{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5163705604234239,
  "eval_steps": 500,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.314144600741565,
      "epoch": 0.005163705604234238,
      "grad_norm": 4.290224552154541,
      "learning_rate": 1.5254237288135596e-06,
      "loss": 4.511758422851562,
      "mean_token_accuracy": 0.3633482086937875,
      "num_tokens": 57862.0,
      "step": 10
    },
    {
      "entropy": 2.3062521956861017,
      "epoch": 0.010327411208468477,
      "grad_norm": 4.981228828430176,
      "learning_rate": 3.2203389830508473e-06,
      "loss": 4.534395980834961,
      "mean_token_accuracy": 0.3670886025764048,
      "num_tokens": 114505.0,
      "step": 20
    },
    {
      "entropy": 2.445302851125598,
      "epoch": 0.015491116812702715,
      "grad_norm": 3.324716091156006,
      "learning_rate": 4.915254237288136e-06,
      "loss": 4.264028549194336,
      "mean_token_accuracy": 0.37419168511405587,
      "num_tokens": 177985.0,
      "step": 30
    },
    {
      "entropy": 2.5100959464907646,
      "epoch": 0.020654822416936953,
      "grad_norm": 2.0952486991882324,
      "learning_rate": 6.610169491525424e-06,
      "loss": 3.9582889556884764,
      "mean_token_accuracy": 0.3963044578209519,
      "num_tokens": 239576.0,
      "step": 40
    },
    {
      "entropy": 2.665686187893152,
      "epoch": 0.025818528021171194,
      "grad_norm": 1.5603235960006714,
      "learning_rate": 8.305084745762712e-06,
      "loss": 3.556003189086914,
      "mean_token_accuracy": 0.4520953473169357,
      "num_tokens": 291640.0,
      "step": 50
    },
    {
      "entropy": 2.741910447925329,
      "epoch": 0.03098223362540543,
      "grad_norm": 0.9505864977836609,
      "learning_rate": 1e-05,
      "loss": 3.2364212036132813,
      "mean_token_accuracy": 0.5039095370098948,
      "num_tokens": 349011.0,
      "step": 60
    },
    {
      "entropy": 2.7771890498697758,
      "epoch": 0.03614593922963967,
      "grad_norm": 0.7826384902000427,
      "learning_rate": 9.999300418283908e-06,
      "loss": 3.075259208679199,
      "mean_token_accuracy": 0.522175510879606,
      "num_tokens": 407532.0,
      "step": 70
    },
    {
      "entropy": 2.6670482574030756,
      "epoch": 0.04130964483387391,
      "grad_norm": 0.7306021451950073,
      "learning_rate": 9.997201868901463e-06,
      "loss": 2.959842300415039,
      "mean_token_accuracy": 0.5392204709351063,
      "num_tokens": 464814.0,
      "step": 80
    },
    {
      "entropy": 2.534086942858994,
      "epoch": 0.04647335043810815,
      "grad_norm": 0.7267245650291443,
      "learning_rate": 9.993704939095376e-06,
      "loss": 2.857519340515137,
      "mean_token_accuracy": 0.5584392322227358,
      "num_tokens": 523494.0,
      "step": 90
    },
    {
      "entropy": 2.48643469568342,
      "epoch": 0.05163705604234239,
      "grad_norm": 0.9745510816574097,
      "learning_rate": 9.988810607420912e-06,
      "loss": 2.770132637023926,
      "mean_token_accuracy": 0.5674901144579053,
      "num_tokens": 578916.0,
      "step": 100
    },
    {
      "entropy": 2.3004672840237617,
      "epoch": 0.056800761646576624,
      "grad_norm": 0.9000684022903442,
      "learning_rate": 9.982520243472044e-06,
      "loss": 2.6292037963867188,
      "mean_token_accuracy": 0.5810430168174208,
      "num_tokens": 640210.0,
      "step": 110
    },
    {
      "entropy": 2.309546428732574,
      "epoch": 0.06196446725081086,
      "grad_norm": 0.9355388283729553,
      "learning_rate": 9.974835607498224e-06,
      "loss": 2.623982620239258,
      "mean_token_accuracy": 0.5811452951282263,
      "num_tokens": 698620.0,
      "step": 120
    },
    {
      "entropy": 2.3742345267906786,
      "epoch": 0.0671281728550451,
