275622,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004928980242066492,
      "loss": 5.5325,
      "mean_token_accuracy": 0.18212261348962783,
      "num_tokens": 17982776.0,
      "step": 8060
    },
    {
      "entropy": 5.688648509979248,
      "epoch": 0.9058235525355197,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004928880131215467,
      "loss": 5.6463,
      "mean_token_accuracy": 0.1749667137861252,
      "num_tokens": 17994460.0,
      "step": 8065
    },
    {
      "entropy": 5.624743270874023,
      "epoch": 0.9063851294434773,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004928779950988053,
      "loss": 5.62,
      "mean_token_accuracy": 0.17174448817968369,
      "num_tokens": 18005744.0,
      "step": 8070
    },
    {
      "entropy": 5.611772775650024,
      "epoch": 0.9069467063514348,
      "grad_norm": 1.4375,
      "learning_rate": 0.000492867970138744,
      "loss": 5.4754,
      "mean_token_accuracy": 0.18351303040981293,
      "num_tokens": 18016517.0,
      "step": 8075
    },
    {
      "entropy": 5.675594282150269,
      "epoch": 0.9075082832593924,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004928579382416821,
      "loss": 5.5975,
      "mean_token_accuracy": 0.16978041380643843,
      "num_tokens": 18027653.0,
      "step": 8080
    },
    {
      "entropy": 5.675237369537354,
      "epoch": 0.9080698601673499,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000492847899407939,
      "loss": 5.636,
      "mean_token_accuracy": 0.17045610696077346,
      "num_tokens": 18038741.0,
      "step": 8085
    },
    {
      "entropy": 5.7510435581207275,
      "epoch": 0.9086314370753075,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004928378536378341,
      "loss": 5.5802,
      "mean_token_accuracy": 0.17184377759695052,
      "num_tokens": 18049188.0,
      "step": 8090
    },
    {
      "entropy": 5.608703708648681,
      "epoch": 0.909193013983265,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004928278009316877,
      "loss": 5.482,
      "mean_token_accuracy": 0.17725870013237,
      "num_tokens": 18059979.0,
      "step": 8095
    },
    {
      "entropy": 5.6458415508270265,
      "epoch": 0.9097545908912226,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004928177412898195,
      "loss": 5.4967,
      "mean_token_accuracy": 0.1831892564892769,
      "num_tokens": 18070712.0,
      "step": 8100
    },
    {
      "entropy": 5.670635318756103,
      "epoch": 0.9103161677991801,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004928076747125499,
      "loss": 5.5632,
      "mean_token_accuracy": 0.18162725418806075,
      "num_tokens": 18081372.0,
      "step": 8105
    },
    {
      "entropy": 5.577960062026977,
      "epoch": 0.9108777447071377,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004927976012001995,
      "loss": 5.4888,
      "mean_token_accuracy": 0.17665131241083146,
      "num_tokens": 18093033.0,
      "step": 8110
    },
    {
      "entropy": 5.555771923065185,
      "epoch": 0.9114393216150952,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004927875207530889,
      "loss": 5.4461,
      "mean_token_accuracy": 0.18108128160238265,
      "num_tokens": 18104808.0,
      "step": 8115
    },
    {
      "entropy": 5.730599784851075,
      "epoch": 0.9120008985230528,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004927774333715394,
      "loss": 5.5846,
      "mean_token_accuracy": 0.17475593835115433,
      "num_tokens": 18115450.0,
      "step": 8120
    },
    {
      "entropy": 5.730729341506958,
      "epoch": 0.9125624754310103,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004927673390558718,
      "loss": 5.6786,
      "mean_token_accuracy": 0.17357307821512222,
      "num_tokens": 18127437.0,
      "step": 8125
    },
    {
      "entropy": 5.705981349945068,
      "epoch": 0.9131240523389679,
      "grad_norm": 1.5,
      "learning_rate": 0.0004927572378064077,
      "loss": 5.5123,
      "mean_token_accuracy": 0.1819301575422287,
      "num_tokens": 18137371.0,
      "step": 8130
    },
    {
      "entropy": 5.618573713302612,
      "epoch": 0.9136856292469253,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004927471296234687,
      "loss": 5.5725,
      "mean_token_accuracy": 0.17275052070617675,
      "num_tokens": 18148332.0,
      "step": 8135
    },
    {
      "entropy": 5.588826656341553,
      "epoch": 0.9142472061548829,
      "grad_norm": 1.5,
      "learning_rate": 0.0004927370145073767,
      "loss": 5.5549,
      "mean_token_accuracy": 0.17383279651403427,
      "num_tokens": 18159918.0,
      "step": 8140
    },
    {
      "entropy": 5.65464768409729,
      "epoch": 0.9148087830628404,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004927268924584536,
      "loss": 5.5406,
      "mean_token_accuracy": 0.18160120993852616,
      "num_tokens": 18170911.0,
      "step": 8145
    },
    {
      "entropy": 5.667963266372681,
      "epoch": 0.915370359970798,
      "grad_norm": 1.375,
      "learning_rate": 0.0004927167634770219,
      "loss": 5.5562,
      "mean_token_accuracy": 0.17828303426504136,
      "num_tokens": 18183089.0,
      "step": 8150
    },
    {
      "entropy": 5.7810358047485355,
      "epoch": 0.9159319368787555,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000492706627563404,
      "loss": 5.8173,
      "mean_token_accuracy": 0.15935783535242082,
      "num_tokens": 18195467.0,
      "step": 8155
    },
    {
      "entropy": 5.740857410430908,
      "epoch": 0.9164935137867131,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004926964847179225,
      "loss": 5.572,
      "mean_token_accuracy": 0.1790512040257454,
      "num_tokens": 18206766.0,
      "step": 8160
    },
    {
      "entropy": 5.679031610488892,
      "epoch": 0.9170550906946706,
      "grad_norm": 1.375,
      "learning_rate": 0.0004926863349409006,
      "loss": 5.5629,
      "mean_token_accuracy": 0.17597046494483948,
      "num_tokens": 18217840.0,
      "step": 8165
    },
    {
      "entropy": 5.649716377258301,
      "epoch": 0.9176166676026282,
      "grad_norm": 2.0,
      "learning_rate": 0.0004926761782326613,
      "loss": 5.5758,
      "mean_token_accuracy": 0.17311379611492156,
      "num_tokens": 18229514.0,
      "step": 8170
    },
    {
      "entropy": 5.5854212760925295,
      "epoch": 0.9181782445105857,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004926660145935282,
      "loss": 5.5559,
      "mean_token_accuracy": 0.1705981135368347,
      "num_tokens": 18241949.0,
      "step": 8175
    },
    {
      "entropy": 5.651837682723999,
