{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5983724269985639,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.14264454692602158,
      "epoch": 0.0023934897079942556,
      "grad_norm": 6.1329498291015625,
      "learning_rate": 1.9985639061752036e-05,
      "loss": 8.6363,
      "mean_token_accuracy": 0.6116024732589722,
      "num_tokens": 77533.0,
      "step": 10
    },
    {
      "entropy": 0.13887472450733185,
      "epoch": 0.004786979415988511,
      "grad_norm": 11.192473411560059,
      "learning_rate": 1.996968246369874e-05,
      "loss": 7.3158,
      "mean_token_accuracy": 0.6413674414157867,
      "num_tokens": 163156.0,
      "step": 20
    },
    {
      "entropy": 0.1928814619779587,
      "epoch": 0.007180469123982767,
      "grad_norm": 14.501362800598145,
      "learning_rate": 1.9953725865645446e-05,
      "loss": 5.8909,
      "mean_token_accuracy": 0.6312316954135895,
      "num_tokens": 247959.0,
      "step": 30
    },
    {
      "entropy": 0.2716422751545906,
      "epoch": 0.009573958831977022,
      "grad_norm": 9.751401901245117,
      "learning_rate": 1.993776926759215e-05,
      "loss": 4.1374,
      "mean_token_accuracy": 0.6447562038898468,
      "num_tokens": 336213.0,
      "step": 40
    },
    {
      "entropy": 0.40589025914669036,
      "epoch": 0.011967448539971278,
      "grad_norm": 5.346670627593994,
      "learning_rate": 1.9921812669538856e-05,
      "loss": 3.0977,
      "mean_token_accuracy": 0.6462277293205261,
      "num_tokens": 418841.0,
      "step": 50
    },
    {
      "entropy": 0.6163419902324676,
      "epoch": 0.014360938247965534,
      "grad_norm": 2.4319427013397217,
      "learning_rate": 1.990585607148556e-05,
      "loss": 2.5241,
      "mean_token_accuracy": 0.6265762448310852,
      "num_tokens": 498137.0,
      "step": 60
    },
    {
      "entropy": 0.8075001418590546,
      "epoch": 0.01675442795595979,
      "grad_norm": 1.5138881206512451,
      "learning_rate": 1.9889899473432266e-05,
      "loss": 2.3184,
      "mean_token_accuracy": 0.6205335795879364,
      "num_tokens": 572657.0,
      "step": 70
    },
    {
      "entropy": 0.9711024582386016,
      "epoch": 0.019147917663954045,
      "grad_norm": 0.987091064453125,
      "learning_rate": 1.987394287537897e-05,
      "loss": 2.0481,
      "mean_token_accuracy": 0.6290253281593323,
      "num_tokens": 658058.0,
      "step": 80
    },
    {
      "entropy": 1.1067683815956115,
      "epoch": 0.0215414073719483,
      "grad_norm": 0.8901429176330566,
      "learning_rate": 1.9857986277325675e-05,
      "loss": 1.8676,
      "mean_token_accuracy": 0.6413214862346649,
      "num_tokens": 738189.0,
      "step": 90
    },
    {
      "entropy": 1.2711581468582154,
      "epoch": 0.023934897079942556,
      "grad_norm": 0.639988362789154,
      "learning_rate": 1.984202967927238e-05,
      "loss": 1.7424,
      "mean_token_accuracy": 0.6481096923351288,
      "num_tokens": 820981.0,
      "step": 100
    },
    {
      "entropy": 1.3605278491973878,
      "epoch": 0.026328386787936812,
      "grad_norm": 0.4999433755874634,
      "learning_rate": 1.9826073081219085e-05,
      "loss": 1.6153,
      "mean_token_accuracy": 0.6676276981830597,
      "num_tokens": 900515.0,
      "step": 110
    },
    {
      "entropy": 1.398988461494446,
      "epoch": 0.028721876495931067,
      "grad_norm": 0.5294203758239746,
      "learning_rate": 1.981011648316579e-05,
      "loss": 1.5254,
      "mean_token_accuracy": 0.6777714610099792,
      "num_tokens": 982502.0,
      "step": 120
    },
    {
      "entropy": 1.317799985408783,
      "epoch": 0.031115366203925323,
      "grad_norm": 0.4618872404098511,
      "learning_rate": 1.9794159885112495e-05,
      "loss": 1.4234,
      "mean_token_accuracy": 0.6953792750835419,
      "num_tokens": 1062881.0,
      "step": 130
    },
    {
      "entropy": 1.3724075078964233,
      "epoch": 0.03350885591191958,
      "grad_norm": 0.4770592153072357,
      "learning_rate": 1.97782032870592e-05,
      "loss": 1.4941,
      "mean_token_accuracy": 0.6818542718887329,
      "num_tokens": 1143353.0,
      "step": 140
    },
    {
      "entropy": 1.3332514762878418,
      "epoch": 0.03590234561991384,
      "grad_norm": 0.39026322960853577,
      "learning_rate": 1.9762246689005908e-05,
      "loss": 1.4662,
      "mean_token_accuracy": 0.68500617146492,
      "num_tokens": 1229053.0,
      "step": 150
    },
    {
      "entropy": 1.1989752531051636,
      "epoch": 0.03829583532790809,
      "grad_norm": 0.3791215121746063,
      "learning_rate": 1.9746290090952613e-05,
      "loss": 1.3299,
      "mean_token_accuracy": 0.7075082182884216,
      "num_tokens": 1313803.0,
      "step": 160
    },
    {
      "entropy": 1.172585916519165,
      "epoch": 0.04068932503590235,
      "grad_norm": 0.36965620517730713,
      "learning_rate": 1.9730333492899317e-05,
      "loss": 1.2716,
      "mean_token_accuracy": 0.7190541923046112,
      "num_tokens": 1401959.0,
      "step": 170
    },
    {
      "entropy": 1.1395128607749938,
      "epoch": 0.0430828147438966,
      "grad_norm": 0.3217441439628601,
      "learning_rate": 1.9714376894846022e-05,
      "loss": 1.2274,
      "mean_token_accuracy": 0.723137640953064,
      "num_tokens": 1478396.0,
      "step": 180
    },
    {
      "entropy": 1.2055456280708312,
      "epoch": 0.04547630445189086,
      "grad_norm": 0.3278598487377167,
      "learning_rate": 1.9698420296792727e-05,
      "loss": 1.2786,
      "mean_token_accuracy": 0.7123579025268555,
      "num_tokens": 1569299.0,
      "step": 190
    },
    {
      "entropy": 1.1748578429222107,
      "epoch": 0.04786979415988511,
      "grad_norm": 0.51478111743927,
      "learning_rate": 1.9682463698739432e-05,
      "loss": 1.2394,
      "mean_token_accuracy": 0.7217173218727112,
      "num_tokens": 1643993.0,
      "step": 200
    },
    {
      "entropy": 1.227055060863495,
      "epoch": 0.05026328386787937,
      "grad_norm": 0.38209617137908936,
      "learning_rate": 1.9666507100686137e-05,
      "loss": 1.2844,
      "mean_token_accuracy": 0.7089298486709594,
      "num_tokens": 1720490.0,
      "step": 210
    },
    {
      "entropy": 1.211005163192749,
      "epoch": 0.052656773575873624,
      "grad_norm": 0.35681259632110596,
      "learning_rate": 1.9650550502632842e-05,
      "loss": 1.2571,
      "mean_token_accuracy": 0.718545663356781,
      "num_tokens": 1796696.0,
      "step": 220
    },
    {
      "entropy": 1.2637357115745544,
      "epoch": 0.05505026328386788,
      "grad_norm": 0.2849113345146179,
      "learning_rate": 1.9634593904579543e-05,
      "loss": 1.321,
      "mean_token_accuracy": 0.7037886202335357,
      "num_tokens": 1882534.0,
      "step": 230
    },
    {
      "entropy": 1.2149217486381532,
      "epoch": 0.057443752991862135,
      "grad_norm": 0.3721363842487335,
      "learning_rate": 1.9618637306526248e-05,
      "loss": 1.264,
      "mean_token_accuracy": 0.7126233696937561,
      "num_tokens": 1966536.0,
      "step": 240
    },
    {
      "entropy": 1.1745641827583313,
      "epoch": 0.059837242699856394,
      "grad_norm": 0.338847279548645,
      "learning_rate": 1.9602680708472953e-05,
      "loss": 1.2195,
      "mean_token_accuracy": 0.7231896340847015,
      "num_tokens": 2052917.0,
      "step": 250
    },
    {
      "entropy": 1.130018949508667,
      "epoch": 0.062230732407850646,
      "grad_norm": 0.3460768163204193,
      "learning_rate": 1.9586724110419658e-05,
      "loss": 1.1682,
      "mean_token_accuracy": 0.7333509624004364,
      "num_tokens": 2133662.0,
      "step": 260
    },
    {
      "entropy": 1.2339953660964966,
      "epoch": 0.0646242221158449,
      "grad_norm": 0.5184815526008606,
      "learning_rate": 1.9570767512366363e-05,
      "loss": 1.2837,
      "mean_token_accuracy": 0.7122360646724701,
      "num_tokens": 2210171.0,
      "step": 270
    },
    {
      "entropy": 1.1672921299934387,
      "epoch": 0.06701771182383916,
      "grad_norm": 0.33685940504074097,
      "learning_rate": 1.955481091431307e-05,
      "loss": 1.1869,
      "mean_token_accuracy": 0.7300077021121979,
      "num_tokens": 2285278.0,
      "step": 280
    },
    {
      "entropy": 1.278721535205841,
      "epoch": 0.06941120153183342,
      "grad_norm": 0.45780664682388306,
      "learning_rate": 1.9538854316259776e-05,
      "loss": 1.3093,
      "mean_token_accuracy": 0.7067837238311767,
      "num_tokens": 2367086.0,
      "step": 290
    },
    {
      "entropy": 1.179729688167572,
      "epoch": 0.07180469123982768,
      "grad_norm": 0.3800657391548157,
      "learning_rate": 1.952289771820648e-05,
      "loss": 1.1871,
      "mean_token_accuracy": 0.7227078795433044,
      "num_tokens": 2452656.0,
      "step": 300
    },
    {
      "entropy": 1.2450039386749268,
      "epoch": 0.07419818094782192,
      "grad_norm": 0.2955397367477417,
      "learning_rate": 1.9506941120153186e-05,
      "loss": 1.2723,
      "mean_token_accuracy": 0.7101258218288422,
      "num_tokens": 2539006.0,
      "step": 310
    },
    {
      "entropy": 1.1539593100547791,