      "grad_norm": 0.9847837090492249,
      "learning_rate": 9.965758849911774e-06,
      "loss": 2.670285224914551,
      "mean_token_accuracy": 0.5726205869577825,
      "num_tokens": 764363.0,
      "step": 130
    },
    {
      "entropy": 2.21920911911875,
      "epoch": 0.07229187845927934,
      "grad_norm": 1.1952255964279175,
      "learning_rate": 9.955292510686156e-06,
      "loss": 2.527632141113281,
      "mean_token_accuracy": 0.5953943591564894,
      "num_tokens": 821339.0,
      "step": 140
    },
    {
      "entropy": 2.1673635978251697,
      "epoch": 0.07745558406351358,
      "grad_norm": 1.1929246187210083,
      "learning_rate": 9.943439518645193e-06,
      "loss": 2.4985078811645507,
      "mean_token_accuracy": 0.6023027936927974,
      "num_tokens": 871877.0,
      "step": 150
    },
    {
      "entropy": 2.211511859484017,
      "epoch": 0.08261928966774781,
      "grad_norm": 1.358289122581482,
      "learning_rate": 9.930203190643491e-06,
      "loss": 2.500803756713867,
      "mean_token_accuracy": 0.5940627983771265,
      "num_tokens": 929202.0,
      "step": 160
    },
    {
      "entropy": 2.1249462071806193,
      "epoch": 0.08778299527198205,
      "grad_norm": 1.2578144073486328,
      "learning_rate": 9.915587230638269e-06,
      "loss": 2.4678808212280274,
      "mean_token_accuracy": 0.6021689807996153,
      "num_tokens": 987967.0,
      "step": 170
    },
    {
      "entropy": 2.0816542433574794,
      "epoch": 0.0929467008762163,
      "grad_norm": 1.4549826383590698,
      "learning_rate": 9.899595728652883e-06,
      "loss": 2.385122871398926,
      "mean_token_accuracy": 0.6153967375867069,
      "num_tokens": 1042757.0,
      "step": 180
    },
    {
      "entropy": 2.01310737170279,
      "epoch": 0.09811040648045054,
      "grad_norm": 1.3826615810394287,
      "learning_rate": 9.882233159632297e-06,
      "loss": 2.281660270690918,
      "mean_token_accuracy": 0.6195336611010134,
      "num_tokens": 1096431.0,
      "step": 190
    },
    {
      "entropy": 2.036235300358385,
      "epoch": 0.10327411208468477,
      "grad_norm": 1.4376295804977417,
      "learning_rate": 9.863504382190838e-06,
      "loss": 2.3184206008911135,
      "mean_token_accuracy": 0.6207408639602363,
      "num_tokens": 1151395.0,
      "step": 200
    },
    {
      "entropy": 2.032619636319578,
      "epoch": 0.10843781768891901,
      "grad_norm": 1.5430495738983154,
      "learning_rate": 9.843414637252615e-06,
      "loss": 2.307057571411133,
      "mean_token_accuracy": 0.6216312819160521,
      "num_tokens": 1206304.0,
      "step": 210
    },
    {
      "entropy": 1.993878381885588,
      "epoch": 0.11360152329315325,
      "grad_norm": 1.6275994777679443,
      "learning_rate": 9.821969546584922e-06,
      "loss": 2.3098770141601563,
      "mean_token_accuracy": 0.6226980553008616,
      "num_tokens": 1263118.0,
      "step": 220
    },
    {
      "entropy": 1.958546225167811,
      "epoch": 0.11876522889738748,
      "grad_norm": 1.6821534633636475,
      "learning_rate": 9.79917511122509e-06,
      "loss": 2.2890209197998046,
      "mean_token_accuracy": 0.6257161197252572,
      "num_tokens": 1321193.0,
      "step": 230
    },
    {
      "entropy": 1.9737237187102437,
      "epoch": 0.12392893450162172,
      "grad_norm": 1.6766833066940308,
      "learning_rate": 9.775037709801206e-06,
      "loss": 2.2315126419067384,
      "mean_token_accuracy": 0.6283347339369356,
      "num_tokens": 1379782.0,
      "step": 240
    },
    {
      "entropy": 1.9693593598902226,
      "epoch": 0.12909264010585597,
      "grad_norm": 1.7196708917617798,
      "learning_rate": 9.749564096747148e-06,
      "loss": 2.2683509826660155,
      "mean_token_accuracy": 0.6241073801182211,
      "num_tokens": 1436026.0,