      "epoch": 0.9187398214185433,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004926558440238247,
      "loss": 5.5314,
      "mean_token_accuracy": 0.17833069413900376,
      "num_tokens": 18252938.0,
      "step": 8180
    },
    {
      "entropy": 5.624422931671143,
      "epoch": 0.9193013983265008,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004926456665238746,
      "loss": 5.479,
      "mean_token_accuracy": 0.18345132768154143,
      "num_tokens": 18263417.0,
      "step": 8185
    },
    {
      "entropy": 5.661493015289307,
      "epoch": 0.9198629752344584,
      "grad_norm": 1.375,
      "learning_rate": 0.0004926354820940023,
      "loss": 5.6485,
      "mean_token_accuracy": 0.17259145379066468,
      "num_tokens": 18275595.0,
      "step": 8190
    },
    {
      "entropy": 5.592373466491699,
      "epoch": 0.9204245521424159,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004926252907345317,
      "loss": 5.4836,
      "mean_token_accuracy": 0.17857701629400252,
      "num_tokens": 18287510.0,
      "step": 8195
    },
    {
      "entropy": 5.717923927307129,
      "epoch": 0.9209861290503735,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004926150924457874,
      "loss": 5.5838,
      "mean_token_accuracy": 0.17386380732059478,
      "num_tokens": 18298764.0,
      "step": 8200
    },
    {
      "entropy": 5.588229608535767,
      "epoch": 0.921547705958331,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004926048872280943,
      "loss": 5.4009,
      "mean_token_accuracy": 0.17802754789590836,
      "num_tokens": 18310427.0,
      "step": 8205
    },
    {
      "entropy": 5.580872583389282,
      "epoch": 0.9221092828662886,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000492594675081777,
      "loss": 5.532,
      "mean_token_accuracy": 0.17346930950880052,
      "num_tokens": 18322174.0,
      "step": 8210
    },
    {
      "entropy": 5.584685850143432,
      "epoch": 0.9226708597742461,
      "grad_norm": 2.8125,
      "learning_rate": 0.000492584456007161,
      "loss": 5.4795,
      "mean_token_accuracy": 0.17690377533435822,
      "num_tokens": 18332372.0,
      "step": 8215
    },
    {
      "entropy": 5.672969627380371,
      "epoch": 0.9232324366822037,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004925742300045716,
      "loss": 5.5459,
      "mean_token_accuracy": 0.1704122766852379,
      "num_tokens": 18343201.0,
      "step": 8220
    },
    {
      "entropy": 5.634353446960449,
      "epoch": 0.9237940135901612,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004925639970743341,
      "loss": 5.5613,
      "mean_token_accuracy": 0.1783763960003853,
      "num_tokens": 18354124.0,
      "step": 8225
    },
    {
      "entropy": 5.626930046081543,
      "epoch": 0.9243555904981187,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004925537572167746,
      "loss": 5.5138,
      "mean_token_accuracy": 0.18148281127214433,
      "num_tokens": 18365148.0,
      "step": 8230
    },
    {
      "entropy": 5.467441701889038,
      "epoch": 0.9249171674060762,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004925435104322192,
      "loss": 5.3908,
      "mean_token_accuracy": 0.18088495582342148,
      "num_tokens": 18376244.0,
      "step": 8235
    },
    {
      "entropy": 5.6002789497375485,
      "epoch": 0.9254787443140338,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000492533256720994,
      "loss": 5.5533,
      "mean_token_accuracy": 0.17563220411539077,
      "num_tokens": 18386728.0,
      "step": 8240
    },
    {
      "entropy": 5.666000461578369,
      "epoch": 0.9260403212219913,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004925229960834256,
      "loss": 5.5451,
      "mean_token_accuracy": 0.17448101937770844,
      "num_tokens": 18397008.0,
      "step": 8245
    },
    {
      "entropy": 5.637777423858642,
      "epoch": 0.9266018981299489,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004925127285198407,
      "loss": 5.5305,
      "mean_token_accuracy": 0.18294473737478256,
      "num_tokens": 18408348.0,
      "step": 8250
    },
    {
      "entropy": 5.5927502632141115,
      "epoch": 0.9271634750379064,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000492502454030566,
      "loss": 5.5562,
      "mean_token_accuracy": 0.18002525418996812,
      "num_tokens": 18419239.0,
      "step": 8255
    },
    {
      "entropy": 5.727730941772461,
      "epoch": 0.927725051945864,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004924921726159288,
      "loss": 5.561,
      "mean_token_accuracy": 0.17431779056787491,
      "num_tokens": 18428918.0,
      "step": 8260
    },
    {
      "entropy": 5.725991296768188,
      "epoch": 0.9282866288538215,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004924818842762565,
      "loss": 5.5523,
      "mean_token_accuracy": 0.1816506192088127,
      "num_tokens": 18440361.0,
      "step": 8265
    },
    {
      "entropy": 5.559592771530151,
      "epoch": 0.9288482057617791,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004924715890118768,
      "loss": 5.4345,
      "mean_token_accuracy": 0.180571249127388,
      "num_tokens": 18450678.0,
      "step": 8270
    },
    {
      "entropy": 5.608138799667358,
      "epoch": 0.9294097826697366,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004924612868231173,
      "loss": 5.6334,
      "mean_token_accuracy": 0.17058404982089997,
      "num_tokens": 18462364.0,
      "step": 8275
    },
    {
      "entropy": 5.645278215408325,
      "epoch": 0.9299713595776942,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004924509777103059,
      "loss": 5.5226,
      "mean_token_accuracy": 0.17628366500139236,
      "num_tokens": 18472896.0,
      "step": 8280
    },
    {
      "entropy": 5.787207078933716,
      "epoch": 0.9305329364856517,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004924406616737712,
      "loss": 5.6589,
      "mean_token_accuracy": 0.16721024066209794,
      "num_tokens": 18484951.0,
      "step": 8285
    },
    {
      "entropy": 5.628914451599121,
      "epoch": 0.9310945133936093,
      "grad_norm": 1.375,
      "learning_rate": 0.0004924303387138415,
      "loss": 5.5925,
      "mean_token_accuracy": 0.174946691095829,
      "num_tokens": 18496006.0,
      "step": 8290
    },
    {
      "entropy": 5.748250484466553,
      "epoch": 0.9316560903015668,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004924200088308455,
      "loss": 5.5748,
      "mean_token_accuracy": 0.17472342401742935,
      "num_tokens": 18508069.0,