      "epoch": 0.07659167065581618,
      "grad_norm": 0.36999160051345825,
      "learning_rate": 1.949098452209989e-05,
      "loss": 1.1753,
      "mean_token_accuracy": 0.7302987039089203,
      "num_tokens": 2619986.0,
      "step": 320
    },
    {
      "entropy": 1.2736897230148316,
      "epoch": 0.07898516036381044,
      "grad_norm": 0.32264092564582825,
      "learning_rate": 1.9475027924046595e-05,
      "loss": 1.2877,
      "mean_token_accuracy": 0.7095052599906921,
      "num_tokens": 2704320.0,
      "step": 330
    },
    {
      "entropy": 1.1782316148281098,
      "epoch": 0.0813786500718047,
      "grad_norm": 0.315851092338562,
      "learning_rate": 1.94590713259933e-05,
      "loss": 1.184,
      "mean_token_accuracy": 0.7285471856594086,
      "num_tokens": 2788595.0,
      "step": 340
    },
    {
      "entropy": 1.180445098876953,
      "epoch": 0.08377213977979894,
      "grad_norm": 0.32032349705696106,
      "learning_rate": 1.9443114727940005e-05,
      "loss": 1.1802,
      "mean_token_accuracy": 0.7264943718910217,
      "num_tokens": 2870216.0,
      "step": 350
    },
    {
      "entropy": 1.1832275629043578,
      "epoch": 0.0861656294877932,
      "grad_norm": 0.33831459283828735,
      "learning_rate": 1.942715812988671e-05,
      "loss": 1.1976,
      "mean_token_accuracy": 0.7220143854618073,
      "num_tokens": 2954044.0,
      "step": 360
    },
    {
      "entropy": 1.2671967267990112,
      "epoch": 0.08855911919578746,
      "grad_norm": 0.33935675024986267,
      "learning_rate": 1.9411201531833415e-05,
      "loss": 1.2961,
      "mean_token_accuracy": 0.7084427297115325,
      "num_tokens": 3031830.0,
      "step": 370
    },
    {
      "entropy": 1.1417800307273864,
      "epoch": 0.09095260890378172,
      "grad_norm": 0.3180742561817169,
      "learning_rate": 1.939524493378012e-05,
      "loss": 1.1531,
      "mean_token_accuracy": 0.7325368881225586,
      "num_tokens": 3110145.0,
      "step": 380
    },
    {
      "entropy": 1.1399097561836242,
      "epoch": 0.09334609861177597,
      "grad_norm": 0.3661193251609802,
      "learning_rate": 1.9379288335726824e-05,
      "loss": 1.1473,
      "mean_token_accuracy": 0.7271141648292542,
      "num_tokens": 3188932.0,
      "step": 390
    },
    {
      "entropy": 1.1624513387680053,
      "epoch": 0.09573958831977022,
      "grad_norm": 0.3623553514480591,
      "learning_rate": 1.936333173767353e-05,
      "loss": 1.1982,
      "mean_token_accuracy": 0.7235555469989776,
      "num_tokens": 3274683.0,
      "step": 400
    },
    {
      "entropy": 1.1469115734100341,
      "epoch": 0.09813307802776448,
      "grad_norm": 0.42979690432548523,
      "learning_rate": 1.9347375139620234e-05,
      "loss": 1.1691,
      "mean_token_accuracy": 0.7302775263786316,
      "num_tokens": 3362254.0,
      "step": 410
    },
    {
      "entropy": 1.2337151288986206,
      "epoch": 0.10052656773575874,
      "grad_norm": 0.35702192783355713,
      "learning_rate": 1.933141854156694e-05,
      "loss": 1.2656,
      "mean_token_accuracy": 0.7132411658763885,
      "num_tokens": 3445794.0,
      "step": 420
    },
    {
      "entropy": 1.106970739364624,
      "epoch": 0.10292005744375299,
      "grad_norm": 0.3846401274204254,
      "learning_rate": 1.9315461943513644e-05,
      "loss": 1.1229,
      "mean_token_accuracy": 0.7359513640403748,
      "num_tokens": 3527793.0,
      "step": 430
    },
    {
      "entropy": 1.1119951069355012,
      "epoch": 0.10531354715174725,
      "grad_norm": 0.40356433391571045,
      "learning_rate": 1.929950534546035e-05,
      "loss": 1.1475,
      "mean_token_accuracy": 0.7323212623596191,
      "num_tokens": 3608380.0,
      "step": 440
    },
    {
      "entropy": 1.1829931259155273,
      "epoch": 0.1077070368597415,
      "grad_norm": 0.3817615211009979,
      "learning_rate": 1.9283548747407054e-05,
      "loss": 1.194,
      "mean_token_accuracy": 0.726850563287735,
      "num_tokens": 3688343.0,
      "step": 450
    },
    {
      "entropy": 1.1868076264858245,
      "epoch": 0.11010052656773577,
      "grad_norm": 0.3502326011657715,
      "learning_rate": 1.926759214935376e-05,
      "loss": 1.2123,
      "mean_token_accuracy": 0.7224819004535675,
      "num_tokens": 3764808.0,
      "step": 460
    },
    {
      "entropy": 1.2451523780822753,
      "epoch": 0.11249401627573001,
      "grad_norm": 0.43315961956977844,
      "learning_rate": 1.9251635551300463e-05,
      "loss": 1.2767,
      "mean_token_accuracy": 0.713153749704361,
      "num_tokens": 3850906.0,
      "step": 470
    },
    {
      "entropy": 1.1267979145050049,
      "epoch": 0.11488750598372427,
      "grad_norm": 0.3926289975643158,
      "learning_rate": 1.923567895324717e-05,
      "loss": 1.1317,
      "mean_token_accuracy": 0.7387243270874023,
      "num_tokens": 3931991.0,
      "step": 480
    },
    {
      "entropy": 1.1515992522239684,
      "epoch": 0.11728099569171853,
      "grad_norm": 0.42953503131866455,
      "learning_rate": 1.9219722355193876e-05,
      "loss": 1.1842,
      "mean_token_accuracy": 0.7252130150794983,
      "num_tokens": 4011051.0,
      "step": 490
    },
    {
      "entropy": 1.138441789150238,
      "epoch": 0.11967448539971279,
      "grad_norm": 0.5039691925048828,
      "learning_rate": 1.920376575714058e-05,
      "loss": 1.1704,
      "mean_token_accuracy": 0.7336141228675842,
      "num_tokens": 4087105.0,
      "step": 500
    },
    {
      "entropy": 1.1398205399513244,
      "epoch": 0.12206797510770703,
      "grad_norm": 0.431519091129303,
      "learning_rate": 1.9187809159087286e-05,
      "loss": 1.1557,
      "mean_token_accuracy": 0.7311812996864319,
      "num_tokens": 4171612.0,
      "step": 510
    },
    {
      "entropy": 1.2045225858688355,
      "epoch": 0.12446146481570129,
      "grad_norm": 0.3644827902317047,
      "learning_rate": 1.917185256103399e-05,
      "loss": 1.2212,
      "mean_token_accuracy": 0.7185820639133453,
      "num_tokens": 4254590.0,
      "step": 520
    },
    {
      "entropy": 1.0943529725074768,
      "epoch": 0.12685495452369555,
      "grad_norm": 0.3654193580150604,
      "learning_rate": 1.9155895962980696e-05,
      "loss": 1.1163,
      "mean_token_accuracy": 0.7362634301185608,
      "num_tokens": 4334608.0,
      "step": 530
    },
    {
      "entropy": 1.113824737071991,
      "epoch": 0.1292484442316898,
      "grad_norm": 0.4767385721206665,
      "learning_rate": 1.91399393649274e-05,
      "loss": 1.1309,
      "mean_token_accuracy": 0.732136869430542,
      "num_tokens": 4421216.0,
      "step": 540
    },
    {
      "entropy": 1.1436684787273408,
      "epoch": 0.13164193393968407,
      "grad_norm": 0.4919729232788086,
      "learning_rate": 1.9123982766874102e-05,
      "loss": 1.1811,
      "mean_token_accuracy": 0.7301218271255493,
      "num_tokens": 4502726.0,
      "step": 550
    },
    {
      "entropy": 1.1604778051376343,
      "epoch": 0.13403542364767831,
      "grad_norm": 0.4208875894546509,
      "learning_rate": 1.9108026168820807e-05,
      "loss": 1.1568,
      "mean_token_accuracy": 0.7322458744049072,
      "num_tokens": 4587371.0,
      "step": 560
    },
    {
      "entropy": 1.1452993869781494,
      "epoch": 0.13642891335567256,
      "grad_norm": 0.47862178087234497,
      "learning_rate": 1.9092069570767512e-05,
      "loss": 1.1664,
      "mean_token_accuracy": 0.7321425914764405,
      "num_tokens": 4670884.0,
      "step": 570
    },
    {
      "entropy": 1.1043135583400727,
      "epoch": 0.13882240306366683,
      "grad_norm": 0.44380509853363037,
      "learning_rate": 1.9076112972714217e-05,
      "loss": 1.1261,
      "mean_token_accuracy": 0.7356089949607849,
      "num_tokens": 4752436.0,
      "step": 580
    },
    {
      "entropy": 1.1199235320091248,
      "epoch": 0.14121589277166108,
      "grad_norm": 0.3798796832561493,
      "learning_rate": 1.906015637466092e-05,
      "loss": 1.1352,
      "mean_token_accuracy": 0.7339802384376526,
      "num_tokens": 4834426.0,
      "step": 590
    },
    {
      "entropy": 1.2029297947883606,
      "epoch": 0.14360938247965535,
      "grad_norm": 0.445160448551178,
      "learning_rate": 1.9044199776607626e-05,
      "loss": 1.2249,
      "mean_token_accuracy": 0.7229050159454345,
      "num_tokens": 4919118.0,
      "step": 600
    },
    {
      "entropy": 1.1414318084716797,
      "epoch": 0.1460028721876496,
      "grad_norm": 0.3705262243747711,
      "learning_rate": 1.9028243178554335e-05,
      "loss": 1.1564,
      "mean_token_accuracy": 0.7287187993526458,
      "num_tokens": 5006195.0,
      "step": 610
    },
    {
      "entropy": 1.1510895013809204,
      "epoch": 0.14839636189564384,
      "grad_norm": 0.3825203776359558,
      "learning_rate": 1.901228658050104e-05,
      "loss": 1.1644,
      "mean_token_accuracy": 0.728028291463852,
      "num_tokens": 5085225.0,
      "step": 620
    },
    {
      "entropy": 1.1984870433807373,
      "epoch": 0.15078985160363811,
      "grad_norm": 0.45023053884506226,
      "learning_rate": 1.8996329982447744e-05,