      "step": 250
    },
    {
      "entropy": 2.006697486527264,
      "epoch": 0.1342563457100902,
      "grad_norm": 1.725196361541748,
      "learning_rate": 9.722761400412496e-06,
      "loss": 2.237827110290527,
      "mean_token_accuracy": 0.618116126768291,
      "num_tokens": 1491038.0,
      "step": 260
    },
    {
      "entropy": 1.9512792754918338,
      "epoch": 0.13942005131432444,
      "grad_norm": 1.7161600589752197,
      "learning_rate": 9.694637121067764e-06,
      "loss": 2.2142683029174806,
      "mean_token_accuracy": 0.630404542479664,
      "num_tokens": 1549020.0,
      "step": 270
    },
    {
      "entropy": 1.911211796849966,
      "epoch": 0.14458375691855868,
      "grad_norm": 1.7496988773345947,
      "learning_rate": 9.6651991288056e-06,
      "loss": 2.1877910614013674,
      "mean_token_accuracy": 0.6333114374428988,
      "num_tokens": 1603318.0,
      "step": 280
    },
    {
      "entropy": 1.9003798129037022,
      "epoch": 0.14974746252279292,
      "grad_norm": 1.8355860710144043,
      "learning_rate": 9.63445566133846e-06,
      "loss": 2.129390525817871,
      "mean_token_accuracy": 0.6355364937335253,
      "num_tokens": 1659733.0,
      "step": 290
    },
    {
      "entropy": 1.8534355744719506,
      "epoch": 0.15491116812702715,
      "grad_norm": 2.051910638809204,
      "learning_rate": 9.602415321693434e-06,
      "loss": 2.175747299194336,
      "mean_token_accuracy": 0.6399295572191477,
      "num_tokens": 1717989.0,
      "step": 300
    },
    {
      "entropy": 1.9231654025614262,
      "epoch": 0.1600748737312614,
      "grad_norm": 2.169067621231079,
      "learning_rate": 9.569087075804842e-06,
      "loss": 2.1539403915405275,
      "mean_token_accuracy": 0.6277882568538189,
      "num_tokens": 1775607.0,
      "step": 310
    },
    {
      "entropy": 1.8140981450676918,
      "epoch": 0.16523857933549563,
      "grad_norm": 2.2599377632141113,
      "learning_rate": 9.534480250005263e-06,
      "loss": 2.035861587524414,
      "mean_token_accuracy": 0.650528599601239,
      "num_tokens": 1828714.0,
      "step": 320
    },
    {
      "entropy": 1.8829220661893487,
      "epoch": 0.17040228493972986,
      "grad_norm": 1.9995423555374146,
      "learning_rate": 9.498604528415731e-06,
      "loss": 2.1237022399902346,
      "mean_token_accuracy": 0.6376794545911253,
      "num_tokens": 1888678.0,
      "step": 330
    },
    {
      "entropy": 1.8207894021645188,
      "epoch": 0.1755659905439641,
      "grad_norm": 1.9134663343429565,
      "learning_rate": 9.461469950235795e-06,
      "loss": 2.097091484069824,
      "mean_token_accuracy": 0.6478853410109877,
      "num_tokens": 1947050.0,
      "step": 340
    },
    {
      "entropy": 1.8442208072170616,
      "epoch": 0.18072969614819834,
      "grad_norm": 1.9808735847473145,
      "learning_rate": 9.423086906934228e-06,
      "loss": 2.088273811340332,
      "mean_token_accuracy": 0.6407218304462731,
      "num_tokens": 2008047.0,
      "step": 350
    },
    {
      "entropy": 1.8381939049810172,
      "epoch": 0.1858934017524326,
      "grad_norm": 2.294248580932617,
      "learning_rate": 9.38346613934115e-06,
      "loss": 2.0682788848876954,
      "mean_token_accuracy": 0.6440740385092795,
      "num_tokens": 2067767.0,
      "step": 360
    },
    {
      "entropy": 1.8192341944202781,
      "epoch": 0.19105710735666684,
      "grad_norm": 2.4948010444641113,
      "learning_rate": 9.342618734642395e-06,
      "loss": 2.0498260498046874,
      "mean_token_accuracy": 0.6429526777938008,
      "num_tokens": 2123254.0,
      "step": 370
    },
    {
      "entropy": 1.8355431320145725,
      "epoch": 0.19622081296090108,
      "grad_norm": 2.3000476360321045,
      "learning_rate": 9.300556123276955e-06,
