heb-hebr-100mb-after-ppt-Dp-100mb-ckpt500_seed3407/checkpoint-6000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6420889293167104,
  "eval_steps": 3000,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 8.172043609619141,
      "epoch": 0.0005350741077639254,
      "grad_norm": 1.0078125,
      "learning_rate": 2e-06,
      "loss": 7.9195,
      "mean_token_accuracy": 0.06807812452316284,
      "num_tokens": 10038.0,
      "step": 5
    },
    {
      "entropy": 8.173345184326172,
      "epoch": 0.0010701482155278507,
      "grad_norm": 1.09375,
      "learning_rate": 4.5e-06,
      "loss": 7.9519,
      "mean_token_accuracy": 0.06761166527867317,
      "num_tokens": 21916.0,
      "step": 10
    },
    {
      "entropy": 8.15230588912964,
      "epoch": 0.0016052223232917758,
      "grad_norm": 1.0234375,
      "learning_rate": 7e-06,
      "loss": 7.8809,
      "mean_token_accuracy": 0.07487182170152665,
      "num_tokens": 33434.0,
      "step": 15
    },
    {
      "entropy": 8.150416851043701,
      "epoch": 0.0021402964310557014,
      "grad_norm": 1.0,
      "learning_rate": 9.5e-06,
      "loss": 7.9178,
      "mean_token_accuracy": 0.07305535562336445,
      "num_tokens": 44216.0,
      "step": 20
    },
    {
      "entropy": 8.19902458190918,
      "epoch": 0.0026753705388196263,
      "grad_norm": 0.90625,
      "learning_rate": 1.2e-05,
      "loss": 7.9382,
      "mean_token_accuracy": 0.07128040939569473,
      "num_tokens": 54874.0,
      "step": 25
    },
    {
      "entropy": 8.148491477966308,
      "epoch": 0.0032104446465835517,
      "grad_norm": 1.015625,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 7.939,
      "mean_token_accuracy": 0.06835630089044571,
      "num_tokens": 65085.0,
      "step": 30
    },
    {
      "entropy": 8.145433521270752,
      "epoch": 0.003745518754347477,
      "grad_norm": 1.015625,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 7.8478,
      "mean_token_accuracy": 0.07239970788359643,
      "num_tokens": 75601.0,
      "step": 35
    },
    {
      "entropy": 8.129260540008545,
      "epoch": 0.004280592862111403,
      "grad_norm": 0.9375,
      "learning_rate": 1.95e-05,
      "loss": 7.8539,
      "mean_token_accuracy": 0.07086838148534298,
      "num_tokens": 86193.0,
      "step": 40
    },
    {
      "entropy": 8.110703659057616,
      "epoch": 0.004815666969875327,
      "grad_norm": 0.9609375,
      "learning_rate": 2.2e-05,
      "loss": 7.868,
      "mean_token_accuracy": 0.07200616635382176,
      "num_tokens": 96957.0,
      "step": 45
    },
    {
      "entropy": 8.097426462173463,
      "epoch": 0.005350741077639253,
      "grad_norm": 0.9453125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 7.8701,
      "mean_token_accuracy": 0.07447938099503518,
      "num_tokens": 108408.0,
      "step": 50
    },
    {
      "entropy": 8.082033777236939,
      "epoch": 0.005885815185403178,
      "grad_norm": 0.82421875,
      "learning_rate": 2.7e-05,
      "loss": 7.8899,
      "mean_token_accuracy": 0.0776177041232586,
      "num_tokens": 119273.0,
      "step": 55
    },
    {
      "entropy": 8.075964164733886,
      "epoch": 0.006420889293167103,
      "grad_norm": 0.859375,
      "learning_rate": 2.95e-05,
      "loss": 7.8943,
      "mean_token_accuracy": 0.07140903994441032,
      "num_tokens": 130817.0,
      "step": 60
    },
    {
      "entropy": 8.11206226348877,
      "epoch": 0.006955963400931029,
      "grad_norm": 0.84765625,
      "learning_rate": 3.2e-05,
      "loss": 7.8565,
      "mean_token_accuracy": 0.07345039658248424,
      "num_tokens": 142599.0,
      "step": 65
    },
    {
      "entropy": 8.122138023376465,
      "epoch": 0.007491037508694954,
      "grad_norm": 0.9375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 7.9386,
      "mean_token_accuracy": 0.07604096084833145,
      "num_tokens": 153617.0,
      "step": 70
    },
    {
      "entropy": 8.110438632965089,
      "epoch": 0.00802611161645888,
      "grad_norm": 0.875,
      "learning_rate": 3.7e-05,
      "loss": 7.9032,
      "mean_token_accuracy": 0.07372927628457546,
      "num_tokens": 165280.0,
      "step": 75
    },
    {
      "entropy": 8.064451265335084,
      "epoch": 0.008561185724222806,
      "grad_norm": 0.81640625,
      "learning_rate": 3.95e-05,
      "loss": 7.9103,
      "mean_token_accuracy": 0.07471717111766338,
      "num_tokens": 176896.0,
      "step": 80
    },
    {
      "entropy": 8.096488285064698,
      "epoch": 0.00909625983198673,
      "grad_norm": 0.84765625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 7.9333,
      "mean_token_accuracy": 0.07502413280308247,
      "num_tokens": 188000.0,
      "step": 85
    },
    {
      "entropy": 8.13657898902893,
      "epoch": 0.009631333939750655,
      "grad_norm": 0.9609375,
      "learning_rate": 4.45e-05,
      "loss": 7.9588,
      "mean_token_accuracy": 0.0756196528673172,
      "num_tokens": 199360.0,
      "step": 90
    },
    {
      "entropy": 8.187862300872803,
      "epoch": 0.01016640804751458,
      "grad_norm": 0.92578125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.939,
      "mean_token_accuracy": 0.07151137851178646,
      "num_tokens": 210214.0,
      "step": 95
    },
    {
      "entropy": 8.11329288482666,
      "epoch": 0.010701482155278505,
      "grad_norm": 0.97265625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 7.8869,
      "mean_token_accuracy": 0.07601158879697323,
      "num_tokens": 220567.0,
      "step": 100
    },
    {
      "entropy": 8.091031312942505,
      "epoch": 0.011236556263042432,
      "grad_norm": 0.9765625,
      "learning_rate": 5.2e-05,
      "loss": 7.8088,
      "mean_token_accuracy": 0.08166341260075569,
      "num_tokens": 230585.0,
      "step": 105
    },
    {
      "entropy": 8.062711715698242,
      "epoch": 0.011771630370806356,
      "grad_norm": 0.9609375,
      "learning_rate": 5.45e-05,
      "loss": 7.8571,
      "mean_token_accuracy": 0.07631958648562431,
      "num_tokens": 241996.0,
      "step": 110
    },
    {
      "entropy": 8.094353532791137,
      "epoch": 0.012306704478570282,
      "grad_norm": 0.94921875,
      "learning_rate": 5.7e-05,
      "loss": 7.9555,
      "mean_token_accuracy": 0.07188590094447136,
      "num_tokens": 253348.0,
      "step": 115
    },
    {
      "entropy": 8.148624420166016,
      "epoch": 0.012841778586334207,
      "grad_norm": 1.03125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 7.889,
      "mean_token_accuracy": 0.07277357392013073,
      "num_tokens": 264239.0,
      "step": 120
    },
    {
      "entropy": 8.139824962615966,
      "epoch": 0.013376852694098133,
      "grad_norm": 0.96875,
      "learning_rate": 6.2e-05,
      "loss": 7.89,
      "mean_token_accuracy": 0.07517567798495292,
      "num_tokens": 274522.0,
      "step": 125
    },
    {
      "entropy": 8.114822006225586,
      "epoch": 0.013911926801862057,
      "grad_norm": 1.1640625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.8113,
      "mean_token_accuracy": 0.07706562504172325,
      "num_tokens": 284492.0,
      "step": 130
    },
    {
      "entropy": 8.067396926879884,
      "epoch": 0.014447000909625984,
      "grad_norm": 1.03125,
      "learning_rate": 6.7e-05,
      "loss": 7.8514,
      "mean_token_accuracy": 0.07260387018322945,
      "num_tokens": 294796.0,
      "step": 135
    },
    {
      "entropy": 8.12016191482544,
      "epoch": 0.014982075017389908,
      "grad_norm": 0.96875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.8603,
      "mean_token_accuracy": 0.06997270435094834,
      "num_tokens": 305252.0,
      "step": 140
    },
    {
      "entropy": 8.123533344268798,
      "epoch": 0.015517149125153834,
      "grad_norm": 0.98828125,
      "learning_rate": 7.2e-05,
      "loss": 7.8396,
      "mean_token_accuracy": 0.07271061241626739,
      "num_tokens": 315505.0,
      "step": 145
    },
    {
      "entropy": 8.117444324493409,
      "epoch": 0.01605222323291776,
      "grad_norm": 0.9296875,
      "learning_rate": 7.45e-05,
      "loss": 7.8922,
      "mean_token_accuracy": 0.07951539009809494,
      "num_tokens": 326997.0,
      "step": 150
    },
    {
      "entropy": 8.110684204101563,
      "epoch": 0.016587297340681685,
      "grad_norm": 0.94921875,
      "learning_rate": 7.7e-05,
      "loss": 7.875,
      "mean_token_accuracy": 0.07424910478293896,
      "num_tokens": 338089.0,
      "step": 155
    },
    {
      "entropy": 8.073831605911256,
      "epoch": 0.01712237144844561,
      "grad_norm": 1.015625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.8353,
      "mean_token_accuracy": 0.07741920799016952,
      "num_tokens": 348536.0,
      "step": 160
    },
    {
      "entropy": 8.16912670135498,
      "epoch": 0.017657445556209534,
      "grad_norm": 0.9609375,
      "learning_rate": 8.2e-05,
      "loss": 7.876,
      "mean_token_accuracy": 0.07565471082925797,
      "num_tokens": 359776.0,
      "step": 165
    },
    {
      "entropy": 8.006576490402221,
      "epoch": 0.01819251966397346,
      "grad_norm": 0.96484375,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.8271,
      "mean_token_accuracy": 0.07974917851388455,
      "num_tokens": 370810.0,
      "step": 170
    },
    {
      "entropy": 8.157261371612549,
      "epoch": 0.018727593771737387,
      "grad_norm": 0.94921875,
      "learning_rate": 8.7e-05,
      "loss": 7.8194,
      "mean_token_accuracy": 0.0741695910692215,
      "num_tokens": 381788.0,
      "step": 175
    },
    {
      "entropy": 8.101666450500488,
      "epoch": 0.01926266787950131,
      "grad_norm": 0.9453125,
      "learning_rate": 8.95e-05,
      "loss": 7.8579,
      "mean_token_accuracy": 0.07514562532305717,
      "num_tokens": 391984.0,
      "step": 180
    },
    {
      "entropy": 8.0212833404541,
      "epoch": 0.019797741987265235,
      "grad_norm": 0.828125,
      "learning_rate": 9.2e-05,
      "loss": 7.8041,
      "mean_token_accuracy": 0.0741691879928112,
      "num_tokens": 403430.0,
      "step": 185
    },
    {
      "entropy": 8.16499195098877,
      "epoch": 0.02033281609502916,
      "grad_norm": 0.98046875,
      "learning_rate": 9.45e-05,
      "loss": 7.8366,
      "mean_token_accuracy": 0.07594205625355244,
      "num_tokens": 414467.0,
      "step": 190
    },
    {
      "entropy": 7.895851373672485,
      "epoch": 0.020867890202793088,
      "grad_norm": 0.9296875,
      "learning_rate": 9.7e-05,
      "loss": 7.8383,
      "mean_token_accuracy": 0.0728034682571888,
      "num_tokens": 425752.0,
      "step": 195
    },
    {
      "entropy": 8.13976879119873,
      "epoch": 0.02140296431055701,
      "grad_norm": 0.9765625,
      "learning_rate": 9.95e-05,
      "loss": 7.8465,
      "mean_token_accuracy": 0.07538696676492691,
      "num_tokens": 435908.0,
      "step": 200
    },
    {
      "entropy": 8.162510967254638,
      "epoch": 0.021938038418320937,
      "grad_norm": 1.015625,
      "learning_rate": 0.000102,
      "loss": 7.8493,
      "mean_token_accuracy": 0.07364911586046219,
      "num_tokens": 446424.0,
      "step": 205
    },
    {
      "entropy": 8.023038148880005,
      "epoch": 0.022473112526084863,
      "grad_norm": 1.03125,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.8633,
      "mean_token_accuracy": 0.07611406669020652,
      "num_tokens": 456854.0,
      "step": 210
    },
    {
      "entropy": 8.120884990692138,
      "epoch": 0.02300818663384879,
      "grad_norm": 1.09375,
      "learning_rate": 0.000107,
      "loss": 7.8487,
      "mean_token_accuracy": 0.07724661827087402,
      "num_tokens": 467899.0,
      "step": 215
    },
    {
      "entropy": 7.984131669998169,
      "epoch": 0.023543260741612712,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001095,
      "loss": 7.7948,
      "mean_token_accuracy": 0.07440953776240349,
      "num_tokens": 478683.0,
      "step": 220
    },
    {
      "entropy": 8.137256622314453,
      "epoch": 0.02407833484937664,
      "grad_norm": 1.0,
      "learning_rate": 0.000112,
      "loss": 7.8224,
      "mean_token_accuracy": 0.07473525032401085,
      "num_tokens": 488745.0,
      "step": 225
    },
    {
      "entropy": 8.094204759597778,
      "epoch": 0.024613408957140565,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001145,
      "loss": 7.8853,
      "mean_token_accuracy": 0.07554894238710404,
      "num_tokens": 499448.0,
      "step": 230
    },
    {
      "entropy": 8.142967748641968,
      "epoch": 0.02514848306490449,
      "grad_norm": 1.0,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.8528,
      "mean_token_accuracy": 0.07399334162473678,
      "num_tokens": 510148.0,
      "step": 235
    },
    {
      "entropy": 8.069922924041748,
      "epoch": 0.025683557172668414,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.8017,
      "mean_token_accuracy": 0.07167132794857026,
      "num_tokens": 522093.0,
      "step": 240
    },
    {
      "entropy": 8.040627574920654,
      "epoch": 0.02621863128043234,
      "grad_norm": 1.03125,
      "learning_rate": 0.000122,
      "loss": 7.801,
      "mean_token_accuracy": 0.07531477734446526,
      "num_tokens": 532598.0,
      "step": 245
    },
    {
      "entropy": 8.111019945144653,
      "epoch": 0.026753705388196266,
      "grad_norm": 1.125,
      "learning_rate": 0.0001245,
      "loss": 7.8009,
      "mean_token_accuracy": 0.07633137851953506,
      "num_tokens": 542652.0,
      "step": 250
    },
    {
      "entropy": 8.048424530029298,
      "epoch": 0.027288779495960192,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000127,
      "loss": 7.8313,
      "mean_token_accuracy": 0.07268497720360756,
      "num_tokens": 552526.0,
      "step": 255
    },
    {
      "entropy": 8.08319902420044,
      "epoch": 0.027823853603724115,
      "grad_norm": 0.9375,
      "learning_rate": 0.0001295,
      "loss": 7.855,
      "mean_token_accuracy": 0.07812338769435882,
      "num_tokens": 563143.0,
      "step": 260
    },
    {
      "entropy": 8.129153060913087,
      "epoch": 0.02835892771148804,
      "grad_norm": 1.0,
      "learning_rate": 0.000132,
      "loss": 7.8048,
      "mean_token_accuracy": 0.07328564301133156,
      "num_tokens": 573814.0,
      "step": 265
    },
    {
      "entropy": 7.936620903015137,
      "epoch": 0.028894001819251967,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.7963,
      "mean_token_accuracy": 0.08088002502918243,
      "num_tokens": 585367.0,
      "step": 270
    },
    {
      "entropy": 8.25247220993042,
      "epoch": 0.02942907592701589,
      "grad_norm": 2.421875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.8577,
      "mean_token_accuracy": 0.07499300055205822,
      "num_tokens": 595902.0,
      "step": 275
    },
    {
      "entropy": 7.961127281188965,
      "epoch": 0.029964150034779816,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001395,
      "loss": 7.7677,
      "mean_token_accuracy": 0.07093185968697072,
      "num_tokens": 607178.0,
      "step": 280
    },
    {
      "entropy": 8.058710098266602,
      "epoch": 0.030499224142543743,
      "grad_norm": 1.0,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.7653,
      "mean_token_accuracy": 0.07953990697860717,
      "num_tokens": 618604.0,
      "step": 285
    },
    {
      "entropy": 7.963685512542725,
      "epoch": 0.03103429825030767,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001445,
      "loss": 7.8304,
      "mean_token_accuracy": 0.07993247732520103,
      "num_tokens": 628111.0,
      "step": 290
    },
    {
      "entropy": 8.22060260772705,
      "epoch": 0.03156937235807159,
      "grad_norm": 1.03125,
      "learning_rate": 0.000147,
      "loss": 7.8365,
      "mean_token_accuracy": 0.08120491243898868,
      "num_tokens": 639358.0,
      "step": 295
    },
    {
      "entropy": 7.96451735496521,
      "epoch": 0.03210444646583552,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001495,
      "loss": 7.7261,
      "mean_token_accuracy": 0.07790916599333286,
      "num_tokens": 649504.0,
      "step": 300
    },
    {
      "entropy": 8.008040952682496,
      "epoch": 0.032639520573599444,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000152,
      "loss": 7.7134,
      "mean_token_accuracy": 0.07555579096078872,
      "num_tokens": 659924.0,
      "step": 305
    },
    {
      "entropy": 8.075908136367797,
      "epoch": 0.03317459468136337,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.8166,
      "mean_token_accuracy": 0.07576014548540115,
      "num_tokens": 671120.0,
      "step": 310
    },
    {
      "entropy": 8.010063886642456,
      "epoch": 0.033709668789127296,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000157,
      "loss": 7.8239,
      "mean_token_accuracy": 0.07674261815845966,
      "num_tokens": 681956.0,
      "step": 315
    },
    {
      "entropy": 8.071939706802368,
      "epoch": 0.03424474289689122,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0001595,
      "loss": 7.8005,
      "mean_token_accuracy": 0.07449028007686138,
      "num_tokens": 693226.0,
      "step": 320
    },
    {
      "entropy": 8.040113973617554,
      "epoch": 0.03477981700465514,
      "grad_norm": 1.125,
      "learning_rate": 0.000162,
      "loss": 7.813,
      "mean_token_accuracy": 0.08225274756550789,
      "num_tokens": 704691.0,
      "step": 325
    },
    {
      "entropy": 8.035579442977905,
      "epoch": 0.03531489111241907,
      "grad_norm": 1.015625,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.787,
      "mean_token_accuracy": 0.07182234078645706,
      "num_tokens": 716000.0,
      "step": 330
    },
    {
      "entropy": 8.042577505111694,
      "epoch": 0.035849965220182994,
      "grad_norm": 1.0,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.7025,
      "mean_token_accuracy": 0.08112649023532867,
      "num_tokens": 726723.0,
      "step": 335
    },
    {
      "entropy": 7.864040803909302,
      "epoch": 0.03638503932794692,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.7372,
      "mean_token_accuracy": 0.08113703206181526,
      "num_tokens": 736740.0,
      "step": 340
    },
    {
      "entropy": 8.075887537002563,
      "epoch": 0.03692011343571085,
      "grad_norm": 2.0,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.7682,
      "mean_token_accuracy": 0.07558154836297035,
      "num_tokens": 747567.0,
      "step": 345
    },
    {
      "entropy": 8.016151714324952,
      "epoch": 0.03745518754347477,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.7141,
      "mean_token_accuracy": 0.07801906056702138,
      "num_tokens": 757937.0,
      "step": 350
    },
    {
      "entropy": 7.936876964569092,
      "epoch": 0.0379902616512387,
      "grad_norm": 1.140625,
      "learning_rate": 0.000177,
      "loss": 7.7553,
      "mean_token_accuracy": 0.07931130304932595,
      "num_tokens": 768003.0,
      "step": 355
    },
    {
      "entropy": 8.079256439208985,
      "epoch": 0.03852533575900262,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0001795,
      "loss": 7.7858,
      "mean_token_accuracy": 0.07208012118935585,
      "num_tokens": 779098.0,
      "step": 360
    },
    {
      "entropy": 7.9655650615692135,
      "epoch": 0.039060409866766545,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000182,
      "loss": 7.7239,
      "mean_token_accuracy": 0.07492073476314545,
      "num_tokens": 790002.0,
      "step": 365
    },
    {
      "entropy": 8.07114543914795,
      "epoch": 0.03959548397453047,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001845,
      "loss": 7.7535,
      "mean_token_accuracy": 0.07890913225710391,
      "num_tokens": 800444.0,
      "step": 370
    },
    {
      "entropy": 7.981046056747436,
      "epoch": 0.0401305580822944,
      "grad_norm": 1.140625,
      "learning_rate": 0.000187,
      "loss": 7.6963,
      "mean_token_accuracy": 0.08140306100249291,
      "num_tokens": 810912.0,
      "step": 375
    },
    {
      "entropy": 7.970801067352295,
      "epoch": 0.04066563219005832,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0001895,
      "loss": 7.7453,
      "mean_token_accuracy": 0.07234299220144749,
      "num_tokens": 822809.0,
      "step": 380
    },
    {
      "entropy": 8.04731183052063,
      "epoch": 0.04120070629782225,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000192,
      "loss": 7.7301,
      "mean_token_accuracy": 0.08183165267109871,
      "num_tokens": 833362.0,
      "step": 385
    },
    {
      "entropy": 7.919857549667358,
      "epoch": 0.041735780405586176,
      "grad_norm": 1.109375,
      "learning_rate": 0.0001945,
      "loss": 7.7339,
      "mean_token_accuracy": 0.07847344167530537,
      "num_tokens": 844239.0,
      "step": 390
    },
    {
      "entropy": 7.939679098129273,
      "epoch": 0.0422708545133501,
      "grad_norm": 1.078125,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.7468,
      "mean_token_accuracy": 0.07599632740020752,
      "num_tokens": 855178.0,
      "step": 395
    },
    {
      "entropy": 8.02885913848877,
      "epoch": 0.04280592862111402,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.7205,
      "mean_token_accuracy": 0.07995647490024567,
      "num_tokens": 866120.0,
      "step": 400
    },
    {
      "entropy": 7.912197017669678,
      "epoch": 0.04334100272887795,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000202,
      "loss": 7.7431,
      "mean_token_accuracy": 0.08108401894569398,
      "num_tokens": 876569.0,
      "step": 405
    },
    {
      "entropy": 8.096347093582153,
      "epoch": 0.043876076836641874,
      "grad_norm": 1.078125,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.7897,
      "mean_token_accuracy": 0.07478848099708557,
      "num_tokens": 887580.0,
      "step": 410
    },
    {
      "entropy": 7.903347396850586,
      "epoch": 0.0444111509444058,
      "grad_norm": 1.140625,
      "learning_rate": 0.000207,
      "loss": 7.6753,
      "mean_token_accuracy": 0.07557728588581085,
      "num_tokens": 897922.0,
      "step": 415
    },
    {
      "entropy": 7.95268235206604,
      "epoch": 0.044946225052169726,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002095,
      "loss": 7.6564,
      "mean_token_accuracy": 0.08323289379477501,
      "num_tokens": 908522.0,
      "step": 420
    },
    {
      "entropy": 7.964509391784668,
      "epoch": 0.04548129915993365,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000212,
      "loss": 7.6832,
      "mean_token_accuracy": 0.07836998216807842,
      "num_tokens": 918253.0,
      "step": 425
    },
    {
      "entropy": 7.942329549789429,
      "epoch": 0.04601637326769758,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0002145,
      "loss": 7.7008,
      "mean_token_accuracy": 0.08039597496390342,
      "num_tokens": 929150.0,
      "step": 430
    },
    {
      "entropy": 7.961909246444702,
      "epoch": 0.0465514473754615,
      "grad_norm": 1.03125,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.7507,
      "mean_token_accuracy": 0.08034233450889587,
      "num_tokens": 941041.0,
      "step": 435
    },
    {
      "entropy": 7.871124219894409,
      "epoch": 0.047086521483225424,
      "grad_norm": 1.671875,
      "learning_rate": 0.0002195,
      "loss": 7.6639,
      "mean_token_accuracy": 0.08978669568896294,
      "num_tokens": 952714.0,
      "step": 440
    },
    {
      "entropy": 7.911958646774292,
      "epoch": 0.04762159559098935,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000222,
      "loss": 7.7413,
      "mean_token_accuracy": 0.0767839141190052,
      "num_tokens": 964330.0,
      "step": 445
    },
    {
      "entropy": 7.94795036315918,
      "epoch": 0.04815666969875328,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002245,
      "loss": 7.6848,
      "mean_token_accuracy": 0.07758989408612252,
      "num_tokens": 975257.0,
      "step": 450
    },
    {
      "entropy": 7.941993951797485,
      "epoch": 0.0486917438065172,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.6847,
      "mean_token_accuracy": 0.08070862591266632,
      "num_tokens": 986335.0,
      "step": 455
    },
    {
      "entropy": 7.828902101516723,
      "epoch": 0.04922681791428113,
      "grad_norm": 0.96875,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.6039,
      "mean_token_accuracy": 0.08260082677006722,
      "num_tokens": 997521.0,
      "step": 460
    },
    {
      "entropy": 7.859019327163696,
      "epoch": 0.049761892022045055,
      "grad_norm": 1.078125,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.642,
      "mean_token_accuracy": 0.07671640515327453,
      "num_tokens": 1008383.0,
      "step": 465
    },
    {
      "entropy": 7.908183193206787,
      "epoch": 0.05029696612980898,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.7061,
      "mean_token_accuracy": 0.08019287064671517,
      "num_tokens": 1020552.0,
      "step": 470
    },
    {
      "entropy": 7.827936267852783,
      "epoch": 0.0508320402375729,
      "grad_norm": 1.0625,
      "learning_rate": 0.000237,
      "loss": 7.6924,
      "mean_token_accuracy": 0.07920794561505318,
      "num_tokens": 1031127.0,
      "step": 475
    },
    {
      "entropy": 7.8519299030303955,
      "epoch": 0.05136711434533683,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002395,
      "loss": 7.6252,
      "mean_token_accuracy": 0.0852290503680706,
      "num_tokens": 1043209.0,
      "step": 480
    },
    {
      "entropy": 8.008257150650024,
      "epoch": 0.05190218845310075,
      "grad_norm": 1.046875,
      "learning_rate": 0.000242,
      "loss": 7.7358,
      "mean_token_accuracy": 0.07626549825072289,
      "num_tokens": 1054032.0,
      "step": 485
    },
    {
      "entropy": 7.757750940322876,
      "epoch": 0.05243726256086468,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002445,
      "loss": 7.5824,
      "mean_token_accuracy": 0.08520738780498505,
      "num_tokens": 1064565.0,
      "step": 490
    },
    {
      "entropy": 7.949021720886231,
      "epoch": 0.052972336668628606,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000247,
      "loss": 7.7047,
      "mean_token_accuracy": 0.08036523126065731,
      "num_tokens": 1076352.0,
      "step": 495
    },
    {
      "entropy": 7.884907054901123,
      "epoch": 0.05350741077639253,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002495,
      "loss": 7.6239,
      "mean_token_accuracy": 0.0878230832517147,
      "num_tokens": 1087697.0,
      "step": 500
    },
    {
      "entropy": 7.8866808891296385,
      "epoch": 0.05404248488415646,
      "grad_norm": 1.03125,
      "learning_rate": 0.000252,
      "loss": 7.6108,
      "mean_token_accuracy": 0.08426605761051179,
      "num_tokens": 1098056.0,
      "step": 505
    },
    {
      "entropy": 7.793342256546021,
      "epoch": 0.054577558991920384,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0002545,
      "loss": 7.5685,
      "mean_token_accuracy": 0.08895699828863143,
      "num_tokens": 1109219.0,
      "step": 510
    },
    {
      "entropy": 7.788063955307007,
      "epoch": 0.055112633099684304,
      "grad_norm": 1.078125,
      "learning_rate": 0.000257,
      "loss": 7.6909,
      "mean_token_accuracy": 0.07860136553645133,
      "num_tokens": 1120248.0,
      "step": 515
    },
    {
      "entropy": 7.8371082782745365,
      "epoch": 0.05564770720744823,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002595,
      "loss": 7.6891,
      "mean_token_accuracy": 0.08039635419845581,
      "num_tokens": 1131278.0,
      "step": 520
    },
    {
      "entropy": 7.913965082168579,
      "epoch": 0.056182781315212156,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000262,
      "loss": 7.6912,
      "mean_token_accuracy": 0.07047672085464,
      "num_tokens": 1143147.0,
      "step": 525
    },
    {
      "entropy": 7.861634349822998,
      "epoch": 0.05671785542297608,
      "grad_norm": 0.96875,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.6163,
      "mean_token_accuracy": 0.07919415943324566,
      "num_tokens": 1154316.0,
      "step": 530
    },
    {
      "entropy": 7.795156002044678,
      "epoch": 0.05725292953074001,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.6284,
      "mean_token_accuracy": 0.08119702488183975,
      "num_tokens": 1164696.0,
      "step": 535
    },
    {
      "entropy": 7.818248176574707,
      "epoch": 0.057788003638503935,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.6079,
      "mean_token_accuracy": 0.08769951537251472,
      "num_tokens": 1176129.0,
      "step": 540
    },
    {
      "entropy": 7.891226720809937,
      "epoch": 0.05832307774626786,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.6067,
      "mean_token_accuracy": 0.0797014981508255,
      "num_tokens": 1186499.0,
      "step": 545
    },
    {
      "entropy": 7.75864052772522,
      "epoch": 0.05885815185403178,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002745,
      "loss": 7.6008,
      "mean_token_accuracy": 0.078534334897995,
      "num_tokens": 1196841.0,
      "step": 550
    },
    {
      "entropy": 7.817751502990722,
      "epoch": 0.059393225961795706,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000277,
      "loss": 7.5394,
      "mean_token_accuracy": 0.08645424395799636,
      "num_tokens": 1207647.0,
      "step": 555
    },
    {
      "entropy": 7.717403888702393,
      "epoch": 0.05992830006955963,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002795,
      "loss": 7.5438,
      "mean_token_accuracy": 0.08563198633491993,
      "num_tokens": 1217072.0,
      "step": 560
    },
    {
      "entropy": 7.8630136013031,
      "epoch": 0.06046337417732356,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.673,
      "mean_token_accuracy": 0.07517069838941097,
      "num_tokens": 1228564.0,
      "step": 565
    },
    {
      "entropy": 7.755182695388794,
      "epoch": 0.060998448285087485,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002845,
      "loss": 7.6117,
      "mean_token_accuracy": 0.08110473416745663,
      "num_tokens": 1238755.0,
      "step": 570
    },
    {
      "entropy": 7.747598361968994,
      "epoch": 0.06153352239285141,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000287,
      "loss": 7.5684,
      "mean_token_accuracy": 0.08383116647601127,
      "num_tokens": 1248968.0,
      "step": 575
    },
    {
      "entropy": 7.79936990737915,
      "epoch": 0.06206859650061534,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002895,
      "loss": 7.5403,
      "mean_token_accuracy": 0.08267647698521614,
      "num_tokens": 1259256.0,
      "step": 580
    },
    {
      "entropy": 7.712107229232788,
      "epoch": 0.06260367060837926,
      "grad_norm": 1.1875,
      "learning_rate": 0.000292,
      "loss": 7.5666,
      "mean_token_accuracy": 0.08285431116819382,
      "num_tokens": 1270469.0,
      "step": 585
    },
    {
      "entropy": 7.765189361572266,
      "epoch": 0.06313874471614318,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002945,
      "loss": 7.5491,
      "mean_token_accuracy": 0.08401528745889664,
      "num_tokens": 1280558.0,
      "step": 590
    },
    {
      "entropy": 7.807211065292359,
      "epoch": 0.06367381882390712,
      "grad_norm": 1.171875,
      "learning_rate": 0.000297,
      "loss": 7.6139,
      "mean_token_accuracy": 0.08002614453434945,
      "num_tokens": 1291372.0,
      "step": 595
    },
    {
      "entropy": 7.829832220077515,
      "epoch": 0.06420889293167104,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002995,
      "loss": 7.6538,
      "mean_token_accuracy": 0.07477352321147919,
      "num_tokens": 1302521.0,
      "step": 600
    },
    {
      "entropy": 7.637760925292969,
      "epoch": 0.06474396703943497,
      "grad_norm": 1.0,
      "learning_rate": 0.000302,
      "loss": 7.4977,
      "mean_token_accuracy": 0.08267719894647599,
      "num_tokens": 1313632.0,
      "step": 605
    },
    {
      "entropy": 7.775080394744873,
      "epoch": 0.06527904114719889,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003045,
      "loss": 7.5881,
      "mean_token_accuracy": 0.08696693032979966,
      "num_tokens": 1325247.0,
      "step": 610
    },
    {
      "entropy": 7.754818916320801,
      "epoch": 0.06581411525496281,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000307,
      "loss": 7.5746,
      "mean_token_accuracy": 0.08667751327157021,
      "num_tokens": 1336026.0,
      "step": 615
    },
    {
      "entropy": 7.712781190872192,
      "epoch": 0.06634918936272674,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003095,
      "loss": 7.502,
      "mean_token_accuracy": 0.08229328319430351,
      "num_tokens": 1346920.0,
      "step": 620
    },
    {
      "entropy": 7.714415168762207,
      "epoch": 0.06688426347049066,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000312,
      "loss": 7.6459,
      "mean_token_accuracy": 0.08195903673768043,
      "num_tokens": 1359752.0,
      "step": 625
    },
    {
      "entropy": 7.818537712097168,
      "epoch": 0.06741933757825459,
      "grad_norm": 1.0,
      "learning_rate": 0.0003145,
      "loss": 7.6415,
      "mean_token_accuracy": 0.08333790749311447,
      "num_tokens": 1370874.0,
      "step": 630
    },
    {
      "entropy": 7.740803337097168,
      "epoch": 0.06795441168601851,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000317,
      "loss": 7.5492,
      "mean_token_accuracy": 0.0865517497062683,
      "num_tokens": 1382316.0,
      "step": 635
    },
    {
      "entropy": 7.679195737838745,
      "epoch": 0.06848948579378245,
      "grad_norm": 1.125,
      "learning_rate": 0.0003195,
      "loss": 7.5851,
      "mean_token_accuracy": 0.07738500684499741,
      "num_tokens": 1394656.0,
      "step": 640
    },
    {
      "entropy": 7.756368350982666,
      "epoch": 0.06902455990154636,
      "grad_norm": 1.09375,
      "learning_rate": 0.000322,
      "loss": 7.5499,
      "mean_token_accuracy": 0.08601778075098991,
      "num_tokens": 1405061.0,
      "step": 645
    },
    {
      "entropy": 7.646595811843872,
      "epoch": 0.06955963400931028,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.5231,
      "mean_token_accuracy": 0.08854705616831779,
      "num_tokens": 1415573.0,
      "step": 650
    },
    {
      "entropy": 7.6589337348937985,
      "epoch": 0.07009470811707422,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.5524,
      "mean_token_accuracy": 0.08328185752034187,
      "num_tokens": 1426103.0,
      "step": 655
    },
    {
      "entropy": 7.728953456878662,
      "epoch": 0.07062978222483814,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.5573,
      "mean_token_accuracy": 0.07903586179018021,
      "num_tokens": 1437256.0,
      "step": 660
    },
    {
      "entropy": 7.691567277908325,
      "epoch": 0.07116485633260207,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.521,
      "mean_token_accuracy": 0.08464005514979363,
      "num_tokens": 1446975.0,
      "step": 665
    },
    {
      "entropy": 7.705724573135376,
      "epoch": 0.07169993044036599,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.5353,
      "mean_token_accuracy": 0.080365489423275,
      "num_tokens": 1457751.0,
      "step": 670
    },
    {
      "entropy": 7.651171350479126,
      "epoch": 0.07223500454812992,
      "grad_norm": 1.046875,
      "learning_rate": 0.000337,
      "loss": 7.5179,
      "mean_token_accuracy": 0.08747594058513641,
      "num_tokens": 1468186.0,
      "step": 675
    },
    {
      "entropy": 7.646878385543824,
      "epoch": 0.07277007865589384,
      "grad_norm": 0.875,
      "learning_rate": 0.0003395,
      "loss": 7.5674,
      "mean_token_accuracy": 0.08277356252074242,
      "num_tokens": 1479780.0,
      "step": 680
    },
    {
      "entropy": 7.6683413028717045,
      "epoch": 0.07330515276365776,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000342,
      "loss": 7.4726,
      "mean_token_accuracy": 0.0866861030459404,
      "num_tokens": 1490565.0,
      "step": 685
    },
    {
      "entropy": 7.719349336624146,
      "epoch": 0.0738402268714217,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.5349,
      "mean_token_accuracy": 0.08366219028830528,
      "num_tokens": 1501552.0,
      "step": 690
    },
    {
      "entropy": 7.626680374145508,
      "epoch": 0.07437530097918561,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000347,
      "loss": 7.5036,
      "mean_token_accuracy": 0.07987239845097065,
      "num_tokens": 1513311.0,
      "step": 695
    },
    {
      "entropy": 7.682998895645142,
      "epoch": 0.07491037508694955,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003495,
      "loss": 7.5551,
      "mean_token_accuracy": 0.08537785261869431,
      "num_tokens": 1524142.0,
      "step": 700
    },
    {
      "entropy": 7.614016437530518,
      "epoch": 0.07544544919471347,
      "grad_norm": 1.0625,
      "learning_rate": 0.000352,
      "loss": 7.5286,
      "mean_token_accuracy": 0.08181034475564956,
      "num_tokens": 1535571.0,
      "step": 705
    },
    {
      "entropy": 7.77970175743103,
      "epoch": 0.0759805233024774,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003545,
      "loss": 7.5527,
      "mean_token_accuracy": 0.08348672389984131,
      "num_tokens": 1546528.0,
      "step": 710
    },
    {
      "entropy": 7.565948343276977,
      "epoch": 0.07651559741024132,
      "grad_norm": 0.9375,
      "learning_rate": 0.000357,
      "loss": 7.5634,
      "mean_token_accuracy": 0.0795982949435711,
      "num_tokens": 1557264.0,
      "step": 715
    },
    {
      "entropy": 7.680731344223022,
      "epoch": 0.07705067151800524,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003595,
      "loss": 7.4939,
      "mean_token_accuracy": 0.0882859081029892,
      "num_tokens": 1566773.0,
      "step": 720
    },
    {
      "entropy": 7.545563220977783,
      "epoch": 0.07758574562576917,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000362,
      "loss": 7.4698,
      "mean_token_accuracy": 0.09191560298204422,
      "num_tokens": 1578173.0,
      "step": 725
    },
    {
      "entropy": 7.6455999374389645,
      "epoch": 0.07812081973353309,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003645,
      "loss": 7.4611,
      "mean_token_accuracy": 0.0855359174311161,
      "num_tokens": 1588537.0,
      "step": 730
    },
    {
      "entropy": 7.49965648651123,
      "epoch": 0.07865589384129702,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000367,
      "loss": 7.4594,
      "mean_token_accuracy": 0.09296399280428887,
      "num_tokens": 1599159.0,
      "step": 735
    },
    {
      "entropy": 7.620386981964112,
      "epoch": 0.07919096794906094,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003695,
      "loss": 7.5168,
      "mean_token_accuracy": 0.0834181286394596,
      "num_tokens": 1610378.0,
      "step": 740
    },
    {
      "entropy": 7.585515832901001,
      "epoch": 0.07972604205682488,
      "grad_norm": 1.1875,
      "learning_rate": 0.000372,
      "loss": 7.5421,
      "mean_token_accuracy": 0.08486890234053135,
      "num_tokens": 1621215.0,
      "step": 745
    },
    {
      "entropy": 7.641656970977783,
      "epoch": 0.0802611161645888,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003745,
      "loss": 7.5105,
      "mean_token_accuracy": 0.08722797259688378,
      "num_tokens": 1631170.0,
      "step": 750
    },
    {
      "entropy": 7.653715991973877,
      "epoch": 0.08079619027235273,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000377,
      "loss": 7.5106,
      "mean_token_accuracy": 0.07990852892398834,
      "num_tokens": 1641897.0,
      "step": 755
    },
    {
      "entropy": 7.6051116466522215,
      "epoch": 0.08133126438011665,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003795,
      "loss": 7.5798,
      "mean_token_accuracy": 0.08361625298857689,
      "num_tokens": 1653592.0,
      "step": 760
    },
    {
      "entropy": 7.5156501770019535,
      "epoch": 0.08186633848788057,
      "grad_norm": 1.03125,
      "learning_rate": 0.000382,
      "loss": 7.5134,
      "mean_token_accuracy": 0.0794034518301487,
      "num_tokens": 1665760.0,
      "step": 765
    },
    {
      "entropy": 7.652586936950684,
      "epoch": 0.0824014125956445,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003845,
      "loss": 7.4808,
      "mean_token_accuracy": 0.09275959730148316,
      "num_tokens": 1676211.0,
      "step": 770
    },
    {
      "entropy": 7.628318643569946,
      "epoch": 0.08293648670340842,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.5125,
      "mean_token_accuracy": 0.08556769341230393,
      "num_tokens": 1686405.0,
      "step": 775
    },
    {
      "entropy": 7.617849349975586,
      "epoch": 0.08347156081117235,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.5086,
      "mean_token_accuracy": 0.08250572569668294,
      "num_tokens": 1697340.0,
      "step": 780
    },
    {
      "entropy": 7.555404567718506,