      "loss": 1.2109,
      "mean_token_accuracy": 0.7187324345111847,
      "num_tokens": 5166008.0,
      "step": 630
    },
    {
      "entropy": 1.0969820857048034,
      "epoch": 0.15318334131163236,
      "grad_norm": 0.4414844810962677,
      "learning_rate": 1.898037338439445e-05,
      "loss": 1.1029,
      "mean_token_accuracy": 0.7383518695831299,
      "num_tokens": 5244644.0,
      "step": 640
    },
    {
      "entropy": 1.1387382984161376,
      "epoch": 0.1555768310196266,
      "grad_norm": 0.5475702285766602,
      "learning_rate": 1.8964416786341154e-05,
      "loss": 1.1626,
      "mean_token_accuracy": 0.7259968459606171,
      "num_tokens": 5328328.0,
      "step": 650
    },
    {
      "entropy": 1.1580329537391663,
      "epoch": 0.15797032072762088,
      "grad_norm": 0.504128634929657,
      "learning_rate": 1.894846018828786e-05,
      "loss": 1.1581,
      "mean_token_accuracy": 0.7270308196544647,
      "num_tokens": 5409787.0,
      "step": 660
    },
    {
      "entropy": 1.1716572701931,
      "epoch": 0.16036381043561512,
      "grad_norm": 0.44866034388542175,
      "learning_rate": 1.8932503590234564e-05,
      "loss": 1.1903,
      "mean_token_accuracy": 0.7269732773303985,
      "num_tokens": 5485086.0,
      "step": 670
    },
    {
      "entropy": 1.1826915740966797,
      "epoch": 0.1627573001436094,
      "grad_norm": 0.5264368057250977,
      "learning_rate": 1.891654699218127e-05,
      "loss": 1.2096,
      "mean_token_accuracy": 0.7235718369483948,
      "num_tokens": 5560967.0,
      "step": 680
    },
    {
      "entropy": 1.1117774069309234,
      "epoch": 0.16515078985160364,
      "grad_norm": 0.3255954086780548,
      "learning_rate": 1.8900590394127974e-05,
      "loss": 1.1182,
      "mean_token_accuracy": 0.7342773914337158,
      "num_tokens": 5647128.0,
      "step": 690
    },
    {
      "entropy": 1.1096196174621582,
      "epoch": 0.1675442795595979,
      "grad_norm": 0.4005047380924225,
      "learning_rate": 1.888463379607468e-05,
      "loss": 1.114,
      "mean_token_accuracy": 0.737242442369461,
      "num_tokens": 5728810.0,
      "step": 700
    },
    {
      "entropy": 1.1606119513511657,
      "epoch": 0.16993776926759216,
      "grad_norm": 0.4264616072177887,
      "learning_rate": 1.8868677198021383e-05,
      "loss": 1.1786,
      "mean_token_accuracy": 0.7281650364398956,
      "num_tokens": 5810170.0,
      "step": 710
    },
    {
      "entropy": 1.112518960237503,
      "epoch": 0.1723312589755864,
      "grad_norm": 0.4914409816265106,
      "learning_rate": 1.8852720599968088e-05,
      "loss": 1.1331,
      "mean_token_accuracy": 0.7338595867156983,
      "num_tokens": 5891497.0,
      "step": 720
    },
    {
      "entropy": 1.1373264729976653,
      "epoch": 0.17472474868358065,
      "grad_norm": 0.45702746510505676,
      "learning_rate": 1.8836764001914793e-05,
      "loss": 1.166,
      "mean_token_accuracy": 0.7291199922561645,
      "num_tokens": 5968740.0,
      "step": 730
    },
    {
      "entropy": 1.1402511954307557,
      "epoch": 0.17711823839157492,
      "grad_norm": 0.43656495213508606,
      "learning_rate": 1.8820807403861498e-05,
      "loss": 1.1371,
      "mean_token_accuracy": 0.7335954666137695,
      "num_tokens": 6043075.0,
      "step": 740
    },
    {
      "entropy": 1.1422988772392273,
      "epoch": 0.17951172809956917,
      "grad_norm": 0.38513317704200745,
      "learning_rate": 1.8804850805808203e-05,
      "loss": 1.1664,
      "mean_token_accuracy": 0.7298018157482147,
      "num_tokens": 6129707.0,
      "step": 750
    },
    {
      "entropy": 1.08004851937294,
      "epoch": 0.18190521780756344,
      "grad_norm": 0.4508156478404999,
      "learning_rate": 1.8788894207754908e-05,
      "loss": 1.086,
      "mean_token_accuracy": 0.740784329175949,
      "num_tokens": 6205785.0,
      "step": 760
    },
    {
      "entropy": 1.1543840169906616,
      "epoch": 0.18429870751555769,
      "grad_norm": 0.4366827607154846,
      "learning_rate": 1.8772937609701612e-05,
      "loss": 1.1706,
      "mean_token_accuracy": 0.7310328662395478,
      "num_tokens": 6283981.0,
      "step": 770
    },
    {
      "entropy": 1.1501765966415405,
      "epoch": 0.18669219722355193,
      "grad_norm": 0.4135178327560425,
      "learning_rate": 1.8756981011648317e-05,
      "loss": 1.1577,
      "mean_token_accuracy": 0.730017501115799,
      "num_tokens": 6365464.0,
      "step": 780
    },
    {
      "entropy": 1.114385050535202,
      "epoch": 0.1890856869315462,
      "grad_norm": 0.4548529088497162,
      "learning_rate": 1.8741024413595022e-05,
      "loss": 1.1328,
      "mean_token_accuracy": 0.7312047243118286,
      "num_tokens": 6455090.0,
      "step": 790
    },
    {
      "entropy": 1.150389266014099,
      "epoch": 0.19147917663954045,
      "grad_norm": 0.5956313014030457,
      "learning_rate": 1.8725067815541727e-05,
      "loss": 1.1584,
      "mean_token_accuracy": 0.7284776091575622,
      "num_tokens": 6539862.0,
      "step": 800
    },
    {
      "entropy": 1.1701642751693726,
      "epoch": 0.1938726663475347,
      "grad_norm": 0.47608885169029236,
      "learning_rate": 1.8709111217488435e-05,
      "loss": 1.1786,
      "mean_token_accuracy": 0.7253443360328674,
      "num_tokens": 6619025.0,
      "step": 810
    },
    {
      "entropy": 1.173157000541687,
      "epoch": 0.19626615605552897,
      "grad_norm": 0.4210651218891144,
      "learning_rate": 1.869315461943514e-05,
      "loss": 1.1904,
      "mean_token_accuracy": 0.7227234840393066,
      "num_tokens": 6703630.0,
      "step": 820
    },
    {
      "entropy": 1.1706862688064574,
      "epoch": 0.1986596457635232,
      "grad_norm": 0.7524719834327698,
      "learning_rate": 1.8677198021381845e-05,
      "loss": 1.1905,
      "mean_token_accuracy": 0.7265216827392578,
      "num_tokens": 6783967.0,
      "step": 830
    },
    {
      "entropy": 1.1662927329540254,
      "epoch": 0.20105313547151749,
      "grad_norm": 0.5453492403030396,
      "learning_rate": 1.866124142332855e-05,
      "loss": 1.1699,
      "mean_token_accuracy": 0.7259662568569183,
      "num_tokens": 6870805.0,
      "step": 840
    },
    {
      "entropy": 1.1356736063957213,
      "epoch": 0.20344662517951173,
      "grad_norm": 0.44763392210006714,
      "learning_rate": 1.8645284825275255e-05,
      "loss": 1.1501,
      "mean_token_accuracy": 0.7307538330554962,
      "num_tokens": 6956283.0,
      "step": 850
    },
    {
      "entropy": 1.1026572406291961,
      "epoch": 0.20584011488750598,
      "grad_norm": 0.422983855009079,
      "learning_rate": 1.862932822722196e-05,
      "loss": 1.1267,
      "mean_token_accuracy": 0.7333247244358063,
      "num_tokens": 7034837.0,
      "step": 860
    },
    {
      "entropy": 1.1422979474067687,
      "epoch": 0.20823360459550025,
      "grad_norm": 0.4250401258468628,
      "learning_rate": 1.861337162916866e-05,
      "loss": 1.1449,
      "mean_token_accuracy": 0.7318553447723388,
      "num_tokens": 7117834.0,
      "step": 870
    },
    {
      "entropy": 1.0403048276901246,
      "epoch": 0.2106270943034945,
      "grad_norm": 0.4412606358528137,
      "learning_rate": 1.8597415031115366e-05,
      "loss": 1.0463,
      "mean_token_accuracy": 0.7481624603271484,
      "num_tokens": 7196257.0,
      "step": 880
    },
    {
      "entropy": 1.1827841997146606,
      "epoch": 0.21302058401148874,
      "grad_norm": 0.4300619661808014,
      "learning_rate": 1.858145843306207e-05,
      "loss": 1.1906,
      "mean_token_accuracy": 0.723038399219513,
      "num_tokens": 7277903.0,
      "step": 890
    },
    {
      "entropy": 1.1414884567260741,
      "epoch": 0.215414073719483,
      "grad_norm": 0.4286167323589325,
      "learning_rate": 1.8565501835008776e-05,
      "loss": 1.1554,
      "mean_token_accuracy": 0.7324569940567016,
      "num_tokens": 7362480.0,
      "step": 900
    },
    {
      "entropy": 1.0790764927864074,
      "epoch": 0.21780756342747726,
      "grad_norm": 0.5234811902046204,
      "learning_rate": 1.854954523695548e-05,
      "loss": 1.0969,
      "mean_token_accuracy": 0.7386617481708526,
      "num_tokens": 7440012.0,
      "step": 910
    },
    {
      "entropy": 1.1535210132598877,
      "epoch": 0.22020105313547153,
      "grad_norm": 0.37154629826545715,
      "learning_rate": 1.8533588638902185e-05,
      "loss": 1.166,
      "mean_token_accuracy": 0.7290057837963104,
      "num_tokens": 7525930.0,
      "step": 920
    },
    {
      "entropy": 1.1074342250823974,
      "epoch": 0.22259454284346578,
      "grad_norm": 0.46574974060058594,
      "learning_rate": 1.851763204084889e-05,
      "loss": 1.1132,
      "mean_token_accuracy": 0.73510040640831,
      "num_tokens": 7612026.0,
      "step": 930
    },
    {
      "entropy": 1.0925402879714965,
      "epoch": 0.22498803255146002,
      "grad_norm": 0.5284107327461243,
      "learning_rate": 1.85016754427956e-05,
      "loss": 1.0982,
      "mean_token_accuracy": 0.7392546772956848,
      "num_tokens": 7695379.0,
      "step": 940