      "loss": 2.0951522827148437,
      "mean_token_accuracy": 0.6433267280459404,
      "num_tokens": 2182713.0,
      "step": 380
    },
    {
      "entropy": 1.7422058419324458,
      "epoch": 0.2013845185651353,
      "grad_norm": 2.5333197116851807,
      "learning_rate": 9.257290075738365e-06,
      "loss": 2.0272647857666017,
      "mean_token_accuracy": 0.6579547147266567,
      "num_tokens": 2237830.0,
      "step": 390
    },
    {
      "entropy": 1.835691255517304,
      "epoch": 0.20654822416936955,
      "grad_norm": 2.446470260620117,
      "learning_rate": 9.212832699280942e-06,
      "loss": 2.0740291595458986,
      "mean_token_accuracy": 0.6418645719066263,
      "num_tokens": 2297593.0,
      "step": 400
    },
    {
      "entropy": 1.7601716231554747,
      "epoch": 0.21171192977360379,
      "grad_norm": 2.3453147411346436,
      "learning_rate": 9.16719643453177e-06,
      "loss": 2.0224067687988283,
      "mean_token_accuracy": 0.6526825280860067,
      "num_tokens": 2358420.0,
      "step": 410
    },
    {
      "entropy": 1.7842545576393605,
      "epoch": 0.21687563537783802,
      "grad_norm": 2.343998908996582,
      "learning_rate": 9.120394052009412e-06,
      "loss": 2.0284013748168945,
      "mean_token_accuracy": 0.6491527157835663,
      "num_tokens": 2413411.0,
      "step": 420
    },
    {
      "entropy": 1.8035912558436393,
      "epoch": 0.22203934098207226,
      "grad_norm": 2.4624032974243164,
      "learning_rate": 9.072438648550304e-06,
      "loss": 2.0193674087524416,
      "mean_token_accuracy": 0.64464613776654,
      "num_tokens": 2474489.0,
      "step": 430
    },
    {
      "entropy": 1.770039344392717,
      "epoch": 0.2272030465863065,
      "grad_norm": 2.3914804458618164,
      "learning_rate": 9.023343643643821e-06,
      "loss": 1.9908281326293946,
      "mean_token_accuracy": 0.6467450003139674,
      "num_tokens": 2534478.0,
      "step": 440
    },
    {
      "entropy": 1.7397029646672308,
      "epoch": 0.23236675219054073,
      "grad_norm": 2.477668285369873,
      "learning_rate": 8.973122775677078e-06,
      "loss": 1.9814176559448242,
      "mean_token_accuracy": 0.6576142696663737,
      "num_tokens": 2588626.0,
      "step": 450
    },
    {
      "entropy": 1.7738654158078133,
      "epoch": 0.23753045779477497,
      "grad_norm": 2.0178322792053223,
      "learning_rate": 8.921790098090477e-06,
      "loss": 1.9442926406860352,
      "mean_token_accuracy": 0.6512955661863089,
      "num_tokens": 2647237.0,
      "step": 460
    },
    {
      "entropy": 1.7735189321450888,
      "epoch": 0.2426941633990092,
      "grad_norm": 2.549083948135376,
      "learning_rate": 8.869359975445085e-06,
      "loss": 1.9767679214477538,
      "mean_token_accuracy": 0.6500832958146929,
      "num_tokens": 2703821.0,
      "step": 470
    },
    {
      "entropy": 1.7094212627038359,
      "epoch": 0.24785786900324344,
      "grad_norm": 2.312490224838257,
      "learning_rate": 8.815847079402972e-06,
      "loss": 1.9500492095947266,
      "mean_token_accuracy": 0.6588690262287855,
      "num_tokens": 2760705.0,
      "step": 480
    },
    {
      "entropy": 1.6987796766683458,
      "epoch": 0.2530215746074777,
      "grad_norm": 2.1077768802642822,
      "learning_rate": 8.761266384621599e-06,
      "loss": 1.8798688888549804,
      "mean_token_accuracy": 0.6636769481934607,
      "num_tokens": 2816194.0,
      "step": 490
    },
    {
      "entropy": 1.771947745513171,
      "epoch": 0.25818528021171194,
      "grad_norm": 2.3342082500457764,
      "learning_rate": 8.705633164563413e-06,
      "loss": 1.9463973999023438,
      "mean_token_accuracy": 0.6494354791007936,
      "num_tokens": 2875579.0,
      "step": 500
    },
    {
      "entropy": 1.6847813020460307,