      "epoch": 0.08400663491893627,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.4003,
      "mean_token_accuracy": 0.09449514746665955,
      "num_tokens": 1707250.0,
      "step": 785
    },
    {
      "entropy": 7.466612195968628,
      "epoch": 0.0845417090267002,
      "grad_norm": 1.375,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.3975,
      "mean_token_accuracy": 0.08475929126143456,
      "num_tokens": 1718333.0,
      "step": 790
    },
    {
      "entropy": 7.5835918426513675,
      "epoch": 0.08507678313446412,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.4962,
      "mean_token_accuracy": 0.0763777457177639,
      "num_tokens": 1730520.0,
      "step": 795
    },
    {
      "entropy": 7.55741515159607,
      "epoch": 0.08561185724222804,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003995,
      "loss": 7.5552,
      "mean_token_accuracy": 0.07963615767657757,
      "num_tokens": 1741898.0,
      "step": 800
    },
    {
      "entropy": 7.54046630859375,
      "epoch": 0.08614693134999198,
      "grad_norm": 1.0625,
      "learning_rate": 0.000402,
      "loss": 7.4987,
      "mean_token_accuracy": 0.08333209455013275,
      "num_tokens": 1751671.0,
      "step": 805
    },
    {
      "entropy": 7.5205401420593265,
      "epoch": 0.0866820054577559,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004045,
      "loss": 7.5163,
      "mean_token_accuracy": 0.0879331462085247,
      "num_tokens": 1762076.0,
      "step": 810
    },
    {
      "entropy": 7.609332609176636,
      "epoch": 0.08721707956551983,
      "grad_norm": 1.0,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.4735,
      "mean_token_accuracy": 0.08462397307157517,
      "num_tokens": 1772666.0,
      "step": 815
    },
    {
      "entropy": 7.601525020599365,
      "epoch": 0.08775215367328375,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004095,
      "loss": 7.5034,
      "mean_token_accuracy": 0.08532712645828724,
      "num_tokens": 1783621.0,
      "step": 820
    },
    {
      "entropy": 7.5773783206939695,
      "epoch": 0.08828722778104768,
      "grad_norm": 1.5625,
      "learning_rate": 0.000412,
      "loss": 7.5739,
      "mean_token_accuracy": 0.0806800290942192,
      "num_tokens": 1794654.0,
      "step": 825
    },
    {
      "entropy": 7.463850975036621,
      "epoch": 0.0888223018888116,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004145,
      "loss": 7.4682,
      "mean_token_accuracy": 0.08854547813534737,
      "num_tokens": 1805853.0,
      "step": 830
    },
    {
      "entropy": 7.505795669555664,
      "epoch": 0.08935737599657552,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000417,
      "loss": 7.4489,
      "mean_token_accuracy": 0.08519805371761321,
      "num_tokens": 1817006.0,
      "step": 835
    },
    {
      "entropy": 7.583885574340821,
      "epoch": 0.08989245010433945,
      "grad_norm": 1.0,
      "learning_rate": 0.0004195,
      "loss": 7.4355,
      "mean_token_accuracy": 0.08483640179038048,
      "num_tokens": 1827166.0,
      "step": 840
    },
    {
      "entropy": 7.5850004196167,
      "epoch": 0.09042752421210337,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000422,
      "loss": 7.525,
      "mean_token_accuracy": 0.07825911939144134,
      "num_tokens": 1837182.0,
      "step": 845
    },
    {
      "entropy": 7.491214847564697,
      "epoch": 0.0909625983198673,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004245,
      "loss": 7.4868,
      "mean_token_accuracy": 0.08787625655531883,
      "num_tokens": 1849034.0,
      "step": 850
    },
    {
      "entropy": 7.437751960754395,
      "epoch": 0.09149767242763122,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000427,
      "loss": 7.3616,
      "mean_token_accuracy": 0.09659026935696602,
      "num_tokens": 1859033.0,
      "step": 855
    },
    {
      "entropy": 7.510915803909302,
      "epoch": 0.09203274653539516,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004295,
      "loss": 7.4807,
      "mean_token_accuracy": 0.08575069010257722,
      "num_tokens": 1870274.0,
      "step": 860
    },
    {
      "entropy": 7.390764904022217,
      "epoch": 0.09256782064315908,
      "grad_norm": 1.296875,
      "learning_rate": 0.000432,
      "loss": 7.4171,
      "mean_token_accuracy": 0.09046047702431678,
      "num_tokens": 1881306.0,
      "step": 865
    },
    {
      "entropy": 7.484241676330567,
      "epoch": 0.093102894750923,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004345,
      "loss": 7.4006,
      "mean_token_accuracy": 0.08314777314662933,
      "num_tokens": 1892464.0,
      "step": 870
    },
    {
      "entropy": 7.537802791595459,
      "epoch": 0.09363796885868693,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000437,
      "loss": 7.4881,
      "mean_token_accuracy": 0.08111159168183804,
      "num_tokens": 1903743.0,
      "step": 875
    },
    {
      "entropy": 7.457689571380615,
      "epoch": 0.09417304296645085,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004395,
      "loss": 7.4545,
      "mean_token_accuracy": 0.08882787078619003,
      "num_tokens": 1915253.0,
      "step": 880
    },
    {
      "entropy": 7.526367330551148,
      "epoch": 0.09470811707421478,
      "grad_norm": 1.453125,
      "learning_rate": 0.000442,
      "loss": 7.3926,
      "mean_token_accuracy": 0.0867061398923397,
      "num_tokens": 1926113.0,
      "step": 885
    },
    {
      "entropy": 7.4687563419342045,
      "epoch": 0.0952431911819787,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004445,
      "loss": 7.5099,
      "mean_token_accuracy": 0.08852507770061493,
      "num_tokens": 1937718.0,
      "step": 890
    },
    {
      "entropy": 7.511999654769897,
      "epoch": 0.09577826528974263,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000447,
      "loss": 7.5,
      "mean_token_accuracy": 0.0858601987361908,
      "num_tokens": 1948307.0,
      "step": 895
    },
    {
      "entropy": 7.451641511917114,
      "epoch": 0.09631333939750655,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.3121,
      "mean_token_accuracy": 0.09499122053384781,
      "num_tokens": 1957861.0,
      "step": 900
    },
    {
      "entropy": 7.375934410095215,
      "epoch": 0.09684841350527049,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.353,
      "mean_token_accuracy": 0.09795267134904861,
      "num_tokens": 1967984.0,
      "step": 905
    },
    {
      "entropy": 7.425767183303833,
      "epoch": 0.0973834876130344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.344,
      "mean_token_accuracy": 0.09382129013538361,
      "num_tokens": 1978608.0,
      "step": 910
    },
    {
      "entropy": 7.466197538375854,
      "epoch": 0.09791856172079832,
      "grad_norm": 1.0,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.3934,
      "mean_token_accuracy": 0.08854326158761978,
      "num_tokens": 1989355.0,
      "step": 915
    },
    {
      "entropy": 7.344980144500733,
      "epoch": 0.09845363582856226,
      "grad_norm": 1.0,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.3994,
      "mean_token_accuracy": 0.09236646220088005,
      "num_tokens": 1999754.0,
      "step": 920
    },
    {
      "entropy": 7.47080078125,
      "epoch": 0.09898870993632618,
      "grad_norm": 1.046875,
      "learning_rate": 0.000462,
      "loss": 7.3879,
      "mean_token_accuracy": 0.08912354037165642,
      "num_tokens": 2010877.0,
      "step": 925
    },
    {
      "entropy": 7.541212749481201,
      "epoch": 0.09952378404409011,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004645,
      "loss": 7.4925,
      "mean_token_accuracy": 0.08468883112072945,
      "num_tokens": 2022301.0,
      "step": 930
    },
    {
      "entropy": 7.4023819923400875,
      "epoch": 0.10005885815185403,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000467,
      "loss": 7.3971,
      "mean_token_accuracy": 0.08642457351088524,
      "num_tokens": 2033537.0,
      "step": 935
    },
    {
      "entropy": 7.3656556606292725,
      "epoch": 0.10059393225961796,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004695,
      "loss": 7.3013,
      "mean_token_accuracy": 0.08984294459223748,
      "num_tokens": 2044360.0,
      "step": 940
    },
    {
      "entropy": 7.446710872650146,
      "epoch": 0.10112900636738188,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000472,
      "loss": 7.441,
      "mean_token_accuracy": 0.08957641422748566,
      "num_tokens": 2055000.0,
      "step": 945
    },
    {
      "entropy": 7.370997333526612,
      "epoch": 0.1016640804751458,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004745,
      "loss": 7.4239,
      "mean_token_accuracy": 0.09220219105482101,
      "num_tokens": 2064807.0,
      "step": 950
    },
    {
      "entropy": 7.488531112670898,
      "epoch": 0.10219915458290973,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000477,
      "loss": 7.3926,
      "mean_token_accuracy": 0.08401189967989922,
      "num_tokens": 2075618.0,
      "step": 955
    },
    {
      "entropy": 7.366859388351441,
      "epoch": 0.10273422869067365,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004795,
      "loss": 7.3898,
      "mean_token_accuracy": 0.08558554649353027,
      "num_tokens": 2086226.0,
      "step": 960
    },
    {
      "entropy": 7.435228681564331,
      "epoch": 0.10326930279843759,
      "grad_norm": 0.953125,
      "learning_rate": 0.000482,
      "loss": 7.3806,
      "mean_token_accuracy": 0.08150073923170567,
      "num_tokens": 2097222.0,
      "step": 965
    },
    {
      "entropy": 7.409606075286865,
      "epoch": 0.1038043769062015,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004845,
      "loss": 7.422,
      "mean_token_accuracy": 0.090543332695961,
      "num_tokens": 2108919.0,
      "step": 970
    },
    {
      "entropy": 7.371023559570313,
      "epoch": 0.10433945101396544,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000487,
      "loss": 7.3811,
      "mean_token_accuracy": 0.08817952051758766,
      "num_tokens": 2120215.0,
      "step": 975
    },
    {
      "entropy": 7.42234091758728,
      "epoch": 0.10487452512172936,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004895,
      "loss": 7.4002,
      "mean_token_accuracy": 0.08778398633003234,
      "num_tokens": 2131503.0,
      "step": 980
    },
    {
      "entropy": 7.365479326248169,
      "epoch": 0.10540959922949328,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000492,
      "loss": 7.3376,
      "mean_token_accuracy": 0.09631536826491356,
      "num_tokens": 2141551.0,
      "step": 985
    },
    {
      "entropy": 7.368793535232544,
      "epoch": 0.10594467333725721,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004945,
      "loss": 7.3454,
      "mean_token_accuracy": 0.09466473385691643,
      "num_tokens": 2151821.0,
      "step": 990
    },
    {
      "entropy": 7.307546615600586,
      "epoch": 0.10647974744502113,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000497,
      "loss": 7.3927,
      "mean_token_accuracy": 0.08924503847956658,
      "num_tokens": 2162474.0,
      "step": 995
    },
    {
      "entropy": 7.453886079788208,
      "epoch": 0.10701482155278506,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995,
      "loss": 7.4497,
      "mean_token_accuracy": 0.09008113294839859,
      "num_tokens": 2173985.0,
      "step": 1000
    },
    {
      "entropy": 7.3950518608093265,
      "epoch": 0.10754989566054898,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999999979210101,
      "loss": 7.4669,
      "mean_token_accuracy": 0.08625381216406822,
      "num_tokens": 2186432.0,
      "step": 1005
    },
    {
      "entropy": 7.386218738555908,
      "epoch": 0.10808496976831292,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999999894751138,
      "loss": 7.331,
      "mean_token_accuracy": 0.08911918997764587,
      "num_tokens": 2197085.0,
      "step": 1010
    },
    {
      "entropy": 7.326195859909058,
      "epoch": 0.10862004387607684,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999999745323744,
      "loss": 7.3106,
      "mean_token_accuracy": 0.09575477167963982,
      "num_tokens": 2207162.0,
      "step": 1015
    },
    {
      "entropy": 7.389324140548706,
      "epoch": 0.10915511798384077,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999999530927923,
      "loss": 7.4269,
      "mean_token_accuracy": 0.08239098042249679,
      "num_tokens": 2218714.0,
      "step": 1020
    },
    {
      "entropy": 7.396574306488037,
      "epoch": 0.10969019209160469,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999999251563682,
      "loss": 7.3741,
      "mean_token_accuracy": 0.08852500319480897,
      "num_tokens": 2229711.0,
      "step": 1025
    },
    {
      "entropy": 7.334770202636719,
      "epoch": 0.11022526619936861,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999998907231029,
      "loss": 7.368,
      "mean_token_accuracy": 0.08981707468628883,
      "num_tokens": 2239885.0,
      "step": 1030
    },
    {
      "entropy": 7.3592136859893795,
      "epoch": 0.11076034030713254,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999998497929974,
      "loss": 7.3731,
      "mean_token_accuracy": 0.08999298885464668,
      "num_tokens": 2250221.0,
      "step": 1035
    },
    {
      "entropy": 7.336800050735474,
      "epoch": 0.11129541441489646,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999998023660527,
      "loss": 7.4528,
      "mean_token_accuracy": 0.0816170632839203,
      "num_tokens": 2261059.0,
      "step": 1040
    },
    {
      "entropy": 7.3705222606658936,
      "epoch": 0.11183048852266039,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999997484422705,
      "loss": 7.2994,
      "mean_token_accuracy": 0.09296035394072533,
      "num_tokens": 2271174.0,
      "step": 1045
    },
    {
      "entropy": 7.307314395904541,
      "epoch": 0.11236556263042431,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999996880216521,
      "loss": 7.402,
      "mean_token_accuracy": 0.08877327442169189,
      "num_tokens": 2282431.0,
      "step": 1050
    },
    {
      "entropy": 7.3447693347930905,
      "epoch": 0.11290063673818825,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999996211041994,
      "loss": 7.3696,
      "mean_token_accuracy": 0.09118950441479683,
      "num_tokens": 2292891.0,
      "step": 1055
    },
    {
      "entropy": 7.394657135009766,
      "epoch": 0.11343571084595216,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999995476899141,
      "loss": 7.3946,
      "mean_token_accuracy": 0.08678853958845138,
      "num_tokens": 2303764.0,
      "step": 1060
    },
    {
      "entropy": 7.24360089302063,
      "epoch": 0.11397078495371608,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999994677787987,
      "loss": 7.306,
      "mean_token_accuracy": 0.09111010134220124,
      "num_tokens": 2314172.0,
      "step": 1065
    },
    {
      "entropy": 7.3992757320404055,
      "epoch": 0.11450585906148002,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999993813708551,
      "loss": 7.2911,
      "mean_token_accuracy": 0.09467030167579651,
      "num_tokens": 2324848.0,
      "step": 1070
    },
    {
      "entropy": 7.23645977973938,
      "epoch": 0.11504093316924394,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499999288466086,
      "loss": 7.3134,
      "mean_token_accuracy": 0.08993211090564728,
      "num_tokens": 2336601.0,
      "step": 1075
    },
    {
      "entropy": 7.271498298645019,
      "epoch": 0.11557600727700787,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999991890644941,
      "loss": 7.2982,
      "mean_token_accuracy": 0.08822696208953858,
      "num_tokens": 2347281.0,
      "step": 1080
    },
    {
      "entropy": 7.264888286590576,
      "epoch": 0.11611108138477179,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999990831660822,
      "loss": 7.3075,
      "mean_token_accuracy": 0.09646312147378922,
      "num_tokens": 2358199.0,
      "step": 1085
    },
    {
      "entropy": 7.390944719314575,
      "epoch": 0.11664615549253572,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999989707708534,
      "loss": 7.3665,
      "mean_token_accuracy": 0.08797665126621723,
      "num_tokens": 2369092.0,
      "step": 1090
    },
    {
      "entropy": 7.266194009780884,
      "epoch": 0.11718122960029964,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999988518788111,
      "loss": 7.346,
      "mean_token_accuracy": 0.09059268310666084,
      "num_tokens": 2378849.0,
      "step": 1095
    },
    {
      "entropy": 7.295655584335327,
      "epoch": 0.11771630370806356,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999987264899583,
      "loss": 7.3429,
      "mean_token_accuracy": 0.09276795759797096,
      "num_tokens": 2389587.0,
      "step": 1100
    },
    {
      "entropy": 7.434128189086914,
      "epoch": 0.1182513778158275,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499998594604299,
      "loss": 7.3326,
      "mean_token_accuracy": 0.08941550850868225,
      "num_tokens": 2400442.0,
      "step": 1105
    },
    {
      "entropy": 7.190779304504394,
      "epoch": 0.11878645192359141,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000499998456221837,
      "loss": 7.2916,
      "mean_token_accuracy": 0.09264838397502899,
      "num_tokens": 2410771.0,
      "step": 1110
    },
    {
      "entropy": 7.338149070739746,
      "epoch": 0.11932152603135535,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999983113425762,
      "loss": 7.3261,
      "mean_token_accuracy": 0.09488410204648971,
      "num_tokens": 2421112.0,
      "step": 1115
    },
    {
      "entropy": 7.299089956283569,
      "epoch": 0.11985660013911927,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999981599665207,
      "loss": 7.3102,
      "mean_token_accuracy": 0.0882981389760971,
      "num_tokens": 2431891.0,
      "step": 1120
    },
    {
      "entropy": 7.262867403030396,
      "epoch": 0.1203916742468832,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999980020936748,
      "loss": 7.2716,
      "mean_token_accuracy": 0.09344939067959786,
      "num_tokens": 2442987.0,
      "step": 1125
    },
    {
      "entropy": 7.208923196792602,
      "epoch": 0.12092674835464712,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999978377240434,
      "loss": 7.2906,
      "mean_token_accuracy": 0.09081464558839798,
      "num_tokens": 2452703.0,
      "step": 1130
    },
    {
      "entropy": 7.347393655776978,
      "epoch": 0.12146182246241104,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499997666857631,
      "loss": 7.3421,
      "mean_token_accuracy": 0.0911066085100174,
      "num_tokens": 2462975.0,
      "step": 1135
    },
    {
      "entropy": 7.26486177444458,
      "epoch": 0.12199689657017497,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999974894944426,
      "loss": 7.2973,
      "mean_token_accuracy": 0.08943878263235092,
      "num_tokens": 2473853.0,
      "step": 1140
    },
    {
      "entropy": 7.349594688415527,
      "epoch": 0.12253197067793889,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999973056344832,
      "loss": 7.2744,
      "mean_token_accuracy": 0.08951021954417229,
      "num_tokens": 2485633.0,
      "step": 1145
    },
    {
      "entropy": 7.180385971069336,
      "epoch": 0.12306704478570282,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999971152777583,
      "loss": 7.2294,
      "mean_token_accuracy": 0.08965703323483468,
      "num_tokens": 2496922.0,
      "step": 1150
    },
    {
      "entropy": 7.369343852996826,
      "epoch": 0.12360211889346674,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999969184242733,
      "loss": 7.2917,
      "mean_token_accuracy": 0.09407783150672913,
      "num_tokens": 2507890.0,
      "step": 1155
    },
    {
      "entropy": 7.187175321578979,
      "epoch": 0.12413719300123068,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499996715074034,
      "loss": 7.2939,
      "mean_token_accuracy": 0.09616836681962013,
      "num_tokens": 2519749.0,
      "step": 1160
    },
    {
      "entropy": 7.24845552444458,
      "epoch": 0.1246722671089946,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999965052270461,
      "loss": 7.3137,
      "mean_token_accuracy": 0.09240455776453019,
      "num_tokens": 2530666.0,
      "step": 1165
    },
    {
      "entropy": 7.2619400978088375,
      "epoch": 0.12520734121675853,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999962888833157,
      "loss": 7.3139,
      "mean_token_accuracy": 0.09363392889499664,
      "num_tokens": 2541594.0,
      "step": 1170
    },
    {
      "entropy": 7.283105421066284,
      "epoch": 0.12574241532452243,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999960660428491,
      "loss": 7.2992,
      "mean_token_accuracy": 0.09548846408724784,
      "num_tokens": 2552330.0,
      "step": 1175
    },
    {
      "entropy": 7.361569499969482,
      "epoch": 0.12627748943228637,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999958367056526,
      "loss": 7.3528,
      "mean_token_accuracy": 0.08479058220982552,
      "num_tokens": 2564601.0,
      "step": 1180
    },
    {
      "entropy": 7.248473691940307,
      "epoch": 0.1268125635400503,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499995600871733,
      "loss": 7.3185,
      "mean_token_accuracy": 0.09062025472521781,
      "num_tokens": 2575539.0,
      "step": 1185
    },
    {
      "entropy": 7.264167356491089,
      "epoch": 0.12734763764781423,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999953585410971,
      "loss": 7.2642,
      "mean_token_accuracy": 0.09182353541254998,
      "num_tokens": 2586635.0,
      "step": 1190
    },
    {
      "entropy": 7.278759431838989,
      "epoch": 0.12788271175557814,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999951097137518,
      "loss": 7.2785,
      "mean_token_accuracy": 0.09316747561097145,
      "num_tokens": 2596814.0,
      "step": 1195
    },
    {
      "entropy": 7.198076295852661,
      "epoch": 0.12841778586334207,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999948543897044,
      "loss": 7.3038,
      "mean_token_accuracy": 0.08896985203027725,
      "num_tokens": 2608087.0,
      "step": 1200
    },
    {
      "entropy": 7.3914577007293705,
      "epoch": 0.128952859971106,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999945925689621,
      "loss": 7.3115,
      "mean_token_accuracy": 0.09337142929434776,
      "num_tokens": 2619377.0,
      "step": 1205
    },
    {
      "entropy": 7.158187961578369,
      "epoch": 0.12948793407886994,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999943242515325,
      "loss": 7.2166,
      "mean_token_accuracy": 0.09437886625528336,
      "num_tokens": 2629089.0,
      "step": 1210
    },
    {
      "entropy": 7.2671790599823,
      "epoch": 0.13002300818663384,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999940494374236,
      "loss": 7.27,
      "mean_token_accuracy": 0.09492316246032714,
      "num_tokens": 2639743.0,
      "step": 1215
    },
    {
      "entropy": 7.342409324645996,
      "epoch": 0.13055808229439778,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499993768126643,
      "loss": 7.3257,
      "mean_token_accuracy": 0.09084919095039368,
      "num_tokens": 2650631.0,
      "step": 1220
    },
    {
      "entropy": 7.165309810638428,
      "epoch": 0.1310931564021617,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499993480319199,
      "loss": 7.266,
      "mean_token_accuracy": 0.09525621384382248,
      "num_tokens": 2660197.0,
      "step": 1225
    },
    {
      "entropy": 7.2947486400604244,
      "epoch": 0.13162823050992561,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999931860150999,
      "loss": 7.2486,
      "mean_token_accuracy": 0.09175238981842995,
      "num_tokens": 2670162.0,
      "step": 1230
    },
    {
      "entropy": 7.265993690490722,
      "epoch": 0.13216330461768955,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999928852143541,
      "loss": 7.246,
      "mean_token_accuracy": 0.09490691721439362,
      "num_tokens": 2681233.0,
      "step": 1235
    },
    {
      "entropy": 7.14599871635437,
      "epoch": 0.13269837872545348,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999925779169703,
      "loss": 7.229,
      "mean_token_accuracy": 0.09406721666455269,
      "num_tokens": 2692145.0,
      "step": 1240
    },
    {
      "entropy": 7.240669584274292,
      "epoch": 0.13323345283321741,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999922641229575,
      "loss": 7.2451,
      "mean_token_accuracy": 0.0951724499464035,
      "num_tokens": 2703192.0,
      "step": 1245
    },
    {
      "entropy": 7.287291526794434,
      "epoch": 0.13376852694098132,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999919438323248,
      "loss": 7.2557,
      "mean_token_accuracy": 0.09338495507836342,
      "num_tokens": 2714804.0,
      "step": 1250
    },
    {
      "entropy": 7.189889049530029,
      "epoch": 0.13430360104874525,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999916170450812,
      "loss": 7.2268,
      "mean_token_accuracy": 0.09526428878307343,
      "num_tokens": 2724340.0,
      "step": 1255
    },
    {
      "entropy": 7.2332368850708,
      "epoch": 0.13483867515650919,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999912837612364,
      "loss": 7.2267,
      "mean_token_accuracy": 0.09549205154180526,
      "num_tokens": 2735767.0,
      "step": 1260
    },
    {
      "entropy": 7.299641370773315,
      "epoch": 0.1353737492642731,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999909439807997,
      "loss": 7.2781,
      "mean_token_accuracy": 0.08812019154429436,
      "num_tokens": 2747658.0,
      "step": 1265
    },
    {
      "entropy": 7.161581182479859,
      "epoch": 0.13590882337203702,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999905977037813,
      "loss": 7.1398,
      "mean_token_accuracy": 0.09851738065481186,
      "num_tokens": 2757901.0,
      "step": 1270
    },
    {
      "entropy": 7.185445737838745,
      "epoch": 0.13644389747980096,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999902449301909,
      "loss": 7.2005,
      "mean_token_accuracy": 0.09269871562719345,
      "num_tokens": 2768945.0,
      "step": 1275
    },
    {
      "entropy": 7.128801107406616,
      "epoch": 0.1369789715875649,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004999898856600387,
      "loss": 7.1859,
      "mean_token_accuracy": 0.10609000697731971,
      "num_tokens": 2780276.0,
      "step": 1280
    },
    {
      "entropy": 7.273564100265503,
      "epoch": 0.1375140456953288,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999895198933354,
      "loss": 7.2574,
      "mean_token_accuracy": 0.09712079018354416,
      "num_tokens": 2791183.0,
      "step": 1285
    },
    {
      "entropy": 7.189338302612304,
      "epoch": 0.13804911980309273,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999891476300911,
      "loss": 7.1645,
      "mean_token_accuracy": 0.09277313947677612,
      "num_tokens": 2801928.0,
      "step": 1290
    },
    {
      "entropy": 7.217353820800781,
      "epoch": 0.13858419391085666,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499988768870317,
      "loss": 7.2484,
      "mean_token_accuracy": 0.08946470692753791,
      "num_tokens": 2813415.0,
      "step": 1295
    },
    {
      "entropy": 7.241245889663697,
      "epoch": 0.13911926801862057,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999883836140236,
      "loss": 7.2526,
      "mean_token_accuracy": 0.09227527827024459,
      "num_tokens": 2825288.0,
      "step": 1300
    },
    {
      "entropy": 7.2402081966400145,
      "epoch": 0.1396543421263845,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999879918612224,
      "loss": 7.2119,
      "mean_token_accuracy": 0.09847836419939995,
      "num_tokens": 2836298.0,
      "step": 1305
    },
    {
      "entropy": 7.167675733566284,
      "epoch": 0.14018941623414843,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999875936119245,
      "loss": 7.1996,
      "mean_token_accuracy": 0.09501123502850532,
      "num_tokens": 2846367.0,
      "step": 1310
    },
    {
      "entropy": 7.157606840133667,
      "epoch": 0.14072449034191237,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999871888661415,
      "loss": 7.2085,
      "mean_token_accuracy": 0.09355056807398796,
      "num_tokens": 2856832.0,
      "step": 1315
    },
    {
      "entropy": 7.234083890914917,
      "epoch": 0.14125956444967627,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499986777623885,
      "loss": 7.2303,
      "mean_token_accuracy": 0.08869453743100167,
      "num_tokens": 2867954.0,
      "step": 1320
    },
    {
      "entropy": 7.287561988830566,
      "epoch": 0.1417946385574402,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999863598851669,
      "loss": 7.2555,
      "mean_token_accuracy": 0.09643668159842492,
      "num_tokens": 2878714.0,
      "step": 1325
    },
    {
      "entropy": 6.991106748580933,
      "epoch": 0.14232971266520414,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004999859356499992,
      "loss": 7.1981,
      "mean_token_accuracy": 0.10334166139364243,
      "num_tokens": 2890697.0,
      "step": 1330
    },
    {
      "entropy": 7.246985292434692,
      "epoch": 0.14286478677296804,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999855049183943,
      "loss": 7.2216,
      "mean_token_accuracy": 0.0913633793592453,
      "num_tokens": 2901468.0,
      "step": 1335
    },
    {
      "entropy": 7.159916067123413,
      "epoch": 0.14339986088073198,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999850676903646,
      "loss": 7.1861,
      "mean_token_accuracy": 0.09884981662034989,
      "num_tokens": 2913466.0,
      "step": 1340
    },
    {
      "entropy": 7.178111553192139,
      "epoch": 0.1439349349884959,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999846239659226,
      "loss": 7.1578,
      "mean_token_accuracy": 0.1024281658232212,
      "num_tokens": 2923638.0,
      "step": 1345
    },
    {
      "entropy": 7.1597270488739015,
      "epoch": 0.14447000909625984,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999841737450812,
      "loss": 7.2306,
      "mean_token_accuracy": 0.09842249751091003,
      "num_tokens": 2934790.0,
      "step": 1350
    },
    {
      "entropy": 7.1113499164581295,
      "epoch": 0.14500508320402375,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999837170278535,
      "loss": 7.179,
      "mean_token_accuracy": 0.09384954497218131,
      "num_tokens": 2946320.0,
      "step": 1355
    },
    {
      "entropy": 7.278202247619629,
      "epoch": 0.14554015731178768,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999832538142526,
      "loss": 7.1427,
      "mean_token_accuracy": 0.09324340000748635,
      "num_tokens": 2957077.0,
      "step": 1360
    },
    {
      "entropy": 7.021162271499634,
      "epoch": 0.14607523141955162,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999827841042917,
      "loss": 7.1726,
      "mean_token_accuracy": 0.0902546539902687,
      "num_tokens": 2967286.0,
      "step": 1365
    },
    {
      "entropy": 7.243258142471314,
      "epoch": 0.14661030552731552,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004999823078979846,
      "loss": 7.2087,
      "mean_token_accuracy": 0.09563320577144623,
      "num_tokens": 2979516.0,
      "step": 1370
    },
    {
      "entropy": 7.189844799041748,
      "epoch": 0.14714537963507945,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999818251953449,
      "loss": 7.3292,
      "mean_token_accuracy": 0.09014673680067062,
      "num_tokens": 2991594.0,
      "step": 1375
    },
    {
      "entropy": 7.227779912948608,
      "epoch": 0.1476804537428434,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999813359963867,
      "loss": 7.2036,
      "mean_token_accuracy": 0.09412026777863503,
      "num_tokens": 3002487.0,
      "step": 1380
    },
    {
      "entropy": 7.201776075363159,
      "epoch": 0.14821552785060732,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999808403011241,
      "loss": 7.2109,
      "mean_token_accuracy": 0.0977674312889576,
      "num_tokens": 3013197.0,
      "step": 1385
    },
    {
      "entropy": 7.144187164306641,
      "epoch": 0.14875060195837123,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999803381095712,
      "loss": 7.2453,
      "mean_token_accuracy": 0.09458124712109565,
      "num_tokens": 3024512.0,
      "step": 1390
    },
    {
      "entropy": 7.235658025741577,
      "epoch": 0.14928567606613516,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999798294217428,
      "loss": 7.2111,
      "mean_token_accuracy": 0.09474866688251496,
      "num_tokens": 3035621.0,
      "step": 1395
    },
    {
      "entropy": 7.163389682769775,
      "epoch": 0.1498207501738991,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999793142376533,
      "loss": 7.1779,
      "mean_token_accuracy": 0.09404783025383949,
      "num_tokens": 3046492.0,
      "step": 1400
    },
    {
      "entropy": 7.159232759475708,
      "epoch": 0.150355824281663,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999787925573177,
      "loss": 7.2202,
      "mean_token_accuracy": 0.09497016146779061,
      "num_tokens": 3057715.0,
      "step": 1405
    },
    {
      "entropy": 7.300630807876587,
      "epoch": 0.15089089838942693,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999782643807513,
      "loss": 7.1824,
      "mean_token_accuracy": 0.09338649213314057,
      "num_tokens": 3068344.0,
      "step": 1410
    },
    {
      "entropy": 7.0725761413574215,
      "epoch": 0.15142597249719086,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999777297079689,
      "loss": 7.157,
      "mean_token_accuracy": 0.09506282061338425,
      "num_tokens": 3079576.0,
      "step": 1415
    },
    {
      "entropy": 7.227912521362304,
      "epoch": 0.1519610466049548,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999771885389863,
      "loss": 7.1408,
      "mean_token_accuracy": 0.09675386250019073,
      "num_tokens": 3090201.0,
      "step": 1420
    },
    {
      "entropy": 7.07097864151001,
      "epoch": 0.1524961207127187,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999766408738189,
      "loss": 7.1071,
      "mean_token_accuracy": 0.10306708961725235,
      "num_tokens": 3099824.0,
      "step": 1425
    },
    {
      "entropy": 7.119161462783813,
      "epoch": 0.15303119482048264,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999760867124827,
      "loss": 7.1967,
      "mean_token_accuracy": 0.09370788037776948,
      "num_tokens": 3109948.0,
      "step": 1430
    },
    {
      "entropy": 7.2363121032714846,
      "epoch": 0.15356626892824657,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999755260549937,
      "loss": 7.1778,
      "mean_token_accuracy": 0.10212581753730773,
      "num_tokens": 3121492.0,
      "step": 1435
    },
    {
      "entropy": 7.081883907318115,
      "epoch": 0.15410134303601047,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004999749589013677,
      "loss": 6.9771,
      "mean_token_accuracy": 0.11389621570706368,
      "num_tokens": 3131990.0,
      "step": 1440
    },
    {
      "entropy": 7.00419282913208,
      "epoch": 0.1546364171437744,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999743852516217,
      "loss": 7.146,
      "mean_token_accuracy": 0.09606294631958008,
      "num_tokens": 3142567.0,
      "step": 1445
    },
    {
      "entropy": 7.194999074935913,
      "epoch": 0.15517149125153834,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004999738051057717,
      "loss": 7.1512,
      "mean_token_accuracy": 0.09972617551684379,
      "num_tokens": 3153083.0,
      "step": 1450
    },
    {
      "entropy": 7.111984634399414,
      "epoch": 0.15570656535930227,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999732184638347,
      "loss": 7.1537,
      "mean_token_accuracy": 0.0935523934662342,
      "num_tokens": 3163985.0,
      "step": 1455
    },
    {
      "entropy": 7.149391937255859,
      "epoch": 0.15624163946706618,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999726253258278,
      "loss": 7.1476,
      "mean_token_accuracy": 0.09505921602249146,
      "num_tokens": 3175259.0,
      "step": 1460
    },
    {
      "entropy": 7.14017391204834,
      "epoch": 0.1567767135748301,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999720256917678,
      "loss": 7.1135,
      "mean_token_accuracy": 0.10178006291389466,
      "num_tokens": 3185487.0,
      "step": 1465
    },
    {
      "entropy": 7.029543256759643,
      "epoch": 0.15731178768259405,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999714195616723,
      "loss": 7.0917,
      "mean_token_accuracy": 0.10098810642957687,
      "num_tokens": 3196200.0,
      "step": 1470
    },
    {
      "entropy": 7.233420753479004,
      "epoch": 0.15784686179035795,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999708069355586,
      "loss": 7.2247,
      "mean_token_accuracy": 0.08620018288493156,
      "num_tokens": 3207675.0,
      "step": 1475
    },
    {
      "entropy": 7.065043163299561,
      "epoch": 0.15838193589812188,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999701878134445,
      "loss": 7.0881,
      "mean_token_accuracy": 0.09926359876990318,
      "num_tokens": 3218117.0,
      "step": 1480
    },
    {
      "entropy": 7.1511718273162845,
      "epoch": 0.15891701000588582,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999695621953477,
      "loss": 7.1542,
      "mean_token_accuracy": 0.10217127278447151,
      "num_tokens": 3229208.0,
      "step": 1485
    },
    {
      "entropy": 7.146858739852905,
      "epoch": 0.15945208411364975,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999689300812866,
      "loss": 7.051,
      "mean_token_accuracy": 0.10721841007471085,
      "num_tokens": 3239586.0,
      "step": 1490
    },
    {
      "entropy": 7.119947052001953,
      "epoch": 0.15998715822141366,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999682914712791,
      "loss": 7.132,
      "mean_token_accuracy": 0.10072574242949486,
      "num_tokens": 3250103.0,
      "step": 1495
    },
    {
      "entropy": 7.145537424087524,
      "epoch": 0.1605222323291776,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999676463653439,
      "loss": 7.1263,
      "mean_token_accuracy": 0.1042477697134018,
      "num_tokens": 3261844.0,
      "step": 1500
    },
    {
      "entropy": 7.1070863723754885,
      "epoch": 0.16105730643694152,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999669947634996,
      "loss": 7.0823,
      "mean_token_accuracy": 0.10051160603761673,
      "num_tokens": 3272073.0,
      "step": 1505
    },
    {
      "entropy": 7.099180889129639,
      "epoch": 0.16159238054470546,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999663366657647,
      "loss": 7.0711,
      "mean_token_accuracy": 0.09962419345974922,
      "num_tokens": 3281725.0,
      "step": 1510
    },
    {
      "entropy": 7.10269718170166,
      "epoch": 0.16212745465246936,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999656720721586,
      "loss": 7.2146,
      "mean_token_accuracy": 0.09675706773996354,
      "num_tokens": 3292708.0,
      "step": 1515
    },
    {
      "entropy": 7.203677415847778,
      "epoch": 0.1626625287602333,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999650009827004,
      "loss": 7.2064,
      "mean_token_accuracy": 0.09477584436535835,
      "num_tokens": 3303548.0,
      "step": 1520
    },
    {
      "entropy": 7.078126525878906,
      "epoch": 0.16319760286799723,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999643233974092,
      "loss": 7.0705,
      "mean_token_accuracy": 0.10342822223901749,
      "num_tokens": 3314261.0,
      "step": 1525
    },
    {
      "entropy": 7.067970943450928,
      "epoch": 0.16373267697576113,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999636393163049,
      "loss": 7.0828,
      "mean_token_accuracy": 0.10165008082985878,
      "num_tokens": 3324833.0,
      "step": 1530
    },
    {
      "entropy": 7.004071617126465,
      "epoch": 0.16426775108352507,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999629487394071,
      "loss": 6.9848,
      "mean_token_accuracy": 0.10750006660819053,
      "num_tokens": 3334790.0,
      "step": 1535
    },
    {
      "entropy": 7.110624504089356,
      "epoch": 0.164802825191289,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999622516667358,
      "loss": 7.1113,
      "mean_token_accuracy": 0.10276103839278221,
      "num_tokens": 3345115.0,
      "step": 1540
    },
    {
      "entropy": 7.02404932975769,