    },
    {
      "entropy": 1.1203206062316895,
      "epoch": 0.2273815222594543,
      "grad_norm": 0.4579569399356842,
      "learning_rate": 1.8485718844742303e-05,
      "loss": 1.137,
      "mean_token_accuracy": 0.7334638714790345,
      "num_tokens": 7777583.0,
      "step": 950
    },
    {
      "entropy": 1.098641002178192,
      "epoch": 0.22977501196744854,
      "grad_norm": 0.43879202008247375,
      "learning_rate": 1.8469762246689008e-05,
      "loss": 1.1177,
      "mean_token_accuracy": 0.7360766649246215,
      "num_tokens": 7852209.0,
      "step": 960
    },
    {
      "entropy": 1.0046986997127534,
      "epoch": 0.23216850167544278,
      "grad_norm": 0.49049076437950134,
      "learning_rate": 1.8453805648635713e-05,
      "loss": 1.006,
      "mean_token_accuracy": 0.7516354858875275,
      "num_tokens": 7935280.0,
      "step": 970
    },
    {
      "entropy": 1.158735889196396,
      "epoch": 0.23456199138343706,
      "grad_norm": 0.4842958450317383,
      "learning_rate": 1.8437849050582418e-05,
      "loss": 1.1701,
      "mean_token_accuracy": 0.7276633143424988,
      "num_tokens": 8016426.0,
      "step": 980
    },
    {
      "entropy": 1.1089605867862702,
      "epoch": 0.2369554810914313,
      "grad_norm": 0.4378778636455536,
      "learning_rate": 1.8421892452529123e-05,
      "loss": 1.1193,
      "mean_token_accuracy": 0.7394879758358002,
      "num_tokens": 8101358.0,
      "step": 990
    },
    {
      "entropy": 1.1790334105491638,
      "epoch": 0.23934897079942558,
      "grad_norm": 0.3953637182712555,
      "learning_rate": 1.8405935854475828e-05,
      "loss": 1.1922,
      "mean_token_accuracy": 0.7241828083992005,
      "num_tokens": 8182849.0,
      "step": 1000
    },
    {
      "entropy": 1.060250174999237,
      "epoch": 0.24174246050741982,
      "grad_norm": 0.5230845808982849,
      "learning_rate": 1.8389979256422532e-05,
      "loss": 1.0676,
      "mean_token_accuracy": 0.7449507713317871,
      "num_tokens": 8261297.0,
      "step": 1010
    },
    {
      "entropy": 1.1369280517101288,
      "epoch": 0.24413595021541407,
      "grad_norm": 0.460395872592926,
      "learning_rate": 1.8374022658369237e-05,
      "loss": 1.1408,
      "mean_token_accuracy": 0.7317798137664795,
      "num_tokens": 8346814.0,
      "step": 1020
    },
    {
      "entropy": 1.1554837226867676,
      "epoch": 0.24652943992340834,
      "grad_norm": 0.49037691950798035,
      "learning_rate": 1.8358066060315942e-05,
      "loss": 1.1736,
      "mean_token_accuracy": 0.7271105825901032,
      "num_tokens": 8432664.0,
      "step": 1030
    },
    {
      "entropy": 1.0856216609477998,
      "epoch": 0.24892292963140258,
      "grad_norm": 0.5764221549034119,
      "learning_rate": 1.8342109462262647e-05,
      "loss": 1.0928,
      "mean_token_accuracy": 0.7418696939945221,
      "num_tokens": 8510307.0,
      "step": 1040
    },
    {
      "entropy": 1.1276116609573363,
      "epoch": 0.25131641933939686,
      "grad_norm": 0.4118581712245941,
      "learning_rate": 1.8326152864209352e-05,
      "loss": 1.1538,
      "mean_token_accuracy": 0.7313674926757813,
      "num_tokens": 8592223.0,
      "step": 1050
    },
    {
      "entropy": 1.1677935779094697,
      "epoch": 0.2537099090473911,
      "grad_norm": 0.5074272155761719,
      "learning_rate": 1.8310196266156057e-05,
      "loss": 1.1762,
      "mean_token_accuracy": 0.7262083709239959,
      "num_tokens": 8671470.0,
      "step": 1060
    },
    {
      "entropy": 1.1634832084178925,
      "epoch": 0.25610339875538535,
      "grad_norm": 0.5505145788192749,
      "learning_rate": 1.829423966810276e-05,
      "loss": 1.1685,
      "mean_token_accuracy": 0.7255993783473969,
      "num_tokens": 8748843.0,
      "step": 1070
    },
    {
      "entropy": 1.1505753695964813,
      "epoch": 0.2584968884633796,
      "grad_norm": 0.41146954894065857,
      "learning_rate": 1.8278283070049466e-05,
      "loss": 1.1591,
      "mean_token_accuracy": 0.7274834334850311,
      "num_tokens": 8828379.0,
      "step": 1080
    },
    {
      "entropy": 1.112020456790924,
      "epoch": 0.26089037817137384,
      "grad_norm": 0.615053653717041,
      "learning_rate": 1.826232647199617e-05,
      "loss": 1.1312,
      "mean_token_accuracy": 0.7352388620376586,
      "num_tokens": 8914585.0,
      "step": 1090
    },
    {
      "entropy": 1.143721067905426,
      "epoch": 0.26328386787936814,
      "grad_norm": 0.5205114483833313,
      "learning_rate": 1.8246369873942876e-05,
      "loss": 1.1547,
      "mean_token_accuracy": 0.7253860235214233,
      "num_tokens": 8998808.0,
      "step": 1100
    },
    {
      "entropy": 1.0884106397628783,
      "epoch": 0.2656773575873624,
      "grad_norm": 0.5011736750602722,
      "learning_rate": 1.823041327588958e-05,
      "loss": 1.0933,
      "mean_token_accuracy": 0.738983541727066,
      "num_tokens": 9082538.0,
      "step": 1110
    },
    {
      "entropy": 1.1211643397808075,
      "epoch": 0.26807084729535663,
      "grad_norm": 0.5459955930709839,
      "learning_rate": 1.8214456677836286e-05,
      "loss": 1.1216,
      "mean_token_accuracy": 0.7338890194892883,
      "num_tokens": 9160021.0,
      "step": 1120
    },
    {
      "entropy": 1.0946745038032533,
      "epoch": 0.2704643370033509,
      "grad_norm": 0.4615417420864105,
      "learning_rate": 1.8198500079782994e-05,
      "loss": 1.112,
      "mean_token_accuracy": 0.7379445254802703,
      "num_tokens": 9244536.0,
      "step": 1130
    },
    {
      "entropy": 1.0817080974578857,
      "epoch": 0.2728578267113451,
      "grad_norm": 0.5138433575630188,
      "learning_rate": 1.81825434817297e-05,
      "loss": 1.0912,
      "mean_token_accuracy": 0.7364392399787902,
      "num_tokens": 9326725.0,
      "step": 1140
    },
    {
      "entropy": 1.124049973487854,
      "epoch": 0.2752513164193394,
      "grad_norm": 1.1013171672821045,
      "learning_rate": 1.8166586883676404e-05,
      "loss": 1.1361,
      "mean_token_accuracy": 0.7332423567771912,
      "num_tokens": 9400056.0,
      "step": 1150
    },
    {
      "entropy": 1.227579951286316,
      "epoch": 0.27764480612733367,
      "grad_norm": 0.4477737545967102,
      "learning_rate": 1.815063028562311e-05,
      "loss": 1.2311,
      "mean_token_accuracy": 0.7192757904529572,
      "num_tokens": 9477840.0,
      "step": 1160
    },
    {
      "entropy": 1.0108287751674652,
      "epoch": 0.2800382958353279,
      "grad_norm": 0.5158432722091675,
      "learning_rate": 1.8134673687569814e-05,
      "loss": 1.0087,
      "mean_token_accuracy": 0.7559822559356689,
      "num_tokens": 9555470.0,
      "step": 1170
    },
    {
      "entropy": 1.0718107998371125,
      "epoch": 0.28243178554332216,
      "grad_norm": 0.5826597213745117,
      "learning_rate": 1.811871708951652e-05,
      "loss": 1.1079,
      "mean_token_accuracy": 0.7415627300739288,
      "num_tokens": 9635793.0,
      "step": 1180
    },
    {
      "entropy": 1.1753923535346984,
      "epoch": 0.2848252752513164,
      "grad_norm": 0.5123017430305481,
      "learning_rate": 1.810276049146322e-05,
      "loss": 1.1678,
      "mean_token_accuracy": 0.7245855867862702,
      "num_tokens": 9718455.0,
      "step": 1190
    },
    {
      "entropy": 1.0536331057548523,
      "epoch": 0.2872187649593107,
      "grad_norm": 0.515906035900116,
      "learning_rate": 1.8086803893409925e-05,
      "loss": 1.0539,
      "mean_token_accuracy": 0.746714061498642,
      "num_tokens": 9798227.0,
      "step": 1200
    },
    {
      "entropy": 1.0995701193809508,
      "epoch": 0.28961225466730495,
      "grad_norm": 0.5043840408325195,
      "learning_rate": 1.807084729535663e-05,
      "loss": 1.1141,
      "mean_token_accuracy": 0.7365080714225769,
      "num_tokens": 9881529.0,
      "step": 1210
    },
    {
      "entropy": 1.1067807972431183,
      "epoch": 0.2920057443752992,
      "grad_norm": 0.5044357776641846,
      "learning_rate": 1.8054890697303335e-05,
      "loss": 1.114,
      "mean_token_accuracy": 0.7387550234794616,
      "num_tokens": 9962408.0,
      "step": 1220
    },
    {
      "entropy": 1.1084092378616333,
      "epoch": 0.29439923408329344,
      "grad_norm": 0.539025604724884,
      "learning_rate": 1.803893409925004e-05,
      "loss": 1.1234,
      "mean_token_accuracy": 0.7368146061897278,
      "num_tokens": 10047556.0,
      "step": 1230
    },
    {
      "entropy": 1.1232216238975525,
      "epoch": 0.2967927237912877,
      "grad_norm": 0.5782031416893005,
      "learning_rate": 1.8022977501196744e-05,
      "loss": 1.118,
      "mean_token_accuracy": 0.7328577399253845,
      "num_tokens": 10134677.0,
      "step": 1240
    },
    {
      "entropy": 1.0230436742305755,
      "epoch": 0.29918621349928193,
      "grad_norm": 0.5231150388717651,
      "learning_rate": 1.800702090314345e-05,
      "loss": 1.0415,
      "mean_token_accuracy": 0.7487457931041718,
      "num_tokens": 10213876.0,
      "step": 1250
    },
    {
      "entropy": 1.173479026556015,
      "epoch": 0.30157970320727623,