      "epoch": 0.2633489858159462,
      "grad_norm": 2.528355121612549,
      "learning_rate": 8.648962987221837e-06,
      "loss": 1.8783418655395507,
      "mean_token_accuracy": 0.6657532411627471,
      "num_tokens": 2935829.0,
      "step": 510
    },
    {
      "entropy": 1.7175002202391625,
      "epoch": 0.2685126914201804,
      "grad_norm": 2.8453621864318848,
      "learning_rate": 8.591271710764839e-06,
      "loss": 1.9445688247680664,
      "mean_token_accuracy": 0.66049327487126,
      "num_tokens": 2998604.0,
      "step": 520
    },
    {
      "entropy": 1.6917494175955654,
      "epoch": 0.27367639702441465,
      "grad_norm": 2.5570693016052246,
      "learning_rate": 8.532575479097294e-06,
      "loss": 1.9342041015625,
      "mean_token_accuracy": 0.6653022102080286,
      "num_tokens": 3057855.0,
      "step": 530
    },
    {
      "entropy": 1.6881575733423233,
      "epoch": 0.2788401026286489,
      "grad_norm": 2.789048433303833,
      "learning_rate": 8.472890717343391e-06,
      "loss": 1.9400022506713868,
      "mean_token_accuracy": 0.6622203260660171,
      "num_tokens": 3117202.0,
      "step": 540
    },
    {
      "entropy": 1.6826916345395149,
      "epoch": 0.2840038082328831,
      "grad_norm": 2.3692917823791504,
      "learning_rate": 8.412234127250353e-06,
      "loss": 1.9006750106811523,
      "mean_token_accuracy": 0.6645334579981863,
      "num_tokens": 3172362.0,
      "step": 550
    },
    {
      "entropy": 1.766061083227396,
      "epoch": 0.28916751383711736,
      "grad_norm": 2.456371545791626,
      "learning_rate": 8.350622682514735e-06,
      "loss": 1.921757125854492,
      "mean_token_accuracy": 0.6515732076019048,
      "num_tokens": 3232155.0,
      "step": 560
    },
    {
      "entropy": 1.6578089219518006,
      "epoch": 0.2943312194413516,
      "grad_norm": 2.7018346786499023,
      "learning_rate": 8.288073624032634e-06,
      "loss": 1.8441638946533203,
      "mean_token_accuracy": 0.6680269423872233,
      "num_tokens": 3286329.0,
      "step": 570
    },
    {
      "entropy": 1.654622305277735,
      "epoch": 0.29949492504558584,
      "grad_norm": 4.346279621124268,
      "learning_rate": 8.224604455075115e-06,
      "loss": 1.8432125091552733,
      "mean_token_accuracy": 0.6731441116891801,
      "num_tokens": 3342818.0,
      "step": 580
    },
    {
      "entropy": 1.6787990959361196,
      "epoch": 0.3046586306498201,
      "grad_norm": 3.129047393798828,
      "learning_rate": 8.160232936390239e-06,
      "loss": 1.8712753295898437,
      "mean_token_accuracy": 0.6667228513397276,
      "num_tokens": 3400376.0,
      "step": 590
    },
    {
      "entropy": 1.6542711967602373,
      "epoch": 0.3098223362540543,
      "grad_norm": 2.696822166442871,
      "learning_rate": 8.094977081233006e-06,
      "loss": 1.8849586486816405,
      "mean_token_accuracy": 0.6719589663669467,
      "num_tokens": 3456541.0,
      "step": 600
    },
    {
      "entropy": 1.6884654670953751,
      "epoch": 0.31498604185828855,
      "grad_norm": 2.8305869102478027,
      "learning_rate": 8.02885515032467e-06,
      "loss": 1.8814594268798828,
      "mean_token_accuracy": 0.6616028868593276,
      "num_tokens": 3517505.0,
      "step": 610
    },
    {
      "entropy": 1.6523817678913475,
      "epoch": 0.3201497474625228,
      "grad_norm": 2.9480113983154297,
      "learning_rate": 7.961885646742793e-06,
      "loss": 1.8499563217163086,
      "mean_token_accuracy": 0.6699530448764562,
      "num_tokens": 3576873.0,
      "step": 620
    },
    {
      "entropy": 1.6591516112908722,
      "epoch": 0.325313453066757,
      "grad_norm": 2.808748722076416,
      "learning_rate": 7.894087310743468e-06,
      "loss": 1.8640859603881836,
      "mean_token_accuracy": 0.6685048970393836,