      "epoch": 0.16533789929905293,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000499961548098311,
      "loss": 7.0019,
      "mean_token_accuracy": 0.10630428940057754,
      "num_tokens": 3356187.0,
      "step": 1545
    },
    {
      "entropy": 7.009617757797241,
      "epoch": 0.16587297340681684,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999608380341533,
      "loss": 7.0332,
      "mean_token_accuracy": 0.10402404069900513,
      "num_tokens": 3367218.0,
      "step": 1550
    },
    {
      "entropy": 7.075384426116943,
      "epoch": 0.16640804751458077,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999601214742829,
      "loss": 7.1838,
      "mean_token_accuracy": 0.09452657997608185,
      "num_tokens": 3377557.0,
      "step": 1555
    },
    {
      "entropy": 7.236601543426514,
      "epoch": 0.1669431216223447,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999593984187206,
      "loss": 7.1688,
      "mean_token_accuracy": 0.09885648638010025,
      "num_tokens": 3387402.0,
      "step": 1560
    },
    {
      "entropy": 7.009947204589844,
      "epoch": 0.1674781957301086,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999586688674872,
      "loss": 7.0329,
      "mean_token_accuracy": 0.10211480408906937,
      "num_tokens": 3397924.0,
      "step": 1565
    },
    {
      "entropy": 7.1467907428741455,
      "epoch": 0.16801326983787254,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499957932820604,
      "loss": 7.1262,
      "mean_token_accuracy": 0.09135372191667557,
      "num_tokens": 3408622.0,
      "step": 1570
    },
    {
      "entropy": 7.128115224838257,
      "epoch": 0.16854834394563648,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499957190278092,
      "loss": 7.0847,
      "mean_token_accuracy": 0.0968036450445652,
      "num_tokens": 3419692.0,
      "step": 1575
    },
    {
      "entropy": 6.991016530990601,
      "epoch": 0.1690834180534004,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999564412399728,
      "loss": 7.1449,
      "mean_token_accuracy": 0.09633105620741844,
      "num_tokens": 3430986.0,
      "step": 1580
    },
    {
      "entropy": 7.10005898475647,
      "epoch": 0.1696184921611643,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499955685706268,
      "loss": 7.0656,
      "mean_token_accuracy": 0.10031602829694748,
      "num_tokens": 3442211.0,
      "step": 1585
    },
    {
      "entropy": 6.937238311767578,
      "epoch": 0.17015356626892825,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999549236769993,
      "loss": 7.0491,
      "mean_token_accuracy": 0.10472746640443802,
      "num_tokens": 3453585.0,
      "step": 1590
    },
    {
      "entropy": 7.097245359420777,
      "epoch": 0.17068864037669218,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999541551521888,
      "loss": 7.0967,
      "mean_token_accuracy": 0.09863503649830818,
      "num_tokens": 3464156.0,
      "step": 1595
    },
    {
      "entropy": 7.113105726242066,
      "epoch": 0.17122371448445609,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999533801318587,
      "loss": 7.1512,
      "mean_token_accuracy": 0.09326506778597832,
      "num_tokens": 3474853.0,
      "step": 1600
    },
    {
      "entropy": 7.1461693286895756,
      "epoch": 0.17175878859222002,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999525986160313,
      "loss": 7.1285,
      "mean_token_accuracy": 0.10076401606202126,
      "num_tokens": 3486841.0,
      "step": 1605
    },
    {
      "entropy": 7.064342927932739,
      "epoch": 0.17229386269998395,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999518106047293,
      "loss": 7.14,
      "mean_token_accuracy": 0.09933508709073066,
      "num_tokens": 3498735.0,
      "step": 1610
    },
    {
      "entropy": 7.0867256164550785,
      "epoch": 0.17282893680774788,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999510160979754,
      "loss": 7.1682,
      "mean_token_accuracy": 0.0990536592900753,
      "num_tokens": 3508829.0,
      "step": 1615
    },
    {
      "entropy": 7.060147380828857,
      "epoch": 0.1733640109155118,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999502150957925,
      "loss": 7.0744,
      "mean_token_accuracy": 0.09967830777168274,
      "num_tokens": 3519465.0,
      "step": 1620
    },
    {
      "entropy": 7.124733781814575,
      "epoch": 0.17389908502327572,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999494075982037,
      "loss": 7.05,
      "mean_token_accuracy": 0.1039630189538002,
      "num_tokens": 3530401.0,
      "step": 1625
    },
    {
      "entropy": 7.01870551109314,
      "epoch": 0.17443415913103966,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999485936052324,
      "loss": 7.1136,
      "mean_token_accuracy": 0.10137991532683373,
      "num_tokens": 3541136.0,
      "step": 1630
    },
    {
      "entropy": 7.156218004226685,
      "epoch": 0.17496923323880356,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499947773116902,
      "loss": 7.0954,
      "mean_token_accuracy": 0.09797013625502586,
      "num_tokens": 3552525.0,
      "step": 1635
    },
    {
      "entropy": 7.006080770492554,
      "epoch": 0.1755043073465675,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999469461332365,
      "loss": 7.0628,
      "mean_token_accuracy": 0.1024592712521553,
      "num_tokens": 3563028.0,
      "step": 1640
    },
    {
      "entropy": 7.124646234512329,
      "epoch": 0.17603938145433143,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999461126542592,
      "loss": 7.0201,
      "mean_token_accuracy": 0.1019650749862194,
      "num_tokens": 3573354.0,
      "step": 1645
    },
    {
      "entropy": 6.836950874328613,
      "epoch": 0.17657445556209536,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999452726799947,
      "loss": 7.01,
      "mean_token_accuracy": 0.09837546199560165,
      "num_tokens": 3585576.0,
      "step": 1650
    },
    {
      "entropy": 7.190705299377441,
      "epoch": 0.17710952966985927,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999444262104671,
      "loss": 7.1133,
      "mean_token_accuracy": 0.09539903774857521,
      "num_tokens": 3596478.0,
      "step": 1655
    },
    {
      "entropy": 7.0374476432800295,
      "epoch": 0.1776446037776232,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999435732457007,
      "loss": 7.0623,
      "mean_token_accuracy": 0.09882038906216621,
      "num_tokens": 3608014.0,
      "step": 1660
    },
    {
      "entropy": 7.125629949569702,
      "epoch": 0.17817967788538713,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999427137857203,
      "loss": 7.1265,
      "mean_token_accuracy": 0.09716841951012611,
      "num_tokens": 3620120.0,
      "step": 1665
    },
    {
      "entropy": 7.016817283630371,
      "epoch": 0.17871475199315104,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999418478305506,
      "loss": 7.0394,
      "mean_token_accuracy": 0.1033493585884571,
      "num_tokens": 3632578.0,
      "step": 1670
    },
    {
      "entropy": 7.062435150146484,
      "epoch": 0.17924982610091497,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999409753802167,
      "loss": 7.0509,
      "mean_token_accuracy": 0.10514670312404632,
      "num_tokens": 3643245.0,
      "step": 1675
    },
    {
      "entropy": 7.039481115341187,
      "epoch": 0.1797849002086789,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999400964347437,
      "loss": 7.055,
      "mean_token_accuracy": 0.10101916566491127,
      "num_tokens": 3655832.0,
      "step": 1680
    },
    {
      "entropy": 7.071658515930176,
      "epoch": 0.18031997431644284,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999392109941571,
      "loss": 7.0264,
      "mean_token_accuracy": 0.10084607303142548,
      "num_tokens": 3666566.0,
      "step": 1685
    },
    {
      "entropy": 7.078416633605957,
      "epoch": 0.18085504842420674,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999383190584822,
      "loss": 7.077,
      "mean_token_accuracy": 0.09968247041106224,
      "num_tokens": 3677572.0,
      "step": 1690
    },
    {
      "entropy": 6.993009233474732,
      "epoch": 0.18139012253197068,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999374206277451,
      "loss": 6.9841,
      "mean_token_accuracy": 0.10505180954933166,
      "num_tokens": 3687116.0,
      "step": 1695
    },
    {
      "entropy": 7.02902774810791,
      "epoch": 0.1819251966397346,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999365157019717,
      "loss": 7.1006,
      "mean_token_accuracy": 0.10133999511599541,
      "num_tokens": 3697961.0,
      "step": 1700
    },
    {
      "entropy": 7.110463190078735,
      "epoch": 0.18246027074749852,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999356042811878,
      "loss": 7.0024,
      "mean_token_accuracy": 0.1060384027659893,
      "num_tokens": 3708834.0,
      "step": 1705
    },
    {
      "entropy": 6.942103624343872,
      "epoch": 0.18299534485526245,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00049993468636542,
      "loss": 7.0757,
      "mean_token_accuracy": 0.10011870935559272,
      "num_tokens": 3719835.0,
      "step": 1710
    },
    {
      "entropy": 7.059339475631714,
      "epoch": 0.18353041896302638,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999337619546948,
      "loss": 6.9647,
      "mean_token_accuracy": 0.10536348894238472,
      "num_tokens": 3730408.0,
      "step": 1715
    },
    {
      "entropy": 6.996431350708008,
      "epoch": 0.18406549307079031,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999328310490387,
      "loss": 7.0514,
      "mean_token_accuracy": 0.10348244979977608,
      "num_tokens": 3741257.0,
      "step": 1720
    },
    {
      "entropy": 7.094743442535401,
      "epoch": 0.18460056717855422,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999318936484789,
      "loss": 7.0466,
      "mean_token_accuracy": 0.10182052180171013,
      "num_tokens": 3752615.0,
      "step": 1725
    },
    {
      "entropy": 6.988143110275269,
      "epoch": 0.18513564128631815,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999309497530422,
      "loss": 7.0771,
      "mean_token_accuracy": 0.09553842768073081,
      "num_tokens": 3763656.0,
      "step": 1730
    },
    {
      "entropy": 6.9603125095367435,
      "epoch": 0.1856707153940821,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999299993627558,
      "loss": 7.0461,
      "mean_token_accuracy": 0.10526920706033707,
      "num_tokens": 3775257.0,
      "step": 1735
    },
    {
      "entropy": 6.999796199798584,
      "epoch": 0.186205789501846,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999290424776475,
      "loss": 7.0041,
      "mean_token_accuracy": 0.10799972862005233,
      "num_tokens": 3786094.0,
      "step": 1740
    },
    {
      "entropy": 7.1879129886627195,
      "epoch": 0.18674086360960993,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999280790977445,
      "loss": 7.0932,
      "mean_token_accuracy": 0.09330181926488876,
      "num_tokens": 3796672.0,
      "step": 1745
    },
    {
      "entropy": 7.052556037902832,
      "epoch": 0.18727593771737386,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499927109223075,
      "loss": 7.0792,
      "mean_token_accuracy": 0.10074590295553207,
      "num_tokens": 3807910.0,
      "step": 1750
    },
    {
      "entropy": 7.00290265083313,
      "epoch": 0.1878110118251378,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999261328536667,
      "loss": 7.0367,
      "mean_token_accuracy": 0.10434340015053749,
      "num_tokens": 3820792.0,
      "step": 1755
    },
    {
      "entropy": 6.971473217010498,
      "epoch": 0.1883460859329017,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999251499895479,
      "loss": 6.9408,
      "mean_token_accuracy": 0.10545228272676468,
      "num_tokens": 3831126.0,
      "step": 1760
    },
    {
      "entropy": 7.043059158325195,
      "epoch": 0.18888116004066563,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999241606307471,
      "loss": 7.0742,
      "mean_token_accuracy": 0.10196674540638924,
      "num_tokens": 3841923.0,
      "step": 1765
    },
    {
      "entropy": 7.108615827560425,
      "epoch": 0.18941623414842956,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999231647772927,
      "loss": 6.9747,
      "mean_token_accuracy": 0.10711026415228844,
      "num_tokens": 3852504.0,
      "step": 1770
    },
    {
      "entropy": 6.88094048500061,
      "epoch": 0.1899513082561935,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999221624292136,
      "loss": 6.9834,
      "mean_token_accuracy": 0.10533660650253296,
      "num_tokens": 3862761.0,
      "step": 1775
    },
    {
      "entropy": 7.012593936920166,
      "epoch": 0.1904863823639574,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999211535865385,
      "loss": 6.9989,
      "mean_token_accuracy": 0.10127102062106133,
      "num_tokens": 3873662.0,
      "step": 1780
    },
    {
      "entropy": 6.945406341552735,
      "epoch": 0.19102145647172133,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999201382492969,
      "loss": 6.9436,
      "mean_token_accuracy": 0.10270186215639114,
      "num_tokens": 3884835.0,
      "step": 1785
    },
    {
      "entropy": 7.015659761428833,
      "epoch": 0.19155653057948527,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999191164175178,
      "loss": 7.0831,
      "mean_token_accuracy": 0.10108552053570748,
      "num_tokens": 3895073.0,
      "step": 1790
    },
    {
      "entropy": 7.105494022369385,
      "epoch": 0.19209160468724917,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999180880912309,
      "loss": 7.0455,
      "mean_token_accuracy": 0.10267279669642448,
      "num_tokens": 3905440.0,
      "step": 1795
    },
    {
      "entropy": 6.966555213928222,
      "epoch": 0.1926266787950131,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999170532704657,
      "loss": 7.0569,
      "mean_token_accuracy": 0.09851616621017456,
      "num_tokens": 3917135.0,
      "step": 1800
    },
    {
      "entropy": 7.071110677719116,
      "epoch": 0.19316175290277704,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999160119552523,
      "loss": 7.0492,
      "mean_token_accuracy": 0.10489852800965309,
      "num_tokens": 3928247.0,
      "step": 1805
    },
    {
      "entropy": 6.99960880279541,
      "epoch": 0.19369682701054097,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999149641456206,
      "loss": 7.0329,
      "mean_token_accuracy": 0.10273249596357345,
      "num_tokens": 3938609.0,
      "step": 1810
    },
    {
      "entropy": 7.065947103500366,
      "epoch": 0.19423190111830488,
      "grad_norm": 1.0,
      "learning_rate": 0.000499913909841601,
      "loss": 7.0113,
      "mean_token_accuracy": 0.10417041927576065,
      "num_tokens": 3949221.0,
      "step": 1815
    },
    {
      "entropy": 6.938427543640136,
      "epoch": 0.1947669752260688,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999128490432238,
      "loss": 6.9842,
      "mean_token_accuracy": 0.09849274829030037,
      "num_tokens": 3959475.0,
      "step": 1820
    },
    {
      "entropy": 6.9746458530426025,
      "epoch": 0.19530204933383274,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999117817505196,
      "loss": 6.9831,
      "mean_token_accuracy": 0.11270087212324142,
      "num_tokens": 3969638.0,
      "step": 1825
    },
    {
      "entropy": 6.9958281993865965,
      "epoch": 0.19583712344159665,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999107079635194,
      "loss": 7.0404,
      "mean_token_accuracy": 0.10607995018362999,
      "num_tokens": 3980543.0,
      "step": 1830
    },
    {
      "entropy": 7.199732780456543,
      "epoch": 0.19637219754936058,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499909627682254,
      "loss": 7.0771,
      "mean_token_accuracy": 0.0923505112528801,
      "num_tokens": 3993518.0,
      "step": 1835
    },
    {
      "entropy": 6.989639043807983,
      "epoch": 0.19690727165712452,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999085409067549,
      "loss": 7.0803,
      "mean_token_accuracy": 0.09985442981123924,
      "num_tokens": 4002940.0,
      "step": 1840
    },
    {
      "entropy": 6.968360519409179,
      "epoch": 0.19744234576488845,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999074476370531,
      "loss": 6.9977,
      "mean_token_accuracy": 0.10686009526252746,
      "num_tokens": 4013982.0,
      "step": 1845
    },
    {
      "entropy": 7.023968601226807,
      "epoch": 0.19797741987265235,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999063478731805,
      "loss": 6.9808,
      "mean_token_accuracy": 0.11021318733692169,
      "num_tokens": 4025607.0,
      "step": 1850
    },
    {
      "entropy": 6.987985277175904,
      "epoch": 0.1985124939804163,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999052416151687,
      "loss": 7.0426,
      "mean_token_accuracy": 0.10169775560498237,
      "num_tokens": 4037467.0,
      "step": 1855
    },
    {
      "entropy": 7.049751329421997,
      "epoch": 0.19904756808818022,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999041288630496,
      "loss": 7.0427,
      "mean_token_accuracy": 0.10394901558756828,
      "num_tokens": 4049146.0,
      "step": 1860
    },
    {
      "entropy": 7.08274736404419,
      "epoch": 0.19958264219594413,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999030096168553,
      "loss": 7.0763,
      "mean_token_accuracy": 0.102626833319664,
      "num_tokens": 4060500.0,
      "step": 1865
    },
    {
      "entropy": 7.0754108905792235,
      "epoch": 0.20011771630370806,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999018838766182,
      "loss": 7.0366,
      "mean_token_accuracy": 0.10197209641337394,
      "num_tokens": 4070553.0,
      "step": 1870
    },
    {
      "entropy": 7.014403486251831,
      "epoch": 0.200652790411472,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499900751642371,
      "loss": 6.9599,
      "mean_token_accuracy": 0.11636236384510994,
      "num_tokens": 4079653.0,
      "step": 1875
    },
    {
      "entropy": 6.94902081489563,
      "epoch": 0.20118786451923593,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998996129141461,
      "loss": 7.051,
      "mean_token_accuracy": 0.10020570904016494,
      "num_tokens": 4089912.0,
      "step": 1880
    },
    {
      "entropy": 7.10851526260376,
      "epoch": 0.20172293862699983,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998984676919764,
      "loss": 7.0157,
      "mean_token_accuracy": 0.09695038646459579,
      "num_tokens": 4100577.0,
      "step": 1885
    },
    {
      "entropy": 6.972286224365234,
      "epoch": 0.20225801273476376,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998973159758952,
      "loss": 7.0356,
      "mean_token_accuracy": 0.10260565429925919,
      "num_tokens": 4111752.0,
      "step": 1890
    },
    {
      "entropy": 6.941117143630981,
      "epoch": 0.2027930868425277,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998961577659355,
      "loss": 6.9848,
      "mean_token_accuracy": 0.10373990684747696,
      "num_tokens": 4122616.0,
      "step": 1895
    },
    {
      "entropy": 7.100810861587524,
      "epoch": 0.2033281609502916,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998949930621309,
      "loss": 7.1003,
      "mean_token_accuracy": 0.10509847998619079,
      "num_tokens": 4134275.0,
      "step": 1900
    },
    {
      "entropy": 6.981368684768677,
      "epoch": 0.20386323505805554,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998938218645149,
      "loss": 6.9822,
      "mean_token_accuracy": 0.1013522982597351,
      "num_tokens": 4146412.0,
      "step": 1905
    },
    {
      "entropy": 6.951316595077515,
      "epoch": 0.20439830916581947,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998926441731215,
      "loss": 6.9838,
      "mean_token_accuracy": 0.10630400627851486,
      "num_tokens": 4157592.0,
      "step": 1910
    },
    {
      "entropy": 7.005149888992309,
      "epoch": 0.2049333832735834,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998914599879845,
      "loss": 6.9695,
      "mean_token_accuracy": 0.10409662276506423,
      "num_tokens": 4168637.0,
      "step": 1915
    },
    {
      "entropy": 6.942488479614258,
      "epoch": 0.2054684573813473,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998902693091382,
      "loss": 7.0184,
      "mean_token_accuracy": 0.10043711066246033,
      "num_tokens": 4179951.0,
      "step": 1920
    },
    {
      "entropy": 6.960629844665528,
      "epoch": 0.20600353148911124,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998890721366171,
      "loss": 6.901,
      "mean_token_accuracy": 0.10442391037940979,
      "num_tokens": 4189895.0,
      "step": 1925
    },
    {
      "entropy": 6.995759773254394,
      "epoch": 0.20653860559687517,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998878684704556,
      "loss": 6.9624,
      "mean_token_accuracy": 0.10788150206208229,
      "num_tokens": 4199796.0,
      "step": 1930
    },
    {
      "entropy": 6.939257526397705,
      "epoch": 0.20707367970463908,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998866583106884,
      "loss": 6.9973,
      "mean_token_accuracy": 0.10173572450876237,
      "num_tokens": 4210428.0,
      "step": 1935
    },
    {
      "entropy": 7.1006691455841064,
      "epoch": 0.207608753812403,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998854416573506,
      "loss": 7.0458,
      "mean_token_accuracy": 0.10577386990189552,
      "num_tokens": 4221515.0,
      "step": 1940
    },
    {
      "entropy": 6.972071933746338,
      "epoch": 0.20814382792016695,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998842185104772,
      "loss": 7.076,
      "mean_token_accuracy": 0.0994568184018135,
      "num_tokens": 4232580.0,
      "step": 1945
    },
    {
      "entropy": 6.98147554397583,
      "epoch": 0.20867890202793088,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998829888701036,
      "loss": 6.9875,
      "mean_token_accuracy": 0.11094516292214393,
      "num_tokens": 4242871.0,
      "step": 1950
    },
    {
      "entropy": 7.048088407516479,
      "epoch": 0.20921397613569478,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998817527362654,
      "loss": 7.0105,
      "mean_token_accuracy": 0.10091942846775055,
      "num_tokens": 4253419.0,
      "step": 1955
    },
    {
      "entropy": 6.935594701766968,
      "epoch": 0.20974905024345872,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998805101089982,
      "loss": 6.9958,
      "mean_token_accuracy": 0.10472249984741211,
      "num_tokens": 4264327.0,
      "step": 1960
    },
    {
      "entropy": 7.0764631748199465,
      "epoch": 0.21028412435122265,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998792609883378,
      "loss": 6.9836,
      "mean_token_accuracy": 0.1075375609099865,
      "num_tokens": 4274568.0,
      "step": 1965
    },
    {
      "entropy": 6.949454212188721,
      "epoch": 0.21081919845898656,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004998780053743203,
      "loss": 6.9532,
      "mean_token_accuracy": 0.10796868503093719,
      "num_tokens": 4284895.0,
      "step": 1970
    },
    {
      "entropy": 6.989071989059449,
      "epoch": 0.2113542725667505,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998767432669822,
      "loss": 7.0061,
      "mean_token_accuracy": 0.10241716429591179,
      "num_tokens": 4297258.0,
      "step": 1975
    },
    {
      "entropy": 6.966968584060669,
      "epoch": 0.21188934667451442,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998754746663595,
      "loss": 6.9211,
      "mean_token_accuracy": 0.10538342818617821,
      "num_tokens": 4308307.0,
      "step": 1980
    },
    {
      "entropy": 6.958605146408081,
      "epoch": 0.21242442078227836,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998741995724892,
      "loss": 6.9878,
      "mean_token_accuracy": 0.10546612218022347,
      "num_tokens": 4318801.0,
      "step": 1985
    },
    {
      "entropy": 6.953232145309448,
      "epoch": 0.21295949489004226,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998729179854079,
      "loss": 6.956,
      "mean_token_accuracy": 0.10459589958190918,
      "num_tokens": 4329531.0,
      "step": 1990
    },
    {
      "entropy": 6.934867906570434,
      "epoch": 0.2134945689978062,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998716299051527,
      "loss": 6.9678,
      "mean_token_accuracy": 0.102861238270998,
      "num_tokens": 4341203.0,
      "step": 1995
    },
    {
      "entropy": 6.94436354637146,
      "epoch": 0.21402964310557013,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998703353317609,
      "loss": 7.0201,
      "mean_token_accuracy": 0.09982240498065949,
      "num_tokens": 4352426.0,
      "step": 2000
    },
    {
      "entropy": 7.063230895996094,
      "epoch": 0.21456471721333403,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998690342652697,
      "loss": 6.9587,
      "mean_token_accuracy": 0.11106750443577766,
      "num_tokens": 4363413.0,
      "step": 2005
    },
    {
      "entropy": 6.919417858123779,
      "epoch": 0.21509979132109797,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998677267057166,
      "loss": 6.8834,
      "mean_token_accuracy": 0.10697573572397232,
      "num_tokens": 4373470.0,
      "step": 2010
    },
    {
      "entropy": 6.898576736450195,
      "epoch": 0.2156348654288619,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998664126531397,
      "loss": 6.956,
      "mean_token_accuracy": 0.10943344607949257,
      "num_tokens": 4384438.0,
      "step": 2015
    },
    {
      "entropy": 6.99608736038208,
      "epoch": 0.21616993953662583,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998650921075766,
      "loss": 6.9333,
      "mean_token_accuracy": 0.10496605783700944,
      "num_tokens": 4395100.0,
      "step": 2020
    },
    {
      "entropy": 6.930504083633423,
      "epoch": 0.21670501364438974,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998637650690656,
      "loss": 7.0482,
      "mean_token_accuracy": 0.10081744119524956,
      "num_tokens": 4407131.0,
      "step": 2025
    },
    {
      "entropy": 6.987749719619751,
      "epoch": 0.21724008775215367,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499862431537645,
      "loss": 6.8874,
      "mean_token_accuracy": 0.10455133765935898,
      "num_tokens": 4417701.0,
      "step": 2030
    },
    {
      "entropy": 6.9407751083374025,
      "epoch": 0.2177751618599176,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998610915133533,
      "loss": 6.9543,
      "mean_token_accuracy": 0.0995742492377758,
      "num_tokens": 4429309.0,
      "step": 2035
    },
    {
      "entropy": 6.948079586029053,
      "epoch": 0.21831023596768154,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998597449962292,
      "loss": 7.0144,
      "mean_token_accuracy": 0.10309403762221336,
      "num_tokens": 4440850.0,
      "step": 2040
    },
    {
      "entropy": 6.931986236572266,
      "epoch": 0.21884531007544544,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998583919863115,
      "loss": 6.9959,
      "mean_token_accuracy": 0.10206221267580987,
      "num_tokens": 4451995.0,
      "step": 2045
    },
    {
      "entropy": 6.9314284324646,
      "epoch": 0.21938038418320938,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004998570324836393,
      "loss": 6.8959,
      "mean_token_accuracy": 0.10956206992268562,
      "num_tokens": 4461916.0,
      "step": 2050
    },
    {
      "entropy": 6.981660795211792,
      "epoch": 0.2199154582909733,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499855666488252,
      "loss": 6.9973,
      "mean_token_accuracy": 0.10489473640918731,
      "num_tokens": 4471603.0,
      "step": 2055
    },
    {
      "entropy": 7.026265430450439,
      "epoch": 0.22045053239873721,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998542940001889,
      "loss": 7.0251,
      "mean_token_accuracy": 0.10392524302005768,
      "num_tokens": 4483166.0,
      "step": 2060
    },
    {
      "entropy": 6.931446123123169,
      "epoch": 0.22098560650650115,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998529150194895,
      "loss": 6.9344,
      "mean_token_accuracy": 0.1028307020664215,
      "num_tokens": 4493824.0,
      "step": 2065
    },
    {
      "entropy": 6.929507732391357,
      "epoch": 0.22152068061426508,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998515295461939,
      "loss": 6.8511,
      "mean_token_accuracy": 0.1250505678355694,
      "num_tokens": 4505674.0,
      "step": 2070
    },
    {
      "entropy": 6.889703559875488,
      "epoch": 0.22205575472202901,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499850137580342,
      "loss": 6.942,
      "mean_token_accuracy": 0.10855949223041535,
      "num_tokens": 4516516.0,
      "step": 2075
    },
    {
      "entropy": 6.92522497177124,
      "epoch": 0.22259082882979292,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998487391219739,
      "loss": 6.92,
      "mean_token_accuracy": 0.0990898996591568,
      "num_tokens": 4527811.0,
      "step": 2080
    },
    {
      "entropy": 6.947738695144653,
      "epoch": 0.22312590293755685,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998473341711301,
      "loss": 6.8778,
      "mean_token_accuracy": 0.10222716778516769,
      "num_tokens": 4538988.0,
      "step": 2085
    },
    {
      "entropy": 6.9012058734893795,
      "epoch": 0.22366097704532079,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998459227278512,
      "loss": 6.9047,
      "mean_token_accuracy": 0.10971350893378258,
      "num_tokens": 4550049.0,
      "step": 2090
    },
    {
      "entropy": 6.96103801727295,
      "epoch": 0.2241960511530847,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998445047921777,
      "loss": 7.0301,
      "mean_token_accuracy": 0.1055637203156948,
      "num_tokens": 4559784.0,
      "step": 2095
    },
    {
      "entropy": 6.954623889923096,
      "epoch": 0.22473112526084862,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998430803641509,
      "loss": 6.8853,
      "mean_token_accuracy": 0.10650338679552078,
      "num_tokens": 4569982.0,
      "step": 2100
    },
    {
      "entropy": 6.876595544815063,
      "epoch": 0.22526619936861256,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998416494438118,
      "loss": 6.8835,
      "mean_token_accuracy": 0.1076424852013588,
      "num_tokens": 4581117.0,
      "step": 2105
    },
    {
      "entropy": 6.9081587314605715,
      "epoch": 0.2258012734763765,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998402120312014,
      "loss": 6.9568,
      "mean_token_accuracy": 0.11032719910144806,
      "num_tokens": 4592012.0,
      "step": 2110
    },
    {
      "entropy": 6.976634693145752,
      "epoch": 0.2263363475841404,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998387681263616,
      "loss": 6.9954,
      "mean_token_accuracy": 0.10546118691563607,
      "num_tokens": 4602591.0,
      "step": 2115
    },
    {
      "entropy": 6.943795967102051,
      "epoch": 0.22687142169190433,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499837317729334,
      "loss": 6.9311,
      "mean_token_accuracy": 0.10475296303629875,
      "num_tokens": 4613418.0,
      "step": 2120
    },
    {
      "entropy": 6.856068897247314,
      "epoch": 0.22740649579966826,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998358608401603,
      "loss": 6.8447,
      "mean_token_accuracy": 0.11353197693824768,
      "num_tokens": 4625156.0,
      "step": 2125
    },
    {
      "entropy": 6.97535138130188,
      "epoch": 0.22794156990743217,
      "grad_norm": 1.0,
      "learning_rate": 0.000499834397458883,
      "loss": 6.8859,
      "mean_token_accuracy": 0.11059412583708764,
      "num_tokens": 4634523.0,
      "step": 2130
    },
    {
      "entropy": 6.910711145401001,
      "epoch": 0.2284766440151961,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998329275855438,
      "loss": 6.8784,
      "mean_token_accuracy": 0.11305185407400131,
      "num_tokens": 4646292.0,
      "step": 2135
    },
    {
      "entropy": 6.910568618774414,
      "epoch": 0.22901171812296003,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998314512201855,
      "loss": 6.87,
      "mean_token_accuracy": 0.09877351820468902,
      "num_tokens": 4656417.0,
      "step": 2140
    },
    {
      "entropy": 6.880194187164307,
      "epoch": 0.22954679223072397,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998299683628507,
      "loss": 6.8746,
      "mean_token_accuracy": 0.10891946405172348,
      "num_tokens": 4666933.0,
      "step": 2145
    },
    {
      "entropy": 6.9828877449035645,
      "epoch": 0.23008186633848787,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998284790135821,
      "loss": 7.0497,
      "mean_token_accuracy": 0.09953607842326165,
      "num_tokens": 4678482.0,
      "step": 2150
    },
    {
      "entropy": 6.981028890609741,
      "epoch": 0.2306169404462518,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998269831724228,
      "loss": 6.8943,
      "mean_token_accuracy": 0.1074309766292572,
      "num_tokens": 4688454.0,
      "step": 2155
    },
    {
      "entropy": 6.829613304138183,
      "epoch": 0.23115201455401574,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004998254808394159,
      "loss": 6.931,
      "mean_token_accuracy": 0.1070957824587822,
      "num_tokens": 4699545.0,
      "step": 2160
    },
    {
      "entropy": 7.053047943115234,
      "epoch": 0.23168708866177964,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998239720146048,
      "loss": 6.9387,
      "mean_token_accuracy": 0.10415649190545082,
      "num_tokens": 4711149.0,
      "step": 2165
    },
    {
      "entropy": 6.836380195617676,
      "epoch": 0.23222216276954358,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998224566980332,
      "loss": 6.9489,
      "mean_token_accuracy": 0.10446546524763108,
      "num_tokens": 4722651.0,
      "step": 2170
    },
    {
      "entropy": 6.930455112457276,
      "epoch": 0.2327572368773075,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998209348897447,
      "loss": 6.9384,
      "mean_token_accuracy": 0.09837790131568909,
      "num_tokens": 4733883.0,
      "step": 2175
    },
    {
      "entropy": 7.116912317276001,
      "epoch": 0.23329231098507144,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998194065897833,
      "loss": 6.9102,
      "mean_token_accuracy": 0.10532031208276749,
      "num_tokens": 4744517.0,
      "step": 2180
    },
    {
      "entropy": 6.833180570602417,
      "epoch": 0.23382738509283535,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499817871798193,
      "loss": 6.926,
      "mean_token_accuracy": 0.1177787572145462,
      "num_tokens": 4755481.0,
      "step": 2185
    },
    {
      "entropy": 6.943826532363891,
      "epoch": 0.23436245920059928,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998163305150185,
      "loss": 6.8539,
      "mean_token_accuracy": 0.1105301819741726,
      "num_tokens": 4765535.0,
      "step": 2190
    },
    {
      "entropy": 6.894761371612549,
      "epoch": 0.23489753330836322,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998147827403038,
      "loss": 6.9646,
      "mean_token_accuracy": 0.104603161662817,
      "num_tokens": 4777045.0,
      "step": 2195
    },
    {
      "entropy": 6.95879225730896,
      "epoch": 0.23543260741612712,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499813228474094,
      "loss": 6.8388,
      "mean_token_accuracy": 0.11467055454850197,
      "num_tokens": 4787958.0,
      "step": 2200
    },
    {
      "entropy": 6.825476264953613,
      "epoch": 0.23596768152389105,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998116677164338,
      "loss": 6.8831,
      "mean_token_accuracy": 0.1111849419772625,
      "num_tokens": 4798502.0,
      "step": 2205
    },
    {
      "entropy": 6.885318279266357,
      "epoch": 0.236502755631655,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998101004673682,
      "loss": 6.9021,
      "mean_token_accuracy": 0.11082184463739395,
      "num_tokens": 4809179.0,
      "step": 2210
    },
    {
      "entropy": 7.027739429473877,
      "epoch": 0.23703782973941892,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998085267269426,
      "loss": 6.9699,
      "mean_token_accuracy": 0.10047761201858521,
      "num_tokens": 4819803.0,
      "step": 2215
    },
    {
      "entropy": 6.962998962402343,
      "epoch": 0.23757290384718283,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998069464952025,
      "loss": 6.9336,
      "mean_token_accuracy": 0.10305302366614341,
      "num_tokens": 4831226.0,
      "step": 2220
    },
    {
      "entropy": 6.906156587600708,
      "epoch": 0.23810797795494676,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998053597721932,
      "loss": 6.9565,
      "mean_token_accuracy": 0.10782697647809983,
      "num_tokens": 4842168.0,
      "step": 2225
    },
    {
      "entropy": 6.957432651519776,
      "epoch": 0.2386430520627107,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998037665579608,
      "loss": 6.9202,
      "mean_token_accuracy": 0.11156222894787789,
      "num_tokens": 4852655.0,
      "step": 2230
    },
    {
      "entropy": 6.845734071731568,
      "epoch": 0.2391781261704746,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998021668525513,
      "loss": 6.8226,
      "mean_token_accuracy": 0.11052219197154045,
      "num_tokens": 4862164.0,
      "step": 2235
    },
    {
      "entropy": 6.954825830459595,
      "epoch": 0.23971320027823853,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998005606560107,
      "loss": 6.9551,
      "mean_token_accuracy": 0.10994420796632767,
      "num_tokens": 4874233.0,
      "step": 2240
    },
    {
      "entropy": 6.940128040313721,
      "epoch": 0.24024827438600246,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997989479683856,
      "loss": 6.9501,
      "mean_token_accuracy": 0.10953364297747611,
      "num_tokens": 4885948.0,
      "step": 2245
    },
    {
      "entropy": 6.895329713821411,
      "epoch": 0.2407833484937664,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997973287897224,
      "loss": 6.9204,
      "mean_token_accuracy": 0.10798448026180267,
      "num_tokens": 4897128.0,
      "step": 2250
    },
    {
      "entropy": 6.940074491500854,
      "epoch": 0.2413184226015303,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997957031200681,
      "loss": 6.8887,
      "mean_token_accuracy": 0.11226024478673935,
      "num_tokens": 4907699.0,
      "step": 2255
    },
    {
      "entropy": 6.78710708618164,
      "epoch": 0.24185349670929424,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997940709594692,
      "loss": 6.8369,
      "mean_token_accuracy": 0.10901893302798271,
      "num_tokens": 4918772.0,
      "step": 2260
    },
    {
      "entropy": 7.022173166275024,
      "epoch": 0.24238857081705817,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997924323079733,
      "loss": 6.8807,
      "mean_token_accuracy": 0.10904730930924415,
      "num_tokens": 4930046.0,
      "step": 2265
    },
    {
      "entropy": 6.846764373779297,
      "epoch": 0.24292364492482207,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997907871656273,
      "loss": 6.9125,
      "mean_token_accuracy": 0.11078384071588517,
      "num_tokens": 4940609.0,
      "step": 2270
    },
    {
      "entropy": 6.931916522979736,
      "epoch": 0.243458719032586,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997891355324791,
      "loss": 6.9172,
      "mean_token_accuracy": 0.10807178467512131,
      "num_tokens": 4951339.0,
      "step": 2275
    },
    {
      "entropy": 6.923230695724487,
      "epoch": 0.24399379314034994,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997874774085761,
      "loss": 6.8348,
      "mean_token_accuracy": 0.10518774688243866,
      "num_tokens": 4961083.0,
      "step": 2280
    },
    {
      "entropy": 6.841183996200561,
      "epoch": 0.24452886724811387,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997858127939662,
      "loss": 6.8342,
      "mean_token_accuracy": 0.11167632862925529,
      "num_tokens": 4971399.0,
      "step": 2285
    },
    {
      "entropy": 6.936826086044311,
      "epoch": 0.24506394135587778,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997841416886976,
      "loss": 6.9039,
      "mean_token_accuracy": 0.10855253934860229,
      "num_tokens": 4982429.0,
      "step": 2290
    },
    {
      "entropy": 6.965252494812011,
      "epoch": 0.2455990154636417,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997824640928186,