      "grad_norm": 0.5498600006103516,
      "learning_rate": 1.7991064305090157e-05,
      "loss": 1.1829,
      "mean_token_accuracy": 0.7260289549827575,
      "num_tokens": 10292884.0,
      "step": 1260
    },
    {
      "entropy": 1.1466880917549134,
      "epoch": 0.3039731929152705,
      "grad_norm": 0.46151983737945557,
      "learning_rate": 1.7975107707036862e-05,
      "loss": 1.1394,
      "mean_token_accuracy": 0.7284374833106995,
      "num_tokens": 10377210.0,
      "step": 1270
    },
    {
      "entropy": 1.0677096903324128,
      "epoch": 0.3063666826232647,
      "grad_norm": 0.5466307997703552,
      "learning_rate": 1.7959151108983567e-05,
      "loss": 1.0858,
      "mean_token_accuracy": 0.7383745372295379,
      "num_tokens": 10459196.0,
      "step": 1280
    },
    {
      "entropy": 1.109909290075302,
      "epoch": 0.30876017233125896,
      "grad_norm": 0.6939959526062012,
      "learning_rate": 1.7943194510930272e-05,
      "loss": 1.1233,
      "mean_token_accuracy": 0.7348092496395111,
      "num_tokens": 10541884.0,
      "step": 1290
    },
    {
      "entropy": 1.0882596015930175,
      "epoch": 0.3111536620392532,
      "grad_norm": 0.48167234659194946,
      "learning_rate": 1.7927237912876977e-05,
      "loss": 1.0874,
      "mean_token_accuracy": 0.7419361233711242,
      "num_tokens": 10624600.0,
      "step": 1300
    },
    {
      "entropy": 1.0614607214927674,
      "epoch": 0.3135471517472475,
      "grad_norm": 0.4936775863170624,
      "learning_rate": 1.791128131482368e-05,
      "loss": 1.0661,
      "mean_token_accuracy": 0.740717351436615,
      "num_tokens": 10702252.0,
      "step": 1310
    },
    {
      "entropy": 1.0592714846134186,
      "epoch": 0.31594064145524176,
      "grad_norm": 0.8975985646247864,
      "learning_rate": 1.7895324716770387e-05,
      "loss": 1.086,
      "mean_token_accuracy": 0.7410467028617859,
      "num_tokens": 10774660.0,
      "step": 1320
    },
    {
      "entropy": 1.1287354588508607,
      "epoch": 0.318334131163236,
      "grad_norm": 0.8397033214569092,
      "learning_rate": 1.787936811871709e-05,
      "loss": 1.1272,
      "mean_token_accuracy": 0.7296611070632935,
      "num_tokens": 10859939.0,
      "step": 1330
    },
    {
      "entropy": 1.0429444849491118,
      "epoch": 0.32072762087123025,
      "grad_norm": 0.4758578836917877,
      "learning_rate": 1.7863411520663796e-05,
      "loss": 1.0455,
      "mean_token_accuracy": 0.7479507088661194,
      "num_tokens": 10945756.0,
      "step": 1340
    },
    {
      "entropy": 1.0388957977294921,
      "epoch": 0.3231211105792245,
      "grad_norm": 0.48702606558799744,
      "learning_rate": 1.78474549226105e-05,
      "loss": 1.0541,
      "mean_token_accuracy": 0.7459675371646881,
      "num_tokens": 11025527.0,
      "step": 1350
    },
    {
      "entropy": 1.1503496289253234,
      "epoch": 0.3255146002872188,
      "grad_norm": 0.464547336101532,
      "learning_rate": 1.7831498324557206e-05,
      "loss": 1.1437,
      "mean_token_accuracy": 0.7314624607563018,
      "num_tokens": 11106478.0,
      "step": 1360
    },
    {
      "entropy": 1.116293078660965,
      "epoch": 0.32790808999521304,
      "grad_norm": 0.49953344464302063,
      "learning_rate": 1.781554172650391e-05,
      "loss": 1.1364,
      "mean_token_accuracy": 0.7377607464790344,
      "num_tokens": 11187937.0,
      "step": 1370
    },
    {
      "entropy": 1.1121998131275177,
      "epoch": 0.3303015797032073,
      "grad_norm": 0.5883446335792542,
      "learning_rate": 1.7799585128450616e-05,
      "loss": 1.1227,
      "mean_token_accuracy": 0.7302426040172577,
      "num_tokens": 11274073.0,
      "step": 1380
    },
    {
      "entropy": 1.0967606782913208,
      "epoch": 0.3326950694112015,
      "grad_norm": 0.5348601341247559,
      "learning_rate": 1.778362853039732e-05,
      "loss": 1.1049,
      "mean_token_accuracy": 0.7353415608406066,
      "num_tokens": 11358174.0,
      "step": 1390
    },
    {
      "entropy": 1.2084548234939576,
      "epoch": 0.3350885591191958,
      "grad_norm": 0.5608245730400085,
      "learning_rate": 1.7767671932344025e-05,
      "loss": 1.2129,
      "mean_token_accuracy": 0.7224825620651245,
      "num_tokens": 11449700.0,
      "step": 1400
    },
    {
      "entropy": 1.0699814438819886,
      "epoch": 0.33748204882719,
      "grad_norm": 0.5325828790664673,
      "learning_rate": 1.775171533429073e-05,
      "loss": 1.0801,
      "mean_token_accuracy": 0.74050452709198,
      "num_tokens": 11542684.0,
      "step": 1410
    },
    {
      "entropy": 1.0742067635059356,
      "epoch": 0.3398755385351843,
      "grad_norm": 0.5795136094093323,
      "learning_rate": 1.7735758736237435e-05,
      "loss": 1.0895,
      "mean_token_accuracy": 0.7408529937267303,
      "num_tokens": 11624818.0,
      "step": 1420
    },
    {
      "entropy": 1.1562271177768708,
      "epoch": 0.34226902824317856,
      "grad_norm": 0.501658022403717,
      "learning_rate": 1.771980213818414e-05,
      "loss": 1.1557,
      "mean_token_accuracy": 0.7292845904827118,
      "num_tokens": 11703713.0,
      "step": 1430
    },
    {
      "entropy": 1.1149949967861175,
      "epoch": 0.3446625179511728,
      "grad_norm": 0.5859589576721191,
      "learning_rate": 1.7703845540130845e-05,
      "loss": 1.1314,
      "mean_token_accuracy": 0.7346829295158386,
      "num_tokens": 11784387.0,
      "step": 1440
    },
    {
      "entropy": 1.1393958926200867,
      "epoch": 0.34705600765916705,
      "grad_norm": 0.6459059715270996,
      "learning_rate": 1.768788894207755e-05,
      "loss": 1.1481,
      "mean_token_accuracy": 0.7314362764358521,
      "num_tokens": 11864013.0,
      "step": 1450
    },
    {
      "entropy": 1.1156118392944336,
      "epoch": 0.3494494973671613,
      "grad_norm": 0.5335090160369873,
      "learning_rate": 1.7671932344024258e-05,
      "loss": 1.1327,
      "mean_token_accuracy": 0.7320908427238464,
      "num_tokens": 11943448.0,
      "step": 1460
    },
    {
      "entropy": 1.1601254105567933,
      "epoch": 0.3518429870751556,
      "grad_norm": 0.5655401945114136,
      "learning_rate": 1.7655975745970963e-05,
      "loss": 1.1687,
      "mean_token_accuracy": 0.7263648569583893,
      "num_tokens": 12027368.0,
      "step": 1470
    },
    {
      "entropy": 1.0911331295967102,
      "epoch": 0.35423647678314985,
      "grad_norm": 0.5665701031684875,
      "learning_rate": 1.7640019147917668e-05,
      "loss": 1.093,
      "mean_token_accuracy": 0.7388573408126831,
      "num_tokens": 12103515.0,
      "step": 1480
    },
    {
      "entropy": 1.0512312710285188,
      "epoch": 0.3566299664911441,
      "grad_norm": 0.5473869442939758,
      "learning_rate": 1.7624062549864373e-05,
      "loss": 1.0725,
      "mean_token_accuracy": 0.7439163744449615,
      "num_tokens": 12190253.0,
      "step": 1490
    },
    {
      "entropy": 1.10569007396698,
      "epoch": 0.35902345619913834,
      "grad_norm": 0.5401135087013245,
      "learning_rate": 1.7608105951811077e-05,
      "loss": 1.1048,
      "mean_token_accuracy": 0.7423075079917908,
      "num_tokens": 12270470.0,
      "step": 1500
    },
    {
      "entropy": 1.1149644315242768,
      "epoch": 0.3614169459071326,
      "grad_norm": 0.5855125784873962,
      "learning_rate": 1.759214935375778e-05,
      "loss": 1.1207,
      "mean_token_accuracy": 0.7331575155258179,
      "num_tokens": 12347521.0,
      "step": 1510
    },
    {
      "entropy": 1.1357326567173005,
      "epoch": 0.3638104356151269,
      "grad_norm": 0.6087114214897156,
      "learning_rate": 1.7576192755704484e-05,
      "loss": 1.1527,
      "mean_token_accuracy": 0.731180077791214,
      "num_tokens": 12433888.0,
      "step": 1520
    },
    {
      "entropy": 1.13399178981781,
      "epoch": 0.3662039253231211,
      "grad_norm": 0.559622049331665,
      "learning_rate": 1.756023615765119e-05,
      "loss": 1.1448,
      "mean_token_accuracy": 0.726129674911499,
      "num_tokens": 12514766.0,
      "step": 1530
    },
    {
      "entropy": 1.140819215774536,
      "epoch": 0.36859741503111537,
      "grad_norm": 0.5918214321136475,
      "learning_rate": 1.7544279559597893e-05,
      "loss": 1.1442,
      "mean_token_accuracy": 0.7298588156700134,
      "num_tokens": 12595540.0,
      "step": 1540
    },
    {
      "entropy": 1.0882207870483398,
      "epoch": 0.3709909047391096,
      "grad_norm": 0.5943206548690796,
      "learning_rate": 1.7528322961544598e-05,
      "loss": 1.0974,
      "mean_token_accuracy": 0.7369262516498566,
      "num_tokens": 12685850.0,
      "step": 1550
    },
    {
      "entropy": 1.049841034412384,
      "epoch": 0.37338439444710386,
      "grad_norm": 0.6741575598716736,
      "learning_rate": 1.7512366363491303e-05,
      "loss": 1.0546,
      "mean_token_accuracy": 0.7472926259040833,
      "num_tokens": 12765335.0,
      "step": 1560
    },
    {
      "entropy": 1.1476511836051941,
      "epoch": 0.3757778841550981,
      "grad_norm": 0.5215249061584473,
      "learning_rate": 1.7496409765438008e-05,
      "loss": 1.1574,