      "num_tokens": 3635972.0,
      "step": 630
    },
    {
      "entropy": 1.6725056597962975,
      "epoch": 0.33047715867099126,
      "grad_norm": 2.86258602142334,
      "learning_rate": 7.825479114517197e-06,
      "loss": 1.8851621627807618,
      "mean_token_accuracy": 0.6656569407321513,
      "num_tokens": 3695665.0,
      "step": 640
    },
    {
      "entropy": 1.7031017780303954,
      "epoch": 0.3356408642752255,
      "grad_norm": 2.7420291900634766,
      "learning_rate": 7.756080256879837e-06,
      "loss": 1.9014200210571288,
      "mean_token_accuracy": 0.6594709505327045,
      "num_tokens": 3756825.0,
      "step": 650
    },
    {
      "entropy": 1.6275493199937046,
      "epoch": 0.34080456987945973,
      "grad_norm": 2.4571967124938965,
      "learning_rate": 7.685910157900158e-06,
      "loss": 1.8154626846313477,
      "mean_token_accuracy": 0.6747128195129335,
      "num_tokens": 3810252.0,
      "step": 660
    },
    {
      "entropy": 1.6585977658629418,
      "epoch": 0.34596827548369397,
      "grad_norm": 2.930661678314209,
      "learning_rate": 7.614988453465469e-06,
      "loss": 1.8117719650268556,
      "mean_token_accuracy": 0.6702397681772709,
      "num_tokens": 3867849.0,
      "step": 670
    },
    {
      "entropy": 1.613425246719271,
      "epoch": 0.3511319810879282,
      "grad_norm": 2.6743385791778564,
      "learning_rate": 7.5433349897868445e-06,
      "loss": 1.8200990676879882,
      "mean_token_accuracy": 0.6785018056631088,
      "num_tokens": 3923617.0,
      "step": 680
    },
    {
      "entropy": 1.6036935173906386,
      "epoch": 0.35629568669216244,
      "grad_norm": 3.227348804473877,
      "learning_rate": 7.470969817845518e-06,
      "loss": 1.8284185409545899,
      "mean_token_accuracy": 0.677451788727194,
      "num_tokens": 3982408.0,
      "step": 690
    },
    {
      "entropy": 1.6516794299706816,
      "epoch": 0.3614593922963967,
      "grad_norm": 2.5348830223083496,
      "learning_rate": 7.397913187781962e-06,
      "loss": 1.7846315383911133,
      "mean_token_accuracy": 0.667611647490412,
      "num_tokens": 4042044.0,
      "step": 700
    },
    {
      "entropy": 1.6284298423677683,
      "epoch": 0.36662309790063097,
      "grad_norm": 2.7191591262817383,
      "learning_rate": 7.324185543229226e-06,
      "loss": 1.7883028030395507,
      "mean_token_accuracy": 0.6704605929553509,
      "num_tokens": 4101051.0,
      "step": 710
    },
    {
      "entropy": 1.6191234907135368,
      "epoch": 0.3717868035048652,
      "grad_norm": 3.107022285461426,
      "learning_rate": 7.249807515592149e-06,
      "loss": 1.8238798141479493,
      "mean_token_accuracy": 0.6791106869466603,
      "num_tokens": 4155970.0,
      "step": 720
    },
    {
      "entropy": 1.608538744226098,
      "epoch": 0.37695050910909944,
      "grad_norm": 2.7626278400421143,
      "learning_rate": 7.174799918274018e-06,
      "loss": 1.8144886016845703,
      "mean_token_accuracy": 0.6752197206951678,
      "num_tokens": 4214152.0,
      "step": 730
    },
    {
      "entropy": 1.5898057647049426,
      "epoch": 0.3821142147133337,
      "grad_norm": 3.050821304321289,
      "learning_rate": 7.099183740852296e-06,
      "loss": 1.7754302978515626,
      "mean_token_accuracy": 0.6882044620811939,
      "num_tokens": 4268661.0,
      "step": 740
    },
    {
      "entropy": 1.589722546003759,
      "epoch": 0.3872779203175679,
      "grad_norm": 2.4175822734832764,
      "learning_rate": 7.022980143205046e-06,
      "loss": 1.7560914993286132,
      "mean_token_accuracy": 0.6823293645866215,
      "num_tokens": 4326013.0,
      "step": 750
    },
    {
      "entropy": 1.617487209662795,
      "epoch": 0.39244162592180215,
      "grad_norm": 3.3936614990234375,