      "loss": 6.9709,
      "mean_token_accuracy": 0.11127365306019783,
      "num_tokens": 4993706.0,
      "step": 2295
    },
    {
      "entropy": 6.93140139579773,
      "epoch": 0.24613408957140565,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997807800063774,
      "loss": 6.856,
      "mean_token_accuracy": 0.11392047181725502,
      "num_tokens": 5005411.0,
      "step": 2300
    },
    {
      "entropy": 6.895580244064331,
      "epoch": 0.24666916367916958,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499779089429423,
      "loss": 6.9879,
      "mean_token_accuracy": 0.10498436391353608,
      "num_tokens": 5016011.0,
      "step": 2305
    },
    {
      "entropy": 6.937919092178345,
      "epoch": 0.24720423778693348,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997773923620037,
      "loss": 6.946,
      "mean_token_accuracy": 0.10605266690254211,
      "num_tokens": 5027841.0,
      "step": 2310
    },
    {
      "entropy": 7.0313104629516605,
      "epoch": 0.24773931189469742,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997756888041689,
      "loss": 6.9343,
      "mean_token_accuracy": 0.10023647099733353,
      "num_tokens": 5038640.0,
      "step": 2315
    },
    {
      "entropy": 6.847754049301147,
      "epoch": 0.24827438600246135,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997739787559677,
      "loss": 6.8665,
      "mean_token_accuracy": 0.11019023731350899,
      "num_tokens": 5050423.0,
      "step": 2320
    },
    {
      "entropy": 6.873458576202393,
      "epoch": 0.24880946011022526,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997722622174494,
      "loss": 6.9018,
      "mean_token_accuracy": 0.10377348437905312,
      "num_tokens": 5060654.0,
      "step": 2325
    },
    {
      "entropy": 6.909244632720947,
      "epoch": 0.2493445342179892,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997705391886636,
      "loss": 6.8115,
      "mean_token_accuracy": 0.11297802552580834,
      "num_tokens": 5070165.0,
      "step": 2330
    },
    {
      "entropy": 6.870820236206055,
      "epoch": 0.24987960832575312,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997688096696601,
      "loss": 6.8627,
      "mean_token_accuracy": 0.10569036602973939,
      "num_tokens": 5080937.0,
      "step": 2335
    },
    {
      "entropy": 6.823880338668824,
      "epoch": 0.25041468243351706,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997670736604888,
      "loss": 6.818,
      "mean_token_accuracy": 0.11626443266868591,
      "num_tokens": 5090955.0,
      "step": 2340
    },
    {
      "entropy": 6.957943868637085,
      "epoch": 0.250949756541281,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997653311611998,
      "loss": 6.9679,
      "mean_token_accuracy": 0.10086922571063042,
      "num_tokens": 5102201.0,
      "step": 2345
    },
    {
      "entropy": 6.938482141494751,
      "epoch": 0.25148483064904487,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997635821718435,
      "loss": 6.8881,
      "mean_token_accuracy": 0.11047171801328659,
      "num_tokens": 5112235.0,
      "step": 2350
    },
    {
      "entropy": 6.816356945037842,
      "epoch": 0.2520199047568088,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004997618266924704,
      "loss": 6.7549,
      "mean_token_accuracy": 0.1067312315106392,
      "num_tokens": 5122360.0,
      "step": 2355
    },
    {
      "entropy": 6.923428773880005,
      "epoch": 0.25255497886457273,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499760064723131,
      "loss": 6.9306,
      "mean_token_accuracy": 0.10218076333403588,
      "num_tokens": 5133084.0,
      "step": 2360
    },
    {
      "entropy": 6.894084119796753,
      "epoch": 0.25309005297233667,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997582962638764,
      "loss": 6.8677,
      "mean_token_accuracy": 0.10082524418830871,
      "num_tokens": 5144495.0,
      "step": 2365
    },
    {
      "entropy": 6.919527053833008,
      "epoch": 0.2536251270801006,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997565213147575,
      "loss": 6.8834,
      "mean_token_accuracy": 0.10920164585113526,
      "num_tokens": 5154824.0,
      "step": 2370
    },
    {
      "entropy": 6.910584592819214,
      "epoch": 0.25416020118786453,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997547398758257,
      "loss": 6.8554,
      "mean_token_accuracy": 0.10880272537469864,
      "num_tokens": 5165061.0,
      "step": 2375
    },
    {
      "entropy": 6.875441551208496,
      "epoch": 0.25469527529562846,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997529519471323,
      "loss": 6.8352,
      "mean_token_accuracy": 0.11507040783762931,
      "num_tokens": 5174964.0,
      "step": 2380
    },
    {
      "entropy": 6.84457483291626,
      "epoch": 0.25523034940339234,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997511575287291,
      "loss": 6.883,
      "mean_token_accuracy": 0.10771025717258453,
      "num_tokens": 5185458.0,
      "step": 2385
    },
    {
      "entropy": 6.926706743240357,
      "epoch": 0.2557654235111563,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997493566206677,
      "loss": 6.8622,
      "mean_token_accuracy": 0.10995001047849655,
      "num_tokens": 5196193.0,
      "step": 2390
    },
    {
      "entropy": 6.884949541091919,
      "epoch": 0.2563004976189202,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997475492230003,
      "loss": 6.9753,
      "mean_token_accuracy": 0.10006448924541474,
      "num_tokens": 5208541.0,
      "step": 2395
    },
    {
      "entropy": 6.940023374557495,
      "epoch": 0.25683557172668414,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000499745735335779,
      "loss": 6.8102,
      "mean_token_accuracy": 0.10769248157739639,
      "num_tokens": 5218567.0,
      "step": 2400
    },
    {
      "entropy": 6.903067255020142,
      "epoch": 0.2573706458344481,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997439149590561,
      "loss": 6.9019,
      "mean_token_accuracy": 0.11087262630462646,
      "num_tokens": 5228293.0,
      "step": 2405
    },
    {
      "entropy": 6.8589808464050295,
      "epoch": 0.257905719942212,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997420880928843,
      "loss": 6.8513,
      "mean_token_accuracy": 0.11243690699338912,
      "num_tokens": 5238880.0,
      "step": 2410
    },
    {
      "entropy": 6.952676296234131,
      "epoch": 0.25844079404997594,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997402547373162,
      "loss": 6.9499,
      "mean_token_accuracy": 0.10374770835041999,
      "num_tokens": 5250104.0,
      "step": 2415
    },
    {
      "entropy": 7.00459303855896,
      "epoch": 0.2589758681577399,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997384148924049,
      "loss": 6.9771,
      "mean_token_accuracy": 0.10417036786675453,
      "num_tokens": 5260355.0,
      "step": 2420
    },
    {
      "entropy": 6.921153402328491,
      "epoch": 0.25951094226550375,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997365685582035,
      "loss": 6.9007,
      "mean_token_accuracy": 0.10504551604390144,
      "num_tokens": 5272117.0,
      "step": 2425
    },
    {
      "entropy": 6.933792638778686,
      "epoch": 0.2600460163732677,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997347157347651,
      "loss": 6.9844,
      "mean_token_accuracy": 0.10128821805119514,
      "num_tokens": 5284583.0,
      "step": 2430
    },
    {
      "entropy": 6.901539659500122,
      "epoch": 0.2605810904810316,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997328564221435,
      "loss": 6.766,
      "mean_token_accuracy": 0.12240908816456794,
      "num_tokens": 5295846.0,
      "step": 2435
    },
    {
      "entropy": 6.857765197753906,
      "epoch": 0.26111616458879555,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997309906203922,
      "loss": 6.9367,
      "mean_token_accuracy": 0.10391849651932716,
      "num_tokens": 5307802.0,
      "step": 2440
    },
    {
      "entropy": 7.027815103530884,
      "epoch": 0.2616512386965595,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997291183295652,
      "loss": 6.8509,
      "mean_token_accuracy": 0.10834813043475151,
      "num_tokens": 5318263.0,
      "step": 2445
    },
    {
      "entropy": 6.920684576034546,
      "epoch": 0.2621863128043234,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997272395497164,
      "loss": 6.909,
      "mean_token_accuracy": 0.10439276918768883,
      "num_tokens": 5328555.0,
      "step": 2450
    },
    {
      "entropy": 6.829323577880859,
      "epoch": 0.26272138691208735,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997253542809002,
      "loss": 6.8587,
      "mean_token_accuracy": 0.10975515842437744,
      "num_tokens": 5339954.0,
      "step": 2455
    },
    {
      "entropy": 6.98386926651001,
      "epoch": 0.26325646101985123,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997234625231709,
      "loss": 6.9196,
      "mean_token_accuracy": 0.10977022200822831,
      "num_tokens": 5350919.0,
      "step": 2460
    },
    {
      "entropy": 6.865667390823364,
      "epoch": 0.26379153512761516,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997215642765832,
      "loss": 6.8833,
      "mean_token_accuracy": 0.11174416020512581,
      "num_tokens": 5361844.0,
      "step": 2465
    },
    {
      "entropy": 6.927380132675171,
      "epoch": 0.2643266092353791,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499719659541192,
      "loss": 6.9123,
      "mean_token_accuracy": 0.10714837461709976,
      "num_tokens": 5372026.0,
      "step": 2470
    },
    {
      "entropy": 6.941423130035401,
      "epoch": 0.26486168334314303,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997177483170521,
      "loss": 6.9514,
      "mean_token_accuracy": 0.10840447247028351,
      "num_tokens": 5382965.0,
      "step": 2475
    },
    {
      "entropy": 6.8894938945770265,
      "epoch": 0.26539675745090696,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997158306042188,
      "loss": 6.9196,
      "mean_token_accuracy": 0.10659607201814651,
      "num_tokens": 5393738.0,
      "step": 2480
    },
    {
      "entropy": 6.842194032669068,
      "epoch": 0.2659318315586709,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004997139064027476,
      "loss": 6.7755,
      "mean_token_accuracy": 0.1153898686170578,
      "num_tokens": 5404059.0,
      "step": 2485
    },
    {
      "entropy": 6.849250745773316,
      "epoch": 0.26646690566643483,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004997119757126938,
      "loss": 6.8197,
      "mean_token_accuracy": 0.11082051619887352,
      "num_tokens": 5414152.0,
      "step": 2490
    },
    {
      "entropy": 6.916332626342774,
      "epoch": 0.2670019797741987,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997100385341133,
      "loss": 6.9818,
      "mean_token_accuracy": 0.10668149292469024,
      "num_tokens": 5425227.0,
      "step": 2495
    },
    {
      "entropy": 6.95630989074707,
      "epoch": 0.26753705388196264,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997080948670621,
      "loss": 6.8222,
      "mean_token_accuracy": 0.10940428376197815,
      "num_tokens": 5435716.0,
      "step": 2500
    },
    {
      "entropy": 6.869187307357788,
      "epoch": 0.26807212798972657,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997061447115962,
      "loss": 6.8925,
      "mean_token_accuracy": 0.10535158514976502,
      "num_tokens": 5446519.0,
      "step": 2505
    },
    {
      "entropy": 6.927534055709839,
      "epoch": 0.2686072020974905,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997041880677719,
      "loss": 6.8813,
      "mean_token_accuracy": 0.10491925999522209,
      "num_tokens": 5457496.0,
      "step": 2510
    },
    {
      "entropy": 6.855731201171875,
      "epoch": 0.26914227620525444,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997022249356457,
      "loss": 6.7804,
      "mean_token_accuracy": 0.11086295917630196,
      "num_tokens": 5469059.0,
      "step": 2515
    },
    {
      "entropy": 6.904585552215576,
      "epoch": 0.26967735031301837,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004997002553152745,
      "loss": 6.873,
      "mean_token_accuracy": 0.10860639363527298,
      "num_tokens": 5480324.0,
      "step": 2520
    },
    {
      "entropy": 6.874744558334351,
      "epoch": 0.2702124244207823,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996982792067148,
      "loss": 6.8983,
      "mean_token_accuracy": 0.10033328384160996,
      "num_tokens": 5491513.0,
      "step": 2525
    },
    {
      "entropy": 6.809928035736084,
      "epoch": 0.2707474985285462,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499696296610024,
      "loss": 6.8005,
      "mean_token_accuracy": 0.10953768789768219,
      "num_tokens": 5502414.0,
      "step": 2530
    },
    {
      "entropy": 6.870344495773315,
      "epoch": 0.2712825726363101,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996943075252592,
      "loss": 6.9431,
      "mean_token_accuracy": 0.10491918995976449,
      "num_tokens": 5512962.0,
      "step": 2535
    },
    {
      "entropy": 6.90476393699646,
      "epoch": 0.27181764674407405,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996923119524777,
      "loss": 6.7908,
      "mean_token_accuracy": 0.11395590081810951,
      "num_tokens": 5524648.0,
      "step": 2540
    },
    {
      "entropy": 6.79902868270874,
      "epoch": 0.272352720851838,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996903098917375,
      "loss": 6.8741,
      "mean_token_accuracy": 0.10129198431968689,
      "num_tokens": 5536158.0,
      "step": 2545
    },
    {
      "entropy": 6.882397985458374,
      "epoch": 0.2728877949596019,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499688301343096,
      "loss": 6.787,
      "mean_token_accuracy": 0.11079020276665688,
      "num_tokens": 5547374.0,
      "step": 2550
    },
    {
      "entropy": 6.9354483127594,
      "epoch": 0.27342286906736585,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996862863066114,
      "loss": 6.8667,
      "mean_token_accuracy": 0.10993925333023072,
      "num_tokens": 5559212.0,
      "step": 2555
    },
    {
      "entropy": 6.883192300796509,
      "epoch": 0.2739579431751298,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996842647823419,
      "loss": 6.8512,
      "mean_token_accuracy": 0.1091854490339756,
      "num_tokens": 5569302.0,
      "step": 2560
    },
    {
      "entropy": 6.83465051651001,
      "epoch": 0.27449301728289366,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996822367703458,
      "loss": 6.837,
      "mean_token_accuracy": 0.10728960186243057,
      "num_tokens": 5581688.0,
      "step": 2565
    },
    {
      "entropy": 6.896716356277466,
      "epoch": 0.2750280913906576,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004996802022706817,
      "loss": 6.7679,
      "mean_token_accuracy": 0.11490985676646233,
      "num_tokens": 5591376.0,
      "step": 2570
    },
    {
      "entropy": 6.771258687973022,
      "epoch": 0.2755631654984215,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996781612834083,
      "loss": 6.8363,
      "mean_token_accuracy": 0.11146841421723366,
      "num_tokens": 5601510.0,
      "step": 2575
    },
    {
      "entropy": 6.9605508804321286,
      "epoch": 0.27609823960618546,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996761138085846,
      "loss": 6.9076,
      "mean_token_accuracy": 0.10790455341339111,
      "num_tokens": 5611436.0,
      "step": 2580
    },
    {
      "entropy": 6.874959659576416,
      "epoch": 0.2766333137139494,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004996740598462697,
      "loss": 6.8889,
      "mean_token_accuracy": 0.10777795240283013,
      "num_tokens": 5620470.0,
      "step": 2585
    },
    {
      "entropy": 6.786574125289917,
      "epoch": 0.2771683878217133,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004996719993965229,
      "loss": 6.7929,
      "mean_token_accuracy": 0.11289525181055068,
      "num_tokens": 5630740.0,
      "step": 2590
    },
    {
      "entropy": 6.820392370223999,
      "epoch": 0.27770346192947726,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996699324594036,
      "loss": 6.8056,
      "mean_token_accuracy": 0.11668010428547859,
      "num_tokens": 5641138.0,
      "step": 2595
    },
    {
      "entropy": 6.923859643936157,
      "epoch": 0.27823853603724114,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996678590349717,
      "loss": 6.925,
      "mean_token_accuracy": 0.1053420752286911,
      "num_tokens": 5652474.0,
      "step": 2600
    },
    {
      "entropy": 6.940411186218261,
      "epoch": 0.27877361014500507,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004996657791232868,
      "loss": 6.8298,
      "mean_token_accuracy": 0.11531029716134071,
      "num_tokens": 5663178.0,
      "step": 2605
    },
    {
      "entropy": 6.810340118408203,
      "epoch": 0.279308684252769,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996636927244092,
      "loss": 6.8511,
      "mean_token_accuracy": 0.10380723103880882,
      "num_tokens": 5673565.0,
      "step": 2610
    },
    {
      "entropy": 6.986609363555909,
      "epoch": 0.27984375836053293,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499661599838399,
      "loss": 6.8882,
      "mean_token_accuracy": 0.10501667782664299,
      "num_tokens": 5684651.0,
      "step": 2615
    },
    {
      "entropy": 6.8562994480133055,
      "epoch": 0.28037883246829687,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996595004653167,
      "loss": 6.7776,
      "mean_token_accuracy": 0.10820791944861412,
      "num_tokens": 5696081.0,
      "step": 2620
    },
    {
      "entropy": 6.8485575199127195,
      "epoch": 0.2809139065760608,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996573946052229,
      "loss": 6.8207,
      "mean_token_accuracy": 0.11107847765088082,
      "num_tokens": 5706830.0,
      "step": 2625
    },
    {
      "entropy": 6.877283096313477,
      "epoch": 0.28144898068382473,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996552822581783,
      "loss": 6.8511,
      "mean_token_accuracy": 0.11357574313879013,
      "num_tokens": 5717261.0,
      "step": 2630
    },
    {
      "entropy": 6.85238208770752,
      "epoch": 0.2819840547915886,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000499653163424244,
      "loss": 6.8112,
      "mean_token_accuracy": 0.10760234370827675,
      "num_tokens": 5727280.0,
      "step": 2635
    },
    {
      "entropy": 6.920324659347534,
      "epoch": 0.28251912889935255,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004996510381034814,
      "loss": 6.8794,
      "mean_token_accuracy": 0.10340792760252952,
      "num_tokens": 5737485.0,
      "step": 2640
    },
    {
      "entropy": 6.889347553253174,
      "epoch": 0.2830542030071165,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996489062959515,
      "loss": 6.8485,
      "mean_token_accuracy": 0.10494941174983978,
      "num_tokens": 5747683.0,
      "step": 2645
    },
    {
      "entropy": 6.7954267978668215,
      "epoch": 0.2835892771148804,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996467680017159,
      "loss": 6.8968,
      "mean_token_accuracy": 0.10509655699133873,
      "num_tokens": 5758148.0,
      "step": 2650
    },
    {
      "entropy": 6.890285778045654,
      "epoch": 0.28412435122264434,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996446232208365,
      "loss": 6.8374,
      "mean_token_accuracy": 0.11090912595391274,
      "num_tokens": 5769109.0,
      "step": 2655
    },
    {
      "entropy": 6.86578254699707,
      "epoch": 0.2846594253304083,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996424719533753,
      "loss": 6.8432,
      "mean_token_accuracy": 0.10779658704996109,
      "num_tokens": 5780344.0,
      "step": 2660
    },
    {
      "entropy": 6.840816926956177,
      "epoch": 0.2851944994381722,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004996403141993941,
      "loss": 6.6683,
      "mean_token_accuracy": 0.12086173370480538,
      "num_tokens": 5790167.0,
      "step": 2665
    },
    {
      "entropy": 6.881586742401123,
      "epoch": 0.2857295735459361,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996381499589555,
      "loss": 6.8347,
      "mean_token_accuracy": 0.10883153080940247,
      "num_tokens": 5800076.0,
      "step": 2670
    },
    {
      "entropy": 6.91086368560791,
      "epoch": 0.2862646476537,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996359792321219,
      "loss": 6.7659,
      "mean_token_accuracy": 0.11435504630208015,
      "num_tokens": 5810547.0,
      "step": 2675
    },
    {
      "entropy": 6.8048162937164305,
      "epoch": 0.28679972176146395,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499633802018956,
      "loss": 6.8252,
      "mean_token_accuracy": 0.11114172041416168,
      "num_tokens": 5821210.0,
      "step": 2680
    },
    {
      "entropy": 6.8217119693756105,
      "epoch": 0.2873347958692279,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996316183195204,
      "loss": 6.8542,
      "mean_token_accuracy": 0.10784100592136384,
      "num_tokens": 5832361.0,
      "step": 2685
    },
    {
      "entropy": 6.871018123626709,
      "epoch": 0.2878698699769918,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004996294281338786,
      "loss": 6.836,
      "mean_token_accuracy": 0.10880450755357743,
      "num_tokens": 5843242.0,
      "step": 2690
    },
    {
      "entropy": 6.958883285522461,
      "epoch": 0.28840494408475575,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004996272314620936,
      "loss": 6.8922,
      "mean_token_accuracy": 0.10425070151686669,
      "num_tokens": 5855163.0,
      "step": 2695
    },
    {
      "entropy": 6.799886226654053,
      "epoch": 0.2889400181925197,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004996250283042288,
      "loss": 6.8522,
      "mean_token_accuracy": 0.11065445616841316,
      "num_tokens": 5866908.0,
      "step": 2700
    },
    {
      "entropy": 6.870400524139404,
      "epoch": 0.28947509230028357,
      "grad_norm": 1.25,
      "learning_rate": 0.000499622818660348,
      "loss": 6.8662,
      "mean_token_accuracy": 0.10945025980472564,
      "num_tokens": 5877342.0,
      "step": 2705
    },
    {
      "entropy": 6.890243864059448,
      "epoch": 0.2900101664080475,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996206025305147,
      "loss": 6.8234,
      "mean_token_accuracy": 0.10922683775424957,
      "num_tokens": 5887960.0,
      "step": 2710
    },
    {
      "entropy": 6.841482019424438,
      "epoch": 0.29054524051581143,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004996183799147932,
      "loss": 6.8,
      "mean_token_accuracy": 0.10991375595331192,
      "num_tokens": 5899887.0,
      "step": 2715
    },
    {
      "entropy": 6.88977279663086,
      "epoch": 0.29108031462357536,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996161508132475,
      "loss": 6.8117,
      "mean_token_accuracy": 0.1017048753798008,
      "num_tokens": 5911022.0,
      "step": 2720
    },
    {
      "entropy": 6.925865507125854,
      "epoch": 0.2916153887313393,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499613915225942,
      "loss": 6.8908,
      "mean_token_accuracy": 0.11075199842453003,
      "num_tokens": 5922780.0,
      "step": 2725
    },
    {
      "entropy": 6.832954025268554,
      "epoch": 0.29215046283910323,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004996116731529412,
      "loss": 6.8937,
      "mean_token_accuracy": 0.11031706780195236,
      "num_tokens": 5934112.0,
      "step": 2730
    },
    {
      "entropy": 6.870949840545654,
      "epoch": 0.29268553694686716,
      "grad_norm": 1.390625,
      "learning_rate": 0.00049960942459431,
      "loss": 6.8451,
      "mean_token_accuracy": 0.10947609469294547,
      "num_tokens": 5944386.0,
      "step": 2735
    },
    {
      "entropy": 6.836471223831177,
      "epoch": 0.29322061105463104,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004996071695501132,
      "loss": 6.7915,
      "mean_token_accuracy": 0.1112366646528244,
      "num_tokens": 5955552.0,
      "step": 2740
    },
    {
      "entropy": 6.866098165512085,
      "epoch": 0.293755685162395,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996049080204159,
      "loss": 6.7982,
      "mean_token_accuracy": 0.1079338662326336,
      "num_tokens": 5966790.0,
      "step": 2745
    },
    {
      "entropy": 6.922070741653442,
      "epoch": 0.2942907592701589,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996026400052835,
      "loss": 6.7898,
      "mean_token_accuracy": 0.11280101761221886,
      "num_tokens": 5977704.0,
      "step": 2750
    },
    {
      "entropy": 6.826333141326904,
      "epoch": 0.29482583337792284,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996003655047814,
      "loss": 6.8131,
      "mean_token_accuracy": 0.11185725182294845,
      "num_tokens": 5989221.0,
      "step": 2755
    },
    {
      "entropy": 6.800469446182251,
      "epoch": 0.2953609074856868,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995980845189753,
      "loss": 6.8493,
      "mean_token_accuracy": 0.10706186145544053,
      "num_tokens": 6000906.0,
      "step": 2760
    },
    {
      "entropy": 6.925901269912719,
      "epoch": 0.2958959815934507,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995957970479312,
      "loss": 6.861,
      "mean_token_accuracy": 0.10586799681186676,
      "num_tokens": 6012039.0,
      "step": 2765
    },
    {
      "entropy": 6.811472463607788,
      "epoch": 0.29643105570121464,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995935030917148,
      "loss": 6.7508,
      "mean_token_accuracy": 0.1096365749835968,
      "num_tokens": 6024383.0,
      "step": 2770
    },
    {
      "entropy": 6.902992296218872,
      "epoch": 0.2969661298089785,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995912026503926,
      "loss": 6.8175,
      "mean_token_accuracy": 0.10661071166396141,
      "num_tokens": 6034904.0,
      "step": 2775
    },
    {
      "entropy": 6.844438695907593,
      "epoch": 0.29750120391674245,
      "grad_norm": 1.59375,
      "learning_rate": 0.000499588895724031,
      "loss": 6.8281,
      "mean_token_accuracy": 0.1071818120777607,
      "num_tokens": 6044449.0,
      "step": 2780
    },
    {
      "entropy": 6.934893274307251,
      "epoch": 0.2980362780245064,
      "grad_norm": 1.375,
      "learning_rate": 0.0004995865823126968,
      "loss": 6.8406,
      "mean_token_accuracy": 0.10777448788285256,
      "num_tokens": 6054465.0,
      "step": 2785
    },
    {
      "entropy": 6.792693328857422,
      "epoch": 0.2985713521322703,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004995842624164564,
      "loss": 6.8585,
      "mean_token_accuracy": 0.10810620859265327,
      "num_tokens": 6065539.0,
      "step": 2790
    },
    {
      "entropy": 6.892361211776733,
      "epoch": 0.29910642624003425,
      "grad_norm": 1.515625,
      "learning_rate": 0.000499581936035377,
      "loss": 6.8215,
      "mean_token_accuracy": 0.10853223651647567,
      "num_tokens": 6076135.0,
      "step": 2795
    },
    {
      "entropy": 6.8402428150177,
      "epoch": 0.2996415003477982,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995796031695257,
      "loss": 6.8311,
      "mean_token_accuracy": 0.11217288449406623,
      "num_tokens": 6086604.0,
      "step": 2800
    },
    {
      "entropy": 6.84694619178772,
      "epoch": 0.3001765744555621,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00049957726381897,
      "loss": 6.8437,
      "mean_token_accuracy": 0.10916305035352707,
      "num_tokens": 6095672.0,
      "step": 2805
    },
    {
      "entropy": 6.85554347038269,
      "epoch": 0.300711648563326,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995749179837773,
      "loss": 6.7047,
      "mean_token_accuracy": 0.11797482892870903,
      "num_tokens": 6105652.0,
      "step": 2810
    },
    {
      "entropy": 6.773930311203003,
      "epoch": 0.30124672267108993,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995725656640155,
      "loss": 6.7678,
      "mean_token_accuracy": 0.11047781035304069,
      "num_tokens": 6116930.0,
      "step": 2815
    },
    {
      "entropy": 6.79344072341919,
      "epoch": 0.30178179677885386,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004995702068597524,
      "loss": 6.7643,
      "mean_token_accuracy": 0.11538932919502258,
      "num_tokens": 6126659.0,
      "step": 2820
    },
    {
      "entropy": 6.878014659881591,
      "epoch": 0.3023168708866178,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995678415710561,
      "loss": 6.8063,
      "mean_token_accuracy": 0.10656528547406197,
      "num_tokens": 6137227.0,
      "step": 2825
    },
    {
      "entropy": 6.896840524673462,
      "epoch": 0.3028519449943817,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000499565469797995,
      "loss": 6.8635,
      "mean_token_accuracy": 0.1080358736217022,
      "num_tokens": 6147800.0,
      "step": 2830
    },
    {
      "entropy": 6.800867891311645,
      "epoch": 0.30338701910214566,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995630915406374,
      "loss": 6.7026,
      "mean_token_accuracy": 0.11739906817674636,
      "num_tokens": 6157671.0,
      "step": 2835
    },
    {
      "entropy": 6.8740137100219725,
      "epoch": 0.3039220932099096,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995607067990521,
      "loss": 6.843,
      "mean_token_accuracy": 0.10825344622135162,
      "num_tokens": 6168651.0,
      "step": 2840
    },
    {
      "entropy": 6.845859146118164,
      "epoch": 0.30445716731767347,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000499558315573308,
      "loss": 6.8321,
      "mean_token_accuracy": 0.10733111053705216,
      "num_tokens": 6179925.0,
      "step": 2845
    },
    {
      "entropy": 6.755564451217651,
      "epoch": 0.3049922414254374,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004995559178634741,
      "loss": 6.8108,
      "mean_token_accuracy": 0.11236448213458061,
      "num_tokens": 6190541.0,
      "step": 2850
    },
    {
      "entropy": 6.9959348201751705,
      "epoch": 0.30552731553320134,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995535136696196,
      "loss": 6.86,
      "mean_token_accuracy": 0.101121586561203,
      "num_tokens": 6201491.0,
      "step": 2855
    },
    {
      "entropy": 6.768679523468018,
      "epoch": 0.30606238964096527,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499551102991814,
      "loss": 6.669,
      "mean_token_accuracy": 0.1128725491464138,
      "num_tokens": 6212443.0,
      "step": 2860
    },
    {
      "entropy": 6.8172203540802006,
      "epoch": 0.3065974637487292,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995486858301268,
      "loss": 6.8775,
      "mean_token_accuracy": 0.10187524557113647,
      "num_tokens": 6222623.0,
      "step": 2865
    },
    {
      "entropy": 6.841892099380493,
      "epoch": 0.30713253785649314,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995462621846279,
      "loss": 6.7649,
      "mean_token_accuracy": 0.10936387330293655,
      "num_tokens": 6233712.0,
      "step": 2870
    },
    {
      "entropy": 6.847410821914673,
      "epoch": 0.30766761196425707,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004995438320553871,
      "loss": 6.8175,
      "mean_token_accuracy": 0.10677930638194084,
      "num_tokens": 6244254.0,
      "step": 2875
    },
    {
      "entropy": 6.818604850769043,
      "epoch": 0.30820268607202095,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995413954424748,
      "loss": 6.7178,
      "mean_token_accuracy": 0.11558739468455315,
      "num_tokens": 6254202.0,
      "step": 2880
    },
    {
      "entropy": 6.813026285171508,
      "epoch": 0.3087377601797849,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995389523459612,
      "loss": 6.8018,
      "mean_token_accuracy": 0.10679139941930771,
      "num_tokens": 6264496.0,
      "step": 2885
    },
    {
      "entropy": 6.809697580337525,
      "epoch": 0.3092728342875488,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499536502765917,
      "loss": 6.8077,
      "mean_token_accuracy": 0.11276716142892837,
      "num_tokens": 6275835.0,
      "step": 2890
    },
    {
      "entropy": 6.855095958709716,
      "epoch": 0.30980790839531275,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004995340467024128,
      "loss": 6.7478,
      "mean_token_accuracy": 0.11199733167886734,
      "num_tokens": 6286805.0,
      "step": 2895
    },
    {
      "entropy": 6.821068143844604,
      "epoch": 0.3103429825030767,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995315841555195,
      "loss": 6.758,
      "mean_token_accuracy": 0.10581445917487145,
      "num_tokens": 6298444.0,
      "step": 2900
    },
    {
      "entropy": 6.7351460456848145,
      "epoch": 0.3108780566108406,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004995291151253083,
      "loss": 6.7087,
      "mean_token_accuracy": 0.1133358970284462,
      "num_tokens": 6308259.0,
      "step": 2905
    },
    {
      "entropy": 6.783745813369751,
      "epoch": 0.31141313071860455,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995266396118505,
      "loss": 6.7765,
      "mean_token_accuracy": 0.10394766032695771,
      "num_tokens": 6320058.0,
      "step": 2910
    },
    {
      "entropy": 6.834055233001709,
      "epoch": 0.3119482048263684,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995241576152174,
      "loss": 6.746,
      "mean_token_accuracy": 0.10653191581368446,
      "num_tokens": 6331203.0,
      "step": 2915
    },
    {
      "entropy": 6.847873878479004,
      "epoch": 0.31248327893413236,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499521669135481,
      "loss": 6.865,
      "mean_token_accuracy": 0.10627863183617592,
      "num_tokens": 6342027.0,
      "step": 2920
    },
    {
      "entropy": 6.82856764793396,
      "epoch": 0.3130183530418963,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995191741727128,
      "loss": 6.7631,
      "mean_token_accuracy": 0.11083006113767624,
      "num_tokens": 6352580.0,
      "step": 2925
    },
    {
      "entropy": 6.842922592163086,
      "epoch": 0.3135534271496602,
      "grad_norm": 1.125,
      "learning_rate": 0.000499516672726985,
      "loss": 6.7884,
      "mean_token_accuracy": 0.11612819582223892,
      "num_tokens": 6363997.0,
      "step": 2930
    },
    {
      "entropy": 6.759811305999756,
      "epoch": 0.31408850125742416,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995141647983699,
      "loss": 6.8297,
      "mean_token_accuracy": 0.1118119165301323,
      "num_tokens": 6375284.0,
      "step": 2935
    },
    {
      "entropy": 6.876916217803955,
      "epoch": 0.3146235753651881,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995116503869398,
      "loss": 6.7239,
      "mean_token_accuracy": 0.11943885236978531,
      "num_tokens": 6386698.0,
      "step": 2940
    },
    {
      "entropy": 6.8405262470245365,
      "epoch": 0.315158649472952,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995091294927673,
      "loss": 6.8008,
      "mean_token_accuracy": 0.10860055312514305,
      "num_tokens": 6397237.0,
      "step": 2945
    },
    {
      "entropy": 6.879640722274781,
      "epoch": 0.3156937235807159,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995066021159253,
      "loss": 6.9112,
      "mean_token_accuracy": 0.10427958816289902,
      "num_tokens": 6408266.0,
      "step": 2950
    },
    {
      "entropy": 6.7258257389068605,
      "epoch": 0.31622879768847983,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995040682564867,
      "loss": 6.7992,
      "mean_token_accuracy": 0.1148921512067318,
      "num_tokens": 6419563.0,
      "step": 2955
    },
    {
      "entropy": 6.841591644287109,
      "epoch": 0.31676387179624377,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995015279145247,
      "loss": 6.7586,
      "mean_token_accuracy": 0.11175140738487244,
      "num_tokens": 6430220.0,
      "step": 2960
    },
    {
      "entropy": 6.83943738937378,
      "epoch": 0.3172989459040077,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994989810901126,
      "loss": 6.7692,
      "mean_token_accuracy": 0.11018451079726219,
      "num_tokens": 6440157.0,
      "step": 2965
    },
    {
      "entropy": 6.7851283073425295,
      "epoch": 0.31783402001177163,
      "grad_norm": 1.25,
      "learning_rate": 0.0004994964277833239,
      "loss": 6.8115,
      "mean_token_accuracy": 0.1146237537264824,
      "num_tokens": 6449511.0,
      "step": 2970
    },
    {
      "entropy": 6.883852767944336,
      "epoch": 0.31836909411953557,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994938679942324,
      "loss": 6.8141,
      "mean_token_accuracy": 0.10362748950719833,
      "num_tokens": 6460608.0,
      "step": 2975
    },
    {
      "entropy": 6.832978010177612,
      "epoch": 0.3189041682272995,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004994913017229121,
      "loss": 6.8057,
      "mean_token_accuracy": 0.11033113375306129,
      "num_tokens": 6472737.0,
      "step": 2980
    },
    {
      "entropy": 6.801235628128052,
      "epoch": 0.31943924233506343,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994887289694371,
      "loss": 6.7104,
      "mean_token_accuracy": 0.11842553317546844,
      "num_tokens": 6482993.0,
      "step": 2985
    },
    {
      "entropy": 6.89878888130188,
      "epoch": 0.3199743164428273,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994861497338814,
      "loss": 6.8957,
      "mean_token_accuracy": 0.10716022476553917,
      "num_tokens": 6494530.0,
      "step": 2990
    },
    {
      "entropy": 6.804651165008545,
      "epoch": 0.32050939055059124,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994835640163198,
      "loss": 6.7251,
      "mean_token_accuracy": 0.10995796993374825,
      "num_tokens": 6504691.0,
      "step": 2995
    },
    {
      "entropy": 6.851648616790771,
      "epoch": 0.3210444646583552,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994809718168267,
      "loss": 6.7702,
      "mean_token_accuracy": 0.11220109388232231,
      "num_tokens": 6514924.0,
      "step": 3000
    },
    {
      "epoch": 0.3210444646583552,
      "eval_entropy": 6.663155538502408,
      "eval_loss": 6.872208595275879,
      "eval_mean_token_accuracy": 0.11419496110206184,
      "eval_num_tokens": 6514924.0,
      "eval_runtime": 22.687,
      "eval_samples_per_second": 1308.193,
      "eval_steps_per_second": 163.53,
      "step": 3000
    },
    {
      "entropy": 6.745563745498657,
      "epoch": 0.3215795387661191,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994783731354771,
      "loss": 6.7117,
      "mean_token_accuracy": 0.11273391768336297,
      "num_tokens": 6526097.0,
      "step": 3005
    },
    {
      "entropy": 6.84506573677063,
      "epoch": 0.32211461287388304,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994757679723461,
      "loss": 6.798,
      "mean_token_accuracy": 0.10792320370674133,
      "num_tokens": 6536384.0,
      "step": 3010
    },
    {
      "entropy": 6.854998588562012,
      "epoch": 0.322649686981647,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994731563275088,
      "loss": 6.8462,
      "mean_token_accuracy": 0.10304803922772407,
      "num_tokens": 6547161.0,
      "step": 3015
    },
    {
      "entropy": 6.727174854278564,
      "epoch": 0.3231847610894109,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994705382010406,
      "loss": 6.7479,
      "mean_token_accuracy": 0.11188101395964622,
      "num_tokens": 6556751.0,
      "step": 3020
    },
    {
      "entropy": 6.803781652450562,
      "epoch": 0.3237198351971748,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994679135930174,
      "loss": 6.8008,
      "mean_token_accuracy": 0.11748359650373459,
      "num_tokens": 6567655.0,
      "step": 3025
    },
    {
      "entropy": 6.928503513336182,
      "epoch": 0.3242549093049387,
      "grad_norm": 1.375,
      "learning_rate": 0.0004994652825035144,
      "loss": 6.8405,
      "mean_token_accuracy": 0.10758018866181374,
      "num_tokens": 6579127.0,
      "step": 3030
    },
    {
      "entropy": 6.875131845474243,
      "epoch": 0.32478998341270265,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004994626449326081,
      "loss": 6.8049,
      "mean_token_accuracy": 0.11168428286910057,
      "num_tokens": 6590779.0,
      "step": 3035
    },