      "step": 8295
    },
    {
      "entropy": 5.7464500904083256,
      "epoch": 0.9322176672095244,
      "grad_norm": 1.390625,
      "learning_rate": 0.000492409672025112,
      "loss": 5.6889,
      "mean_token_accuracy": 0.16791522353887559,
      "num_tokens": 18519746.0,
      "step": 8300
    },
    {
      "entropy": 5.742476892471314,
      "epoch": 0.9327792441174819,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004923993282969703,
      "loss": 5.592,
      "mean_token_accuracy": 0.17625355273485183,
      "num_tokens": 18530249.0,
      "step": 8305
    },
    {
      "entropy": 5.6794898986816404,
      "epoch": 0.9333408210254395,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004923889776467495,
      "loss": 5.613,
      "mean_token_accuracy": 0.17979559898376465,
      "num_tokens": 18543306.0,
      "step": 8310
    },
    {
      "entropy": 5.634688186645508,
      "epoch": 0.933902397933397,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004923786200747795,
      "loss": 5.6575,
      "mean_token_accuracy": 0.17098734974861146,
      "num_tokens": 18555250.0,
      "step": 8315
    },
    {
      "entropy": 5.6697102069854735,
      "epoch": 0.9344639748413546,
      "grad_norm": 1.40625,
      "learning_rate": 0.00049236825558139,
      "loss": 5.402,
      "mean_token_accuracy": 0.1727229595184326,
      "num_tokens": 18566378.0,
      "step": 8320
    },
    {
      "entropy": 5.67743182182312,
      "epoch": 0.9350255517493121,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004923578841669108,
      "loss": 5.6108,
      "mean_token_accuracy": 0.17351700663566588,
      "num_tokens": 18576827.0,
      "step": 8325
    },
    {
      "entropy": 5.616914176940918,
      "epoch": 0.9355871286572696,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004923475058316723,
      "loss": 5.5397,
      "mean_token_accuracy": 0.17626303136348725,
      "num_tokens": 18589209.0,
      "step": 8330
    },
    {
      "entropy": 5.6495646953582765,
      "epoch": 0.9361487055652271,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004923371205760049,
      "loss": 5.5529,
      "mean_token_accuracy": 0.17969396114349365,
      "num_tokens": 18599367.0,
      "step": 8335
    },
    {
      "entropy": 5.666165113449097,
      "epoch": 0.9367102824731847,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004923267284002394,
      "loss": 5.5904,
      "mean_token_accuracy": 0.1729588732123375,
      "num_tokens": 18610505.0,
      "step": 8340
    },
    {
      "entropy": 5.692822408676148,
      "epoch": 0.9372718593811422,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004923163293047065,
      "loss": 5.5945,
      "mean_token_accuracy": 0.17593003809452057,
      "num_tokens": 18622616.0,
      "step": 8345
    },
    {
      "entropy": 5.594743490219116,
      "epoch": 0.9378334362890998,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004923059232897374,
      "loss": 5.4359,
      "mean_token_accuracy": 0.18015130013227462,
      "num_tokens": 18634127.0,
      "step": 8350
    },
    {
      "entropy": 5.634627008438111,
      "epoch": 0.9383950131970573,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004922955103556633,
      "loss": 5.6466,
      "mean_token_accuracy": 0.17177557200193405,
      "num_tokens": 18646196.0,
      "step": 8355
    },
    {
      "entropy": 5.645306730270386,
      "epoch": 0.9389565901050149,
      "grad_norm": 1.296875,
      "learning_rate": 0.000492285090502816,
      "loss": 5.552,
      "mean_token_accuracy": 0.1770355746150017,
      "num_tokens": 18656911.0,
      "step": 8360
    },
    {
      "entropy": 5.7078851699829105,
      "epoch": 0.9395181670129724,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004922746637315272,
      "loss": 5.5278,
      "mean_token_accuracy": 0.1755846068263054,
      "num_tokens": 18667553.0,
      "step": 8365
    },
    {
      "entropy": 5.620309591293335,
      "epoch": 0.94007974392093,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004922642300421288,
      "loss": 5.5011,
      "mean_token_accuracy": 0.1790532335639,
      "num_tokens": 18678087.0,
      "step": 8370
    },
    {
      "entropy": 5.6581525802612305,
      "epoch": 0.9406413208288875,
      "grad_norm": 1.484375,
      "learning_rate": 0.000492253789434953,
      "loss": 5.6158,
      "mean_token_accuracy": 0.1750505179166794,
      "num_tokens": 18689946.0,
      "step": 8375
    },
    {
      "entropy": 5.625391101837158,
      "epoch": 0.9412028977368451,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004922433419103323,
      "loss": 5.5269,
      "mean_token_accuracy": 0.18149401843547822,
      "num_tokens": 18700944.0,
      "step": 8380
    },
    {
      "entropy": 5.687411165237426,
      "epoch": 0.9417644746448026,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004922328874685993,
      "loss": 5.6265,
      "mean_token_accuracy": 0.17350946068763734,
      "num_tokens": 18712901.0,
      "step": 8385
    },
    {
      "entropy": 5.6175501346588135,
      "epoch": 0.9423260515527602,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004922224261100869,
      "loss": 5.4589,
      "mean_token_accuracy": 0.17959638684988022,
      "num_tokens": 18724322.0,
      "step": 8390
    },
    {
      "entropy": 5.593675136566162,
      "epoch": 0.9428876284607177,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004922119578351283,
      "loss": 5.5165,
      "mean_token_accuracy": 0.17416234016418458,
      "num_tokens": 18735315.0,
      "step": 8395
    },
    {
      "entropy": 5.61961555480957,
      "epoch": 0.9434492053686753,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004922014826440566,
      "loss": 5.5541,
      "mean_token_accuracy": 0.18015410602092743,
      "num_tokens": 18747497.0,
      "step": 8400
    },
    {
      "entropy": 5.691785049438477,
      "epoch": 0.9440107822766328,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004921910005372056,
      "loss": 5.5891,
      "mean_token_accuracy": 0.1704226091504097,
      "num_tokens": 18758835.0,
      "step": 8405
    },
    {
      "entropy": 5.671810340881348,
      "epoch": 0.9445723591845904,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004921805115149088,
      "loss": 5.483,
      "mean_token_accuracy": 0.18468721210956573,
      "num_tokens": 18770325.0,
      "step": 8410
    },
    {
      "entropy": 5.520982933044434,
      "epoch": 0.9451339360925479,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004921700155775002,
      "loss": 5.3773,
      "mean_token_accuracy": 0.18612601608037949,