      "learning_rate": 6.946210449589714e-06,
      "loss": 1.7842348098754883,
      "mean_token_accuracy": 0.6780483150854707,
      "num_tokens": 4383920.0,
      "step": 760
    },
    {
      "entropy": 1.533230573683977,
      "epoch": 0.3976053315260364,
      "grad_norm": 2.9665133953094482,
      "learning_rate": 6.868896142675903e-06,
      "loss": 1.710638427734375,
      "mean_token_accuracy": 0.6923438510857522,
      "num_tokens": 4439586.0,
      "step": 770
    },
    {
      "entropy": 1.5793293181806802,
      "epoch": 0.4027690371302706,
      "grad_norm": 3.2671494483947754,
      "learning_rate": 6.791058857533814e-06,
      "loss": 1.7711971282958985,
      "mean_token_accuracy": 0.6793887334875762,
      "num_tokens": 4499436.0,
      "step": 780
    },
    {
      "entropy": 1.6055224198848008,
      "epoch": 0.40793274273450486,
      "grad_norm": 3.0365958213806152,
      "learning_rate": 6.712720375580057e-06,
      "loss": 1.7774797439575196,
      "mean_token_accuracy": 0.6797658964060247,
      "num_tokens": 4560396.0,
      "step": 790
    },
    {
      "entropy": 1.5976737398654222,
      "epoch": 0.4130964483387391,
      "grad_norm": 3.591005802154541,
      "learning_rate": 6.633902618482484e-06,
      "loss": 1.799213981628418,
      "mean_token_accuracy": 0.6778102850541472,
      "num_tokens": 4615989.0,
      "step": 800
    },
    {
      "entropy": 1.5854384305886924,
      "epoch": 0.41826015394297333,
      "grad_norm": 3.1233737468719482,
      "learning_rate": 6.554627642025807e-06,
      "loss": 1.7758049011230468,
      "mean_token_accuracy": 0.6774546066299081,
      "num_tokens": 4672312.0,
      "step": 810
    },
    {
      "entropy": 1.600548622570932,
      "epoch": 0.42342385954720757,
      "grad_norm": 2.7607760429382324,
      "learning_rate": 6.474917629939652e-06,
      "loss": 1.76180419921875,
      "mean_token_accuracy": 0.6780235351994633,
      "num_tokens": 4732253.0,
      "step": 820
    },
    {
      "entropy": 1.5646572889760137,
      "epoch": 0.4285875651514418,
      "grad_norm": 2.9936444759368896,
      "learning_rate": 6.394794887690838e-06,
      "loss": 1.7622745513916016,
      "mean_token_accuracy": 0.6872309047728777,
      "num_tokens": 4790066.0,
      "step": 830
    },
    {
      "entropy": 1.5807129088789225,
      "epoch": 0.43375127075567604,
      "grad_norm": 3.0497989654541016,
      "learning_rate": 6.314281836241573e-06,
      "loss": 1.7779891967773438,
      "mean_token_accuracy": 0.6838998203165829,
      "num_tokens": 4845690.0,
      "step": 840
    },
    {
      "entropy": 1.5713144928216933,
      "epoch": 0.4389149763599103,
      "grad_norm": 3.0101318359375,
      "learning_rate": 6.233401005775339e-06,
      "loss": 1.7724430084228515,
      "mean_token_accuracy": 0.6852510153315962,
      "num_tokens": 4906514.0,
      "step": 850
    },
    {
      "entropy": 1.5716465872712433,
      "epoch": 0.4440786819641445,
      "grad_norm": 3.062901258468628,
      "learning_rate": 6.1521750293922035e-06,
      "loss": 1.728415298461914,
      "mean_token_accuracy": 0.6828448900021613,
      "num_tokens": 4962657.0,
      "step": 860
    },
    {
      "entropy": 1.5911185608245433,
      "epoch": 0.44924238756837875,
      "grad_norm": 3.1436402797698975,
      "learning_rate": 6.070626636775349e-06,
      "loss": 1.7529178619384767,
      "mean_token_accuracy": 0.6815405933186411,
      "num_tokens": 5019475.0,
      "step": 870
    },
    {
      "entropy": 1.6022390400990845,
      "epoch": 0.454406093172613,
      "grad_norm": 3.270408868789673,
      "learning_rate": 5.988778647830554e-06,
      "loss": 1.7645608901977539,
      "mean_token_accuracy": 0.6775381861254572,
      "num_tokens": 5074009.0,
      "step": 880