    {
      "entropy": 6.774374866485596,
      "epoch": 0.3253250575204666,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994600008803745,
      "loss": 6.7335,
      "mean_token_accuracy": 0.11798494532704354,
      "num_tokens": 6600849.0,
      "step": 3040
    },
    {
      "entropy": 6.798848724365234,
      "epoch": 0.3258601316282305,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004994573503468899,
      "loss": 6.7321,
      "mean_token_accuracy": 0.11672844365239143,
      "num_tokens": 6610633.0,
      "step": 3045
    },
    {
      "entropy": 6.771644926071167,
      "epoch": 0.32639520573599445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994546933322308,
      "loss": 6.7739,
      "mean_token_accuracy": 0.11045937165617943,
      "num_tokens": 6621327.0,
      "step": 3050
    },
    {
      "entropy": 6.874581432342529,
      "epoch": 0.3269302798437584,
      "grad_norm": 1.25,
      "learning_rate": 0.000499452029836474,
      "loss": 6.7295,
      "mean_token_accuracy": 0.11313018202781677,
      "num_tokens": 6632267.0,
      "step": 3055
    },
    {
      "entropy": 6.786725950241089,
      "epoch": 0.32746535395152226,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004994493598596963,
      "loss": 6.7551,
      "mean_token_accuracy": 0.10917075648903847,
      "num_tokens": 6642714.0,
      "step": 3060
    },
    {
      "entropy": 6.825065422058105,
      "epoch": 0.3280004280592862,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004994466834019749,
      "loss": 6.7567,
      "mean_token_accuracy": 0.11459867134690285,
      "num_tokens": 6652353.0,
      "step": 3065
    },
    {
      "entropy": 6.794483184814453,
      "epoch": 0.32853550216705013,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994440004633871,
      "loss": 6.7356,
      "mean_token_accuracy": 0.11090541109442711,
      "num_tokens": 6663133.0,
      "step": 3070
    },
    {
      "entropy": 6.734937572479248,
      "epoch": 0.32907057627481406,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004994413110440103,
      "loss": 6.7047,
      "mean_token_accuracy": 0.11492902711033821,
      "num_tokens": 6673984.0,
      "step": 3075
    },
    {
      "entropy": 6.844591569900513,
      "epoch": 0.329605650382578,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004994386151439222,
      "loss": 6.7803,
      "mean_token_accuracy": 0.10733171626925468,
      "num_tokens": 6686171.0,
      "step": 3080
    },
    {
      "entropy": 6.93147349357605,
      "epoch": 0.33014072449034193,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004994359127632006,
      "loss": 6.8192,
      "mean_token_accuracy": 0.1099952794611454,
      "num_tokens": 6696609.0,
      "step": 3085
    },
    {
      "entropy": 6.821014881134033,
      "epoch": 0.33067579859810586,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004994332039019235,
      "loss": 6.763,
      "mean_token_accuracy": 0.11340149566531181,
      "num_tokens": 6706188.0,
      "step": 3090
    },
    {
      "entropy": 6.705182504653931,
      "epoch": 0.33121087270586974,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994304885601694,
      "loss": 6.7418,
      "mean_token_accuracy": 0.10889102444052697,
      "num_tokens": 6716276.0,
      "step": 3095
    },
    {
      "entropy": 6.740193796157837,
      "epoch": 0.3317459468136337,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994277667380162,
      "loss": 6.7062,
      "mean_token_accuracy": 0.11886503547430038,
      "num_tokens": 6727176.0,
      "step": 3100
    },
    {
      "entropy": 6.819032144546509,
      "epoch": 0.3322810209213976,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004994250384355429,
      "loss": 6.7953,
      "mean_token_accuracy": 0.11346929892897606,
      "num_tokens": 6737768.0,
      "step": 3105
    },
    {
      "entropy": 6.8038139820098875,
      "epoch": 0.33281609502916154,
      "grad_norm": 1.375,
      "learning_rate": 0.0004994223036528282,
      "loss": 6.7227,
      "mean_token_accuracy": 0.1112707644701004,
      "num_tokens": 6747962.0,
      "step": 3110
    },
    {
      "entropy": 6.7923671245574955,
      "epoch": 0.3333511691369255,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000499419562389951,
      "loss": 6.7364,
      "mean_token_accuracy": 0.11729706302285195,
      "num_tokens": 6758400.0,
      "step": 3115
    },
    {
      "entropy": 6.8117434024810795,
      "epoch": 0.3338862432446894,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994168146469904,
      "loss": 6.7423,
      "mean_token_accuracy": 0.11491606682538986,
      "num_tokens": 6768855.0,
      "step": 3120
    },
    {
      "entropy": 6.806019639968872,
      "epoch": 0.33442131735245334,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994140604240259,
      "loss": 6.7074,
      "mean_token_accuracy": 0.10914952158927918,
      "num_tokens": 6778656.0,
      "step": 3125
    },
    {
      "entropy": 6.780620765686035,
      "epoch": 0.3349563914602172,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994112997211369,
      "loss": 6.7966,
      "mean_token_accuracy": 0.11377846822142601,
      "num_tokens": 6789960.0,
      "step": 3130
    },
    {
      "entropy": 6.753642463684082,
      "epoch": 0.33549146556798115,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004994085325384031,
      "loss": 6.7635,
      "mean_token_accuracy": 0.11045785918831826,
      "num_tokens": 6800188.0,
      "step": 3135
    },
    {
      "entropy": 6.871442890167236,
      "epoch": 0.3360265396757451,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004994057588759045,
      "loss": 6.722,
      "mean_token_accuracy": 0.11737764328718185,
      "num_tokens": 6811032.0,
      "step": 3140
    },
    {
      "entropy": 6.7277820110321045,
      "epoch": 0.336561613783509,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994029787337212,
      "loss": 6.736,
      "mean_token_accuracy": 0.11023216396570205,
      "num_tokens": 6823060.0,
      "step": 3145
    },
    {
      "entropy": 6.802412271499634,
      "epoch": 0.33709668789127295,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004994001921119333,
      "loss": 6.8219,
      "mean_token_accuracy": 0.1107473075389862,
      "num_tokens": 6833559.0,
      "step": 3150
    },
    {
      "entropy": 6.853548192977906,
      "epoch": 0.3376317619990369,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004993973990106215,
      "loss": 6.7542,
      "mean_token_accuracy": 0.11447353884577752,
      "num_tokens": 6843657.0,
      "step": 3155
    },
    {
      "entropy": 6.792013549804688,
      "epoch": 0.3381668361068008,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004993945994298662,
      "loss": 6.702,
      "mean_token_accuracy": 0.10676191747188568,
      "num_tokens": 6853615.0,
      "step": 3160
    },
    {
      "entropy": 6.795974826812744,
      "epoch": 0.3387019102145647,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993917933697484,
      "loss": 6.8142,
      "mean_token_accuracy": 0.1132157288491726,
      "num_tokens": 6865256.0,
      "step": 3165
    },
    {
      "entropy": 6.959893465042114,
      "epoch": 0.3392369843223286,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993889808303493,
      "loss": 6.8499,
      "mean_token_accuracy": 0.11078018173575402,
      "num_tokens": 6875358.0,
      "step": 3170
    },
    {
      "entropy": 6.717615413665771,
      "epoch": 0.33977205843009256,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993861618117498,
      "loss": 6.7472,
      "mean_token_accuracy": 0.11126919835805893,
      "num_tokens": 6886584.0,
      "step": 3175
    },
    {
      "entropy": 6.791712093353271,
      "epoch": 0.3403071325378565,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993833363140314,
      "loss": 6.6961,
      "mean_token_accuracy": 0.11922059878706932,
      "num_tokens": 6896958.0,
      "step": 3180
    },
    {
      "entropy": 6.775498533248902,
      "epoch": 0.3408422066456204,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004993805043372756,
      "loss": 6.7571,
      "mean_token_accuracy": 0.11386067494750023,
      "num_tokens": 6907944.0,
      "step": 3185
    },
    {
      "entropy": 6.860631656646729,
      "epoch": 0.34137728075338436,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993776658815644,
      "loss": 6.8337,
      "mean_token_accuracy": 0.10652636811137199,
      "num_tokens": 6919486.0,
      "step": 3190
    },
    {
      "entropy": 6.90429196357727,
      "epoch": 0.3419123548611483,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004993748209469796,
      "loss": 6.8248,
      "mean_token_accuracy": 0.11483817845582962,
      "num_tokens": 6930650.0,
      "step": 3195
    },
    {
      "entropy": 6.78099422454834,
      "epoch": 0.34244742896891217,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004993719695336033,
      "loss": 6.7568,
      "mean_token_accuracy": 0.10781802982091904,
      "num_tokens": 6941519.0,
      "step": 3200
    },
    {
      "entropy": 6.730162286758423,
      "epoch": 0.3429825030766761,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004993691116415179,
      "loss": 6.8018,
      "mean_token_accuracy": 0.11166835129261017,
      "num_tokens": 6952952.0,
      "step": 3205
    },
    {
      "entropy": 6.847195672988891,
      "epoch": 0.34351757718444004,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499366247270806,
      "loss": 6.7639,
      "mean_token_accuracy": 0.11426469087600707,
      "num_tokens": 6962867.0,
      "step": 3210
    },
    {
      "entropy": 6.813245677947998,
      "epoch": 0.34405265129220397,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993633764215502,
      "loss": 6.7098,
      "mean_token_accuracy": 0.11524691209197044,
      "num_tokens": 6973434.0,
      "step": 3215
    },
    {
      "entropy": 6.814610481262207,
      "epoch": 0.3445877253999679,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004993604990938336,
      "loss": 6.708,
      "mean_token_accuracy": 0.12139374017715454,
      "num_tokens": 6984161.0,
      "step": 3220
    },
    {
      "entropy": 6.788215398788452,
      "epoch": 0.34512279950773184,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993576152877389,
      "loss": 6.6851,
      "mean_token_accuracy": 0.11097052693367004,
      "num_tokens": 6994700.0,
      "step": 3225
    },
    {
      "entropy": 6.797618675231933,
      "epoch": 0.34565787361549577,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993547250033496,
      "loss": 6.8513,
      "mean_token_accuracy": 0.1027825728058815,
      "num_tokens": 7006497.0,
      "step": 3230
    },
    {
      "entropy": 6.857847785949707,
      "epoch": 0.34619294772325965,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993518282407493,
      "loss": 6.764,
      "mean_token_accuracy": 0.11577967777848244,
      "num_tokens": 7016415.0,
      "step": 3235
    },
    {
      "entropy": 6.754112482070923,
      "epoch": 0.3467280218310236,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004993489250000213,
      "loss": 6.7178,
      "mean_token_accuracy": 0.1137387789785862,
      "num_tokens": 7026610.0,
      "step": 3240
    },
    {
      "entropy": 6.782814884185791,
      "epoch": 0.3472630959387875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993460152812497,
      "loss": 6.7111,
      "mean_token_accuracy": 0.11138368397951126,
      "num_tokens": 7038236.0,
      "step": 3245
    },
    {
      "entropy": 6.791642427444458,
      "epoch": 0.34779817004655145,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993430990845184,
      "loss": 6.7648,
      "mean_token_accuracy": 0.11891621202230454,
      "num_tokens": 7049438.0,
      "step": 3250
    },
    {
      "entropy": 6.776367568969727,
      "epoch": 0.3483332441543154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993401764099118,
      "loss": 6.7353,
      "mean_token_accuracy": 0.10871460661292076,
      "num_tokens": 7061001.0,
      "step": 3255
    },
    {
      "entropy": 6.793423700332641,
      "epoch": 0.3488683182620793,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993372472575139,
      "loss": 6.7209,
      "mean_token_accuracy": 0.11650840044021607,
      "num_tokens": 7071328.0,
      "step": 3260
    },
    {
      "entropy": 6.850569820404052,
      "epoch": 0.34940339236984325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993343116274096,
      "loss": 6.9068,
      "mean_token_accuracy": 0.10371987372636796,
      "num_tokens": 7082597.0,
      "step": 3265
    },
    {
      "entropy": 6.807454252243042,
      "epoch": 0.3499384664776071,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993313695196836,
      "loss": 6.7332,
      "mean_token_accuracy": 0.11512698754668235,
      "num_tokens": 7093463.0,
      "step": 3270
    },
    {
      "entropy": 6.7837098121643065,
      "epoch": 0.35047354058537106,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993284209344208,
      "loss": 6.7997,
      "mean_token_accuracy": 0.1145312875509262,
      "num_tokens": 7104350.0,
      "step": 3275
    },
    {
      "entropy": 6.870277261734008,
      "epoch": 0.351008614693135,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004993254658717065,
      "loss": 6.7369,
      "mean_token_accuracy": 0.1199392855167389,
      "num_tokens": 7114851.0,
      "step": 3280
    },
    {
      "entropy": 6.812195873260498,
      "epoch": 0.3515436888008989,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993225043316257,
      "loss": 6.7939,
      "mean_token_accuracy": 0.10890180021524429,
      "num_tokens": 7126440.0,
      "step": 3285
    },
    {
      "entropy": 6.865800571441651,
      "epoch": 0.35207876290866286,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004993195363142643,
      "loss": 6.8197,
      "mean_token_accuracy": 0.1106312945485115,
      "num_tokens": 7137884.0,
      "step": 3290
    },
    {
      "entropy": 6.832308912277222,
      "epoch": 0.3526138370164268,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993165618197076,
      "loss": 6.7948,
      "mean_token_accuracy": 0.10431528687477112,
      "num_tokens": 7147930.0,
      "step": 3295
    },
    {
      "entropy": 6.803113508224487,
      "epoch": 0.3531489111241907,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993135808480418,
      "loss": 6.6503,
      "mean_token_accuracy": 0.1201249398291111,
      "num_tokens": 7159200.0,
      "step": 3300
    },
    {
      "entropy": 6.816599941253662,
      "epoch": 0.3536839852319546,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004993105933993529,
      "loss": 6.7626,
      "mean_token_accuracy": 0.11429173201322555,
      "num_tokens": 7169601.0,
      "step": 3305
    },
    {
      "entropy": 6.66412386894226,
      "epoch": 0.35421905933971853,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004993075994737271,
      "loss": 6.6744,
      "mean_token_accuracy": 0.11934652402997017,
      "num_tokens": 7180465.0,
      "step": 3310
    },
    {
      "entropy": 6.818277883529663,
      "epoch": 0.35475413344748247,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004993045990712508,
      "loss": 6.708,
      "mean_token_accuracy": 0.11447355002164841,
      "num_tokens": 7190603.0,
      "step": 3315
    },
    {
      "entropy": 6.783639717102051,
      "epoch": 0.3552892075552464,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004993015921920109,
      "loss": 6.6647,
      "mean_token_accuracy": 0.11898905709385872,
      "num_tokens": 7201017.0,
      "step": 3320
    },
    {
      "entropy": 6.625533580780029,
      "epoch": 0.35582428166301033,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499298578836094,
      "loss": 6.6605,
      "mean_token_accuracy": 0.11563685759902001,
      "num_tokens": 7210908.0,
      "step": 3325
    },
    {
      "entropy": 6.863593101501465,
      "epoch": 0.35635935577077427,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499295559003587,
      "loss": 6.7213,
      "mean_token_accuracy": 0.11133797988295555,
      "num_tokens": 7221306.0,
      "step": 3330
    },
    {
      "entropy": 6.822593641281128,
      "epoch": 0.3568944298785382,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992925326945774,
      "loss": 6.8278,
      "mean_token_accuracy": 0.10899438485503196,
      "num_tokens": 7233076.0,
      "step": 3335
    },
    {
      "entropy": 6.80994987487793,
      "epoch": 0.3574295039863021,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004992894999091524,
      "loss": 6.8078,
      "mean_token_accuracy": 0.11141066178679467,
      "num_tokens": 7244703.0,
      "step": 3340
    },
    {
      "entropy": 6.87962965965271,
      "epoch": 0.357964578094066,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004992864606473996,
      "loss": 6.7946,
      "mean_token_accuracy": 0.10520617961883545,
      "num_tokens": 7255438.0,
      "step": 3345
    },
    {
      "entropy": 6.798658418655395,
      "epoch": 0.35849965220182994,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004992834149094069,
      "loss": 6.7395,
      "mean_token_accuracy": 0.11447354853153228,
      "num_tokens": 7265720.0,
      "step": 3350
    },
    {
      "entropy": 6.858418321609497,
      "epoch": 0.3590347263095939,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004992803626952619,
      "loss": 6.793,
      "mean_token_accuracy": 0.11047556102275849,
      "num_tokens": 7277791.0,
      "step": 3355
    },
    {
      "entropy": 6.7222527980804445,
      "epoch": 0.3595698004173578,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000499277304005053,
      "loss": 6.6585,
      "mean_token_accuracy": 0.12052367404103279,
      "num_tokens": 7288521.0,
      "step": 3360
    },
    {
      "entropy": 6.798248815536499,
      "epoch": 0.36010487452512174,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004992742388388686,
      "loss": 6.8421,
      "mean_token_accuracy": 0.10391957610845566,
      "num_tokens": 7299298.0,
      "step": 3365
    },
    {
      "entropy": 6.781761646270752,
      "epoch": 0.3606399486328857,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004992711671967968,
      "loss": 6.6749,
      "mean_token_accuracy": 0.11205682978034019,
      "num_tokens": 7310271.0,
      "step": 3370
    },
    {
      "entropy": 6.819436931610108,
      "epoch": 0.36117502274064955,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992680890789268,
      "loss": 6.7204,
      "mean_token_accuracy": 0.10891408175230026,
      "num_tokens": 7321112.0,
      "step": 3375
    },
    {
      "entropy": 6.711510848999024,
      "epoch": 0.3617100968484135,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004992650044853471,
      "loss": 6.6925,
      "mean_token_accuracy": 0.11342240199446678,
      "num_tokens": 7331976.0,
      "step": 3380
    },
    {
      "entropy": 6.83953218460083,
      "epoch": 0.3622451709561774,
      "grad_norm": 1.265625,
      "learning_rate": 0.000499261913416147,
      "loss": 6.8077,
      "mean_token_accuracy": 0.11410307884216309,
      "num_tokens": 7342800.0,
      "step": 3385
    },
    {
      "entropy": 6.794235801696777,
      "epoch": 0.36278024506394135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992588158714157,
      "loss": 6.6276,
      "mean_token_accuracy": 0.11011371910572051,
      "num_tokens": 7354213.0,
      "step": 3390
    },
    {
      "entropy": 6.776007080078125,
      "epoch": 0.3633153191717053,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004992557118512424,
      "loss": 6.7167,
      "mean_token_accuracy": 0.11738533824682236,
      "num_tokens": 7363250.0,
      "step": 3395
    },
    {
      "entropy": 6.75283145904541,
      "epoch": 0.3638503932794692,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004992526013557171,
      "loss": 6.7369,
      "mean_token_accuracy": 0.11341856718063355,
      "num_tokens": 7374185.0,
      "step": 3400
    },
    {
      "entropy": 6.711459302902222,
      "epoch": 0.36438546738723315,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992494843849294,
      "loss": 6.7078,
      "mean_token_accuracy": 0.10516770109534264,
      "num_tokens": 7384579.0,
      "step": 3405
    },
    {
      "entropy": 6.900410556793213,
      "epoch": 0.36492054149499703,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004992463609389694,
      "loss": 6.7332,
      "mean_token_accuracy": 0.1097166284918785,
      "num_tokens": 7395277.0,
      "step": 3410
    },
    {
      "entropy": 6.7915020942687985,
      "epoch": 0.36545561560276096,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992432310179272,
      "loss": 6.7261,
      "mean_token_accuracy": 0.11375608071684837,
      "num_tokens": 7406688.0,
      "step": 3415
    },
    {
      "entropy": 6.689106082916259,
      "epoch": 0.3659906897105249,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004992400946218932,
      "loss": 6.7707,
      "mean_token_accuracy": 0.11034525856375695,
      "num_tokens": 7417378.0,
      "step": 3420
    },
    {
      "entropy": 6.916859579086304,
      "epoch": 0.36652576381828883,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004992369517509581,
      "loss": 6.8004,
      "mean_token_accuracy": 0.11413674354553223,
      "num_tokens": 7427675.0,
      "step": 3425
    },
    {
      "entropy": 6.825582122802734,
      "epoch": 0.36706083792605276,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004992338024052123,
      "loss": 6.8317,
      "mean_token_accuracy": 0.10687674582004547,
      "num_tokens": 7438083.0,
      "step": 3430
    },
    {
      "entropy": 6.752775859832764,
      "epoch": 0.3675959120338167,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992306465847473,
      "loss": 6.6768,
      "mean_token_accuracy": 0.11751338317990304,
      "num_tokens": 7449051.0,
      "step": 3435
    },
    {
      "entropy": 6.752591276168824,
      "epoch": 0.36813098614158063,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004992274842896536,
      "loss": 6.7025,
      "mean_token_accuracy": 0.11567284539341927,
      "num_tokens": 7459074.0,
      "step": 3440
    },
    {
      "entropy": 6.787220621109009,
      "epoch": 0.3686660602493445,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004992243155200229,
      "loss": 6.7755,
      "mean_token_accuracy": 0.11534915417432785,
      "num_tokens": 7470472.0,
      "step": 3445
    },
    {
      "entropy": 6.828776693344116,
      "epoch": 0.36920113435710844,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004992211402759466,
      "loss": 6.7396,
      "mean_token_accuracy": 0.1147568017244339,
      "num_tokens": 7481477.0,
      "step": 3450
    },
    {
      "entropy": 6.774149703979492,
      "epoch": 0.3697362084648724,
      "grad_norm": 1.25,
      "learning_rate": 0.0004992179585575164,
      "loss": 6.7033,
      "mean_token_accuracy": 0.1158942274749279,
      "num_tokens": 7491906.0,
      "step": 3455
    },
    {
      "entropy": 6.795584964752197,
      "epoch": 0.3702712825726363,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000499214770364824,
      "loss": 6.6307,
      "mean_token_accuracy": 0.10889941975474357,
      "num_tokens": 7502447.0,
      "step": 3460
    },
    {
      "entropy": 6.701265859603882,
      "epoch": 0.37080635668040024,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004992115756979617,
      "loss": 6.6362,
      "mean_token_accuracy": 0.12357409447431564,
      "num_tokens": 7513505.0,
      "step": 3465
    },
    {
      "entropy": 6.767413330078125,
      "epoch": 0.3713414307881642,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992083745570218,
      "loss": 6.6523,
      "mean_token_accuracy": 0.12134870514273643,
      "num_tokens": 7524884.0,
      "step": 3470
    },
    {
      "entropy": 6.713840341567993,
      "epoch": 0.3718765048959281,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004992051669420964,
      "loss": 6.7316,
      "mean_token_accuracy": 0.11824962720274926,
      "num_tokens": 7535942.0,
      "step": 3475
    },
    {
      "entropy": 6.757349061965942,
      "epoch": 0.372411579003692,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004992019528532785,
      "loss": 6.7058,
      "mean_token_accuracy": 0.11440639570355415,
      "num_tokens": 7545506.0,
      "step": 3480
    },
    {
      "entropy": 6.827877569198608,
      "epoch": 0.3729466531114559,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004991987322906605,
      "loss": 6.7843,
      "mean_token_accuracy": 0.10873871818184852,
      "num_tokens": 7555328.0,
      "step": 3485
    },
    {
      "entropy": 6.809679365158081,
      "epoch": 0.37348172721921985,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004991955052543357,
      "loss": 6.6898,
      "mean_token_accuracy": 0.11306628957390785,
      "num_tokens": 7565762.0,
      "step": 3490
    },
    {
      "entropy": 6.832235860824585,
      "epoch": 0.3740168013269838,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004991922717443972,
      "loss": 6.701,
      "mean_token_accuracy": 0.12100865170359612,
      "num_tokens": 7575884.0,
      "step": 3495
    },
    {
      "entropy": 6.80165286064148,
      "epoch": 0.3745518754347477,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004991890317609384,
      "loss": 6.7252,
      "mean_token_accuracy": 0.11258633807301521,
      "num_tokens": 7586100.0,
      "step": 3500
    },
    {
      "entropy": 6.787338972091675,
      "epoch": 0.37508694954251165,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004991857853040528,
      "loss": 6.7466,
      "mean_token_accuracy": 0.1115113340318203,
      "num_tokens": 7597137.0,
      "step": 3505
    },
    {
      "entropy": 6.7378333568572994,
      "epoch": 0.3756220236502756,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004991825323738341,
      "loss": 6.7824,
      "mean_token_accuracy": 0.11462030112743378,
      "num_tokens": 7607640.0,
      "step": 3510
    },
    {
      "entropy": 6.825198125839234,
      "epoch": 0.3761570977580395,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004991792729703763,
      "loss": 6.8399,
      "mean_token_accuracy": 0.11375791877508164,
      "num_tokens": 7618319.0,
      "step": 3515
    },
    {
      "entropy": 6.861372232437134,
      "epoch": 0.3766921718658034,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004991760070937734,
      "loss": 6.7157,
      "mean_token_accuracy": 0.11241295039653779,
      "num_tokens": 7628350.0,
      "step": 3520
    },
    {
      "entropy": 6.775536441802979,
      "epoch": 0.3772272459735673,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004991727347441199,
      "loss": 6.7374,
      "mean_token_accuracy": 0.10838376954197884,
      "num_tokens": 7637857.0,
      "step": 3525
    },
    {
      "entropy": 6.840644407272339,
      "epoch": 0.37776232008133126,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991694559215102,
      "loss": 6.7405,
      "mean_token_accuracy": 0.11169720366597176,
      "num_tokens": 7648242.0,
      "step": 3530
    },
    {
      "entropy": 6.813172006607056,
      "epoch": 0.3782973941890952,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499166170626039,
      "loss": 6.7742,
      "mean_token_accuracy": 0.11039445921778679,
      "num_tokens": 7660565.0,
      "step": 3535
    },
    {
      "entropy": 6.728032398223877,
      "epoch": 0.3788324682968591,
      "grad_norm": 1.8203125,
      "learning_rate": 0.000499162878857801,
      "loss": 6.7546,
      "mean_token_accuracy": 0.11566273272037506,
      "num_tokens": 7671600.0,
      "step": 3540
    },
    {
      "entropy": 6.7928694725036625,
      "epoch": 0.37936754240462306,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991595806168915,
      "loss": 6.7712,
      "mean_token_accuracy": 0.11043274328112603,
      "num_tokens": 7682453.0,
      "step": 3545
    },
    {
      "entropy": 6.808826303482055,
      "epoch": 0.379902616512387,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991562759034055,
      "loss": 6.7161,
      "mean_token_accuracy": 0.11529388427734374,
      "num_tokens": 7693490.0,
      "step": 3550
    },
    {
      "entropy": 6.77369909286499,
      "epoch": 0.38043769062015087,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004991529647174386,
      "loss": 6.6585,
      "mean_token_accuracy": 0.11225939691066741,
      "num_tokens": 7703478.0,
      "step": 3555
    },
    {
      "entropy": 6.753496932983398,
      "epoch": 0.3809727647279148,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004991496470590864,
      "loss": 6.7506,
      "mean_token_accuracy": 0.11137450858950615,
      "num_tokens": 7714647.0,
      "step": 3560
    },
    {
      "entropy": 6.797580623626709,
      "epoch": 0.38150783883567874,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004991463229284445,
      "loss": 6.6918,
      "mean_token_accuracy": 0.11793707832694053,
      "num_tokens": 7725528.0,
      "step": 3565
    },
    {
      "entropy": 6.728217697143554,
      "epoch": 0.38204291294344267,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004991429923256091,
      "loss": 6.7626,
      "mean_token_accuracy": 0.1149514652788639,
      "num_tokens": 7737708.0,
      "step": 3570
    },
    {
      "entropy": 6.858790016174316,
      "epoch": 0.3825779870512066,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991396552506762,
      "loss": 6.7393,
      "mean_token_accuracy": 0.11154627427458763,
      "num_tokens": 7749093.0,
      "step": 3575
    },
    {
      "entropy": 6.721363496780396,
      "epoch": 0.38311306115897054,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991363117037423,
      "loss": 6.6723,
      "mean_token_accuracy": 0.11780490428209305,
      "num_tokens": 7760024.0,
      "step": 3580
    },
    {
      "entropy": 6.753328323364258,
      "epoch": 0.38364813526673447,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004991329616849039,
      "loss": 6.726,
      "mean_token_accuracy": 0.11707820892333984,
      "num_tokens": 7770801.0,
      "step": 3585
    },
    {
      "entropy": 6.811793375015259,
      "epoch": 0.38418320937449835,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004991296051942578,
      "loss": 6.7764,
      "mean_token_accuracy": 0.11302950009703636,
      "num_tokens": 7782257.0,
      "step": 3590
    },
    {
      "entropy": 6.777999687194824,
      "epoch": 0.3847182834822623,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004991262422319008,
      "loss": 6.7062,
      "mean_token_accuracy": 0.11388157904148102,
      "num_tokens": 7793016.0,
      "step": 3595
    },
    {
      "entropy": 6.741271257400513,
      "epoch": 0.3852533575900262,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00049912287279793,
      "loss": 6.6244,
      "mean_token_accuracy": 0.12548631727695464,
      "num_tokens": 7802207.0,
      "step": 3600
    },
    {
      "entropy": 6.803160715103149,
      "epoch": 0.38578843169779015,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004991194968924429,
      "loss": 6.7667,
      "mean_token_accuracy": 0.11139355152845383,
      "num_tokens": 7813487.0,
      "step": 3605
    },
    {
      "entropy": 6.820051670074463,
      "epoch": 0.3863235058055541,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004991161145155367,
      "loss": 6.7436,
      "mean_token_accuracy": 0.10898175165057182,
      "num_tokens": 7824587.0,
      "step": 3610
    },
    {
      "entropy": 6.810724067687988,
      "epoch": 0.386858579913318,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991127256673091,
      "loss": 6.7258,
      "mean_token_accuracy": 0.11021277531981469,
      "num_tokens": 7836240.0,
      "step": 3615
    },
    {
      "entropy": 6.730466079711914,
      "epoch": 0.38739365402108195,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004991093303478582,
      "loss": 6.6835,
      "mean_token_accuracy": 0.1120126262307167,
      "num_tokens": 7847014.0,
      "step": 3620
    },
    {
      "entropy": 6.777944612503052,
      "epoch": 0.3879287281288458,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991059285572818,
      "loss": 6.7889,
      "mean_token_accuracy": 0.11452787667512894,
      "num_tokens": 7857899.0,
      "step": 3625
    },
    {
      "entropy": 6.834170627593994,
      "epoch": 0.38846380223660976,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991025202956781,
      "loss": 6.6747,
      "mean_token_accuracy": 0.11366800889372826,
      "num_tokens": 7868987.0,
      "step": 3630
    },
    {
      "entropy": 6.789688396453857,
      "epoch": 0.3889988763443737,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004990991055631457,
      "loss": 6.7962,
      "mean_token_accuracy": 0.10578811541199684,
      "num_tokens": 7879547.0,
      "step": 3635
    },
    {
      "entropy": 6.7377862453460695,
      "epoch": 0.3895339504521376,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004990956843597832,
      "loss": 6.6631,
      "mean_token_accuracy": 0.11802548319101333,
      "num_tokens": 7889824.0,
      "step": 3640
    },
    {
      "entropy": 6.69925103187561,
      "epoch": 0.39006902455990156,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004990922566856892,
      "loss": 6.6561,
      "mean_token_accuracy": 0.11325121521949769,
      "num_tokens": 7901156.0,
      "step": 3645
    },
    {
      "entropy": 6.809516572952271,
      "epoch": 0.3906040986676655,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004990888225409629,
      "loss": 6.6966,
      "mean_token_accuracy": 0.11288304328918457,
      "num_tokens": 7911695.0,
      "step": 3650
    },
    {
      "entropy": 6.678881454467773,
      "epoch": 0.3911391727754294,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004990853819257033,
      "loss": 6.6831,
      "mean_token_accuracy": 0.11762734055519104,
      "num_tokens": 7923559.0,
      "step": 3655
    },
    {
      "entropy": 6.829882717132568,
      "epoch": 0.3916742468831933,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004990819348400097,
      "loss": 6.7028,
      "mean_token_accuracy": 0.11330192908644676,
      "num_tokens": 7935792.0,
      "step": 3660
    },
    {
      "entropy": 6.827629613876343,
      "epoch": 0.39220932099095723,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004990784812839818,
      "loss": 6.7259,
      "mean_token_accuracy": 0.11621317267417908,
      "num_tokens": 7946987.0,
      "step": 3665
    },
    {
      "entropy": 6.687596178054809,
      "epoch": 0.39274439509872117,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004990750212577192,
      "loss": 6.6242,
      "mean_token_accuracy": 0.11732003539800644,
      "num_tokens": 7957160.0,
      "step": 3670
    },
    {
      "entropy": 6.75628023147583,
      "epoch": 0.3932794692064851,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004990715547613219,
      "loss": 6.7267,
      "mean_token_accuracy": 0.10887964740395546,
      "num_tokens": 7968833.0,
      "step": 3675
    },
    {
      "entropy": 6.86474928855896,
      "epoch": 0.39381454331424903,
      "grad_norm": 1.125,
      "learning_rate": 0.0004990680817948898,
      "loss": 6.7746,
      "mean_token_accuracy": 0.10795791894197464,
      "num_tokens": 7980661.0,
      "step": 3680
    },
    {
      "entropy": 6.82999620437622,
      "epoch": 0.39434961742201297,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004990646023585234,
      "loss": 6.6961,
      "mean_token_accuracy": 0.11407504081726075,
      "num_tokens": 7992677.0,
      "step": 3685
    },
    {
      "entropy": 6.724775314331055,
      "epoch": 0.3948846915297769,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004990611164523231,
      "loss": 6.6708,
      "mean_token_accuracy": 0.11982596963644028,
      "num_tokens": 8002796.0,
      "step": 3690
    },
    {
      "entropy": 6.676330137252807,
      "epoch": 0.3954197656375408,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004990576240763896,
      "loss": 6.7051,
      "mean_token_accuracy": 0.1177534967660904,
      "num_tokens": 8012779.0,
      "step": 3695
    },
    {
      "entropy": 6.781534337997437,
      "epoch": 0.3959548397453047,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990541252308237,
      "loss": 6.692,
      "mean_token_accuracy": 0.11586613357067108,
      "num_tokens": 8023442.0,
      "step": 3700
    },
    {
      "entropy": 6.77800612449646,
      "epoch": 0.39648991385306864,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004990506199157263,
      "loss": 6.7827,
      "mean_token_accuracy": 0.1080534890294075,
      "num_tokens": 8035163.0,
      "step": 3705
    },
    {
      "entropy": 6.80181851387024,
      "epoch": 0.3970249879608326,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004990471081311988,
      "loss": 6.7055,
      "mean_token_accuracy": 0.11133532226085663,
      "num_tokens": 8045330.0,
      "step": 3710
    },
    {
      "entropy": 6.811718845367432,
      "epoch": 0.3975600620685965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990435898773426,
      "loss": 6.7351,
      "mean_token_accuracy": 0.11524785310029984,
      "num_tokens": 8056154.0,
      "step": 3715
    },
    {
      "entropy": 6.784503698348999,
      "epoch": 0.39809513617636044,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004990400651542592,
      "loss": 6.6841,
      "mean_token_accuracy": 0.11801928132772446,
      "num_tokens": 8066101.0,
      "step": 3720
    },
    {
      "entropy": 6.792222738265991,
      "epoch": 0.3986302102841244,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004990365339620504,
      "loss": 6.7648,
      "mean_token_accuracy": 0.10933328941464424,
      "num_tokens": 8077249.0,
      "step": 3725
    },
    {
      "entropy": 6.7601371765136715,
      "epoch": 0.39916528439188825,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000499032996300818,
      "loss": 6.6969,
      "mean_token_accuracy": 0.11813873648643494,
      "num_tokens": 8088532.0,
      "step": 3730
    },
    {
      "entropy": 6.811507701873779,
      "epoch": 0.3997003584996522,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004990294521706645,
      "loss": 6.7442,
      "mean_token_accuracy": 0.1067180298268795,
      "num_tokens": 8099541.0,
      "step": 3735
    },
    {
      "entropy": 6.733289003372192,
      "epoch": 0.4002354326074161,
      "grad_norm": 1.4375,
      "learning_rate": 0.000499025901571692,
      "loss": 6.7865,
      "mean_token_accuracy": 0.11544861868023873,
      "num_tokens": 8110532.0,
      "step": 3740
    },
    {
      "entropy": 6.8296427726745605,
      "epoch": 0.40077050671518005,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990223445040031,
      "loss": 6.7143,
      "mean_token_accuracy": 0.12142985686659813,
      "num_tokens": 8120991.0,
      "step": 3745
    },
    {
      "entropy": 6.763740015029907,
      "epoch": 0.401305580822944,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990187809677004,
      "loss": 6.6187,
      "mean_token_accuracy": 0.11583676040172577,
      "num_tokens": 8130737.0,
      "step": 3750
    },
    {
      "entropy": 6.621914052963257,
      "epoch": 0.4018406549307079,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004990152109628869,
      "loss": 6.6428,
      "mean_token_accuracy": 0.11762692779302597,
      "num_tokens": 8142315.0,
      "step": 3755
    },
    {
      "entropy": 6.790521764755249,
      "epoch": 0.40237572903847185,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004990116344896657,
      "loss": 6.7367,
      "mean_token_accuracy": 0.10920806154608727,
      "num_tokens": 8153618.0,
      "step": 3760
    },
    {
      "entropy": 6.795371150970459,
      "epoch": 0.40291080314623573,
      "grad_norm": 1.015625,
      "learning_rate": 0.00049900805154814,
      "loss": 6.681,
      "mean_token_accuracy": 0.11820271760225295,
      "num_tokens": 8164304.0,
      "step": 3765
    },
    {
      "entropy": 6.709203910827637,
      "epoch": 0.40344587725399966,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990044621384134,
      "loss": 6.6729,
      "mean_token_accuracy": 0.11573547199368477,
      "num_tokens": 8175289.0,
      "step": 3770
    },
    {
      "entropy": 6.818386125564575,
      "epoch": 0.4039809513617636,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004990008662605893,
      "loss": 6.7335,
      "mean_token_accuracy": 0.1082306794822216,
      "num_tokens": 8186274.0,
      "step": 3775
    },
    {
      "entropy": 6.820314836502075,
      "epoch": 0.40451602546952753,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004989972639147717,
      "loss": 6.6704,
      "mean_token_accuracy": 0.11876931563019752,
      "num_tokens": 8196380.0,
      "step": 3780
    },
    {
      "entropy": 6.734559679031372,