      "num_tokens": 18781204.0,
      "step": 8415
    },
    {
      "entropy": 5.526636171340942,
      "epoch": 0.9456955130005055,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004921595127253142,
      "loss": 5.4905,
      "mean_token_accuracy": 0.18205440640449524,
      "num_tokens": 18791899.0,
      "step": 8420
    },
    {
      "entropy": 5.603455257415772,
      "epoch": 0.9462570899084629,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004921490029586852,
      "loss": 5.4779,
      "mean_token_accuracy": 0.1817940965294838,
      "num_tokens": 18803043.0,
      "step": 8425
    },
    {
      "entropy": 5.60422306060791,
      "epoch": 0.9468186668164205,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004921384862779475,
      "loss": 5.4969,
      "mean_token_accuracy": 0.17535765916109086,
      "num_tokens": 18814316.0,
      "step": 8430
    },
    {
      "entropy": 5.586996269226074,
      "epoch": 0.947380243724378,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004921279626834363,
      "loss": 5.5614,
      "mean_token_accuracy": 0.17469020187854767,
      "num_tokens": 18825490.0,
      "step": 8435
    },
    {
      "entropy": 5.758854627609253,
      "epoch": 0.9479418206323356,
      "grad_norm": 1.5,
      "learning_rate": 0.0004921174321754865,
      "loss": 5.6497,
      "mean_token_accuracy": 0.1785335808992386,
      "num_tokens": 18836479.0,
      "step": 8440
    },
    {
      "entropy": 5.655997276306152,
      "epoch": 0.9485033975402931,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004921068947544335,
      "loss": 5.5424,
      "mean_token_accuracy": 0.17748990803956985,
      "num_tokens": 18848634.0,
      "step": 8445
    },
    {
      "entropy": 5.593111944198609,
      "epoch": 0.9490649744482507,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004920963504206128,
      "loss": 5.5003,
      "mean_token_accuracy": 0.18238631188869475,
      "num_tokens": 18858451.0,
      "step": 8450
    },
    {
      "entropy": 5.757823705673218,
      "epoch": 0.9496265513562082,
      "grad_norm": 1.578125,
      "learning_rate": 0.00049208579917436,
      "loss": 5.6392,
      "mean_token_accuracy": 0.16730133891105653,
      "num_tokens": 18869561.0,
      "step": 8455
    },
    {
      "entropy": 5.777918672561645,
      "epoch": 0.9501881282641658,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004920752410160113,
      "loss": 5.6417,
      "mean_token_accuracy": 0.16759925112128257,
      "num_tokens": 18880232.0,
      "step": 8460
    },
    {
      "entropy": 5.7202564716339115,
      "epoch": 0.9507497051721233,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004920646759459026,
      "loss": 5.6543,
      "mean_token_accuracy": 0.17059721201658248,
      "num_tokens": 18892562.0,
      "step": 8465
    },
    {
      "entropy": 5.640997695922851,
      "epoch": 0.9513112820800809,
      "grad_norm": 2.125,
      "learning_rate": 0.0004920541039643705,
      "loss": 5.5844,
      "mean_token_accuracy": 0.17161750495433808,
      "num_tokens": 18903566.0,
      "step": 8470
    },
    {
      "entropy": 5.633873176574707,
      "epoch": 0.9518728589880384,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004920435250717515,
      "loss": 5.5761,
      "mean_token_accuracy": 0.17739747911691667,
      "num_tokens": 18914619.0,
      "step": 8475
    },
    {
      "entropy": 5.76814079284668,
      "epoch": 0.952434435895996,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004920329392683826,
      "loss": 5.6495,
      "mean_token_accuracy": 0.17262647897005082,
      "num_tokens": 18927039.0,
      "step": 8480
    },
    {
      "entropy": 5.777394104003906,
      "epoch": 0.9529960128039535,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004920223465546006,
      "loss": 5.7291,
      "mean_token_accuracy": 0.16832094341516496,
      "num_tokens": 18940449.0,
      "step": 8485
    },
    {
      "entropy": 5.695525121688843,
      "epoch": 0.9535575897119111,
      "grad_norm": 1.625,
      "learning_rate": 0.0004920117469307429,
      "loss": 5.6045,
      "mean_token_accuracy": 0.17119804620742798,
      "num_tokens": 18950798.0,
      "step": 8490
    },
    {
      "entropy": 5.686978483200074,
      "epoch": 0.9541191666198686,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004920011403971471,
      "loss": 5.4587,
      "mean_token_accuracy": 0.17551997900009156,
      "num_tokens": 18960958.0,
      "step": 8495
    },
    {
      "entropy": 5.588903331756592,
      "epoch": 0.9546807435278262,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004919905269541508,
      "loss": 5.4933,
      "mean_token_accuracy": 0.17877306193113326,
      "num_tokens": 18971890.0,
      "step": 8500
    },
    {
      "entropy": 5.560519123077393,
      "epoch": 0.9552423204357837,
      "grad_norm": 1.78125,
      "learning_rate": 0.000491979906602092,
      "loss": 5.4282,
      "mean_token_accuracy": 0.18534162938594817,
      "num_tokens": 18982309.0,
      "step": 8505
    },
    {
      "entropy": 5.67816572189331,
      "epoch": 0.9558038973437413,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004919692793413089,
      "loss": 5.6482,
      "mean_token_accuracy": 0.16850887835025788,
      "num_tokens": 18993648.0,
      "step": 8510
    },
    {
      "entropy": 5.667968082427978,
      "epoch": 0.9563654742516988,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004919586451721396,
      "loss": 5.4243,
      "mean_token_accuracy": 0.18356293588876724,
      "num_tokens": 19003613.0,
      "step": 8515
    },
    {
      "entropy": 5.685746192932129,
      "epoch": 0.9569270511596563,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004919480040949229,
      "loss": 5.5428,
      "mean_token_accuracy": 0.17324323505163192,
      "num_tokens": 19014720.0,
      "step": 8520
    },
    {
      "entropy": 5.643367195129395,
      "epoch": 0.9574886280676138,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004919373561099977,
      "loss": 5.5788,
      "mean_token_accuracy": 0.17302146852016448,
      "num_tokens": 19026635.0,
      "step": 8525
    },
    {
      "entropy": 5.673974132537841,
      "epoch": 0.9580502049755714,
      "grad_norm": 1.6875,
      "learning_rate": 0.000491926701217703,
      "loss": 5.5794,
      "mean_token_accuracy": 0.1681892067193985,
      "num_tokens": 19037433.0,
      "step": 8530
    },
    {
      "entropy": 5.6716055393219,
      "epoch": 0.9586117818835289,
      "grad_norm": 2.375,
      "learning_rate": 0.0004919160394183779,