      "mean_token_accuracy": 0.7313296914100647,
      "num_tokens": 12843331.0,
      "step": 1570
    },
    {
      "entropy": 1.1424257516860963,
      "epoch": 0.3781713738630924,
      "grad_norm": 0.42422813177108765,
      "learning_rate": 1.7480453167384713e-05,
      "loss": 1.1468,
      "mean_token_accuracy": 0.7322757124900818,
      "num_tokens": 12926479.0,
      "step": 1580
    },
    {
      "entropy": 1.1224781215190887,
      "epoch": 0.38056486357108665,
      "grad_norm": 0.5699655413627625,
      "learning_rate": 1.746449656933142e-05,
      "loss": 1.1348,
      "mean_token_accuracy": 0.7340908408164978,
      "num_tokens": 13008395.0,
      "step": 1590
    },
    {
      "entropy": 1.1242433786392212,
      "epoch": 0.3829583532790809,
      "grad_norm": 0.5160660743713379,
      "learning_rate": 1.7448539971278126e-05,
      "loss": 1.1415,
      "mean_token_accuracy": 0.7333247721195221,
      "num_tokens": 13092858.0,
      "step": 1600
    },
    {
      "entropy": 1.093923270702362,
      "epoch": 0.38535184298707514,
      "grad_norm": 0.7464497089385986,
      "learning_rate": 1.743258337322483e-05,
      "loss": 1.0964,
      "mean_token_accuracy": 0.7392175436019898,
      "num_tokens": 13174733.0,
      "step": 1610
    },
    {
      "entropy": 1.10646812915802,
      "epoch": 0.3877453326950694,
      "grad_norm": 0.5711216330528259,
      "learning_rate": 1.7416626775171536e-05,
      "loss": 1.1139,
      "mean_token_accuracy": 0.7354251325130463,
      "num_tokens": 13261918.0,
      "step": 1620
    },
    {
      "entropy": 1.041162359714508,
      "epoch": 0.3901388224030637,
      "grad_norm": 0.6076639294624329,
      "learning_rate": 1.740067017711824e-05,
      "loss": 1.0475,
      "mean_token_accuracy": 0.7487479627132416,
      "num_tokens": 13338814.0,
      "step": 1630
    },
    {
      "entropy": 1.0727667212486267,
      "epoch": 0.39253231211105793,
      "grad_norm": 0.46674367785453796,
      "learning_rate": 1.7384713579064945e-05,
      "loss": 1.0752,
      "mean_token_accuracy": 0.7416320621967316,
      "num_tokens": 13424594.0,
      "step": 1640
    },
    {
      "entropy": 1.080169862508774,
      "epoch": 0.3949258018190522,
      "grad_norm": 3.994786262512207,
      "learning_rate": 1.736875698101165e-05,
      "loss": 1.0911,
      "mean_token_accuracy": 0.7385109722614288,
      "num_tokens": 13500390.0,
      "step": 1650
    },
    {
      "entropy": 1.0480771601200103,
      "epoch": 0.3973192915270464,
      "grad_norm": 0.5100842118263245,
      "learning_rate": 1.7352800382958355e-05,
      "loss": 1.0525,
      "mean_token_accuracy": 0.7464718580245971,
      "num_tokens": 13582647.0,
      "step": 1660
    },
    {
      "entropy": 1.0785650134086608,
      "epoch": 0.39971278123504067,
      "grad_norm": 0.5472614765167236,
      "learning_rate": 1.733684378490506e-05,
      "loss": 1.0833,
      "mean_token_accuracy": 0.7412976503372193,
      "num_tokens": 13671194.0,
      "step": 1670
    },
    {
      "entropy": 1.0455707967281342,
      "epoch": 0.40210627094303497,
      "grad_norm": 2.5722084045410156,
      "learning_rate": 1.7320887186851765e-05,
      "loss": 1.0528,
      "mean_token_accuracy": 0.7453463256359101,
      "num_tokens": 13753751.0,
      "step": 1680
    },
    {
      "entropy": 1.0688288927078247,
      "epoch": 0.4044997606510292,
      "grad_norm": 0.5685151219367981,
      "learning_rate": 1.730493058879847e-05,
      "loss": 1.0803,
      "mean_token_accuracy": 0.7423784971237183,
      "num_tokens": 13836887.0,
      "step": 1690
    },
    {
      "entropy": 1.0849794149398804,
      "epoch": 0.40689325035902346,
      "grad_norm": 0.7111828923225403,
      "learning_rate": 1.7288973990745175e-05,
      "loss": 1.0816,
      "mean_token_accuracy": 0.7410798668861389,
      "num_tokens": 13911144.0,
      "step": 1700
    },
    {
      "entropy": 1.0746318578720093,
      "epoch": 0.4092867400670177,
      "grad_norm": 0.47197631001472473,
      "learning_rate": 1.727301739269188e-05,
      "loss": 1.0958,
      "mean_token_accuracy": 0.7405313014984131,
      "num_tokens": 13994518.0,
      "step": 1710
    },
    {
      "entropy": 1.1007722556591033,
      "epoch": 0.41168022977501195,
      "grad_norm": 0.6988211274147034,
      "learning_rate": 1.7257060794638584e-05,
      "loss": 1.1122,
      "mean_token_accuracy": 0.7361564874649048,
      "num_tokens": 14077901.0,
      "step": 1720
    },
    {
      "entropy": 1.1553027510643006,
      "epoch": 0.4140737194830062,
      "grad_norm": 0.5971831679344177,
      "learning_rate": 1.724110419658529e-05,
      "loss": 1.169,
      "mean_token_accuracy": 0.7279999256134033,
      "num_tokens": 14160956.0,
      "step": 1730
    },
    {
      "entropy": 1.1167118430137635,
      "epoch": 0.4164672091910005,
      "grad_norm": 0.6331872344017029,
      "learning_rate": 1.7225147598531994e-05,
      "loss": 1.106,
      "mean_token_accuracy": 0.7389692306518555,
      "num_tokens": 14246870.0,
      "step": 1740
    },
    {
      "entropy": 1.07000989317894,
      "epoch": 0.41886069889899474,
      "grad_norm": 0.6542571783065796,
      "learning_rate": 1.72091910004787e-05,
      "loss": 1.0985,
      "mean_token_accuracy": 0.7383829712867737,
      "num_tokens": 14325211.0,
      "step": 1750
    },
    {
      "entropy": 1.1347588062286378,
      "epoch": 0.421254188606989,
      "grad_norm": 0.6905112862586975,
      "learning_rate": 1.7193234402425404e-05,
      "loss": 1.132,
      "mean_token_accuracy": 0.7278760671615601,
      "num_tokens": 14411713.0,
      "step": 1760
    },
    {
      "entropy": 1.0780755043029786,
      "epoch": 0.42364767831498323,
      "grad_norm": 0.5749602913856506,
      "learning_rate": 1.717727780437211e-05,
      "loss": 1.0921,
      "mean_token_accuracy": 0.7410579800605774,
      "num_tokens": 14496205.0,
      "step": 1770
    },
    {
      "entropy": 1.0782687783241272,
      "epoch": 0.4260411680229775,
      "grad_norm": 0.5854172110557556,
      "learning_rate": 1.7161321206318813e-05,
      "loss": 1.0831,
      "mean_token_accuracy": 0.7435673117637634,
      "num_tokens": 14579678.0,
      "step": 1780
    },
    {
      "entropy": 1.0986156225204469,
      "epoch": 0.4284346577309718,
      "grad_norm": 0.6664464473724365,
      "learning_rate": 1.714536460826552e-05,
      "loss": 1.1006,
      "mean_token_accuracy": 0.7404700219631195,
      "num_tokens": 14661038.0,
      "step": 1790
    },
    {
      "entropy": 1.053427791595459,
      "epoch": 0.430828147438966,
      "grad_norm": 0.6061694622039795,
      "learning_rate": 1.7129408010212227e-05,
      "loss": 1.0693,
      "mean_token_accuracy": 0.7474850475788116,
      "num_tokens": 14738447.0,
      "step": 1800
    },
    {
      "entropy": 1.117923903465271,
      "epoch": 0.43322163714696027,
      "grad_norm": 0.7042562961578369,
      "learning_rate": 1.711345141215893e-05,
      "loss": 1.1162,
      "mean_token_accuracy": 0.7414107978343963,
      "num_tokens": 14821337.0,
      "step": 1810
    },
    {
      "entropy": 1.0853823006153107,
      "epoch": 0.4356151268549545,
      "grad_norm": 0.6966001987457275,
      "learning_rate": 1.7097494814105636e-05,
      "loss": 1.1008,
      "mean_token_accuracy": 0.7390415370464325,
      "num_tokens": 14897475.0,
      "step": 1820
    },
    {
      "entropy": 1.1424123167991638,
      "epoch": 0.43800861656294876,
      "grad_norm": 0.5323998928070068,
      "learning_rate": 1.7081538216052338e-05,
      "loss": 1.1454,
      "mean_token_accuracy": 0.7323163211345672,
      "num_tokens": 14985446.0,
      "step": 1830
    },
    {
      "entropy": 1.042124879360199,
      "epoch": 0.44040210627094306,
      "grad_norm": 0.6735746264457703,
      "learning_rate": 1.7065581617999043e-05,
      "loss": 1.0437,
      "mean_token_accuracy": 0.7483052670955658,
      "num_tokens": 15065538.0,
      "step": 1840
    },
    {
      "entropy": 1.061891108751297,
      "epoch": 0.4427955959789373,
      "grad_norm": 0.7179056406021118,
      "learning_rate": 1.7049625019945747e-05,
      "loss": 1.0768,
      "mean_token_accuracy": 0.7440852046012878,
      "num_tokens": 15145043.0,
      "step": 1850
    },
    {
      "entropy": 1.033814686536789,
      "epoch": 0.44518908568693155,
      "grad_norm": 0.670626163482666,
      "learning_rate": 1.7033668421892452e-05,
      "loss": 1.0314,
      "mean_token_accuracy": 0.7532525777816772,
      "num_tokens": 15221022.0,
      "step": 1860
    },
    {
      "entropy": 1.0369441986083985,
      "epoch": 0.4475825753949258,
      "grad_norm": 0.6771154403686523,
      "learning_rate": 1.7017711823839157e-05,
      "loss": 1.0475,
      "mean_token_accuracy": 0.7451251745223999,
      "num_tokens": 15299009.0,
      "step": 1870
    },
    {
      "entropy": 1.115807294845581,
      "epoch": 0.44997606510292004,
      "grad_norm": 0.622519850730896,
      "learning_rate": 1.7001755225785862e-05,
      "loss": 1.1308,
      "mean_token_accuracy": 0.7329511046409607,
      "num_tokens": 15382425.0,