      "epoch": 0.40505109957729146,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004989936551010646,
      "loss": 6.7339,
      "mean_token_accuracy": 0.11214953884482384,
      "num_tokens": 8207204.0,
      "step": 3785
    },
    {
      "entropy": 6.630817031860351,
      "epoch": 0.4055861736850554,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004989900398195722,
      "loss": 6.6288,
      "mean_token_accuracy": 0.12022271677851677,
      "num_tokens": 8217745.0,
      "step": 3790
    },
    {
      "entropy": 6.71739935874939,
      "epoch": 0.40612124779281933,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004989864180703988,
      "loss": 6.6365,
      "mean_token_accuracy": 0.12018959075212479,
      "num_tokens": 8228171.0,
      "step": 3795
    },
    {
      "entropy": 6.732503509521484,
      "epoch": 0.4066563219005832,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004989827898536491,
      "loss": 6.6778,
      "mean_token_accuracy": 0.11271608099341393,
      "num_tokens": 8239220.0,
      "step": 3800
    },
    {
      "entropy": 6.769748783111572,
      "epoch": 0.40719139600834714,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004989791551694279,
      "loss": 6.7441,
      "mean_token_accuracy": 0.1108921080827713,
      "num_tokens": 8250363.0,
      "step": 3805
    },
    {
      "entropy": 6.749256992340088,
      "epoch": 0.4077264701161111,
      "grad_norm": 1.25,
      "learning_rate": 0.0004989755140178398,
      "loss": 6.6972,
      "mean_token_accuracy": 0.11420515179634094,
      "num_tokens": 8261207.0,
      "step": 3810
    },
    {
      "entropy": 6.714679622650147,
      "epoch": 0.408261544223875,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004989718663989905,
      "loss": 6.5966,
      "mean_token_accuracy": 0.11439496502280236,
      "num_tokens": 8272343.0,
      "step": 3815
    },
    {
      "entropy": 6.751222801208496,
      "epoch": 0.40879661833163894,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989682123129849,
      "loss": 6.6511,
      "mean_token_accuracy": 0.11429359316825867,
      "num_tokens": 8282677.0,
      "step": 3820
    },
    {
      "entropy": 6.628594398498535,
      "epoch": 0.40933169243940287,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004989645517599286,
      "loss": 6.6651,
      "mean_token_accuracy": 0.1157628171145916,
      "num_tokens": 8292753.0,
      "step": 3825
    },
    {
      "entropy": 6.7410119533538815,
      "epoch": 0.4098667665471668,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989608847399275,
      "loss": 6.6493,
      "mean_token_accuracy": 0.11546954065561295,
      "num_tokens": 8302747.0,
      "step": 3830
    },
    {
      "entropy": 6.788758563995361,
      "epoch": 0.4104018406549307,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989572112530871,
      "loss": 6.7205,
      "mean_token_accuracy": 0.10800857022404671,
      "num_tokens": 8314766.0,
      "step": 3835
    },
    {
      "entropy": 6.810330104827881,
      "epoch": 0.4109369147626946,
      "grad_norm": 1.25,
      "learning_rate": 0.0004989535312995139,
      "loss": 6.7574,
      "mean_token_accuracy": 0.11394470632076263,
      "num_tokens": 8325791.0,
      "step": 3840
    },
    {
      "entropy": 6.70790843963623,
      "epoch": 0.41147198887045855,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004989498448793139,
      "loss": 6.669,
      "mean_token_accuracy": 0.11584601849317551,
      "num_tokens": 8335809.0,
      "step": 3845
    },
    {
      "entropy": 6.735694932937622,
      "epoch": 0.4120070629782225,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989461519925935,
      "loss": 6.6678,
      "mean_token_accuracy": 0.11307423189282417,
      "num_tokens": 8347505.0,
      "step": 3850
    },
    {
      "entropy": 6.809847497940064,
      "epoch": 0.4125421370859864,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004989424526394596,
      "loss": 6.6906,
      "mean_token_accuracy": 0.1093782402575016,
      "num_tokens": 8358363.0,
      "step": 3855
    },
    {
      "entropy": 6.754508018493652,
      "epoch": 0.41307721119375035,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004989387468200187,
      "loss": 6.6878,
      "mean_token_accuracy": 0.11631895303726196,
      "num_tokens": 8369076.0,
      "step": 3860
    },
    {
      "entropy": 6.798090124130249,
      "epoch": 0.4136122853015143,
      "grad_norm": 1.28125,
      "learning_rate": 0.000498935034534378,
      "loss": 6.7307,
      "mean_token_accuracy": 0.11500347554683685,
      "num_tokens": 8380071.0,
      "step": 3865
    },
    {
      "entropy": 6.767126131057739,
      "epoch": 0.41414735940927816,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004989313157826446,
      "loss": 6.6987,
      "mean_token_accuracy": 0.11583667397499084,
      "num_tokens": 8390574.0,
      "step": 3870
    },
    {
      "entropy": 6.7329933643341064,
      "epoch": 0.4146824335170421,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000498927590564926,
      "loss": 6.6959,
      "mean_token_accuracy": 0.11751513332128524,
      "num_tokens": 8401154.0,
      "step": 3875
    },
    {
      "entropy": 6.821478986740113,
      "epoch": 0.415217507624806,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004989238588813296,
      "loss": 6.7806,
      "mean_token_accuracy": 0.10575368329882621,
      "num_tokens": 8412113.0,
      "step": 3880
    },
    {
      "entropy": 6.873749494552612,
      "epoch": 0.41575258173256996,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004989201207319634,
      "loss": 6.7507,
      "mean_token_accuracy": 0.1134581632912159,
      "num_tokens": 8422734.0,
      "step": 3885
    },
    {
      "entropy": 6.756129789352417,
      "epoch": 0.4162876558403339,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989163761169351,
      "loss": 6.6987,
      "mean_token_accuracy": 0.11055214703083038,
      "num_tokens": 8433647.0,
      "step": 3890
    },
    {
      "entropy": 6.7640749454498295,
      "epoch": 0.4168227299480978,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004989126250363529,
      "loss": 6.7122,
      "mean_token_accuracy": 0.1113808773458004,
      "num_tokens": 8444726.0,
      "step": 3895
    },
    {
      "entropy": 6.79126296043396,
      "epoch": 0.41735780405586176,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004989088674903251,
      "loss": 6.7312,
      "mean_token_accuracy": 0.10955160930752754,
      "num_tokens": 8455351.0,
      "step": 3900
    },
    {
      "entropy": 6.740010356903076,
      "epoch": 0.41789287816362564,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004989051034789602,
      "loss": 6.7066,
      "mean_token_accuracy": 0.12177284359931946,
      "num_tokens": 8466423.0,
      "step": 3905
    },
    {
      "entropy": 6.71815242767334,
      "epoch": 0.41842795227138957,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000498901333002367,
      "loss": 6.6361,
      "mean_token_accuracy": 0.11695924326777458,
      "num_tokens": 8475841.0,
      "step": 3910
    },
    {
      "entropy": 6.785331583023071,
      "epoch": 0.4189630263791535,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004988975560606541,
      "loss": 6.7534,
      "mean_token_accuracy": 0.11003556698560715,
      "num_tokens": 8487656.0,
      "step": 3915
    },
    {
      "entropy": 6.797838401794434,
      "epoch": 0.41949810048691744,
      "grad_norm": 1.25,
      "learning_rate": 0.0004988937726539308,
      "loss": 6.7546,
      "mean_token_accuracy": 0.10969007089734077,
      "num_tokens": 8497959.0,
      "step": 3920
    },
    {
      "entropy": 6.729758834838867,
      "epoch": 0.42003317459468137,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004988899827823062,
      "loss": 6.6696,
      "mean_token_accuracy": 0.11463452354073525,
      "num_tokens": 8508228.0,
      "step": 3925
    },
    {
      "entropy": 6.794983673095703,
      "epoch": 0.4205682487024453,
      "grad_norm": 1.203125,
      "learning_rate": 0.00049888618644589,
      "loss": 6.717,
      "mean_token_accuracy": 0.1153414785861969,
      "num_tokens": 8518999.0,
      "step": 3930
    },
    {
      "entropy": 6.710626888275146,
      "epoch": 0.42110332281020924,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004988823836447916,
      "loss": 6.7565,
      "mean_token_accuracy": 0.11152829825878144,
      "num_tokens": 8529583.0,
      "step": 3935
    },
    {
      "entropy": 6.845953941345215,
      "epoch": 0.4216383969179731,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004988785743791207,
      "loss": 6.8116,
      "mean_token_accuracy": 0.11097450107336045,
      "num_tokens": 8540069.0,
      "step": 3940
    },
    {
      "entropy": 6.8115967273712155,
      "epoch": 0.42217347102573705,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004988747586489876,
      "loss": 6.6778,
      "mean_token_accuracy": 0.11443818733096123,
      "num_tokens": 8550441.0,
      "step": 3945
    },
    {
      "entropy": 6.702272653579712,
      "epoch": 0.422708545133501,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000498870936454502,
      "loss": 6.6471,
      "mean_token_accuracy": 0.11951410844922065,
      "num_tokens": 8560257.0,
      "step": 3950
    },
    {
      "entropy": 6.736100673675537,
      "epoch": 0.4232436192412649,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988671077957749,
      "loss": 6.7287,
      "mean_token_accuracy": 0.11796722114086151,
      "num_tokens": 8570991.0,
      "step": 3955
    },
    {
      "entropy": 6.852146434783935,
      "epoch": 0.42377869334902885,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004988632726729164,
      "loss": 6.7792,
      "mean_token_accuracy": 0.10974573194980622,
      "num_tokens": 8583668.0,
      "step": 3960
    },
    {
      "entropy": 6.802029323577881,
      "epoch": 0.4243137674567928,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004988594310860374,
      "loss": 6.7134,
      "mean_token_accuracy": 0.11194633021950721,
      "num_tokens": 8593727.0,
      "step": 3965
    },
    {
      "entropy": 6.697398519515991,
      "epoch": 0.4248488415645567,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004988555830352485,
      "loss": 6.7122,
      "mean_token_accuracy": 0.11545829698443413,
      "num_tokens": 8605470.0,
      "step": 3970
    },
    {
      "entropy": 6.822403383255005,
      "epoch": 0.4253839156723206,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004988517285206614,
      "loss": 6.6439,
      "mean_token_accuracy": 0.11119706183671951,
      "num_tokens": 8617176.0,
      "step": 3975
    },
    {
      "entropy": 6.7473792552948,
      "epoch": 0.4259189897800845,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004988478675423871,
      "loss": 6.7288,
      "mean_token_accuracy": 0.11152340099215508,
      "num_tokens": 8627756.0,
      "step": 3980
    },
    {
      "entropy": 6.716528797149659,
      "epoch": 0.42645406388784846,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004988440001005369,
      "loss": 6.6705,
      "mean_token_accuracy": 0.11469498351216316,
      "num_tokens": 8637766.0,
      "step": 3985
    },
    {
      "entropy": 6.825556993484497,
      "epoch": 0.4269891379956124,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004988401261952227,
      "loss": 6.706,
      "mean_token_accuracy": 0.11419713199138641,
      "num_tokens": 8648562.0,
      "step": 3990
    },
    {
      "entropy": 6.837621450424194,
      "epoch": 0.4275242121033763,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004988362458265565,
      "loss": 6.7268,
      "mean_token_accuracy": 0.1081519827246666,
      "num_tokens": 8658393.0,
      "step": 3995
    },
    {
      "entropy": 6.648343229293824,
      "epoch": 0.42805928621114026,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004988323589946499,
      "loss": 6.6506,
      "mean_token_accuracy": 0.115006672590971,
      "num_tokens": 8668365.0,
      "step": 4000
    },
    {
      "entropy": 6.728427124023438,
      "epoch": 0.4285943603189042,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004988284656996156,
      "loss": 6.6856,
      "mean_token_accuracy": 0.11702194884419441,
      "num_tokens": 8679228.0,
      "step": 4005
    },
    {
      "entropy": 6.747385787963867,
      "epoch": 0.42912943442666807,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004988245659415656,
      "loss": 6.5737,
      "mean_token_accuracy": 0.12146354094147682,
      "num_tokens": 8688940.0,
      "step": 4010
    },
    {
      "entropy": 6.764614677429199,
      "epoch": 0.429664508534432,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004988206597206129,
      "loss": 6.7362,
      "mean_token_accuracy": 0.11172115281224251,
      "num_tokens": 8699923.0,
      "step": 4015
    },
    {
      "entropy": 6.70063853263855,
      "epoch": 0.43019958264219593,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00049881674703687,
      "loss": 6.5675,
      "mean_token_accuracy": 0.11938829645514488,
      "num_tokens": 8710445.0,
      "step": 4020
    },
    {
      "entropy": 6.7165913581848145,
      "epoch": 0.43073465674995987,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004988128278904499,
      "loss": 6.7093,
      "mean_token_accuracy": 0.11114189699292183,
      "num_tokens": 8720686.0,
      "step": 4025
    },
    {
      "entropy": 6.800965404510498,
      "epoch": 0.4312697308577238,
      "grad_norm": 1.40625,
      "learning_rate": 0.000498808902281466,
      "loss": 6.7301,
      "mean_token_accuracy": 0.11561328247189522,
      "num_tokens": 8732122.0,
      "step": 4030
    },
    {
      "entropy": 6.702591466903686,
      "epoch": 0.43180480496548773,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004988049702100315,
      "loss": 6.6836,
      "mean_token_accuracy": 0.11544679328799248,
      "num_tokens": 8744122.0,
      "step": 4035
    },
    {
      "entropy": 6.817325067520142,
      "epoch": 0.43233987907325166,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988010316762598,
      "loss": 6.6752,
      "mean_token_accuracy": 0.11280133873224259,
      "num_tokens": 8754758.0,
      "step": 4040
    },
    {
      "entropy": 6.716195344924927,
      "epoch": 0.4328749531810156,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004987970866802649,
      "loss": 6.7162,
      "mean_token_accuracy": 0.10749638751149178,
      "num_tokens": 8766577.0,
      "step": 4045
    },
    {
      "entropy": 6.725743436813355,
      "epoch": 0.4334100272887795,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004987931352221605,
      "loss": 6.6932,
      "mean_token_accuracy": 0.11314602792263032,
      "num_tokens": 8777698.0,
      "step": 4050
    },
    {
      "entropy": 6.784752559661865,
      "epoch": 0.4339451013965434,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004987891773020606,
      "loss": 6.7042,
      "mean_token_accuracy": 0.1157095618546009,
      "num_tokens": 8788637.0,
      "step": 4055
    },
    {
      "entropy": 6.75966739654541,
      "epoch": 0.43448017550430734,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004987852129200799,
      "loss": 6.6904,
      "mean_token_accuracy": 0.12118750438094139,
      "num_tokens": 8799709.0,
      "step": 4060
    },
    {
      "entropy": 6.763680171966553,
      "epoch": 0.4350152496120713,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004987812420763326,
      "loss": 6.6982,
      "mean_token_accuracy": 0.11437618657946587,
      "num_tokens": 8810469.0,
      "step": 4065
    },
    {
      "entropy": 6.7934815883636475,
      "epoch": 0.4355503237198352,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004987772647709333,
      "loss": 6.6646,
      "mean_token_accuracy": 0.1136926494538784,
      "num_tokens": 8821142.0,
      "step": 4070
    },
    {
      "entropy": 6.67820553779602,
      "epoch": 0.43608539782759914,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000498773281003997,
      "loss": 6.6126,
      "mean_token_accuracy": 0.1220873899757862,
      "num_tokens": 8832951.0,
      "step": 4075
    },
    {
      "entropy": 6.693231391906738,
      "epoch": 0.4366204719353631,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004987692907756385,
      "loss": 6.7212,
      "mean_token_accuracy": 0.11040145382285119,
      "num_tokens": 8843579.0,
      "step": 4080
    },
    {
      "entropy": 6.709018087387085,
      "epoch": 0.43715554604312695,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004987652940859733,
      "loss": 6.6616,
      "mean_token_accuracy": 0.1068007379770279,
      "num_tokens": 8854612.0,
      "step": 4085
    },
    {
      "entropy": 6.724940013885498,
      "epoch": 0.4376906201508909,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004987612909351167,
      "loss": 6.6025,
      "mean_token_accuracy": 0.11898693069815636,
      "num_tokens": 8865584.0,
      "step": 4090
    },
    {
      "entropy": 6.678734683990479,
      "epoch": 0.4382256942586548,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004987572813231842,
      "loss": 6.6436,
      "mean_token_accuracy": 0.12159312292933463,
      "num_tokens": 8876662.0,
      "step": 4095
    },
    {
      "entropy": 6.733673048019409,
      "epoch": 0.43876076836641875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004987532652502917,
      "loss": 6.6882,
      "mean_token_accuracy": 0.11322307661175728,
      "num_tokens": 8887249.0,
      "step": 4100
    },
    {
      "entropy": 6.705870485305786,
      "epoch": 0.4392958424741827,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004987492427165551,
      "loss": 6.5221,
      "mean_token_accuracy": 0.1308815859258175,
      "num_tokens": 8898635.0,
      "step": 4105
    },
    {
      "entropy": 6.666793918609619,
      "epoch": 0.4398309165819466,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004987452137220906,
      "loss": 6.6872,
      "mean_token_accuracy": 0.11336872428655624,
      "num_tokens": 8908794.0,
      "step": 4110
    },
    {
      "entropy": 6.823773908615112,
      "epoch": 0.44036599068971055,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004987411782670144,
      "loss": 6.7029,
      "mean_token_accuracy": 0.1101386807858944,
      "num_tokens": 8919015.0,
      "step": 4115
    },
    {
      "entropy": 6.81250901222229,
      "epoch": 0.44090106479747443,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004987371363514431,
      "loss": 6.7245,
      "mean_token_accuracy": 0.1101568505167961,
      "num_tokens": 8929028.0,
      "step": 4120
    },
    {
      "entropy": 6.635428714752197,
      "epoch": 0.44143613890523836,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004987330879754933,
      "loss": 6.6238,
      "mean_token_accuracy": 0.11282940730452537,
      "num_tokens": 8939789.0,
      "step": 4125
    },
    {
      "entropy": 6.688043451309204,
      "epoch": 0.4419712130130023,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004987290331392823,
      "loss": 6.6084,
      "mean_token_accuracy": 0.11892879232764245,
      "num_tokens": 8950657.0,
      "step": 4130
    },
    {
      "entropy": 6.805780267715454,
      "epoch": 0.44250628712076623,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004987249718429269,
      "loss": 6.7204,
      "mean_token_accuracy": 0.11565686762332916,
      "num_tokens": 8960744.0,
      "step": 4135
    },
    {
      "entropy": 6.7055583000183105,
      "epoch": 0.44304136122853016,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004987209040865441,
      "loss": 6.5687,
      "mean_token_accuracy": 0.12038824409246444,
      "num_tokens": 8972411.0,
      "step": 4140
    },
    {
      "entropy": 6.730427551269531,
      "epoch": 0.4435764353362941,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004987168298702519,
      "loss": 6.7317,
      "mean_token_accuracy": 0.1166949637234211,
      "num_tokens": 8982818.0,
      "step": 4145
    },
    {
      "entropy": 6.724998998641968,
      "epoch": 0.44411150944405803,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004987127491941675,
      "loss": 6.6675,
      "mean_token_accuracy": 0.11578278020024299,
      "num_tokens": 8993227.0,
      "step": 4150
    },
    {
      "entropy": 6.677845239639282,
      "epoch": 0.4446465835518219,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004987086620584091,
      "loss": 6.6091,
      "mean_token_accuracy": 0.1162223570048809,
      "num_tokens": 9003796.0,
      "step": 4155
    },
    {
      "entropy": 6.727551651000977,
      "epoch": 0.44518165765958584,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004987045684630944,
      "loss": 6.6369,
      "mean_token_accuracy": 0.1215149775147438,
      "num_tokens": 9014931.0,
      "step": 4160
    },
    {
      "entropy": 6.773304080963134,
      "epoch": 0.44571673176734977,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004987004684083417,
      "loss": 6.695,
      "mean_token_accuracy": 0.11384897977113724,
      "num_tokens": 9025312.0,
      "step": 4165
    },
    {
      "entropy": 6.698511838912964,
      "epoch": 0.4462518058751137,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004986963618942694,
      "loss": 6.6607,
      "mean_token_accuracy": 0.12171419262886048,
      "num_tokens": 9035563.0,
      "step": 4170
    },
    {
      "entropy": 6.79593334197998,
      "epoch": 0.44678687998287764,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004986922489209961,
      "loss": 6.6928,
      "mean_token_accuracy": 0.11618542000651359,
      "num_tokens": 9046065.0,
      "step": 4175
    },
    {
      "entropy": 6.759232664108277,
      "epoch": 0.44732195409064157,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004986881294886406,
      "loss": 6.5975,
      "mean_token_accuracy": 0.12073868811130524,
      "num_tokens": 9056696.0,
      "step": 4180
    },
    {
      "entropy": 6.701839923858643,
      "epoch": 0.4478570281984055,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004986840035973218,
      "loss": 6.6752,
      "mean_token_accuracy": 0.11579028517007828,
      "num_tokens": 9067035.0,
      "step": 4185
    },
    {
      "entropy": 6.638756370544433,
      "epoch": 0.4483921023061694,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004986798712471588,
      "loss": 6.6443,
      "mean_token_accuracy": 0.11565731018781662,
      "num_tokens": 9078253.0,
      "step": 4190
    },
    {
      "entropy": 6.800721168518066,
      "epoch": 0.4489271764139333,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004986757324382709,
      "loss": 6.7076,
      "mean_token_accuracy": 0.10884766504168511,
      "num_tokens": 9088184.0,
      "step": 4195
    },
    {
      "entropy": 6.8079078674316404,
      "epoch": 0.44946225052169725,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004986715871707777,
      "loss": 6.7723,
      "mean_token_accuracy": 0.10660546869039536,
      "num_tokens": 9099321.0,
      "step": 4200
    },
    {
      "entropy": 6.796434831619263,
      "epoch": 0.4499973246294612,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004986674354447988,
      "loss": 6.7953,
      "mean_token_accuracy": 0.11461175382137298,
      "num_tokens": 9111410.0,
      "step": 4205
    },
    {
      "entropy": 6.752979707717896,
      "epoch": 0.4505323987372251,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004986632772604543,
      "loss": 6.6154,
      "mean_token_accuracy": 0.116925810277462,
      "num_tokens": 9121932.0,
      "step": 4210
    },
    {
      "entropy": 6.650112581253052,
      "epoch": 0.45106747284498905,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004986591126178639,
      "loss": 6.7024,
      "mean_token_accuracy": 0.12244073674082756,
      "num_tokens": 9132686.0,
      "step": 4215
    },
    {
      "entropy": 6.689599466323853,
      "epoch": 0.451602546952753,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004986549415171482,
      "loss": 6.636,
      "mean_token_accuracy": 0.1178563304245472,
      "num_tokens": 9143582.0,
      "step": 4220
    },
    {
      "entropy": 6.856327199935913,
      "epoch": 0.45213762106051686,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004986507639584274,
      "loss": 6.7034,
      "mean_token_accuracy": 0.11118230447173119,
      "num_tokens": 9156173.0,
      "step": 4225
    },
    {
      "entropy": 6.729980373382569,
      "epoch": 0.4526726951682808,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004986465799418223,
      "loss": 6.7019,
      "mean_token_accuracy": 0.1158701390028,
      "num_tokens": 9166882.0,
      "step": 4230
    },
    {
      "entropy": 6.663695859909057,
      "epoch": 0.4532077692760447,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004986423894674534,
      "loss": 6.6983,
      "mean_token_accuracy": 0.10844632685184478,
      "num_tokens": 9177525.0,
      "step": 4235
    },
    {
      "entropy": 6.789501523971557,
      "epoch": 0.45374284338380866,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004986381925354422,
      "loss": 6.6582,
      "mean_token_accuracy": 0.11211462765932083,
      "num_tokens": 9188586.0,
      "step": 4240
    },
    {
      "entropy": 6.768905687332153,
      "epoch": 0.4542779174915726,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004986339891459094,
      "loss": 6.5657,
      "mean_token_accuracy": 0.11805166080594062,
      "num_tokens": 9199920.0,
      "step": 4245
    },
    {
      "entropy": 6.551817274093628,
      "epoch": 0.4548129915993365,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004986297792989768,
      "loss": 6.5788,
      "mean_token_accuracy": 0.12117914929986,
      "num_tokens": 9210923.0,
      "step": 4250
    },
    {
      "entropy": 6.800738334655762,
      "epoch": 0.45534806570710046,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004986255629947655,
      "loss": 6.7368,
      "mean_token_accuracy": 0.11400101706385612,
      "num_tokens": 9221459.0,
      "step": 4255
    },
    {
      "entropy": 6.817624425888061,
      "epoch": 0.45588313981486434,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004986213402333978,
      "loss": 6.6856,
      "mean_token_accuracy": 0.11327185034751892,
      "num_tokens": 9231768.0,
      "step": 4260
    },
    {
      "entropy": 6.737683629989624,
      "epoch": 0.45641821392262827,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986171110149951,
      "loss": 6.6689,
      "mean_token_accuracy": 0.11119912713766097,
      "num_tokens": 9242230.0,
      "step": 4265
    },
    {
      "entropy": 6.731328105926513,
      "epoch": 0.4569532880303922,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004986128753396798,
      "loss": 6.5908,
      "mean_token_accuracy": 0.11729067414999009,
      "num_tokens": 9252485.0,
      "step": 4270
    },
    {
      "entropy": 6.63718695640564,
      "epoch": 0.45748836213815613,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004986086332075742,
      "loss": 6.6474,
      "mean_token_accuracy": 0.11773178726434708,
      "num_tokens": 9263014.0,
      "step": 4275
    },
    {
      "entropy": 6.737563610076904,
      "epoch": 0.45802343624592007,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004986043846188006,
      "loss": 6.6907,
      "mean_token_accuracy": 0.12012004256248474,
      "num_tokens": 9273777.0,
      "step": 4280
    },
    {
      "entropy": 6.647010040283203,
      "epoch": 0.458558510353684,
      "grad_norm": 1.359375,
      "learning_rate": 0.000498600129573482,
      "loss": 6.6115,
      "mean_token_accuracy": 0.11131804138422012,
      "num_tokens": 9284738.0,
      "step": 4285
    },
    {
      "entropy": 6.825235176086426,
      "epoch": 0.45909358446144793,
      "grad_norm": 2.1875,
      "learning_rate": 0.000498595868071741,
      "loss": 6.6814,
      "mean_token_accuracy": 0.11572107002139091,
      "num_tokens": 9295921.0,
      "step": 4290
    },
    {
      "entropy": 6.783471202850341,
      "epoch": 0.4596286585692118,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004985916001137006,
      "loss": 6.6691,
      "mean_token_accuracy": 0.11060161292552947,
      "num_tokens": 9306568.0,
      "step": 4295
    },
    {
      "entropy": 6.6858655452728275,
      "epoch": 0.46016373267697575,
      "grad_norm": 1.25,
      "learning_rate": 0.0004985873256994843,
      "loss": 6.607,
      "mean_token_accuracy": 0.114323341101408,
      "num_tokens": 9316468.0,
      "step": 4300
    },
    {
      "entropy": 6.6945489883422855,
      "epoch": 0.4606988067847397,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004985830448292154,
      "loss": 6.6939,
      "mean_token_accuracy": 0.11146686300635338,
      "num_tokens": 9328185.0,
      "step": 4305
    },
    {
      "entropy": 6.783587503433227,
      "epoch": 0.4612338808925036,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004985787575030175,
      "loss": 6.6404,
      "mean_token_accuracy": 0.11596988141536713,
      "num_tokens": 9338591.0,
      "step": 4310
    },
    {
      "entropy": 6.729450464248657,
      "epoch": 0.46176895500026754,
      "grad_norm": 1.25,
      "learning_rate": 0.0004985744637210144,
      "loss": 6.6646,
      "mean_token_accuracy": 0.11485074758529663,
      "num_tokens": 9349286.0,
      "step": 4315
    },
    {
      "entropy": 6.733769798278809,
      "epoch": 0.4623040291080315,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00049857016348333,
      "loss": 6.6445,
      "mean_token_accuracy": 0.11966344341635704,
      "num_tokens": 9359752.0,
      "step": 4320
    },
    {
      "entropy": 6.770653200149536,
      "epoch": 0.4628391032157954,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004985658567900886,
      "loss": 6.7118,
      "mean_token_accuracy": 0.10913585424423218,
      "num_tokens": 9371823.0,
      "step": 4325
    },
    {
      "entropy": 6.616776466369629,
      "epoch": 0.4633741773235593,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004985615436414145,
      "loss": 6.5723,
      "mean_token_accuracy": 0.11298257783055306,
      "num_tokens": 9383127.0,
      "step": 4330
    },
    {
      "entropy": 6.718986940383911,
      "epoch": 0.4639092514313232,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004985572240374321,
      "loss": 6.6758,
      "mean_token_accuracy": 0.11775618195533752,
      "num_tokens": 9394296.0,
      "step": 4335
    },
    {
      "entropy": 6.758514356613159,
      "epoch": 0.46444432553908715,
      "grad_norm": 1.375,
      "learning_rate": 0.0004985528979782665,
      "loss": 6.6659,
      "mean_token_accuracy": 0.11210668310523034,
      "num_tokens": 9405988.0,
      "step": 4340
    },
    {
      "entropy": 6.788887548446655,
      "epoch": 0.4649793996468511,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004985485654640423,
      "loss": 6.662,
      "mean_token_accuracy": 0.11312065199017525,
      "num_tokens": 9417479.0,
      "step": 4345
    },
    {
      "entropy": 6.646952819824219,
      "epoch": 0.465514473754615,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004985442264948847,
      "loss": 6.6233,
      "mean_token_accuracy": 0.12258987501263618,
      "num_tokens": 9428050.0,
      "step": 4350
    },
    {
      "entropy": 6.678357696533203,
      "epoch": 0.46604954786237895,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004985398810709189,
      "loss": 6.5779,
      "mean_token_accuracy": 0.12279872298240661,
      "num_tokens": 9438229.0,
      "step": 4355
    },
    {
      "entropy": 6.694482088088989,
      "epoch": 0.4665846219701429,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004985355291922704,
      "loss": 6.6502,
      "mean_token_accuracy": 0.1130458801984787,
      "num_tokens": 9449611.0,
      "step": 4360
    },
    {
      "entropy": 6.6899035453796385,
      "epoch": 0.46711969607790677,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000498531170859065,
      "loss": 6.6153,
      "mean_token_accuracy": 0.11360270753502846,
      "num_tokens": 9460234.0,
      "step": 4365
    },
    {
      "entropy": 6.768774032592773,
      "epoch": 0.4676547701856707,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004985268060714285,
      "loss": 6.682,
      "mean_token_accuracy": 0.11366551518440246,
      "num_tokens": 9472136.0,
      "step": 4370
    },
    {
      "entropy": 6.685565567016601,
      "epoch": 0.46818984429343463,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004985224348294868,
      "loss": 6.6355,
      "mean_token_accuracy": 0.11784549504518509,
      "num_tokens": 9482602.0,
      "step": 4375
    },
    {
      "entropy": 6.7874668598175045,
      "epoch": 0.46872491840119856,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004985180571333663,
      "loss": 6.6296,
      "mean_token_accuracy": 0.11935302391648292,
      "num_tokens": 9493782.0,
      "step": 4380
    },
    {
      "entropy": 6.696510982513428,
      "epoch": 0.4692599925089625,
      "grad_norm": 1.5,
      "learning_rate": 0.0004985136729831932,
      "loss": 6.6032,
      "mean_token_accuracy": 0.11815939024090767,
      "num_tokens": 9504389.0,
      "step": 4385
    },
    {
      "entropy": 6.680622529983521,
      "epoch": 0.46979506661672643,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004985092823790942,
      "loss": 6.7378,
      "mean_token_accuracy": 0.11527246832847596,
      "num_tokens": 9515166.0,
      "step": 4390
    },
    {
      "entropy": 6.754097366333008,
      "epoch": 0.47033014072449036,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004985048853211961,
      "loss": 6.6354,
      "mean_token_accuracy": 0.1160196453332901,
      "num_tokens": 9525566.0,
      "step": 4395
    },
    {
      "entropy": 6.771543645858765,
      "epoch": 0.47086521483225424,
      "grad_norm": 1.671875,
      "learning_rate": 0.000498500481809626,
      "loss": 6.5404,
      "mean_token_accuracy": 0.12670470625162125,
      "num_tokens": 9536021.0,
      "step": 4400
    },
    {
      "entropy": 6.676743268966675,
      "epoch": 0.4714002889400182,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004984960718445107,
      "loss": 6.6619,
      "mean_token_accuracy": 0.11299687251448631,
      "num_tokens": 9546827.0,
      "step": 4405
    },
    {
      "entropy": 6.7357440948486325,
      "epoch": 0.4719353630477821,
      "grad_norm": 1.5,
      "learning_rate": 0.0004984916554259778,
      "loss": 6.779,
      "mean_token_accuracy": 0.10542291179299354,
      "num_tokens": 9558104.0,
      "step": 4410
    },
    {
      "entropy": 6.807965230941773,
      "epoch": 0.47247043715554604,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004984872325541547,
      "loss": 6.6585,
      "mean_token_accuracy": 0.11875886842608452,
      "num_tokens": 9568487.0,
      "step": 4415
    },
    {
      "entropy": 6.72672324180603,
      "epoch": 0.47300551126331,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004984828032291692,
      "loss": 6.6088,
      "mean_token_accuracy": 0.12393661811947823,
      "num_tokens": 9579509.0,
      "step": 4420
    },
    {
      "entropy": 6.716574716567993,
      "epoch": 0.4735405853710739,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004984783674511492,
      "loss": 6.6388,
      "mean_token_accuracy": 0.11688132956624031,
      "num_tokens": 9589931.0,
      "step": 4425
    },
    {
      "entropy": 6.7539482593536375,
      "epoch": 0.47407565947883784,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004984739252202227,
      "loss": 6.688,
      "mean_token_accuracy": 0.11628134176135063,
      "num_tokens": 9599764.0,
      "step": 4430
    },
    {
      "entropy": 6.757166957855224,
      "epoch": 0.4746107335866017,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000498469476536518,
      "loss": 6.6493,
      "mean_token_accuracy": 0.11907771229743958,
      "num_tokens": 9611145.0,
      "step": 4435
    },
    {
      "entropy": 6.693255519866943,
      "epoch": 0.47514580769436565,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004984650214001636,
      "loss": 6.6372,
      "mean_token_accuracy": 0.11735451370477676,
      "num_tokens": 9621441.0,
      "step": 4440
    },
    {
      "entropy": 6.723459720611572,
      "epoch": 0.4756808818021296,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004984605598112881,
      "loss": 6.6922,
      "mean_token_accuracy": 0.11924364715814591,
      "num_tokens": 9631874.0,
      "step": 4445
    },
    {
      "entropy": 6.672471523284912,
      "epoch": 0.4762159559098935,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004984560917700204,
      "loss": 6.6712,
      "mean_token_accuracy": 0.10813649892807006,
      "num_tokens": 9643006.0,
      "step": 4450
    },
    {
      "entropy": 6.661323070526123,
      "epoch": 0.47675103001765745,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004984516172764894,
      "loss": 6.5508,
      "mean_token_accuracy": 0.12179366350173951,
      "num_tokens": 9653897.0,
      "step": 4455
    },
    {
      "entropy": 6.691922187805176,
      "epoch": 0.4772861041254214,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004984471363308243,
      "loss": 6.6301,
      "mean_token_accuracy": 0.11610539332032203,
      "num_tokens": 9664460.0,
      "step": 4460
    },
    {
      "entropy": 6.757071876525879,
      "epoch": 0.4778211782331853,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004984426489331546,
      "loss": 6.6456,
      "mean_token_accuracy": 0.11981170177459717,
      "num_tokens": 9674534.0,
      "step": 4465
    },
    {
      "entropy": 6.799601793289185,
      "epoch": 0.4783562523409492,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004984381550836097,
      "loss": 6.7187,
      "mean_token_accuracy": 0.11188295409083367,
      "num_tokens": 9684828.0,
      "step": 4470
    },
    {
      "entropy": 6.758711624145508,
      "epoch": 0.47889132644871313,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004984336547823196,
      "loss": 6.6933,
      "mean_token_accuracy": 0.11652329340577125,
      "num_tokens": 9694536.0,
      "step": 4475
    },
    {
      "entropy": 6.68265733718872,
      "epoch": 0.47942640055647706,
      "grad_norm": 1.171875,
      "learning_rate": 0.000498429148029414,
      "loss": 6.5507,
      "mean_token_accuracy": 0.12725870534777642,
      "num_tokens": 9705715.0,
      "step": 4480
    },
    {
      "entropy": 6.687233924865723,
      "epoch": 0.479961474664241,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004984246348250233,
      "loss": 6.6204,
      "mean_token_accuracy": 0.12101843059062958,
      "num_tokens": 9717505.0,
      "step": 4485
    },
    {
      "entropy": 6.6921549320220945,
      "epoch": 0.4804965487720049,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004984201151692775,
      "loss": 6.5105,
      "mean_token_accuracy": 0.11698877215385436,
      "num_tokens": 9727910.0,
      "step": 4490
    },
    {
      "entropy": 6.705784749984741,
      "epoch": 0.48103162287976886,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004984155890623074,
      "loss": 6.6029,
      "mean_token_accuracy": 0.1223738431930542,
      "num_tokens": 9738528.0,
      "step": 4495
    },
    {
      "entropy": 6.68727297782898,
      "epoch": 0.4815666969875328,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004984110565042435,
      "loss": 6.6413,
      "mean_token_accuracy": 0.1166694700717926,
      "num_tokens": 9749744.0,
      "step": 4500
    },
    {
      "entropy": 6.786122751235962,
      "epoch": 0.48210177109529667,
      "grad_norm": 1.375,
      "learning_rate": 0.0004984065174952168,
      "loss": 6.6433,
      "mean_token_accuracy": 0.11737963259220123,
      "num_tokens": 9759968.0,
      "step": 4505
    },
    {
      "entropy": 6.7307963371276855,
      "epoch": 0.4826368452030606,
      "grad_norm": 1.5,
      "learning_rate": 0.0004984019720353582,
      "loss": 6.6591,
      "mean_token_accuracy": 0.11601342111825944,
      "num_tokens": 9769791.0,
      "step": 4510
    },
    {
      "entropy": 6.613153791427612,
      "epoch": 0.48317191931082454,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004983974201247991,
      "loss": 6.6165,
      "mean_token_accuracy": 0.12467638105154037,
      "num_tokens": 9780369.0,
      "step": 4515
    },
    {
      "entropy": 6.6523078918457035,
      "epoch": 0.48370699341858847,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004983928617636709,
      "loss": 6.6717,
      "mean_token_accuracy": 0.1142366997897625,
      "num_tokens": 9791125.0,
      "step": 4520
    },
    {
      "entropy": 6.807734060287475,
      "epoch": 0.4842420675263524,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004983882969521052,
      "loss": 6.6638,
      "mean_token_accuracy": 0.11914790123701095,
      "num_tokens": 9801697.0,
      "step": 4525
    },
    {