      "loss": 5.4346,
      "mean_token_accuracy": 0.18276526629924775,
      "num_tokens": 19047351.0,
      "step": 8535
    },
    {
      "entropy": 5.513965892791748,
      "epoch": 0.9591733587914865,
      "grad_norm": 1.34375,
      "learning_rate": 0.000491905370712362,
      "loss": 5.4747,
      "mean_token_accuracy": 0.1769762381911278,
      "num_tokens": 19058561.0,
      "step": 8540
    },
    {
      "entropy": 5.654763650894165,
      "epoch": 0.959734935699444,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004918946950999949,
      "loss": 5.5741,
      "mean_token_accuracy": 0.16921738386154175,
      "num_tokens": 19069877.0,
      "step": 8545
    },
    {
      "entropy": 5.7251872539520265,
      "epoch": 0.9602965126074016,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004918840125816167,
      "loss": 5.5792,
      "mean_token_accuracy": 0.17826244831085206,
      "num_tokens": 19081962.0,
      "step": 8550
    },
    {
      "entropy": 5.646814775466919,
      "epoch": 0.9608580895153591,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004918733231575673,
      "loss": 5.6367,
      "mean_token_accuracy": 0.1751244530081749,
      "num_tokens": 19092955.0,
      "step": 8555
    },
    {
      "entropy": 5.690008735656738,
      "epoch": 0.9614196664233167,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004918626268281874,
      "loss": 5.4435,
      "mean_token_accuracy": 0.18832328021526337,
      "num_tokens": 19103521.0,
      "step": 8560
    },
    {
      "entropy": 5.521415567398071,
      "epoch": 0.9619812433312742,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004918519235938171,
      "loss": 5.4644,
      "mean_token_accuracy": 0.18721356838941575,
      "num_tokens": 19114190.0,
      "step": 8565
    },
    {
      "entropy": 5.628028392791748,
      "epoch": 0.9625428202392318,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004918412134547976,
      "loss": 5.5215,
      "mean_token_accuracy": 0.1752006560564041,
      "num_tokens": 19123815.0,
      "step": 8570
    },
    {
      "entropy": 5.675377082824707,
      "epoch": 0.9631043971471893,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004918304964114696,
      "loss": 5.6645,
      "mean_token_accuracy": 0.17109055519104005,
      "num_tokens": 19135805.0,
      "step": 8575
    },
    {
      "entropy": 5.735991382598877,
      "epoch": 0.9636659740551469,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004918197724641747,
      "loss": 5.4943,
      "mean_token_accuracy": 0.17655572593212127,
      "num_tokens": 19146656.0,
      "step": 8580
    },
    {
      "entropy": 5.620199251174927,
      "epoch": 0.9642275509631044,
      "grad_norm": 1.28125,
      "learning_rate": 0.000491809041613254,
      "loss": 5.5713,
      "mean_token_accuracy": 0.1704343155026436,
      "num_tokens": 19158271.0,
      "step": 8585
    },
    {
      "entropy": 5.580804252624512,
      "epoch": 0.964789127871062,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004917983038590495,
      "loss": 5.484,
      "mean_token_accuracy": 0.17846604883670808,
      "num_tokens": 19168439.0,
      "step": 8590
    },
    {
      "entropy": 5.6851584911346436,
      "epoch": 0.9653507047790195,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004917875592019028,
      "loss": 5.5721,
      "mean_token_accuracy": 0.17893614768981933,
      "num_tokens": 19179411.0,
      "step": 8595
    },
    {
      "entropy": 5.689495515823364,
      "epoch": 0.9659122816869771,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004917768076421561,
      "loss": 5.5917,
      "mean_token_accuracy": 0.174559023976326,
      "num_tokens": 19189377.0,
      "step": 8600
    },
    {
      "entropy": 5.615097713470459,
      "epoch": 0.9664738585949346,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004917660491801518,
      "loss": 5.5604,
      "mean_token_accuracy": 0.17394228130578995,
      "num_tokens": 19200662.0,
      "step": 8605
    },
    {
      "entropy": 5.648424863815308,
      "epoch": 0.9670354355028922,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004917552838162324,
      "loss": 5.4598,
      "mean_token_accuracy": 0.18265042901039125,
      "num_tokens": 19211535.0,
      "step": 8610
    },
    {
      "entropy": 5.628148460388184,
      "epoch": 0.9675970124108496,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004917445115507408,
      "loss": 5.4433,
      "mean_token_accuracy": 0.18630209118127822,
      "num_tokens": 19223402.0,
      "step": 8615
    },
    {
      "entropy": 5.487207746505737,
      "epoch": 0.9681585893188072,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004917337323840198,
      "loss": 5.5248,
      "mean_token_accuracy": 0.17707892805337905,
      "num_tokens": 19234976.0,
      "step": 8620
    },
    {
      "entropy": 5.468156814575195,
      "epoch": 0.9687201662267647,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004917229463164128,
      "loss": 5.3766,
      "mean_token_accuracy": 0.18640546053647994,
      "num_tokens": 19247328.0,
      "step": 8625
    },
    {
      "entropy": 5.678643655776978,
      "epoch": 0.9692817431347223,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004917121533482631,
      "loss": 5.5186,
      "mean_token_accuracy": 0.17082815170288085,
      "num_tokens": 19259021.0,
      "step": 8630
    },
    {
      "entropy": 5.5514545917510985,
      "epoch": 0.9698433200426798,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004917013534799144,
      "loss": 5.4829,
      "mean_token_accuracy": 0.17271644324064256,
      "num_tokens": 19270021.0,
      "step": 8635
    },
    {
      "entropy": 5.603173446655274,
      "epoch": 0.9704048969506374,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004916905467117104,
      "loss": 5.5294,
      "mean_token_accuracy": 0.18102877885103225,
      "num_tokens": 19281520.0,
      "step": 8640
    },
    {
      "entropy": 5.627556180953979,
      "epoch": 0.9709664738585949,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004916797330439955,
      "loss": 5.4723,
      "mean_token_accuracy": 0.1788710445165634,
      "num_tokens": 19292761.0,
      "step": 8645
    },
    {
      "entropy": 5.644547462463379,
      "epoch": 0.9715280507665525,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004916689124771138,
      "loss": 5.5805,
      "mean_token_accuracy": 0.17524223029613495,
      "num_tokens": 19305022.0,
      "step": 8650
    },
    {
      "entropy": 5.744730758666992,
      "epoch": 0.97208962767451,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004916580850114099,