      "step": 1880
    },
    {
      "entropy": 1.0232145309448242,
      "epoch": 0.4523695548109143,
      "grad_norm": 0.6011036038398743,
      "learning_rate": 1.6985798627732567e-05,
      "loss": 1.034,
      "mean_token_accuracy": 0.7500796377658844,
      "num_tokens": 15464291.0,
      "step": 1890
    },
    {
      "entropy": 1.0658549070358276,
      "epoch": 0.4547630445189086,
      "grad_norm": 5.619943618774414,
      "learning_rate": 1.6969842029679272e-05,
      "loss": 1.0715,
      "mean_token_accuracy": 0.7470685422420502,
      "num_tokens": 15539663.0,
      "step": 1900
    },
    {
      "entropy": 1.137593150138855,
      "epoch": 0.45715653422690283,
      "grad_norm": 0.5702992081642151,
      "learning_rate": 1.6953885431625977e-05,
      "loss": 1.1402,
      "mean_token_accuracy": 0.7303393661975861,
      "num_tokens": 15625502.0,
      "step": 1910
    },
    {
      "entropy": 1.0707310140132904,
      "epoch": 0.4595500239348971,
      "grad_norm": 0.6284565329551697,
      "learning_rate": 1.6937928833572685e-05,
      "loss": 1.0714,
      "mean_token_accuracy": 0.7435721397399903,
      "num_tokens": 15707483.0,
      "step": 1920
    },
    {
      "entropy": 1.122710120677948,
      "epoch": 0.4619435136428913,
      "grad_norm": 0.5178508758544922,
      "learning_rate": 1.692197223551939e-05,
      "loss": 1.1387,
      "mean_token_accuracy": 0.7373745322227478,
      "num_tokens": 15793807.0,
      "step": 1930
    },
    {
      "entropy": 1.0189306616783143,
      "epoch": 0.46433700335088557,
      "grad_norm": 0.7075283527374268,
      "learning_rate": 1.6906015637466095e-05,
      "loss": 1.0254,
      "mean_token_accuracy": 0.7500097513198852,
      "num_tokens": 15876177.0,
      "step": 1940
    },
    {
      "entropy": 1.0296615958213806,
      "epoch": 0.46673049305887987,
      "grad_norm": 0.655680775642395,
      "learning_rate": 1.68900590394128e-05,
      "loss": 1.0336,
      "mean_token_accuracy": 0.7511914551258088,
      "num_tokens": 15965790.0,
      "step": 1950
    },
    {
      "entropy": 1.0647503197193147,
      "epoch": 0.4691239827668741,
      "grad_norm": 0.6232638359069824,
      "learning_rate": 1.6874102441359504e-05,
      "loss": 1.0793,
      "mean_token_accuracy": 0.7438226282596588,
      "num_tokens": 16050434.0,
      "step": 1960
    },
    {
      "entropy": 1.0943642377853393,
      "epoch": 0.47151747247486836,
      "grad_norm": 0.6169235706329346,
      "learning_rate": 1.685814584330621e-05,
      "loss": 1.101,
      "mean_token_accuracy": 0.7388047873973846,
      "num_tokens": 16135646.0,
      "step": 1970
    },
    {
      "entropy": 0.9858994007110595,
      "epoch": 0.4739109621828626,
      "grad_norm": 0.771087110042572,
      "learning_rate": 1.6842189245252914e-05,
      "loss": 0.9905,
      "mean_token_accuracy": 0.7570463836193084,
      "num_tokens": 16207202.0,
      "step": 1980
    },
    {
      "entropy": 1.0939780235290528,
      "epoch": 0.47630445189085685,
      "grad_norm": 1.2727882862091064,
      "learning_rate": 1.682623264719962e-05,
      "loss": 1.1023,
      "mean_token_accuracy": 0.7375462532043457,
      "num_tokens": 16288759.0,
      "step": 1990
    },
    {
      "entropy": 1.0276434659957885,
      "epoch": 0.47869794159885115,
      "grad_norm": 0.6156533360481262,
      "learning_rate": 1.6810276049146324e-05,
      "loss": 1.0344,
      "mean_token_accuracy": 0.7529487133026123,
      "num_tokens": 16376517.0,
      "step": 2000
    },
    {
      "entropy": 1.0436306655406953,
      "epoch": 0.4810914313068454,
      "grad_norm": 0.8076905012130737,
      "learning_rate": 1.679431945109303e-05,
      "loss": 1.0422,
      "mean_token_accuracy": 0.7471736788749694,
      "num_tokens": 16457929.0,
      "step": 2010
    },
    {
      "entropy": 1.0113335013389588,
      "epoch": 0.48348492101483964,
      "grad_norm": 0.5823526978492737,
      "learning_rate": 1.6778362853039733e-05,
      "loss": 1.0238,
      "mean_token_accuracy": 0.7553565919399261,
      "num_tokens": 16544849.0,
      "step": 2020
    },
    {
      "entropy": 1.111458420753479,
      "epoch": 0.4858784107228339,
      "grad_norm": 0.4666674733161926,
      "learning_rate": 1.676240625498644e-05,
      "loss": 1.1183,
      "mean_token_accuracy": 0.7348304867744446,
      "num_tokens": 16631944.0,
      "step": 2030
    },
    {
      "entropy": 1.1391419231891633,
      "epoch": 0.48827190043082813,
      "grad_norm": 0.6024052500724792,
      "learning_rate": 1.6746449656933143e-05,
      "loss": 1.1519,
      "mean_token_accuracy": 0.731666499376297,
      "num_tokens": 16722399.0,
      "step": 2040
    },
    {
      "entropy": 1.0513649940490724,
      "epoch": 0.4906653901388224,
      "grad_norm": 0.8297297954559326,
      "learning_rate": 1.6730493058879848e-05,
      "loss": 1.061,
      "mean_token_accuracy": 0.7470701456069946,
      "num_tokens": 16800897.0,
      "step": 2050
    },
    {
      "entropy": 1.1298478245735168,
      "epoch": 0.4930588798468167,
      "grad_norm": 0.663592517375946,
      "learning_rate": 1.6714536460826553e-05,
      "loss": 1.1243,
      "mean_token_accuracy": 0.7326838731765747,
      "num_tokens": 16875372.0,
      "step": 2060
    },
    {
      "entropy": 1.0739252805709838,
      "epoch": 0.4954523695548109,
      "grad_norm": 0.5974112749099731,
      "learning_rate": 1.6698579862773258e-05,
      "loss": 1.0834,
      "mean_token_accuracy": 0.7434284806251525,
      "num_tokens": 16956515.0,
      "step": 2070
    },
    {
      "entropy": 1.1313821852207184,
      "epoch": 0.49784585926280517,
      "grad_norm": 0.5157338380813599,
      "learning_rate": 1.6682623264719963e-05,
      "loss": 1.15,
      "mean_token_accuracy": 0.7329830229282379,
      "num_tokens": 17032455.0,
      "step": 2080
    },
    {
      "entropy": 1.0915243804454804,
      "epoch": 0.5002393489707995,
      "grad_norm": 0.5727088451385498,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.0871,
      "mean_token_accuracy": 0.7376357674598694,
      "num_tokens": 17111537.0,
      "step": 2090
    },
    {
      "entropy": 1.0860334396362306,
      "epoch": 0.5026328386787937,
      "grad_norm": 1.131248950958252,
      "learning_rate": 1.6650710068613372e-05,
      "loss": 1.0958,
      "mean_token_accuracy": 0.7394561052322388,
      "num_tokens": 17199367.0,
      "step": 2100
    },
    {
      "entropy": 1.0895774960517883,
      "epoch": 0.505026328386788,
      "grad_norm": 0.6308713555335999,
      "learning_rate": 1.6634753470560077e-05,
      "loss": 1.0994,
      "mean_token_accuracy": 0.7413264513015747,
      "num_tokens": 17286069.0,
      "step": 2110
    },
    {
      "entropy": 1.042851161956787,
      "epoch": 0.5074198180947822,
      "grad_norm": 0.6670368909835815,
      "learning_rate": 1.6618796872506785e-05,
      "loss": 1.0481,
      "mean_token_accuracy": 0.7471791267395019,
      "num_tokens": 17365360.0,
      "step": 2120
    },
    {
      "entropy": 1.0485999524593352,
      "epoch": 0.5098133078027764,
      "grad_norm": 0.6198334097862244,
      "learning_rate": 1.660284027445349e-05,
      "loss": 1.0648,
      "mean_token_accuracy": 0.7454506397247315,
      "num_tokens": 17439869.0,
      "step": 2130
    },
    {
      "entropy": 1.1252771377563477,
      "epoch": 0.5122067975107707,
      "grad_norm": 0.8398559093475342,
      "learning_rate": 1.6586883676400195e-05,
      "loss": 1.1281,
      "mean_token_accuracy": 0.731948834657669,
      "num_tokens": 17526143.0,
      "step": 2140
    },
    {
      "entropy": 1.0755485057830811,
      "epoch": 0.5146002872187649,
      "grad_norm": 0.6819499135017395,
      "learning_rate": 1.6570927078346897e-05,
      "loss": 1.0815,
      "mean_token_accuracy": 0.7402812361717224,
      "num_tokens": 17607490.0,
      "step": 2150
    },
    {
      "entropy": 1.0722748279571532,
      "epoch": 0.5169937769267592,
      "grad_norm": 0.8138436675071716,
      "learning_rate": 1.65549704802936e-05,
      "loss": 1.0765,
      "mean_token_accuracy": 0.7441859841346741,
      "num_tokens": 17695308.0,
      "step": 2160
    },
    {
      "entropy": 1.1051388561725617,
      "epoch": 0.5193872666347534,
      "grad_norm": 0.8223373293876648,
      "learning_rate": 1.6539013882240306e-05,
      "loss": 1.1184,
      "mean_token_accuracy": 0.735604465007782,
      "num_tokens": 17779682.0,
      "step": 2170
    },
    {
      "entropy": 1.145215743780136,
      "epoch": 0.5217807563427477,
      "grad_norm": 0.6529273986816406,
      "learning_rate": 1.652305728418701e-05,
      "loss": 1.1329,
      "mean_token_accuracy": 0.7334801733493805,
      "num_tokens": 17859679.0,
      "step": 2180
    },
    {
      "entropy": 0.963028508424759,
      "epoch": 0.524174246050742,
      "grad_norm": 0.6153014898300171,
      "learning_rate": 1.6507100686133716e-05,
      "loss": 0.9805,
      "mean_token_accuracy": 0.7606176972389221,
      "num_tokens": 17935802.0,
      "step": 2190
    },
    {
      "entropy": 1.0729603230953217,
      "epoch": 0.5265677357587363,