      "entropy": 6.810341739654541,
      "epoch": 0.48477714163411634,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004983837256902338,
      "loss": 6.704,
      "mean_token_accuracy": 0.11736872941255569,
      "num_tokens": 9812171.0,
      "step": 4530
    },
    {
      "entropy": 6.725469064712525,
      "epoch": 0.48531221574188027,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004983791479781887,
      "loss": 6.5492,
      "mean_token_accuracy": 0.12601315155625342,
      "num_tokens": 9821596.0,
      "step": 4535
    },
    {
      "entropy": 6.740454387664795,
      "epoch": 0.48584728984964415,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983745638161021,
      "loss": 6.6622,
      "mean_token_accuracy": 0.10746660381555558,
      "num_tokens": 9831634.0,
      "step": 4540
    },
    {
      "entropy": 6.724510765075683,
      "epoch": 0.4863823639574081,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004983699732041064,
      "loss": 6.6345,
      "mean_token_accuracy": 0.11942023038864136,
      "num_tokens": 9841662.0,
      "step": 4545
    },
    {
      "entropy": 6.743528079986572,
      "epoch": 0.486917438065172,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000498365376142334,
      "loss": 6.6175,
      "mean_token_accuracy": 0.11318280696868896,
      "num_tokens": 9853797.0,
      "step": 4550
    },
    {
      "entropy": 6.693184661865234,
      "epoch": 0.48745251217293595,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004983607726309179,
      "loss": 6.6665,
      "mean_token_accuracy": 0.11820111572742462,
      "num_tokens": 9865245.0,
      "step": 4555
    },
    {
      "entropy": 6.745386791229248,
      "epoch": 0.4879875862806999,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004983561626699908,
      "loss": 6.7321,
      "mean_token_accuracy": 0.10730336830019951,
      "num_tokens": 9876207.0,
      "step": 4560
    },
    {
      "entropy": 6.771791172027588,
      "epoch": 0.4885226603884638,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004983515462596859,
      "loss": 6.5706,
      "mean_token_accuracy": 0.11805153340101242,
      "num_tokens": 9887210.0,
      "step": 4565
    },
    {
      "entropy": 6.77590970993042,
      "epoch": 0.48905773449622775,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004983469234001365,
      "loss": 6.7364,
      "mean_token_accuracy": 0.11187749877572059,
      "num_tokens": 9899607.0,
      "step": 4570
    },
    {
      "entropy": 6.614283847808838,
      "epoch": 0.4895928086039917,
      "grad_norm": 1.625,
      "learning_rate": 0.000498342294091476,
      "loss": 6.5198,
      "mean_token_accuracy": 0.12553614974021912,
      "num_tokens": 9909910.0,
      "step": 4575
    },
    {
      "entropy": 6.625487995147705,
      "epoch": 0.49012788271175556,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004983376583338382,
      "loss": 6.6568,
      "mean_token_accuracy": 0.11543531715869904,
      "num_tokens": 9922137.0,
      "step": 4580
    },
    {
      "entropy": 6.735957527160645,
      "epoch": 0.4906629568195195,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004983330161273569,
      "loss": 6.6388,
      "mean_token_accuracy": 0.11887715309858322,
      "num_tokens": 9932514.0,
      "step": 4585
    },
    {
      "entropy": 6.728844404220581,
      "epoch": 0.4911980309272834,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004983283674721662,
      "loss": 6.6983,
      "mean_token_accuracy": 0.1141705684363842,
      "num_tokens": 9943054.0,
      "step": 4590
    },
    {
      "entropy": 6.811689996719361,
      "epoch": 0.49173310503504736,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004983237123684002,
      "loss": 6.6487,
      "mean_token_accuracy": 0.11451672837138176,
      "num_tokens": 9953791.0,
      "step": 4595
    },
    {
      "entropy": 6.690435552597046,
      "epoch": 0.4922681791428113,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004983190508161934,
      "loss": 6.5993,
      "mean_token_accuracy": 0.11774043142795562,
      "num_tokens": 9964782.0,
      "step": 4600
    },
    {
      "entropy": 6.71423020362854,
      "epoch": 0.4928032532505752,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004983143828156804,
      "loss": 6.674,
      "mean_token_accuracy": 0.11906063780188561,
      "num_tokens": 9975843.0,
      "step": 4605
    },
    {
      "entropy": 6.751412296295166,
      "epoch": 0.49333832735833916,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004983097083669959,
      "loss": 6.6477,
      "mean_token_accuracy": 0.12174972966313362,
      "num_tokens": 9987060.0,
      "step": 4610
    },
    {
      "entropy": 6.72445158958435,
      "epoch": 0.49387340146610303,
      "grad_norm": 1.5,
      "learning_rate": 0.0004983050274702751,
      "loss": 6.6159,
      "mean_token_accuracy": 0.11575492694973946,
      "num_tokens": 9997380.0,
      "step": 4615
    },
    {
      "entropy": 6.6969047546386715,
      "epoch": 0.49440847557386697,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004983003401256529,
      "loss": 6.6723,
      "mean_token_accuracy": 0.11978892460465432,
      "num_tokens": 10007955.0,
      "step": 4620
    },
    {
      "entropy": 6.840445518493652,
      "epoch": 0.4949435496816309,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004982956463332647,
      "loss": 6.7401,
      "mean_token_accuracy": 0.11213290095329284,
      "num_tokens": 10017645.0,
      "step": 4625
    },
    {
      "entropy": 6.699553489685059,
      "epoch": 0.49547862378939483,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004982909460932463,
      "loss": 6.6667,
      "mean_token_accuracy": 0.1187513791024685,
      "num_tokens": 10027619.0,
      "step": 4630
    },
    {
      "entropy": 6.6246099948883055,
      "epoch": 0.49601369789715877,
      "grad_norm": 1.203125,
      "learning_rate": 0.000498286239405733,
      "loss": 6.5487,
      "mean_token_accuracy": 0.1270728625357151,
      "num_tokens": 10037551.0,
      "step": 4635
    },
    {
      "entropy": 6.668668556213379,
      "epoch": 0.4965487720049227,
      "grad_norm": 1.203125,
      "learning_rate": 0.000498281526270861,
      "loss": 6.5738,
      "mean_token_accuracy": 0.126234732568264,
      "num_tokens": 10047771.0,
      "step": 4640
    },
    {
      "entropy": 6.833411169052124,
      "epoch": 0.49708384611268663,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004982768066887663,
      "loss": 6.7107,
      "mean_token_accuracy": 0.11156112849712371,
      "num_tokens": 10058771.0,
      "step": 4645
    },
    {
      "entropy": 6.687218379974365,
      "epoch": 0.4976189202204505,
      "grad_norm": 1.375,
      "learning_rate": 0.0004982720806595851,
      "loss": 6.5786,
      "mean_token_accuracy": 0.1209044374525547,
      "num_tokens": 10070318.0,
      "step": 4650
    },
    {
      "entropy": 6.636781406402588,
      "epoch": 0.49815399432821444,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004982673481834541,
      "loss": 6.6068,
      "mean_token_accuracy": 0.11889650225639344,
      "num_tokens": 10081810.0,
      "step": 4655
    },
    {
      "entropy": 6.646404218673706,
      "epoch": 0.4986890684359784,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004982626092605097,
      "loss": 6.6221,
      "mean_token_accuracy": 0.1171889565885067,
      "num_tokens": 10091724.0,
      "step": 4660
    },
    {
      "entropy": 6.777220296859741,
      "epoch": 0.4992241425437423,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004982578638908888,
      "loss": 6.6099,
      "mean_token_accuracy": 0.11882191374897957,
      "num_tokens": 10102510.0,
      "step": 4665
    },
    {
      "entropy": 6.6920019626617435,
      "epoch": 0.49975921665150624,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004982531120747286,
      "loss": 6.6661,
      "mean_token_accuracy": 0.11249755322933197,
      "num_tokens": 10112636.0,
      "step": 4670
    },
    {
      "entropy": 6.652344703674316,
      "epoch": 0.5002942907592701,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000498248353812166,
      "loss": 6.6787,
      "mean_token_accuracy": 0.11986306384205818,
      "num_tokens": 10123562.0,
      "step": 4675
    },
    {
      "entropy": 6.767915964126587,
      "epoch": 0.5008293648670341,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004982435891033387,
      "loss": 6.6374,
      "mean_token_accuracy": 0.1114320233464241,
      "num_tokens": 10134731.0,
      "step": 4680
    },
    {
      "entropy": 6.7398899555206295,
      "epoch": 0.501364438974798,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000498238817948384,
      "loss": 6.6234,
      "mean_token_accuracy": 0.11613459065556526,
      "num_tokens": 10145232.0,
      "step": 4685
    },
    {
      "entropy": 6.673232889175415,
      "epoch": 0.501899513082562,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004982340403474398,
      "loss": 6.5915,
      "mean_token_accuracy": 0.1116428665816784,
      "num_tokens": 10155857.0,
      "step": 4690
    },
    {
      "entropy": 6.683508396148682,
      "epoch": 0.5024345871903259,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004982292563006441,
      "loss": 6.5975,
      "mean_token_accuracy": 0.12228783667087555,
      "num_tokens": 10166145.0,
      "step": 4695
    },
    {
      "entropy": 6.740499687194824,
      "epoch": 0.5029696612980897,
      "grad_norm": 1.234375,
      "learning_rate": 0.000498224465808135,
      "loss": 6.7254,
      "mean_token_accuracy": 0.11173807680606843,
      "num_tokens": 10177086.0,
      "step": 4700
    },
    {
      "entropy": 6.775094699859619,
      "epoch": 0.5035047354058537,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004982196688700509,
      "loss": 6.6056,
      "mean_token_accuracy": 0.11548131629824639,
      "num_tokens": 10187120.0,
      "step": 4705
    },
    {
      "entropy": 6.700378704071045,
      "epoch": 0.5040398095136176,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004982148654865302,
      "loss": 6.6714,
      "mean_token_accuracy": 0.116058349609375,
      "num_tokens": 10197784.0,
      "step": 4710
    },
    {
      "entropy": 6.731067323684693,
      "epoch": 0.5045748836213816,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004982100556577115,
      "loss": 6.5272,
      "mean_token_accuracy": 0.12650172412395477,
      "num_tokens": 10209373.0,
      "step": 4715
    },
    {
      "entropy": 6.75590672492981,
      "epoch": 0.5051099577291455,
      "grad_norm": 1.375,
      "learning_rate": 0.0004982052393837338,
      "loss": 6.662,
      "mean_token_accuracy": 0.12109206095337868,
      "num_tokens": 10220316.0,
      "step": 4720
    },
    {
      "entropy": 6.6641072750091555,
      "epoch": 0.5056450318369095,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004982004166647363,
      "loss": 6.555,
      "mean_token_accuracy": 0.12492926642298699,
      "num_tokens": 10230478.0,
      "step": 4725
    },
    {
      "entropy": 6.577092838287354,
      "epoch": 0.5061801059446733,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004981955875008582,
      "loss": 6.6348,
      "mean_token_accuracy": 0.11583661809563636,
      "num_tokens": 10240641.0,
      "step": 4730
    },
    {
      "entropy": 6.696561050415039,
      "epoch": 0.5067151800524372,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004981907518922387,
      "loss": 6.6138,
      "mean_token_accuracy": 0.12158278226852418,
      "num_tokens": 10250676.0,
      "step": 4735
    },
    {
      "entropy": 6.828497314453125,
      "epoch": 0.5072502541602012,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004981859098390177,
      "loss": 6.6949,
      "mean_token_accuracy": 0.11689818948507309,
      "num_tokens": 10262016.0,
      "step": 4740
    },
    {
      "entropy": 6.777277851104737,
      "epoch": 0.5077853282679651,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004981810613413348,
      "loss": 6.6381,
      "mean_token_accuracy": 0.12088718265295029,
      "num_tokens": 10271985.0,
      "step": 4745
    },
    {
      "entropy": 6.707176303863525,
      "epoch": 0.5083204023757291,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004981762063993302,
      "loss": 6.6092,
      "mean_token_accuracy": 0.12419556155800819,
      "num_tokens": 10282156.0,
      "step": 4750
    },
    {
      "entropy": 6.672820425033569,
      "epoch": 0.5088554764834929,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000498171345013144,
      "loss": 6.6133,
      "mean_token_accuracy": 0.1135848581790924,
      "num_tokens": 10292576.0,
      "step": 4755
    },
    {
      "entropy": 6.647624206542969,
      "epoch": 0.5093905505912569,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004981664771829165,
      "loss": 6.6234,
      "mean_token_accuracy": 0.11965413317084313,
      "num_tokens": 10303648.0,
      "step": 4760
    },
    {
      "entropy": 6.7490908145904545,
      "epoch": 0.5099256246990208,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004981616029087884,
      "loss": 6.5074,
      "mean_token_accuracy": 0.12869232892990112,
      "num_tokens": 10314275.0,
      "step": 4765
    },
    {
      "entropy": 6.696971416473389,
      "epoch": 0.5104606988067847,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004981567221909004,
      "loss": 6.6413,
      "mean_token_accuracy": 0.11142453476786614,
      "num_tokens": 10325386.0,
      "step": 4770
    },
    {
      "entropy": 6.657241535186768,
      "epoch": 0.5109957729145487,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004981518350293935,
      "loss": 6.6256,
      "mean_token_accuracy": 0.117152751237154,
      "num_tokens": 10335041.0,
      "step": 4775
    },
    {
      "entropy": 6.6835949420928955,
      "epoch": 0.5115308470223126,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004981469414244086,
      "loss": 6.5783,
      "mean_token_accuracy": 0.12038632929325103,
      "num_tokens": 10345746.0,
      "step": 4780
    },
    {
      "entropy": 6.611367559432983,
      "epoch": 0.5120659211300765,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000498142041376087,
      "loss": 6.6877,
      "mean_token_accuracy": 0.11743063405156136,
      "num_tokens": 10356995.0,
      "step": 4785
    },
    {
      "entropy": 6.6938282489776615,
      "epoch": 0.5126009952378404,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004981371348845705,
      "loss": 6.5878,
      "mean_token_accuracy": 0.12293134033679962,
      "num_tokens": 10368445.0,
      "step": 4790
    },
    {
      "entropy": 6.774974012374878,
      "epoch": 0.5131360693456044,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004981322219500006,
      "loss": 6.5345,
      "mean_token_accuracy": 0.12403296306729317,
      "num_tokens": 10378256.0,
      "step": 4795
    },
    {
      "entropy": 6.728860235214233,
      "epoch": 0.5136711434533683,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000498127302572519,
      "loss": 6.6166,
      "mean_token_accuracy": 0.12460722997784615,
      "num_tokens": 10388698.0,
      "step": 4800
    },
    {
      "entropy": 6.613707876205444,
      "epoch": 0.5142062175611322,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004981223767522679,
      "loss": 6.6095,
      "mean_token_accuracy": 0.1258344203233719,
      "num_tokens": 10399975.0,
      "step": 4805
    },
    {
      "entropy": 6.650073432922364,
      "epoch": 0.5147412916688962,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004981174444893896,
      "loss": 6.6332,
      "mean_token_accuracy": 0.1165822133421898,
      "num_tokens": 10410480.0,
      "step": 4810
    },
    {
      "entropy": 6.698668336868286,
      "epoch": 0.51527636577666,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004981125057840264,
      "loss": 6.5883,
      "mean_token_accuracy": 0.12364424616098404,
      "num_tokens": 10421023.0,
      "step": 4815
    },
    {
      "entropy": 6.73700180053711,
      "epoch": 0.515811439884424,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000498107560636321,
      "loss": 6.6444,
      "mean_token_accuracy": 0.11720668748021126,
      "num_tokens": 10431152.0,
      "step": 4820
    },
    {
      "entropy": 6.710340929031372,
      "epoch": 0.5163465139921879,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004981026090464161,
      "loss": 6.5545,
      "mean_token_accuracy": 0.11190668419003487,
      "num_tokens": 10441986.0,
      "step": 4825
    },
    {
      "entropy": 6.667047786712646,
      "epoch": 0.5168815880999519,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004980976510144548,
      "loss": 6.6265,
      "mean_token_accuracy": 0.11371301561594009,
      "num_tokens": 10452932.0,
      "step": 4830
    },
    {
      "entropy": 6.670616531372071,
      "epoch": 0.5174166622077158,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004980926865405801,
      "loss": 6.6947,
      "mean_token_accuracy": 0.11335934177041054,
      "num_tokens": 10463411.0,
      "step": 4835
    },
    {
      "entropy": 6.711873292922974,
      "epoch": 0.5179517363154797,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004980877156249354,
      "loss": 6.6047,
      "mean_token_accuracy": 0.12172490879893302,
      "num_tokens": 10474820.0,
      "step": 4840
    },
    {
      "entropy": 6.69803557395935,
      "epoch": 0.5184868104232436,
      "grad_norm": 2.625,
      "learning_rate": 0.0004980827382676643,
      "loss": 6.6062,
      "mean_token_accuracy": 0.12013482972979546,
      "num_tokens": 10485701.0,
      "step": 4845
    },
    {
      "entropy": 6.6993248462677,
      "epoch": 0.5190218845310075,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004980777544689105,
      "loss": 6.5844,
      "mean_token_accuracy": 0.12140351459383965,
      "num_tokens": 10495825.0,
      "step": 4850
    },
    {
      "entropy": 6.6549890518188475,
      "epoch": 0.5195569586387715,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004980727642288178,
      "loss": 6.5532,
      "mean_token_accuracy": 0.12483188882470131,
      "num_tokens": 10505847.0,
      "step": 4855
    },
    {
      "entropy": 6.708909797668457,
      "epoch": 0.5200920327465354,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004980677675475305,
      "loss": 6.6115,
      "mean_token_accuracy": 0.11702346950769424,
      "num_tokens": 10516596.0,
      "step": 4860
    },
    {
      "entropy": 6.7008233070373535,
      "epoch": 0.5206271068542994,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004980627644251926,
      "loss": 6.5554,
      "mean_token_accuracy": 0.12680012062191964,
      "num_tokens": 10528175.0,
      "step": 4865
    },
    {
      "entropy": 6.666421699523926,
      "epoch": 0.5211621809620632,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004980577548619489,
      "loss": 6.6463,
      "mean_token_accuracy": 0.11762543395161629,
      "num_tokens": 10540076.0,
      "step": 4870
    },
    {
      "entropy": 6.704486227035522,
      "epoch": 0.5216972550698272,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004980527388579437,
      "loss": 6.6509,
      "mean_token_accuracy": 0.11489062979817391,
      "num_tokens": 10552025.0,
      "step": 4875
    },
    {
      "entropy": 6.730651092529297,
      "epoch": 0.5222323291775911,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004980477164133221,
      "loss": 6.6459,
      "mean_token_accuracy": 0.11571409627795219,
      "num_tokens": 10562535.0,
      "step": 4880
    },
    {
      "entropy": 6.673344421386719,
      "epoch": 0.522767403285355,
      "grad_norm": 1.4375,
      "learning_rate": 0.000498042687528229,
      "loss": 6.5891,
      "mean_token_accuracy": 0.11916224136948586,
      "num_tokens": 10572989.0,
      "step": 4885
    },
    {
      "entropy": 6.67263822555542,
      "epoch": 0.523302477393119,
      "grad_norm": 2.0,
      "learning_rate": 0.0004980376522028098,
      "loss": 6.593,
      "mean_token_accuracy": 0.12143604904413223,
      "num_tokens": 10583033.0,
      "step": 4890
    },
    {
      "entropy": 6.745864820480347,
      "epoch": 0.5238375515008828,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004980326104372095,
      "loss": 6.6788,
      "mean_token_accuracy": 0.11569171249866486,
      "num_tokens": 10595205.0,
      "step": 4895
    },
    {
      "entropy": 6.741711235046386,
      "epoch": 0.5243726256086468,
      "grad_norm": 1.265625,
      "learning_rate": 0.000498027562231574,
      "loss": 6.5294,
      "mean_token_accuracy": 0.12321098670363426,
      "num_tokens": 10605244.0,
      "step": 4900
    },
    {
      "entropy": 6.6820556640625,
      "epoch": 0.5249076997164107,
      "grad_norm": 1.609375,
      "learning_rate": 0.000498022507586049,
      "loss": 6.6115,
      "mean_token_accuracy": 0.11927070915699005,
      "num_tokens": 10615817.0,
      "step": 4905
    },
    {
      "entropy": 6.739273118972778,
      "epoch": 0.5254427738241747,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004980174465007804,
      "loss": 6.6369,
      "mean_token_accuracy": 0.11548423320055008,
      "num_tokens": 10627160.0,
      "step": 4910
    },
    {
      "entropy": 6.701236152648926,
      "epoch": 0.5259778479319386,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004980123789759145,
      "loss": 6.6288,
      "mean_token_accuracy": 0.12106893360614776,
      "num_tokens": 10637552.0,
      "step": 4915
    },
    {
      "entropy": 6.652713632583618,
      "epoch": 0.5265129220397025,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004980073050115973,
      "loss": 6.6653,
      "mean_token_accuracy": 0.11880994513630867,
      "num_tokens": 10648773.0,
      "step": 4920
    },
    {
      "entropy": 6.774326801300049,
      "epoch": 0.5270479961474664,
      "grad_norm": 1.25,
      "learning_rate": 0.0004980022246079755,
      "loss": 6.648,
      "mean_token_accuracy": 0.11824332997202873,
      "num_tokens": 10659537.0,
      "step": 4925
    },
    {
      "entropy": 6.787177610397339,
      "epoch": 0.5275830702552303,
      "grad_norm": 1.484375,
      "learning_rate": 0.000497997137765196,
      "loss": 6.6894,
      "mean_token_accuracy": 0.11097191125154496,
      "num_tokens": 10671390.0,
      "step": 4930
    },
    {
      "entropy": 6.683742904663086,
      "epoch": 0.5281181443629943,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004979920444834054,
      "loss": 6.5367,
      "mean_token_accuracy": 0.12748999372124672,
      "num_tokens": 10682389.0,
      "step": 4935
    },
    {
      "entropy": 6.614830350875854,
      "epoch": 0.5286532184707582,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004979869447627508,
      "loss": 6.5971,
      "mean_token_accuracy": 0.11691061481833458,
      "num_tokens": 10692400.0,
      "step": 4940
    },
    {
      "entropy": 6.658047914505005,
      "epoch": 0.5291882925785222,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004979818386033795,
      "loss": 6.5312,
      "mean_token_accuracy": 0.12515476047992707,
      "num_tokens": 10702396.0,
      "step": 4945
    },
    {
      "entropy": 6.689503908157349,
      "epoch": 0.5297233666862861,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000497976726005439,
      "loss": 6.6287,
      "mean_token_accuracy": 0.11893323883414268,
      "num_tokens": 10712863.0,
      "step": 4950
    },
    {
      "entropy": 6.741910123825074,
      "epoch": 0.5302584407940499,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004979716069690768,
      "loss": 6.6462,
      "mean_token_accuracy": 0.1182481288909912,
      "num_tokens": 10724234.0,
      "step": 4955
    },
    {
      "entropy": 6.784095907211304,
      "epoch": 0.5307935149018139,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004979664814944409,
      "loss": 6.6831,
      "mean_token_accuracy": 0.11570578664541245,
      "num_tokens": 10734405.0,
      "step": 4960
    },
    {
      "entropy": 6.778741836547852,
      "epoch": 0.5313285890095778,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004979613495816791,
      "loss": 6.6321,
      "mean_token_accuracy": 0.11481416672468185,
      "num_tokens": 10744573.0,
      "step": 4965
    },
    {
      "entropy": 6.718149614334107,
      "epoch": 0.5318636631173418,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004979562112309397,
      "loss": 6.6685,
      "mean_token_accuracy": 0.11846106275916099,
      "num_tokens": 10755555.0,
      "step": 4970
    },
    {
      "entropy": 6.686836433410645,
      "epoch": 0.5323987372251057,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000497951066442371,
      "loss": 6.6784,
      "mean_token_accuracy": 0.11362863406538963,
      "num_tokens": 10766493.0,
      "step": 4975
    },
    {
      "entropy": 6.777250051498413,
      "epoch": 0.5329338113328697,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004979459152161215,
      "loss": 6.6304,
      "mean_token_accuracy": 0.11238612830638886,
      "num_tokens": 10776735.0,
      "step": 4980
    },
    {
      "entropy": 6.693412494659424,
      "epoch": 0.5334688854406335,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004979407575523402,
      "loss": 6.6106,
      "mean_token_accuracy": 0.12066147401928902,
      "num_tokens": 10787500.0,
      "step": 4985
    },
    {
      "entropy": 6.690456914901733,
      "epoch": 0.5340039595483974,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004979355934511757,
      "loss": 6.6421,
      "mean_token_accuracy": 0.1213709406554699,
      "num_tokens": 10798002.0,
      "step": 4990
    },
    {
      "entropy": 6.6947132587432865,
      "epoch": 0.5345390336561614,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004979304229127773,
      "loss": 6.603,
      "mean_token_accuracy": 0.11744242087006569,
      "num_tokens": 10808982.0,
      "step": 4995
    },
    {
      "entropy": 6.681551885604859,
      "epoch": 0.5350741077639253,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004979252459372944,
      "loss": 6.5543,
      "mean_token_accuracy": 0.1188325934112072,
      "num_tokens": 10821619.0,
      "step": 5000
    },
    {
      "entropy": 6.6972753524780275,
      "epoch": 0.5356091818716893,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004979200625248761,
      "loss": 6.5819,
      "mean_token_accuracy": 0.12012385129928589,
      "num_tokens": 10832344.0,
      "step": 5005
    },
    {
      "entropy": 6.714497470855713,
      "epoch": 0.5361442559794531,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004979148726756725,
      "loss": 6.5517,
      "mean_token_accuracy": 0.11546608954668044,
      "num_tokens": 10843149.0,
      "step": 5010
    },
    {
      "entropy": 6.675411415100098,
      "epoch": 0.5366793300872171,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004979096763898333,
      "loss": 6.5895,
      "mean_token_accuracy": 0.12327387034893036,
      "num_tokens": 10854590.0,
      "step": 5015
    },
    {
      "entropy": 6.743090057373047,
      "epoch": 0.537214404194981,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004979044736675084,
      "loss": 6.5735,
      "mean_token_accuracy": 0.1187887541949749,
      "num_tokens": 10865500.0,
      "step": 5020
    },
    {
      "entropy": 6.62291259765625,
      "epoch": 0.5377494783027449,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004978992645088483,
      "loss": 6.6013,
      "mean_token_accuracy": 0.12360128462314605,
      "num_tokens": 10876866.0,
      "step": 5025
    },
    {
      "entropy": 6.608084964752197,
      "epoch": 0.5382845524105089,
      "grad_norm": 1.21875,
      "learning_rate": 0.000497894048914003,
      "loss": 6.5644,
      "mean_token_accuracy": 0.11877960562705994,
      "num_tokens": 10888627.0,
      "step": 5030
    },
    {
      "entropy": 6.748707246780396,
      "epoch": 0.5388196265182728,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004978888268831236,
      "loss": 6.6834,
      "mean_token_accuracy": 0.11199896112084388,
      "num_tokens": 10900992.0,
      "step": 5035
    },
    {
      "entropy": 6.803453779220581,
      "epoch": 0.5393547006260367,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004978835984163606,
      "loss": 6.706,
      "mean_token_accuracy": 0.10869432613253593,
      "num_tokens": 10912323.0,
      "step": 5040
    },
    {
      "entropy": 6.7344237804412845,
      "epoch": 0.5398897747338006,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004978783635138649,
      "loss": 6.668,
      "mean_token_accuracy": 0.11559137850999832,
      "num_tokens": 10923991.0,
      "step": 5045
    },
    {
      "entropy": 6.649737024307251,
      "epoch": 0.5404248488415646,
      "grad_norm": 1.25,
      "learning_rate": 0.0004978731221757878,
      "loss": 6.5077,
      "mean_token_accuracy": 0.12473165020346641,
      "num_tokens": 10934081.0,
      "step": 5050
    },
    {
      "entropy": 6.744383955001831,
      "epoch": 0.5409599229493285,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004978678744022808,
      "loss": 6.5946,
      "mean_token_accuracy": 0.12126863822340965,
      "num_tokens": 10944729.0,
      "step": 5055
    },
    {
      "entropy": 6.704465389251709,
      "epoch": 0.5414949970570924,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000497862620193495,
      "loss": 6.6605,
      "mean_token_accuracy": 0.11823692843317986,
      "num_tokens": 10955708.0,
      "step": 5060
    },
    {
      "entropy": 6.707375478744507,
      "epoch": 0.5420300711648564,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004978573595495826,
      "loss": 6.6073,
      "mean_token_accuracy": 0.11719232201576232,
      "num_tokens": 10966427.0,
      "step": 5065
    },
    {
      "entropy": 6.727742099761963,
      "epoch": 0.5425651452726202,
      "grad_norm": 1.359375,
      "learning_rate": 0.000497852092470695,
      "loss": 6.6438,
      "mean_token_accuracy": 0.11848462522029876,
      "num_tokens": 10977741.0,
      "step": 5070
    },
    {
      "entropy": 6.72758674621582,
      "epoch": 0.5431002193803842,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004978468189569847,
      "loss": 6.5907,
      "mean_token_accuracy": 0.12363087236881257,
      "num_tokens": 10989127.0,
      "step": 5075
    },
    {
      "entropy": 6.623909664154053,
      "epoch": 0.5436352934881481,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004978415390086038,
      "loss": 6.6138,
      "mean_token_accuracy": 0.1132516548037529,
      "num_tokens": 10999791.0,
      "step": 5080
    },
    {
      "entropy": 6.733290815353394,
      "epoch": 0.5441703675959121,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004978362526257047,
      "loss": 6.6768,
      "mean_token_accuracy": 0.11073072254657745,
      "num_tokens": 11010045.0,
      "step": 5085
    },
    {
      "entropy": 6.755488729476928,
      "epoch": 0.544705441703676,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004978309598084403,
      "loss": 6.5901,
      "mean_token_accuracy": 0.11701643690466881,
      "num_tokens": 11020729.0,
      "step": 5090
    },
    {
      "entropy": 6.685477638244629,
      "epoch": 0.5452405158114398,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004978256605569631,
      "loss": 6.6324,
      "mean_token_accuracy": 0.11502022594213486,
      "num_tokens": 11030641.0,
      "step": 5095
    },
    {
      "entropy": 6.73808445930481,
      "epoch": 0.5457755899192038,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004978203548714262,
      "loss": 6.6225,
      "mean_token_accuracy": 0.11652573719620704,
      "num_tokens": 11041533.0,
      "step": 5100
    },
    {
      "entropy": 6.728613567352295,
      "epoch": 0.5463106640269677,
      "grad_norm": 1.265625,
      "learning_rate": 0.000497815042751983,
      "loss": 6.5831,
      "mean_token_accuracy": 0.11883758381009102,
      "num_tokens": 11052170.0,
      "step": 5105
    },
    {
      "entropy": 6.692430257797241,
      "epoch": 0.5468457381347317,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004978097241987868,
      "loss": 6.5626,
      "mean_token_accuracy": 0.12145586088299751,
      "num_tokens": 11062872.0,
      "step": 5110
    },
    {
      "entropy": 6.696591234207153,
      "epoch": 0.5473808122424956,
      "grad_norm": 1.359375,
      "learning_rate": 0.000497804399211991,
      "loss": 6.651,
      "mean_token_accuracy": 0.11530458331108093,
      "num_tokens": 11074199.0,
      "step": 5115
    },
    {
      "entropy": 6.669123220443725,
      "epoch": 0.5479158863502596,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004977990677917495,
      "loss": 6.5981,
      "mean_token_accuracy": 0.12415442764759063,
      "num_tokens": 11083881.0,
      "step": 5120
    },
    {
      "entropy": 6.628405380249023,
      "epoch": 0.5484509604580234,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004977937299382162,
      "loss": 6.6456,
      "mean_token_accuracy": 0.11557613834738731,
      "num_tokens": 11094374.0,
      "step": 5125
    },
    {
      "entropy": 6.700885438919068,
      "epoch": 0.5489860345657873,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004977883856515453,
      "loss": 6.4988,
      "mean_token_accuracy": 0.12920200750231742,
      "num_tokens": 11105174.0,
      "step": 5130
    },
    {
      "entropy": 6.730715036392212,
      "epoch": 0.5495211086735513,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000497783034931891,
      "loss": 6.6494,
      "mean_token_accuracy": 0.11611341387033462,
      "num_tokens": 11117223.0,
      "step": 5135
    },
    {
      "entropy": 6.715358400344849,
      "epoch": 0.5500561827813152,
      "grad_norm": 1.265625,
      "learning_rate": 0.000497777677779408,
      "loss": 6.5769,
      "mean_token_accuracy": 0.12152805551886559,
      "num_tokens": 11127240.0,
      "step": 5140
    },
    {
      "entropy": 6.691393184661865,
      "epoch": 0.5505912568890792,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004977723141942509,
      "loss": 6.5762,
      "mean_token_accuracy": 0.12875020131468773,
      "num_tokens": 11137287.0,
      "step": 5145
    },
    {
      "entropy": 6.662536859512329,
      "epoch": 0.551126330996843,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004977669441765743,
      "loss": 6.7036,
      "mean_token_accuracy": 0.11726266741752625,
      "num_tokens": 11148988.0,
      "step": 5150
    },
    {
      "entropy": 6.754349040985107,
      "epoch": 0.551661405104607,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004977615677265336,
      "loss": 6.6381,
      "mean_token_accuracy": 0.119845050573349,
      "num_tokens": 11159874.0,
      "step": 5155
    },
    {
      "entropy": 6.7734168529510494,
      "epoch": 0.5521964792123709,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000497756184844284,
      "loss": 6.5946,
      "mean_token_accuracy": 0.1255686677992344,
      "num_tokens": 11170703.0,
      "step": 5160
    },
    {
      "entropy": 6.652176094055176,
      "epoch": 0.5527315533201348,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004977507955299809,
      "loss": 6.486,
      "mean_token_accuracy": 0.12536792755126952,
      "num_tokens": 11182230.0,
      "step": 5165
    },
    {
      "entropy": 6.633540678024292,
      "epoch": 0.5532666274278988,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004977453997837797,
      "loss": 6.5519,
      "mean_token_accuracy": 0.11289865449070931,
      "num_tokens": 11192897.0,
      "step": 5170
    },
    {
      "entropy": 6.667209911346435,
      "epoch": 0.5538017015356627,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004977399976058366,
      "loss": 6.6543,
      "mean_token_accuracy": 0.11275542676448821,
      "num_tokens": 11203942.0,
      "step": 5175
    },
    {
      "entropy": 6.737595891952514,
      "epoch": 0.5543367756434266,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004977345889963072,
      "loss": 6.6878,
      "mean_token_accuracy": 0.1157545953989029,
      "num_tokens": 11214361.0,
      "step": 5180
    },
    {
      "entropy": 6.674620866775513,
      "epoch": 0.5548718497511905,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004977291739553479,
      "loss": 6.5811,
      "mean_token_accuracy": 0.12168481573462486,
      "num_tokens": 11225928.0,
      "step": 5185
    },
    {
      "entropy": 6.668534517288208,
      "epoch": 0.5554069238589545,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004977237524831149,
      "loss": 6.5477,
      "mean_token_accuracy": 0.1131792739033699,
      "num_tokens": 11236660.0,
      "step": 5190
    },
    {
      "entropy": 6.674350214004517,
      "epoch": 0.5559419979667184,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004977183245797649,
      "loss": 6.5658,
      "mean_token_accuracy": 0.12445326372981072,
      "num_tokens": 11247272.0,
      "step": 5195
    },
    {
      "entropy": 6.749352884292603,
      "epoch": 0.5564770720744823,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004977128902454547,
      "loss": 6.6736,
      "mean_token_accuracy": 0.11154333502054214,
      "num_tokens": 11258472.0,
      "step": 5200
    },
    {
      "entropy": 6.7011803150177,
      "epoch": 0.5570121461822463,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004977074494803409,
      "loss": 6.5757,
      "mean_token_accuracy": 0.12596716955304146,
      "num_tokens": 11268344.0,
      "step": 5205
    },
    {
      "entropy": 6.710012674331665,
      "epoch": 0.5575472202900101,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004977020022845809,
      "loss": 6.6642,
      "mean_token_accuracy": 0.1116225503385067,
      "num_tokens": 11279931.0,
      "step": 5210
    },
    {
      "entropy": 6.786263656616211,
      "epoch": 0.5580822943977741,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004976965486583318,
      "loss": 6.619,
      "mean_token_accuracy": 0.11329972892999648,
      "num_tokens": 11290595.0,
      "step": 5215
    },
    {
      "entropy": 6.69334135055542,
      "epoch": 0.558617368505538,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004976910886017513,
      "loss": 6.6231,
      "mean_token_accuracy": 0.11819947510957718,
      "num_tokens": 11300774.0,
      "step": 5220
    },
    {
      "entropy": 6.639355039596557,
      "epoch": 0.559152442613302,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004976856221149969,
      "loss": 6.4984,
      "mean_token_accuracy": 0.12909843847155572,
      "num_tokens": 11310288.0,
      "step": 5225
    },
    {
      "entropy": 6.635180282592773,
      "epoch": 0.5596875167210659,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004976801491982263,
      "loss": 6.6073,
      "mean_token_accuracy": 0.1188632071018219,
      "num_tokens": 11321599.0,
      "step": 5230
    },
    {
      "entropy": 6.753577756881714,
      "epoch": 0.5602225908288297,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004976746698515977,
      "loss": 6.6551,
      "mean_token_accuracy": 0.11815480217337608,
      "num_tokens": 11331383.0,
      "step": 5235
    },
    {
      "entropy": 6.7181065559387205,
      "epoch": 0.5607576649365937,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004976691840752694,
      "loss": 6.6056,
      "mean_token_accuracy": 0.1177519403398037,
      "num_tokens": 11342085.0,
      "step": 5240
    },
    {
      "entropy": 6.724043464660644,
      "epoch": 0.5612927390443576,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004976636918693996,
      "loss": 6.6521,
      "mean_token_accuracy": 0.1124167375266552,
      "num_tokens": 11354061.0,
      "step": 5245
    },
    {
      "entropy": 6.7793741703033445,
      "epoch": 0.5618278131521216,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004976581932341471,
      "loss": 6.5874,
      "mean_token_accuracy": 0.1198588714003563,
      "num_tokens": 11364448.0,
      "step": 5250
    },
    {
      "entropy": 6.764813423156738,
      "epoch": 0.5623628872598855,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004976526881696706,
      "loss": 6.5707,
      "mean_token_accuracy": 0.11909611076116562,
      "num_tokens": 11374838.0,
      "step": 5255
    },
    {
      "entropy": 6.598009920120239,
      "epoch": 0.5628979613676495,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004976471766761288,
      "loss": 6.6143,
      "mean_token_accuracy": 0.12120521515607834,
      "num_tokens": 11385310.0,
      "step": 5260
    },
    {
      "entropy": 6.654865312576294,
      "epoch": 0.5634330354754133,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004976416587536812,
      "loss": 6.6152,
      "mean_token_accuracy": 0.11421113684773446,
      "num_tokens": 11396590.0,
      "step": 5265
    },
    {
      "entropy": 6.7150064468383786,
      "epoch": 0.5639681095831772,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004976361344024871,