      "loss": 5.6912,
      "mean_token_accuracy": 0.1697470173239708,
      "num_tokens": 19316162.0,
      "step": 8655
    },
    {
      "entropy": 5.853869438171387,
      "epoch": 0.9726512045824676,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004916472506472286,
      "loss": 5.6904,
      "mean_token_accuracy": 0.16199478805065154,
      "num_tokens": 19327015.0,
      "step": 8660
    },
    {
      "entropy": 5.628166675567627,
      "epoch": 0.9732127814904251,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004916364093849147,
      "loss": 5.5602,
      "mean_token_accuracy": 0.1737251363694668,
      "num_tokens": 19338384.0,
      "step": 8665
    },
    {
      "entropy": 5.600831985473633,
      "epoch": 0.9737743583983827,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004916255612248136,
      "loss": 5.4773,
      "mean_token_accuracy": 0.18174067735671998,
      "num_tokens": 19349846.0,
      "step": 8670
    },
    {
      "entropy": 5.7154128074646,
      "epoch": 0.9743359353063402,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004916147061672706,
      "loss": 5.5172,
      "mean_token_accuracy": 0.17738369554281236,
      "num_tokens": 19361399.0,
      "step": 8675
    },
    {
      "entropy": 5.6050217628479,
      "epoch": 0.9748975122142978,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004916038442126314,
      "loss": 5.4925,
      "mean_token_accuracy": 0.1719199925661087,
      "num_tokens": 19371990.0,
      "step": 8680
    },
    {
      "entropy": 5.553276252746582,
      "epoch": 0.9754590891222553,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004915929753612416,
      "loss": 5.4993,
      "mean_token_accuracy": 0.17748630344867705,
      "num_tokens": 19383135.0,
      "step": 8685
    },
    {
      "entropy": 5.666282701492309,
      "epoch": 0.9760206660302129,
      "grad_norm": 2.625,
      "learning_rate": 0.0004915820996134477,
      "loss": 5.5743,
      "mean_token_accuracy": 0.17527981400489806,
      "num_tokens": 19393942.0,
      "step": 8690
    },
    {
      "entropy": 5.59716305732727,
      "epoch": 0.9765822429381704,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004915712169695956,
      "loss": 5.411,
      "mean_token_accuracy": 0.1805658981204033,
      "num_tokens": 19404255.0,
      "step": 8695
    },
    {
      "entropy": 5.596176338195801,
      "epoch": 0.977143819846128,
      "grad_norm": 1.8515625,
      "learning_rate": 0.000491560327430032,
      "loss": 5.5122,
      "mean_token_accuracy": 0.17859420478343963,
      "num_tokens": 19415425.0,
      "step": 8700
    },
    {
      "entropy": 5.594195079803467,
      "epoch": 0.9777053967540855,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004915494309951035,
      "loss": 5.4588,
      "mean_token_accuracy": 0.18495271801948548,
      "num_tokens": 19426050.0,
      "step": 8705
    },
    {
      "entropy": 5.647588205337525,
      "epoch": 0.9782669736620431,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004915385276651573,
      "loss": 5.5586,
      "mean_token_accuracy": 0.1710285097360611,
      "num_tokens": 19437471.0,
      "step": 8710
    },
    {
      "entropy": 5.6465130805969235,
      "epoch": 0.9788285505700005,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004915276174405403,
      "loss": 5.587,
      "mean_token_accuracy": 0.17624323666095734,
      "num_tokens": 19449569.0,
      "step": 8715
    },
    {
      "entropy": 5.666909074783325,
      "epoch": 0.9793901274779581,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004915167003215999,
      "loss": 5.6053,
      "mean_token_accuracy": 0.17028293907642364,
      "num_tokens": 19461630.0,
      "step": 8720
    },
    {
      "entropy": 5.637751913070678,
      "epoch": 0.9799517043859156,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004915057763086838,
      "loss": 5.5337,
      "mean_token_accuracy": 0.17756096571683883,
      "num_tokens": 19472626.0,
      "step": 8725
    },
    {
      "entropy": 5.57206540107727,
      "epoch": 0.9805132812938732,
      "grad_norm": 1.78125,
      "learning_rate": 0.00049149484540214,
      "loss": 5.5253,
      "mean_token_accuracy": 0.1777624398469925,
      "num_tokens": 19483141.0,
      "step": 8730
    },
    {
      "entropy": 5.650867986679077,
      "epoch": 0.9810748582018307,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004914839076023161,
      "loss": 5.499,
      "mean_token_accuracy": 0.18341919481754304,
      "num_tokens": 19495052.0,
      "step": 8735
    },
    {
      "entropy": 5.702249383926391,
      "epoch": 0.9816364351097883,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004914729629095609,
      "loss": 5.5682,
      "mean_token_accuracy": 0.17288218587636947,
      "num_tokens": 19506911.0,
      "step": 8740
    },
    {
      "entropy": 5.562432861328125,
      "epoch": 0.9821980120177458,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004914620113242224,
      "loss": 5.4369,
      "mean_token_accuracy": 0.18190908581018447,
      "num_tokens": 19517812.0,
      "step": 8745
    },
    {
      "entropy": 5.673620414733887,
      "epoch": 0.9827595889257034,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004914510528466495,
      "loss": 5.5298,
      "mean_token_accuracy": 0.176532019674778,
      "num_tokens": 19529857.0,
      "step": 8750
    },
    {
      "entropy": 5.6598528861999515,
      "epoch": 0.9833211658336609,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004914400874771913,
      "loss": 5.5514,
      "mean_token_accuracy": 0.17500680088996887,
      "num_tokens": 19541082.0,
      "step": 8755
    },
    {
      "entropy": 5.5744036674499515,
      "epoch": 0.9838827427416185,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004914291152161967,
      "loss": 5.4743,
      "mean_token_accuracy": 0.18563976138830185,
      "num_tokens": 19551266.0,
      "step": 8760
    },
    {
      "entropy": 5.674076700210572,
      "epoch": 0.984444319649576,
      "grad_norm": 1.59375,
      "learning_rate": 0.000491418136064015,
      "loss": 5.662,
      "mean_token_accuracy": 0.17542957365512848,
      "num_tokens": 19563381.0,
      "step": 8765
    },
    {
      "entropy": 5.752225589752197,
      "epoch": 0.9850058965575336,
      "grad_norm": 1.421875,
      "learning_rate": 0.000491407150020996,
      "loss": 5.6875,
      "mean_token_accuracy": 0.174013951420784,
      "num_tokens": 19575078.0,
      "step": 8770
    },
    {
      "entropy": 5.6871308326721195,