      "grad_norm": 1.7113527059555054,
      "learning_rate": 1.649114408808042e-05,
      "loss": 1.0881,
      "mean_token_accuracy": 0.741437292098999,
      "num_tokens": 18011548.0,
      "step": 2200
    },
    {
      "entropy": 1.0923176765441895,
      "epoch": 0.5289612254667305,
      "grad_norm": 0.6712521910667419,
      "learning_rate": 1.6475187490027126e-05,
      "loss": 1.0847,
      "mean_token_accuracy": 0.7404011368751526,
      "num_tokens": 18093062.0,
      "step": 2210
    },
    {
      "entropy": 1.0267192900180817,
      "epoch": 0.5313547151747248,
      "grad_norm": 0.8290082216262817,
      "learning_rate": 1.645923089197383e-05,
      "loss": 1.0312,
      "mean_token_accuracy": 0.7513166487216949,
      "num_tokens": 18169463.0,
      "step": 2220
    },
    {
      "entropy": 1.0343119144439696,
      "epoch": 0.533748204882719,
      "grad_norm": 0.8802976608276367,
      "learning_rate": 1.6443274293920535e-05,
      "loss": 1.0378,
      "mean_token_accuracy": 0.7520385146141052,
      "num_tokens": 18245259.0,
      "step": 2230
    },
    {
      "entropy": 1.1066267848014832,
      "epoch": 0.5361416945907133,
      "grad_norm": 0.5967556834220886,
      "learning_rate": 1.642731769586724e-05,
      "loss": 1.1279,
      "mean_token_accuracy": 0.7345916450023651,
      "num_tokens": 18327091.0,
      "step": 2240
    },
    {
      "entropy": 1.1284412324428559,
      "epoch": 0.5385351842987075,
      "grad_norm": 0.687362551689148,
      "learning_rate": 1.641136109781395e-05,
      "loss": 1.1253,
      "mean_token_accuracy": 0.7369785964488983,
      "num_tokens": 18410490.0,
      "step": 2250
    },
    {
      "entropy": 1.0592759251594543,
      "epoch": 0.5409286740067017,
      "grad_norm": 1.7939339876174927,
      "learning_rate": 1.6395404499760653e-05,
      "loss": 1.07,
      "mean_token_accuracy": 0.7426570773124694,
      "num_tokens": 18485447.0,
      "step": 2260
    },
    {
      "entropy": 1.0785628974437713,
      "epoch": 0.543322163714696,
      "grad_norm": 1.1665188074111938,
      "learning_rate": 1.637944790170736e-05,
      "loss": 1.092,
      "mean_token_accuracy": 0.7401333570480346,
      "num_tokens": 18574821.0,
      "step": 2270
    },
    {
      "entropy": 1.0812392115592957,
      "epoch": 0.5457156534226902,
      "grad_norm": 0.8480959534645081,
      "learning_rate": 1.6363491303654063e-05,
      "loss": 1.0826,
      "mean_token_accuracy": 0.7413488388061523,
      "num_tokens": 18653584.0,
      "step": 2280
    },
    {
      "entropy": 1.0245578348636628,
      "epoch": 0.5481091431306845,
      "grad_norm": 0.5554677844047546,
      "learning_rate": 1.6347534705600768e-05,
      "loss": 1.0256,
      "mean_token_accuracy": 0.7527715682983398,
      "num_tokens": 18733982.0,
      "step": 2290
    },
    {
      "entropy": 1.1428353428840636,
      "epoch": 0.5505026328386788,
      "grad_norm": 0.5947005152702332,
      "learning_rate": 1.6331578107547473e-05,
      "loss": 1.1498,
      "mean_token_accuracy": 0.7300264656543731,
      "num_tokens": 18822343.0,
      "step": 2300
    },
    {
      "entropy": 1.084150779247284,
      "epoch": 0.5528961225466731,
      "grad_norm": 0.6712626218795776,
      "learning_rate": 1.6315621509494178e-05,
      "loss": 1.0906,
      "mean_token_accuracy": 0.7372301578521728,
      "num_tokens": 18909625.0,
      "step": 2310
    },
    {
      "entropy": 1.0455949902534485,
      "epoch": 0.5552896122546673,
      "grad_norm": 0.7032175064086914,
      "learning_rate": 1.6299664911440883e-05,
      "loss": 1.0544,
      "mean_token_accuracy": 0.7434102296829224,
      "num_tokens": 18990307.0,
      "step": 2320
    },
    {
      "entropy": 1.1317045211791992,
      "epoch": 0.5576831019626616,
      "grad_norm": 0.7299635410308838,
      "learning_rate": 1.6283708313387587e-05,
      "loss": 1.1315,
      "mean_token_accuracy": 0.7348020315170288,
      "num_tokens": 19065995.0,
      "step": 2330
    },
    {
      "entropy": 1.0656555056571961,
      "epoch": 0.5600765916706558,
      "grad_norm": 0.8328961730003357,
      "learning_rate": 1.6267751715334292e-05,
      "loss": 1.0798,
      "mean_token_accuracy": 0.7406236708164216,
      "num_tokens": 19147268.0,
      "step": 2340
    },
    {
      "entropy": 1.0884394943714142,
      "epoch": 0.5624700813786501,
      "grad_norm": 1.196860909461975,
      "learning_rate": 1.6251795117280997e-05,
      "loss": 1.0987,
      "mean_token_accuracy": 0.7410864651203155,
      "num_tokens": 19232016.0,
      "step": 2350
    },
    {
      "entropy": 1.1457077085971832,
      "epoch": 0.5648635710866443,
      "grad_norm": 0.5326634645462036,
      "learning_rate": 1.6235838519227702e-05,
      "loss": 1.1475,
      "mean_token_accuracy": 0.7297524213790894,
      "num_tokens": 19316268.0,
      "step": 2360
    },
    {
      "entropy": 1.0890101790428162,
      "epoch": 0.5672570607946386,
      "grad_norm": 0.7511252760887146,
      "learning_rate": 1.6219881921174407e-05,
      "loss": 1.1012,
      "mean_token_accuracy": 0.7405222654342651,
      "num_tokens": 19398123.0,
      "step": 2370
    },
    {
      "entropy": 1.013045358657837,
      "epoch": 0.5696505505026328,
      "grad_norm": 0.6571887135505676,
      "learning_rate": 1.6203925323121112e-05,
      "loss": 1.019,
      "mean_token_accuracy": 0.7532030701637268,
      "num_tokens": 19478937.0,
      "step": 2380
    },
    {
      "entropy": 1.0578601121902467,
      "epoch": 0.572044040210627,
      "grad_norm": 0.7799140214920044,
      "learning_rate": 1.6187968725067817e-05,
      "loss": 1.0626,
      "mean_token_accuracy": 0.747768622636795,
      "num_tokens": 19564031.0,
      "step": 2390
    },
    {
      "entropy": 1.0582027733325958,
      "epoch": 0.5744375299186214,
      "grad_norm": 0.9500547051429749,
      "learning_rate": 1.617201212701452e-05,
      "loss": 1.048,
      "mean_token_accuracy": 0.7498915076255799,
      "num_tokens": 19638358.0,
      "step": 2400
    },
    {
      "entropy": 1.0389926135540009,
      "epoch": 0.5768310196266156,
      "grad_norm": 0.650881826877594,
      "learning_rate": 1.6156055528961226e-05,
      "loss": 1.0605,
      "mean_token_accuracy": 0.7405982375144958,
      "num_tokens": 19717962.0,
      "step": 2410
    },
    {
      "entropy": 0.9749106526374817,
      "epoch": 0.5792245093346099,
      "grad_norm": 0.69362872838974,
      "learning_rate": 1.614009893090793e-05,
      "loss": 0.9837,
      "mean_token_accuracy": 0.7568762242794037,
      "num_tokens": 19801772.0,
      "step": 2420
    },
    {
      "entropy": 1.0389947533607482,
      "epoch": 0.5816179990426041,
      "grad_norm": 0.8371346592903137,
      "learning_rate": 1.6124142332854636e-05,
      "loss": 1.0396,
      "mean_token_accuracy": 0.7515539228916168,
      "num_tokens": 19880716.0,
      "step": 2430
    },
    {
      "entropy": 1.0958311080932617,
      "epoch": 0.5840114887505984,
      "grad_norm": 0.7574942111968994,
      "learning_rate": 1.610818573480134e-05,
      "loss": 1.1041,
      "mean_token_accuracy": 0.7367929995059967,
      "num_tokens": 19958372.0,
      "step": 2440
    },
    {
      "entropy": 1.0276429772377014,
      "epoch": 0.5864049784585926,
      "grad_norm": 0.6537715792655945,
      "learning_rate": 1.609222913674805e-05,
      "loss": 1.0355,
      "mean_token_accuracy": 0.7487654328346253,
      "num_tokens": 20038953.0,
      "step": 2450
    },
    {
      "entropy": 1.050876259803772,
      "epoch": 0.5887984681665869,
      "grad_norm": 0.6600715517997742,
      "learning_rate": 1.6076272538694754e-05,
      "loss": 1.0614,
      "mean_token_accuracy": 0.7478480696678161,
      "num_tokens": 20121410.0,
      "step": 2460
    },
    {
      "entropy": 1.0625569701194764,
      "epoch": 0.5911919578745811,
      "grad_norm": 0.7250789403915405,
      "learning_rate": 1.6060315940641456e-05,
      "loss": 1.0555,
      "mean_token_accuracy": 0.7494634866714478,
      "num_tokens": 20201415.0,
      "step": 2470
    },
    {
      "entropy": 1.006362384557724,
      "epoch": 0.5935854475825754,
      "grad_norm": 0.8562338352203369,
      "learning_rate": 1.604435934258816e-05,
      "loss": 1.0178,
      "mean_token_accuracy": 0.7520220935344696,
      "num_tokens": 20283377.0,
      "step": 2480
    },
    {
      "entropy": 1.0587669372558595,
      "epoch": 0.5959789372905696,
      "grad_norm": 1.034611701965332,
      "learning_rate": 1.6028402744534865e-05,
      "loss": 1.0628,
      "mean_token_accuracy": 0.7472517371177674,
      "num_tokens": 20370012.0,
      "step": 2490
    },
    {
      "entropy": 1.0843224465847014,
      "epoch": 0.5983724269985639,
      "grad_norm": 2.8688549995422363,
      "learning_rate": 1.601244614648157e-05,
      "loss": 1.0896,
      "mean_token_accuracy": 0.737509822845459,
      "num_tokens": 20448279.0,
      "step": 2500
    }
  ],
  "logging_steps": 10,
  "max_steps": 12534,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.570634216601551e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}