      "loss": 6.5634,
      "mean_token_accuracy": 0.11905626356601715,
      "num_tokens": 11407218.0,
      "step": 5270
    },
    {
      "entropy": 6.613826513290405,
      "epoch": 0.5645031836909412,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004976306036227056,
      "loss": 6.5861,
      "mean_token_accuracy": 0.11498884409666062,
      "num_tokens": 11417716.0,
      "step": 5275
    },
    {
      "entropy": 6.735283041000367,
      "epoch": 0.5650382577987051,
      "grad_norm": 1.546875,
      "learning_rate": 0.000497625066414497,
      "loss": 6.6699,
      "mean_token_accuracy": 0.11644284650683404,
      "num_tokens": 11428295.0,
      "step": 5280
    },
    {
      "entropy": 6.76004843711853,
      "epoch": 0.5655733319064691,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004976195227780207,
      "loss": 6.5879,
      "mean_token_accuracy": 0.11506432518363,
      "num_tokens": 11439044.0,
      "step": 5285
    },
    {
      "entropy": 6.726730251312256,
      "epoch": 0.566108406014233,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004976139727134371,
      "loss": 6.5437,
      "mean_token_accuracy": 0.1157782182097435,
      "num_tokens": 11450029.0,
      "step": 5290
    },
    {
      "entropy": 6.680373620986939,
      "epoch": 0.566643480121997,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004976084162209062,
      "loss": 6.7158,
      "mean_token_accuracy": 0.10938669890165328,
      "num_tokens": 11461190.0,
      "step": 5295
    },
    {
      "entropy": 6.693241691589355,
      "epoch": 0.5671785542297608,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004976028533005886,
      "loss": 6.5658,
      "mean_token_accuracy": 0.11855396032333373,
      "num_tokens": 11472143.0,
      "step": 5300
    },
    {
      "entropy": 6.718105125427246,
      "epoch": 0.5677136283375247,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004975972839526449,
      "loss": 6.7015,
      "mean_token_accuracy": 0.10698949694633483,
      "num_tokens": 11483745.0,
      "step": 5305
    },
    {
      "entropy": 6.7316608905792235,
      "epoch": 0.5682487024452887,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004975917081772358,
      "loss": 6.5709,
      "mean_token_accuracy": 0.11810784116387367,
      "num_tokens": 11495695.0,
      "step": 5310
    },
    {
      "entropy": 6.704360294342041,
      "epoch": 0.5687837765530526,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004975861259745225,
      "loss": 6.6202,
      "mean_token_accuracy": 0.11814290434122085,
      "num_tokens": 11505576.0,
      "step": 5315
    },
    {
      "entropy": 6.6637914180755615,
      "epoch": 0.5693188506608166,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004975805373446662,
      "loss": 6.4714,
      "mean_token_accuracy": 0.12663986161351204,
      "num_tokens": 11516614.0,
      "step": 5320
    },
    {
      "entropy": 6.678481960296631,
      "epoch": 0.5698539247685804,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000497574942287828,
      "loss": 6.5993,
      "mean_token_accuracy": 0.11633578389883041,
      "num_tokens": 11528032.0,
      "step": 5325
    },
    {
      "entropy": 6.654755592346191,
      "epoch": 0.5703889988763444,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004975693408041697,
      "loss": 6.6168,
      "mean_token_accuracy": 0.11541731208562851,
      "num_tokens": 11538822.0,
      "step": 5330
    },
    {
      "entropy": 6.677396774291992,
      "epoch": 0.5709240729841083,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004975637328938529,
      "loss": 6.6259,
      "mean_token_accuracy": 0.11113707721233368,
      "num_tokens": 11548895.0,
      "step": 5335
    },
    {
      "entropy": 6.687600326538086,
      "epoch": 0.5714591470918722,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004975581185570398,
      "loss": 6.504,
      "mean_token_accuracy": 0.1173204354941845,
      "num_tokens": 11559613.0,
      "step": 5340
    },
    {
      "entropy": 6.692977571487427,
      "epoch": 0.5719942211996362,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004975524977938921,
      "loss": 6.5501,
      "mean_token_accuracy": 0.12145494073629379,
      "num_tokens": 11571092.0,
      "step": 5345
    },
    {
      "entropy": 6.6996602535247805,
      "epoch": 0.5725292953074,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004975468706045722,
      "loss": 6.6804,
      "mean_token_accuracy": 0.12007683664560317,
      "num_tokens": 11581448.0,
      "step": 5350
    },
    {
      "entropy": 6.734434652328491,
      "epoch": 0.573064369415164,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004975412369892429,
      "loss": 6.5773,
      "mean_token_accuracy": 0.12340361326932907,
      "num_tokens": 11592273.0,
      "step": 5355
    },
    {
      "entropy": 6.719641304016113,
      "epoch": 0.5735994435229279,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004975355969480665,
      "loss": 6.5633,
      "mean_token_accuracy": 0.12459224984049797,
      "num_tokens": 11602759.0,
      "step": 5360
    },
    {
      "entropy": 6.641370964050293,
      "epoch": 0.5741345176306919,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004975299504812061,
      "loss": 6.5977,
      "mean_token_accuracy": 0.11472523659467697,
      "num_tokens": 11613977.0,
      "step": 5365
    },
    {
      "entropy": 6.755290269851685,
      "epoch": 0.5746695917384558,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004975242975888246,
      "loss": 6.6451,
      "mean_token_accuracy": 0.11898882836103439,
      "num_tokens": 11625425.0,
      "step": 5370
    },
    {
      "entropy": 6.667579507827758,
      "epoch": 0.5752046658462197,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004975186382710854,
      "loss": 6.5923,
      "mean_token_accuracy": 0.11569623947143555,
      "num_tokens": 11636530.0,
      "step": 5375
    },
    {
      "entropy": 6.69462890625,
      "epoch": 0.5757397399539836,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004975129725281516,
      "loss": 6.5842,
      "mean_token_accuracy": 0.12393845319747925,
      "num_tokens": 11646516.0,
      "step": 5380
    },
    {
      "entropy": 6.668670988082885,
      "epoch": 0.5762748140617475,
      "grad_norm": 1.625,
      "learning_rate": 0.0004975073003601871,
      "loss": 6.579,
      "mean_token_accuracy": 0.1199287973344326,
      "num_tokens": 11657424.0,
      "step": 5385
    },
    {
      "entropy": 6.673134517669678,
      "epoch": 0.5768098881695115,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004975016217673556,
      "loss": 6.5883,
      "mean_token_accuracy": 0.11663650795817375,
      "num_tokens": 11669447.0,
      "step": 5390
    },
    {
      "entropy": 6.716589069366455,
      "epoch": 0.5773449622772754,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000497495936749821,
      "loss": 6.5611,
      "mean_token_accuracy": 0.12475978508591652,
      "num_tokens": 11679612.0,
      "step": 5395
    },
    {
      "entropy": 6.678430557250977,
      "epoch": 0.5778800363850394,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004974902453077474,
      "loss": 6.5548,
      "mean_token_accuracy": 0.12052254527807235,
      "num_tokens": 11689527.0,
      "step": 5400
    },
    {
      "entropy": 6.720280361175537,
      "epoch": 0.5784151104928033,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004974845474412993,
      "loss": 6.5765,
      "mean_token_accuracy": 0.12422782853245735,
      "num_tokens": 11700452.0,
      "step": 5405
    },
    {
      "entropy": 6.608441305160523,
      "epoch": 0.5789501846005671,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004974788431506412,
      "loss": 6.5868,
      "mean_token_accuracy": 0.11873762682080269,
      "num_tokens": 11710902.0,
      "step": 5410
    },
    {
      "entropy": 6.722711229324341,
      "epoch": 0.5794852587083311,
      "grad_norm": 2.375,
      "learning_rate": 0.0004974731324359378,
      "loss": 6.6202,
      "mean_token_accuracy": 0.11096708253026008,
      "num_tokens": 11721901.0,
      "step": 5415
    },
    {
      "entropy": 6.726794672012329,
      "epoch": 0.580020332816095,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004974674152973538,
      "loss": 6.6079,
      "mean_token_accuracy": 0.115871462225914,
      "num_tokens": 11733622.0,
      "step": 5420
    },
    {
      "entropy": 6.751911401748657,
      "epoch": 0.580555406923859,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004974616917350545,
      "loss": 6.6164,
      "mean_token_accuracy": 0.12217370122671127,
      "num_tokens": 11745116.0,
      "step": 5425
    },
    {
      "entropy": 6.629001712799072,
      "epoch": 0.5810904810316229,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004974559617492052,
      "loss": 6.5419,
      "mean_token_accuracy": 0.12381923273205757,
      "num_tokens": 11756203.0,
      "step": 5430
    },
    {
      "entropy": 6.742204999923706,
      "epoch": 0.5816255551393869,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004974502253399712,
      "loss": 6.6065,
      "mean_token_accuracy": 0.11900223791599274,
      "num_tokens": 11766797.0,
      "step": 5435
    },
    {
      "entropy": 6.743202018737793,
      "epoch": 0.5821606292471507,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004974444825075182,
      "loss": 6.6836,
      "mean_token_accuracy": 0.11354267820715905,
      "num_tokens": 11778110.0,
      "step": 5440
    },
    {
      "entropy": 6.611339521408081,
      "epoch": 0.5826957033549146,
      "grad_norm": 1.296875,
      "learning_rate": 0.000497438733252012,
      "loss": 6.5464,
      "mean_token_accuracy": 0.1208656445145607,
      "num_tokens": 11788396.0,
      "step": 5445
    },
    {
      "entropy": 6.740115785598755,
      "epoch": 0.5832307774626786,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004974329775736188,
      "loss": 6.6253,
      "mean_token_accuracy": 0.11759175732731819,
      "num_tokens": 11798691.0,
      "step": 5450
    },
    {
      "entropy": 6.743464374542237,
      "epoch": 0.5837658515704425,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004974272154725044,
      "loss": 6.6862,
      "mean_token_accuracy": 0.11484382227063179,
      "num_tokens": 11808965.0,
      "step": 5455
    },
    {
      "entropy": 6.656184816360474,
      "epoch": 0.5843009256782065,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004974214469488355,
      "loss": 6.5029,
      "mean_token_accuracy": 0.12686796858906746,
      "num_tokens": 11819932.0,
      "step": 5460
    },
    {
      "entropy": 6.753126621246338,
      "epoch": 0.5848359997859703,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004974156720027786,
      "loss": 6.6071,
      "mean_token_accuracy": 0.12075452283024787,
      "num_tokens": 11831332.0,
      "step": 5465
    },
    {
      "entropy": 6.719524621963501,
      "epoch": 0.5853710738937343,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004974098906345004,
      "loss": 6.5902,
      "mean_token_accuracy": 0.11421620100736618,
      "num_tokens": 11840947.0,
      "step": 5470
    },
    {
      "entropy": 6.624907207489014,
      "epoch": 0.5859061480014982,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004974041028441679,
      "loss": 6.5378,
      "mean_token_accuracy": 0.11265920177102089,
      "num_tokens": 11851640.0,
      "step": 5475
    },
    {
      "entropy": 6.7370397567749025,
      "epoch": 0.5864412221092621,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004973983086319481,
      "loss": 6.6519,
      "mean_token_accuracy": 0.11276995167136192,
      "num_tokens": 11863608.0,
      "step": 5480
    },
    {
      "entropy": 6.752982711791992,
      "epoch": 0.5869762962170261,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004973925079980085,
      "loss": 6.5678,
      "mean_token_accuracy": 0.11900977268815041,
      "num_tokens": 11874707.0,
      "step": 5485
    },
    {
      "entropy": 6.644384241104126,
      "epoch": 0.58751137032479,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004973867009425164,
      "loss": 6.6379,
      "mean_token_accuracy": 0.11783983260393142,
      "num_tokens": 11885118.0,
      "step": 5490
    },
    {
      "entropy": 6.724728775024414,
      "epoch": 0.5880464444325539,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004973808874656396,
      "loss": 6.6162,
      "mean_token_accuracy": 0.11670980975031853,
      "num_tokens": 11895859.0,
      "step": 5495
    },
    {
      "entropy": 6.7954737663269045,
      "epoch": 0.5885815185403178,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004973750675675458,
      "loss": 6.6342,
      "mean_token_accuracy": 0.11875612363219261,
      "num_tokens": 11907019.0,
      "step": 5500
    },
    {
      "entropy": 6.744963645935059,
      "epoch": 0.5891165926480818,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004973692412484034,
      "loss": 6.5778,
      "mean_token_accuracy": 0.11970652416348457,
      "num_tokens": 11918257.0,
      "step": 5505
    },
    {
      "entropy": 6.671480941772461,
      "epoch": 0.5896516667558457,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004973634085083802,
      "loss": 6.576,
      "mean_token_accuracy": 0.127632237970829,
      "num_tokens": 11929003.0,
      "step": 5510
    },
    {
      "entropy": 6.736668920516967,
      "epoch": 0.5901867408636096,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000497357569347645,
      "loss": 6.7255,
      "mean_token_accuracy": 0.1138952411711216,
      "num_tokens": 11939786.0,
      "step": 5515
    },
    {
      "entropy": 6.755863952636719,
      "epoch": 0.5907218149713735,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004973517237663661,
      "loss": 6.6304,
      "mean_token_accuracy": 0.11138334199786186,
      "num_tokens": 11951042.0,
      "step": 5520
    },
    {
      "entropy": 6.740656137466431,
      "epoch": 0.5912568890791374,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004973458717647124,
      "loss": 6.5364,
      "mean_token_accuracy": 0.12641868591308594,
      "num_tokens": 11961982.0,
      "step": 5525
    },
    {
      "entropy": 6.707173919677734,
      "epoch": 0.5917919631869014,
      "grad_norm": 1.5625,
      "learning_rate": 0.000497340013342853,
      "loss": 6.5856,
      "mean_token_accuracy": 0.11962274238467216,
      "num_tokens": 11972078.0,
      "step": 5530
    },
    {
      "entropy": 6.688304281234741,
      "epoch": 0.5923270372946653,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000497334148500957,
      "loss": 6.638,
      "mean_token_accuracy": 0.11932587698101997,
      "num_tokens": 11984037.0,
      "step": 5535
    },
    {
      "entropy": 6.709690189361572,
      "epoch": 0.5928621114024293,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004973282772391936,
      "loss": 6.5695,
      "mean_token_accuracy": 0.1173762895166874,
      "num_tokens": 11993688.0,
      "step": 5540
    },
    {
      "entropy": 6.620822906494141,
      "epoch": 0.5933971855101932,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004973223995577324,
      "loss": 6.5427,
      "mean_token_accuracy": 0.11914530023932457,
      "num_tokens": 12003759.0,
      "step": 5545
    },
    {
      "entropy": 6.650865459442139,
      "epoch": 0.593932259617957,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004973165154567432,
      "loss": 6.5773,
      "mean_token_accuracy": 0.1205617792904377,
      "num_tokens": 12015030.0,
      "step": 5550
    },
    {
      "entropy": 6.700285911560059,
      "epoch": 0.594467333725721,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004973106249363959,
      "loss": 6.6351,
      "mean_token_accuracy": 0.11812169402837754,
      "num_tokens": 12026181.0,
      "step": 5555
    },
    {
      "entropy": 6.653229570388794,
      "epoch": 0.5950024078334849,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004973047279968606,
      "loss": 6.4189,
      "mean_token_accuracy": 0.1311015099287033,
      "num_tokens": 12037336.0,
      "step": 5560
    },
    {
      "entropy": 6.671442556381225,
      "epoch": 0.5955374819412489,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004972988246383074,
      "loss": 6.4958,
      "mean_token_accuracy": 0.120445104688406,
      "num_tokens": 12048425.0,
      "step": 5565
    },
    {
      "entropy": 6.553933811187744,
      "epoch": 0.5960725560490128,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004972929148609068,
      "loss": 6.4506,
      "mean_token_accuracy": 0.13666753321886063,
      "num_tokens": 12059814.0,
      "step": 5570
    },
    {
      "entropy": 6.609471130371094,
      "epoch": 0.5966076301567768,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004972869986648296,
      "loss": 6.5223,
      "mean_token_accuracy": 0.12330262139439582,
      "num_tokens": 12069377.0,
      "step": 5575
    },
    {
      "entropy": 6.679553031921387,
      "epoch": 0.5971427042645406,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004972810760502467,
      "loss": 6.65,
      "mean_token_accuracy": 0.1143650807440281,
      "num_tokens": 12080277.0,
      "step": 5580
    },
    {
      "entropy": 6.707303857803344,
      "epoch": 0.5976777783723045,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004972751470173287,
      "loss": 6.5392,
      "mean_token_accuracy": 0.11781578734517098,
      "num_tokens": 12091375.0,
      "step": 5585
    },
    {
      "entropy": 6.6600456714630125,
      "epoch": 0.5982128524800685,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004972692115662473,
      "loss": 6.5835,
      "mean_token_accuracy": 0.11931180357933044,
      "num_tokens": 12103110.0,
      "step": 5590
    },
    {
      "entropy": 6.669634294509888,
      "epoch": 0.5987479265878324,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004972632696971735,
      "loss": 6.6412,
      "mean_token_accuracy": 0.11558156907558441,
      "num_tokens": 12112454.0,
      "step": 5595
    },
    {
      "entropy": 6.636322975158691,
      "epoch": 0.5992830006955964,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004972573214102791,
      "loss": 6.5229,
      "mean_token_accuracy": 0.11954844370484352,
      "num_tokens": 12125140.0,
      "step": 5600
    },
    {
      "entropy": 6.715576267242431,
      "epoch": 0.5998180748033602,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004972513667057358,
      "loss": 6.5519,
      "mean_token_accuracy": 0.11657773405313492,
      "num_tokens": 12136061.0,
      "step": 5605
    },
    {
      "entropy": 6.750255346298218,
      "epoch": 0.6003531489111242,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004972454055837155,
      "loss": 6.595,
      "mean_token_accuracy": 0.12111319378018379,
      "num_tokens": 12146654.0,
      "step": 5610
    },
    {
      "entropy": 6.75370192527771,
      "epoch": 0.6008882230188881,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004972394380443903,
      "loss": 6.5841,
      "mean_token_accuracy": 0.11777229756116867,
      "num_tokens": 12157881.0,
      "step": 5615
    },
    {
      "entropy": 6.691373348236084,
      "epoch": 0.601423297126652,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004972334640879325,
      "loss": 6.615,
      "mean_token_accuracy": 0.11591439619660378,
      "num_tokens": 12168534.0,
      "step": 5620
    },
    {
      "entropy": 6.640426683425903,
      "epoch": 0.601958371234416,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004972274837145147,
      "loss": 6.5329,
      "mean_token_accuracy": 0.1218352235853672,
      "num_tokens": 12178773.0,
      "step": 5625
    },
    {
      "entropy": 6.7092021942138675,
      "epoch": 0.6024934453421799,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004972214969243096,
      "loss": 6.5772,
      "mean_token_accuracy": 0.11726051643490791,
      "num_tokens": 12188819.0,
      "step": 5630
    },
    {
      "entropy": 6.635117769241333,
      "epoch": 0.6030285194499438,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00049721550371749,
      "loss": 6.5502,
      "mean_token_accuracy": 0.12739348784089088,
      "num_tokens": 12199198.0,
      "step": 5635
    },
    {
      "entropy": 6.721414947509766,
      "epoch": 0.6035635935577077,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004972095040942289,
      "loss": 6.5697,
      "mean_token_accuracy": 0.11852023229002953,
      "num_tokens": 12210765.0,
      "step": 5640
    },
    {
      "entropy": 6.656217288970947,
      "epoch": 0.6040986676654717,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004972034980546996,
      "loss": 6.5941,
      "mean_token_accuracy": 0.11481459364295006,
      "num_tokens": 12221067.0,
      "step": 5645
    },
    {
      "entropy": 6.645788288116455,
      "epoch": 0.6046337417732356,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004971974855990755,
      "loss": 6.5999,
      "mean_token_accuracy": 0.12012394964694977,
      "num_tokens": 12232994.0,
      "step": 5650
    },
    {
      "entropy": 6.729011297225952,
      "epoch": 0.6051688158809995,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004971914667275302,
      "loss": 6.5553,
      "mean_token_accuracy": 0.11960238218307495,
      "num_tokens": 12243327.0,
      "step": 5655
    },
    {
      "entropy": 6.600992822647095,
      "epoch": 0.6057038899887635,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004971854414402377,
      "loss": 6.4803,
      "mean_token_accuracy": 0.12356549277901649,
      "num_tokens": 12254249.0,
      "step": 5660
    },
    {
      "entropy": 6.654879474639893,
      "epoch": 0.6062389640965273,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004971794097373716,
      "loss": 6.5167,
      "mean_token_accuracy": 0.11952641755342483,
      "num_tokens": 12264487.0,
      "step": 5665
    },
    {
      "entropy": 6.57205753326416,
      "epoch": 0.6067740382042913,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004971733716191063,
      "loss": 6.5087,
      "mean_token_accuracy": 0.12777614444494248,
      "num_tokens": 12274801.0,
      "step": 5670
    },
    {
      "entropy": 6.634649610519409,
      "epoch": 0.6073091123120552,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004971673270856162,
      "loss": 6.477,
      "mean_token_accuracy": 0.12111672386527061,
      "num_tokens": 12284647.0,
      "step": 5675
    },
    {
      "entropy": 6.6497354984283445,
      "epoch": 0.6078441864198192,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004971612761370756,
      "loss": 6.5947,
      "mean_token_accuracy": 0.12196626737713814,
      "num_tokens": 12295215.0,
      "step": 5680
    },
    {
      "entropy": 6.636915922164917,
      "epoch": 0.6083792605275831,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004971552187736596,
      "loss": 6.503,
      "mean_token_accuracy": 0.12083340361714363,
      "num_tokens": 12305324.0,
      "step": 5685
    },
    {
      "entropy": 6.623749685287476,
      "epoch": 0.6089143346353469,
      "grad_norm": 1.625,
      "learning_rate": 0.0004971491549955427,
      "loss": 6.5314,
      "mean_token_accuracy": 0.12515415251255035,
      "num_tokens": 12316399.0,
      "step": 5690
    },
    {
      "entropy": 6.565249395370484,
      "epoch": 0.6094494087431109,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004971430848029002,
      "loss": 6.4953,
      "mean_token_accuracy": 0.12863782942295074,
      "num_tokens": 12328033.0,
      "step": 5695
    },
    {
      "entropy": 6.62008113861084,
      "epoch": 0.6099844828508748,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004971370081959073,
      "loss": 6.5432,
      "mean_token_accuracy": 0.12434352040290833,
      "num_tokens": 12338962.0,
      "step": 5700
    },
    {
      "entropy": 6.6729292392730715,
      "epoch": 0.6105195569586388,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004971309251747396,
      "loss": 6.5418,
      "mean_token_accuracy": 0.1266559824347496,
      "num_tokens": 12350216.0,
      "step": 5705
    },
    {
      "entropy": 6.681897783279419,
      "epoch": 0.6110546310664027,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004971248357395726,
      "loss": 6.5708,
      "mean_token_accuracy": 0.1206928163766861,
      "num_tokens": 12361277.0,
      "step": 5710
    },
    {
      "entropy": 6.585869026184082,
      "epoch": 0.6115897051741667,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004971187398905821,
      "loss": 6.4779,
      "mean_token_accuracy": 0.12453080341219902,
      "num_tokens": 12373308.0,
      "step": 5715
    },
    {
      "entropy": 6.745548439025879,
      "epoch": 0.6121247792819305,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004971126376279443,
      "loss": 6.7296,
      "mean_token_accuracy": 0.10965547412633896,
      "num_tokens": 12383943.0,
      "step": 5720
    },
    {
      "entropy": 6.67027587890625,
      "epoch": 0.6126598533896944,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004971065289518354,
      "loss": 6.4776,
      "mean_token_accuracy": 0.1167394459247589,
      "num_tokens": 12394575.0,
      "step": 5725
    },
    {
      "entropy": 6.698582744598388,
      "epoch": 0.6131949274974584,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004971004138624315,
      "loss": 6.5898,
      "mean_token_accuracy": 0.11946821883320809,
      "num_tokens": 12405536.0,
      "step": 5730
    },
    {
      "entropy": 6.607072544097901,
      "epoch": 0.6137300016052223,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004970942923599094,
      "loss": 6.4685,
      "mean_token_accuracy": 0.12893888279795646,
      "num_tokens": 12416148.0,
      "step": 5735
    },
    {
      "entropy": 6.632549238204956,
      "epoch": 0.6142650757129863,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004970881644444459,
      "loss": 6.4754,
      "mean_token_accuracy": 0.12886900901794435,
      "num_tokens": 12426182.0,
      "step": 5740
    },
    {
      "entropy": 6.631240129470825,
      "epoch": 0.6148001498207502,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004970820301162178,
      "loss": 6.652,
      "mean_token_accuracy": 0.11639525294303894,
      "num_tokens": 12437548.0,
      "step": 5745
    },
    {
      "entropy": 6.722517490386963,
      "epoch": 0.6153352239285141,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004970758893754022,
      "loss": 6.603,
      "mean_token_accuracy": 0.11473475024104118,
      "num_tokens": 12449720.0,
      "step": 5750
    },
    {
      "entropy": 6.750303125381469,
      "epoch": 0.615870298036278,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004970697422221767,
      "loss": 6.5825,
      "mean_token_accuracy": 0.12193588614463806,
      "num_tokens": 12460666.0,
      "step": 5755
    },
    {
      "entropy": 6.661549139022827,
      "epoch": 0.6164053721440419,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004970635886567185,
      "loss": 6.5831,
      "mean_token_accuracy": 0.11829188615083694,
      "num_tokens": 12471472.0,
      "step": 5760
    },
    {
      "entropy": 6.616854095458985,
      "epoch": 0.6169404462518059,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004970574286792054,
      "loss": 6.5729,
      "mean_token_accuracy": 0.11929405927658081,
      "num_tokens": 12482127.0,
      "step": 5765
    },
    {
      "entropy": 6.641938018798828,
      "epoch": 0.6174755203595698,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004970512622898152,
      "loss": 6.5611,
      "mean_token_accuracy": 0.12291084676980972,
      "num_tokens": 12493680.0,
      "step": 5770
    },
    {
      "entropy": 6.60687289237976,
      "epoch": 0.6180105944673338,
      "grad_norm": 1.65625,
      "learning_rate": 0.000497045089488726,
      "loss": 6.5172,
      "mean_token_accuracy": 0.1259467586874962,
      "num_tokens": 12505838.0,
      "step": 5775
    },
    {
      "entropy": 6.6949209690094,
      "epoch": 0.6185456685750976,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004970389102761161,
      "loss": 6.5121,
      "mean_token_accuracy": 0.12690744772553444,
      "num_tokens": 12516334.0,
      "step": 5780
    },
    {
      "entropy": 6.630453395843506,
      "epoch": 0.6190807426828616,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004970327246521638,
      "loss": 6.5966,
      "mean_token_accuracy": 0.11633171513676643,
      "num_tokens": 12526440.0,
      "step": 5785
    },
    {
      "entropy": 6.615752553939819,
      "epoch": 0.6196158167906255,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004970265326170478,
      "loss": 6.504,
      "mean_token_accuracy": 0.12056778073310852,
      "num_tokens": 12538127.0,
      "step": 5790
    },
    {
      "entropy": 6.725585842132569,
      "epoch": 0.6201508908983894,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004970203341709469,
      "loss": 6.6113,
      "mean_token_accuracy": 0.11749014109373093,
      "num_tokens": 12548459.0,
      "step": 5795
    },
    {
      "entropy": 6.602406692504883,
      "epoch": 0.6206859650061534,
      "grad_norm": 1.34375,
      "learning_rate": 0.00049701412931404,
      "loss": 6.5617,
      "mean_token_accuracy": 0.12014793530106545,
      "num_tokens": 12558989.0,
      "step": 5800
    },
    {
      "entropy": 6.618525218963623,
      "epoch": 0.6212210391139172,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004970079180465064,
      "loss": 6.4752,
      "mean_token_accuracy": 0.12559444829821587,
      "num_tokens": 12569449.0,
      "step": 5805
    },
    {
      "entropy": 6.649777793884278,
      "epoch": 0.6217561132216812,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004970017003685253,
      "loss": 6.5174,
      "mean_token_accuracy": 0.12333801165223121,
      "num_tokens": 12580733.0,
      "step": 5810
    },
    {
      "entropy": 6.639023208618164,
      "epoch": 0.6222911873294451,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004969954762802763,
      "loss": 6.5246,
      "mean_token_accuracy": 0.12096468210220337,
      "num_tokens": 12590869.0,
      "step": 5815
    },
    {
      "entropy": 6.714621353149414,
      "epoch": 0.6228262614372091,
      "grad_norm": 1.546875,
      "learning_rate": 0.000496989245781939,
      "loss": 6.6116,
      "mean_token_accuracy": 0.11883088201284409,
      "num_tokens": 12602109.0,
      "step": 5820
    },
    {
      "entropy": 6.678238773345948,
      "epoch": 0.623361335544973,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004969830088736937,
      "loss": 6.5447,
      "mean_token_accuracy": 0.11637340411543846,
      "num_tokens": 12612234.0,
      "step": 5825
    },
    {
      "entropy": 6.74876914024353,
      "epoch": 0.6238964096527368,
      "grad_norm": 1.40625,
      "learning_rate": 0.00049697676555572,
      "loss": 6.6934,
      "mean_token_accuracy": 0.11071206331253051,
      "num_tokens": 12623252.0,
      "step": 5830
    },
    {
      "entropy": 6.725121641159058,
      "epoch": 0.6244314837605008,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004969705158281985,
      "loss": 6.4462,
      "mean_token_accuracy": 0.12263509407639503,
      "num_tokens": 12632796.0,
      "step": 5835
    },
    {
      "entropy": 6.633543968200684,
      "epoch": 0.6249665578682647,
      "grad_norm": 1.25,
      "learning_rate": 0.0004969642596913095,
      "loss": 6.5394,
      "mean_token_accuracy": 0.11995783001184464,
      "num_tokens": 12643278.0,
      "step": 5840
    },
    {
      "entropy": 6.672138452529907,
      "epoch": 0.6255016319760287,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004969579971452337,
      "loss": 6.6234,
      "mean_token_accuracy": 0.12050508707761765,
      "num_tokens": 12654446.0,
      "step": 5845
    },
    {
      "entropy": 6.699638795852661,
      "epoch": 0.6260367060837926,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004969517281901519,
      "loss": 6.5899,
      "mean_token_accuracy": 0.12274675741791725,
      "num_tokens": 12664281.0,
      "step": 5850
    },
    {
      "entropy": 6.671421384811401,
      "epoch": 0.6265717801915566,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004969454528262451,
      "loss": 6.5693,
      "mean_token_accuracy": 0.11972746700048446,
      "num_tokens": 12675252.0,
      "step": 5855
    },
    {
      "entropy": 6.648418712615967,
      "epoch": 0.6271068542993204,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004969391710536946,
      "loss": 6.5626,
      "mean_token_accuracy": 0.11940335035324097,
      "num_tokens": 12685764.0,
      "step": 5860
    },
    {
      "entropy": 6.590697622299194,
      "epoch": 0.6276419284070843,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004969328828726817,
      "loss": 6.4542,
      "mean_token_accuracy": 0.12306086346507072,
      "num_tokens": 12696974.0,
      "step": 5865
    },
    {
      "entropy": 6.753077173233033,
      "epoch": 0.6281770025148483,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004969265882833879,
      "loss": 6.6187,
      "mean_token_accuracy": 0.11165008768439293,
      "num_tokens": 12708217.0,
      "step": 5870
    },
    {
      "entropy": 6.6730287075042725,
      "epoch": 0.6287120766226122,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004969202872859952,
      "loss": 6.4866,
      "mean_token_accuracy": 0.12286026254296303,
      "num_tokens": 12717840.0,
      "step": 5875
    },
    {
      "entropy": 6.617942667007446,
      "epoch": 0.6292471507303762,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004969139798806854,
      "loss": 6.5918,
      "mean_token_accuracy": 0.11480392217636108,
      "num_tokens": 12728390.0,
      "step": 5880
    },
    {
      "entropy": 6.630872583389282,
      "epoch": 0.6297822248381401,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004969076660676405,
      "loss": 6.5987,
      "mean_token_accuracy": 0.12041830644011497,
      "num_tokens": 12738650.0,
      "step": 5885
    },
    {
      "entropy": 6.697933101654053,
      "epoch": 0.630317298945904,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000496901345847043,
      "loss": 6.591,
      "mean_token_accuracy": 0.11938764452934265,
      "num_tokens": 12748989.0,
      "step": 5890
    },
    {
      "entropy": 6.668829250335693,
      "epoch": 0.6308523730536679,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004968950192190752,
      "loss": 6.5492,
      "mean_token_accuracy": 0.12053480222821236,
      "num_tokens": 12761132.0,
      "step": 5895
    },
    {
      "entropy": 6.63626651763916,
      "epoch": 0.6313874471614318,
      "grad_norm": 1.375,
      "learning_rate": 0.00049688868618392,
      "loss": 6.5544,
      "mean_token_accuracy": 0.12361529469490051,
      "num_tokens": 12771478.0,
      "step": 5900
    },
    {
      "entropy": 6.671739768981934,
      "epoch": 0.6319225212691958,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004968823467417602,
      "loss": 6.6274,
      "mean_token_accuracy": 0.11821751520037652,
      "num_tokens": 12782096.0,
      "step": 5905
    },
    {
      "entropy": 6.6624290466308596,
      "epoch": 0.6324575953769597,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004968760008927788,
      "loss": 6.5292,
      "mean_token_accuracy": 0.12572447210550308,
      "num_tokens": 12792467.0,
      "step": 5910
    },
    {
      "entropy": 6.674447631835937,
      "epoch": 0.6329926694847237,
      "grad_norm": 1.40625,
      "learning_rate": 0.000496869648637159,
      "loss": 6.5279,
      "mean_token_accuracy": 0.12715904042124748,
      "num_tokens": 12803248.0,
      "step": 5915
    },
    {
      "entropy": 6.694916152954102,
      "epoch": 0.6335277435924875,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004968632899750844,
      "loss": 6.5523,
      "mean_token_accuracy": 0.12229961976408958,
      "num_tokens": 12813701.0,
      "step": 5920
    },
    {
      "entropy": 6.736898612976074,
      "epoch": 0.6340628177002515,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004968569249067383,
      "loss": 6.6224,
      "mean_token_accuracy": 0.11488275229930878,
      "num_tokens": 12824945.0,
      "step": 5925
    },
    {
      "entropy": 6.657941913604736,
      "epoch": 0.6345978918080154,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004968505534323047,
      "loss": 6.5681,
      "mean_token_accuracy": 0.11723481565713882,
      "num_tokens": 12836699.0,
      "step": 5930
    },
    {
      "entropy": 6.6175376892089846,
      "epoch": 0.6351329659157794,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004968441755519677,
      "loss": 6.5205,
      "mean_token_accuracy": 0.11750935465097427,
      "num_tokens": 12846905.0,
      "step": 5935
    },
    {
      "entropy": 6.667795610427857,
      "epoch": 0.6356680400235433,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004968377912659112,
      "loss": 6.4846,
      "mean_token_accuracy": 0.12737148776650428,
      "num_tokens": 12856919.0,
      "step": 5940
    },
    {
      "entropy": 6.586252641677857,
      "epoch": 0.6362031141313071,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004968314005743196,
      "loss": 6.4499,
      "mean_token_accuracy": 0.12662240564823152,
      "num_tokens": 12867863.0,
      "step": 5945
    },
    {
      "entropy": 6.576164484024048,
      "epoch": 0.6367381882390711,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004968250034773776,
      "loss": 6.5458,
      "mean_token_accuracy": 0.12098130509257317,
      "num_tokens": 12877877.0,
      "step": 5950
    },
    {
      "entropy": 6.740300512313842,
      "epoch": 0.637273262346835,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004968185999752697,
      "loss": 6.6226,
      "mean_token_accuracy": 0.11211201325058937,
      "num_tokens": 12887954.0,
      "step": 5955
    },
    {
      "entropy": 6.667530632019043,
      "epoch": 0.637808336454599,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004968121900681809,
      "loss": 6.5458,
      "mean_token_accuracy": 0.1223320372402668,
      "num_tokens": 12897744.0,
      "step": 5960
    },
    {
      "entropy": 6.6086818218231205,
      "epoch": 0.6383434105623629,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004968057737562964,
      "loss": 6.6132,
      "mean_token_accuracy": 0.11974837332963943,
      "num_tokens": 12909193.0,
      "step": 5965
    },
    {
      "entropy": 6.65242075920105,
      "epoch": 0.6388784846701269,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004967993510398012,
      "loss": 6.4832,
      "mean_token_accuracy": 0.1264573447406292,
      "num_tokens": 12919879.0,
      "step": 5970
    },
    {
      "entropy": 6.632117652893067,
      "epoch": 0.6394135587778907,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004967929219188809,
      "loss": 6.5361,
      "mean_token_accuracy": 0.12129070311784744,
      "num_tokens": 12930297.0,
      "step": 5975
    },
    {
      "entropy": 6.657332706451416,
      "epoch": 0.6399486328856546,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004967864863937212,
      "loss": 6.5143,
      "mean_token_accuracy": 0.12851046174764633,
      "num_tokens": 12940467.0,
      "step": 5980
    },
    {
      "entropy": 6.6095727443695065,
      "epoch": 0.6404837069934186,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004967800444645079,
      "loss": 6.4992,
      "mean_token_accuracy": 0.11969673112034798,
      "num_tokens": 12950417.0,
      "step": 5985
    },
    {
      "entropy": 6.600736427307129,
      "epoch": 0.6410187811011825,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004967735961314269,
      "loss": 6.461,
      "mean_token_accuracy": 0.12727241218090057,
      "num_tokens": 12960957.0,
      "step": 5990
    },
    {
      "entropy": 6.65483922958374,
      "epoch": 0.6415538552089465,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004967671413946646,
      "loss": 6.5729,
      "mean_token_accuracy": 0.11923076584935188,
      "num_tokens": 12971194.0,
      "step": 5995
    },
    {
      "entropy": 6.6494788646698,
      "epoch": 0.6420889293167104,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004967606802544071,
      "loss": 6.5702,
      "mean_token_accuracy": 0.11753485649824143,
      "num_tokens": 12982479.0,
      "step": 6000
    },
    {
      "epoch": 0.6420889293167104,
      "eval_entropy": 6.531758780877866,
      "eval_loss": 6.6282525062561035,
      "eval_mean_token_accuracy": 0.12331635652315825,
      "eval_num_tokens": 12982479.0,
      "eval_runtime": 22.5385,
      "eval_samples_per_second": 1316.811,
      "eval_steps_per_second": 164.607,
      "step": 6000
    }
  ],
  "logging_steps": 5,
  "max_steps": 93440,
  "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": 1.900850540544e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}