      "epoch": 0.9855674734654911,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004913961570874895,
      "loss": 5.4838,
      "mean_token_accuracy": 0.1798838794231415,
      "num_tokens": 19585273.0,
      "step": 8775
    },
    {
      "entropy": 5.567116832733154,
      "epoch": 0.9861290503734487,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004913851572638453,
      "loss": 5.4983,
      "mean_token_accuracy": 0.17919607013463973,
      "num_tokens": 19594245.0,
      "step": 8780
    },
    {
      "entropy": 5.678967380523682,
      "epoch": 0.9866906272814062,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004913741505504138,
      "loss": 5.5321,
      "mean_token_accuracy": 0.17172959744930266,
      "num_tokens": 19604920.0,
      "step": 8785
    },
    {
      "entropy": 5.7392435550689695,
      "epoch": 0.9872522041893638,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004913631369475454,
      "loss": 5.6313,
      "mean_token_accuracy": 0.17030101865530015,
      "num_tokens": 19616592.0,
      "step": 8790
    },
    {
      "entropy": 5.726884555816651,
      "epoch": 0.9878137810973213,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004913521164555909,
      "loss": 5.5896,
      "mean_token_accuracy": 0.17396857142448424,
      "num_tokens": 19627359.0,
      "step": 8795
    },
    {
      "entropy": 5.6828028678894045,
      "epoch": 0.9883753580052789,
      "grad_norm": 1.859375,
      "learning_rate": 0.000491341089074901,
      "loss": 5.5506,
      "mean_token_accuracy": 0.1802465796470642,
      "num_tokens": 19639685.0,
      "step": 8800
    },
    {
      "entropy": 5.56529130935669,
      "epoch": 0.9889369349132364,
      "grad_norm": 1.8359375,
      "learning_rate": 0.000491330054805827,
      "loss": 5.3945,
      "mean_token_accuracy": 0.18767881393432617,
      "num_tokens": 19650525.0,
      "step": 8805
    },
    {
      "entropy": 5.508602094650269,
      "epoch": 0.9894985118211939,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004913190136487201,
      "loss": 5.4621,
      "mean_token_accuracy": 0.1844386488199234,
      "num_tokens": 19662241.0,
      "step": 8810
    },
    {
      "entropy": 5.670568418502808,
      "epoch": 0.9900600887291514,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004913079656039318,
      "loss": 5.5329,
      "mean_token_accuracy": 0.17866535037755965,
      "num_tokens": 19673752.0,
      "step": 8815
    },
    {
      "entropy": 5.600825834274292,
      "epoch": 0.990621665637109,
      "grad_norm": 1.875,
      "learning_rate": 0.0004912969106718142,
      "loss": 5.5253,
      "mean_token_accuracy": 0.17793213129043578,
      "num_tokens": 19685309.0,
      "step": 8820
    },
    {
      "entropy": 5.5946839332580565,
      "epoch": 0.9911832425450665,
      "grad_norm": 1.46875,
      "learning_rate": 0.000491285848852719,
      "loss": 5.6214,
      "mean_token_accuracy": 0.1743100419640541,
      "num_tokens": 19697391.0,
      "step": 8825
    },
    {
      "entropy": 5.695317173004151,
      "epoch": 0.9917448194530241,
      "grad_norm": 1.5,
      "learning_rate": 0.0004912747801469984,
      "loss": 5.5893,
      "mean_token_accuracy": 0.17411051988601683,
      "num_tokens": 19707426.0,
      "step": 8830
    },
    {
      "entropy": 5.710868787765503,
      "epoch": 0.9923063963609816,
      "grad_norm": 1.65625,
      "learning_rate": 0.000491263704555005,
      "loss": 5.6053,
      "mean_token_accuracy": 0.17150882482528687,
      "num_tokens": 19719800.0,
      "step": 8835
    },
    {
      "entropy": 5.579995489120483,
      "epoch": 0.9928679732689392,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004912526220770913,
      "loss": 5.5116,
      "mean_token_accuracy": 0.17503908276557922,
      "num_tokens": 19731600.0,
      "step": 8840
    },
    {
      "entropy": 5.720822381973266,
      "epoch": 0.9934295501768967,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004912415327136103,
      "loss": 5.5118,
      "mean_token_accuracy": 0.1760959282517433,
      "num_tokens": 19742610.0,
      "step": 8845
    },
    {
      "entropy": 5.776559829711914,
      "epoch": 0.9939911270848543,
      "grad_norm": 1.453125,
      "learning_rate": 0.000491230436464915,
      "loss": 5.721,
      "mean_token_accuracy": 0.1640644207596779,
      "num_tokens": 19754207.0,
      "step": 8850
    },
    {
      "entropy": 5.636538934707642,
      "epoch": 0.9945527039928118,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004912193333313589,
      "loss": 5.4245,
      "mean_token_accuracy": 0.1815497785806656,
      "num_tokens": 19765894.0,
      "step": 8855
    },
    {
      "entropy": 5.662306404113769,
      "epoch": 0.9951142809007694,
      "grad_norm": 2.0,
      "learning_rate": 0.0004912082233132951,
      "loss": 5.5738,
      "mean_token_accuracy": 0.17184434235095977,
      "num_tokens": 19777867.0,
      "step": 8860
    },
    {
      "entropy": 5.616130352020264,
      "epoch": 0.9956758578087269,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004911971064110778,
      "loss": 5.461,
      "mean_token_accuracy": 0.1819846212863922,
      "num_tokens": 19788346.0,
      "step": 8865
    },
    {
      "entropy": 5.745035696029663,
      "epoch": 0.9962374347166845,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004911859826250607,
      "loss": 5.6139,
      "mean_token_accuracy": 0.18041548728942872,
      "num_tokens": 19799752.0,
      "step": 8870
    },
    {
      "entropy": 5.649567937850952,
      "epoch": 0.996799011624642,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004911748519555981,
      "loss": 5.5514,
      "mean_token_accuracy": 0.1762547880411148,
      "num_tokens": 19810387.0,
      "step": 8875
    },
    {
      "entropy": 5.572876071929931,
      "epoch": 0.9973605885325996,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004911637144030444,
      "loss": 5.521,
      "mean_token_accuracy": 0.17834608554840087,
      "num_tokens": 19822092.0,
      "step": 8880
    },
    {
      "entropy": 5.707835960388183,
      "epoch": 0.9979221654405571,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004911525699677541,
      "loss": 5.6279,
      "mean_token_accuracy": 0.16719121038913726,
      "num_tokens": 19833731.0,
      "step": 8885
    },
    {
      "entropy": 5.668497037887573,
      "epoch": 0.9984837423485147,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004911414186500822,
      "loss": 5.5636,
      "mean_token_accuracy": 0.1742876499891281,
      "num_tokens": 19845101.0,
      "step": 8890
    },
    {