    },
    {
      "entropy": 1.5316935462877155,
      "epoch": 0.4595697987768472,
      "grad_norm": 3.3750274181365967,
      "learning_rate": 5.906653966300444e-06,
      "loss": 1.7380414962768556,
      "mean_token_accuracy": 0.6924085746519267,
      "num_tokens": 5131114.0,
      "step": 890
    },
    {
      "entropy": 1.5163973657414318,
      "epoch": 0.46473350438108146,
      "grad_norm": 3.4683239459991455,
      "learning_rate": 5.824275573355278e-06,
      "loss": 1.7235090255737304,
      "mean_token_accuracy": 0.6898601963184774,
      "num_tokens": 5184643.0,
      "step": 900
    },
    {
      "entropy": 1.5578045804984868,
      "epoch": 0.4698972099853157,
      "grad_norm": 2.7193562984466553,
      "learning_rate": 5.741666521162055e-06,
      "loss": 1.7631567001342774,
      "mean_token_accuracy": 0.6847550798207521,
      "num_tokens": 5245294.0,
      "step": 910
    },
    {
      "entropy": 1.5216822805814445,
      "epoch": 0.47506091558954994,
      "grad_norm": 3.321866273880005,
      "learning_rate": 5.658849926433774e-06,
      "loss": 1.7373876571655273,
      "mean_token_accuracy": 0.6913335792720318,
      "num_tokens": 5299386.0,
      "step": 920
    },
    {
      "entropy": 1.5747642540372908,
      "epoch": 0.4802246211937842,
      "grad_norm": 3.5749261379241943,
      "learning_rate": 5.575848963960621e-06,
      "loss": 1.8702735900878906,
      "mean_token_accuracy": 0.6848366302903741,
      "num_tokens": 5359761.0,
      "step": 930
    },
    {
      "entropy": 1.5551650236360728,
      "epoch": 0.4853883267980184,
      "grad_norm": 2.854201555252075,
      "learning_rate": 5.4926868601249e-06,
      "loss": 1.7384683609008789,
      "mean_token_accuracy": 0.6877018769271672,
      "num_tokens": 5419068.0,
      "step": 940
    },
    {
      "entropy": 1.5700966760516166,
      "epoch": 0.49055203240225265,
      "grad_norm": 3.1503734588623047,
      "learning_rate": 5.4093868864015405e-06,
      "loss": 1.7581310272216797,
      "mean_token_accuracy": 0.6791030686348677,
      "num_tokens": 5478815.0,
      "step": 950
    },
    {
      "entropy": 1.613727968931198,
      "epoch": 0.4957157380064869,
      "grad_norm": 2.6259965896606445,
      "learning_rate": 5.325972352845965e-06,
      "loss": 1.7862756729125977,
      "mean_token_accuracy": 0.6786239412613213,
      "num_tokens": 5540701.0,
      "step": 960
    },
    {
      "entropy": 1.5610787151381373,
      "epoch": 0.5008794436107211,
      "grad_norm": 3.689964532852173,
      "learning_rate": 5.24246660157119e-06,
      "loss": 1.7218122482299805,
      "mean_token_accuracy": 0.6857390251941979,
      "num_tokens": 5600139.0,
      "step": 970
    },
    {
      "entropy": 1.5385319583117962,
      "epoch": 0.5060431492149554,
      "grad_norm": 3.3747434616088867,
      "learning_rate": 5.1588930002159255e-06,
      "loss": 1.7435245513916016,
      "mean_token_accuracy": 0.6888230754062533,
      "num_tokens": 5660146.0,
      "step": 980
    },
    {
      "entropy": 1.5208949162624776,
      "epoch": 0.5112068548191896,
      "grad_norm": 3.128319501876831,
      "learning_rate": 5.075274935405554e-06,
      "loss": 1.7131477355957032,
      "mean_token_accuracy": 0.6928804327733815,
      "num_tokens": 5715836.0,
      "step": 990
    },
    {
      "entropy": 1.570724095776677,
      "epoch": 0.5163705604234239,
      "grad_norm": 3.4249417781829834,
      "learning_rate": 4.991635806207788e-06,
      "loss": 1.740692710876465,
      "mean_token_accuracy": 0.6794734395109117,
      "num_tokens": 5775680.0,
      "step": 1000
    }
  ],
  "logging_steps": 10,
  "max_steps": 1937,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 2.4782378380591104e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}