      "entropy": 5.6994829177856445,
      "epoch": 0.9990453192564722,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004911302604503837,
      "loss": 5.6231,
      "mean_token_accuracy": 0.17030643969774245,
      "num_tokens": 19857006.0,
      "step": 8895
    },
    {
      "entropy": 5.674806213378906,
      "epoch": 0.9996068961644298,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004911190953690139,
      "loss": 5.4978,
      "mean_token_accuracy": 0.18147013634443282,
      "num_tokens": 19868007.0,
      "step": 8900
    },
    {
      "entropy": 5.7907646497090655,
      "epoch": 1.0001123153815914,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004911079234063284,
      "loss": 5.4717,
      "mean_token_accuracy": 0.18072778979937235,
      "num_tokens": 19877897.0,
      "step": 8905
    },
    {
      "entropy": 5.650493812561035,
      "epoch": 1.000673892289549,
      "grad_norm": 1.75,
      "learning_rate": 0.0004910967445626826,
      "loss": 5.496,
      "mean_token_accuracy": 0.17796591967344283,
      "num_tokens": 19888449.0,
      "step": 8910
    },
    {
      "entropy": 5.612166070938111,
      "epoch": 1.0012354691975065,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004910855588384327,
      "loss": 5.4678,
      "mean_token_accuracy": 0.17649304419755935,
      "num_tokens": 19898998.0,
      "step": 8915
    },
    {
      "entropy": 5.6738122463226315,
      "epoch": 1.0017970461054642,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004910743662339348,
      "loss": 5.4831,
      "mean_token_accuracy": 0.18333123475313187,
      "num_tokens": 19910369.0,
      "step": 8920
    },
    {
      "entropy": 5.721463680267334,
      "epoch": 1.0023586230134216,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004910631667495452,
      "loss": 5.5219,
      "mean_token_accuracy": 0.1823783427476883,
      "num_tokens": 19921936.0,
      "step": 8925
    },
    {
      "entropy": 5.657253789901733,
      "epoch": 1.0029201999213793,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004910519603856207,
      "loss": 5.5097,
      "mean_token_accuracy": 0.17559026032686234,
      "num_tokens": 19933943.0,
      "step": 8930
    },
    {
      "entropy": 5.494042158126831,
      "epoch": 1.0034817768293367,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004910407471425179,
      "loss": 5.3106,
      "mean_token_accuracy": 0.18985848128795624,
      "num_tokens": 19944377.0,
      "step": 8935
    },
    {
      "entropy": 5.521006155014038,
      "epoch": 1.0040433537372944,
      "grad_norm": 1.6875,
      "learning_rate": 0.000491029527020594,
      "loss": 5.4144,
      "mean_token_accuracy": 0.18066316545009614,
      "num_tokens": 19954865.0,
      "step": 8940
    },
    {
      "entropy": 5.543346166610718,
      "epoch": 1.0046049306452518,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004910183000202061,
      "loss": 5.3318,
      "mean_token_accuracy": 0.19103278070688248,
      "num_tokens": 19965021.0,
      "step": 8945
    },
    {
      "entropy": 5.634443902969361,
      "epoch": 1.0051665075532095,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004910070661417119,
      "loss": 5.4471,
      "mean_token_accuracy": 0.17768314927816392,
      "num_tokens": 19977068.0,
      "step": 8950
    },
    {
      "entropy": 5.5534903049469,
      "epoch": 1.005728084461167,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004909958253854687,
      "loss": 5.474,
      "mean_token_accuracy": 0.17667020857334137,
      "num_tokens": 19987994.0,
      "step": 8955
    },
    {
      "entropy": 5.651862049102784,
      "epoch": 1.0062896613691246,
      "grad_norm": 1.75,
      "learning_rate": 0.0004909845777518347,
      "loss": 5.4939,
      "mean_token_accuracy": 0.17481962591409683,
      "num_tokens": 19998598.0,
      "step": 8960
    },
    {
      "entropy": 5.622123050689697,
      "epoch": 1.006851238277082,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004909733232411681,
      "loss": 5.385,
      "mean_token_accuracy": 0.1836555555462837,
      "num_tokens": 20008881.0,
      "step": 8965
    },
    {
      "entropy": 5.586540174484253,
      "epoch": 1.0074128151850397,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004909620618538271,
      "loss": 5.4177,
      "mean_token_accuracy": 0.17951451539993285,
      "num_tokens": 20019105.0,
      "step": 8970
    },
    {
      "entropy": 5.579048728942871,
      "epoch": 1.007974392092997,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004909507935901702,
      "loss": 5.4769,
      "mean_token_accuracy": 0.17605098336935043,
      "num_tokens": 20031805.0,
      "step": 8975
    },
    {
      "entropy": 5.62233076095581,
      "epoch": 1.0085359690009548,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004909395184505564,
      "loss": 5.4234,
      "mean_token_accuracy": 0.18534507006406784,
      "num_tokens": 20041493.0,
      "step": 8980
    },
    {
      "entropy": 5.607890892028808,
      "epoch": 1.0090975459089122,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004909282364353445,
      "loss": 5.4807,
      "mean_token_accuracy": 0.18110638111829758,
      "num_tokens": 20052489.0,
      "step": 8985
    },
    {
      "entropy": 5.46955451965332,
      "epoch": 1.0096591228168699,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004909169475448938,
      "loss": 5.4328,
      "mean_token_accuracy": 0.1797465994954109,
      "num_tokens": 20064252.0,
      "step": 8990
    },
    {
      "entropy": 5.621907567977905,
      "epoch": 1.0102206997248273,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004909056517795638,
      "loss": 5.3744,
      "mean_token_accuracy": 0.17497411817312242,
      "num_tokens": 20075171.0,
      "step": 8995
    },
    {
      "entropy": 5.624486780166626,
      "epoch": 1.010782276632785,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004908943491397141,
      "loss": 5.376,
      "mean_token_accuracy": 0.18161926418542862,
      "num_tokens": 20085746.0,
      "step": 9000
    },
    {
      "epoch": 1.010782276632785,
      "eval_entropy": 5.459080059872112,
      "eval_loss": 5.578691005706787,
      "eval_mean_token_accuracy": 0.1821541891581993,
      "eval_num_tokens": 20085746.0,
      "eval_runtime": 23.814,
      "eval_samples_per_second": 1302.304,
      "eval_steps_per_second": 162.804,
      "step": 9000
    }
  ],
  "logging_steps": 5,
  "max_steps": 89030,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.1507094368512e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}