{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.46731234866828086,
  "eval_steps": 500,
  "global_step": 5597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008349336227769892,
      "grad_norm": 4.207607746124268,
      "learning_rate": 1.1061797752808989e-06,
      "loss": 1.5395,
      "step": 10
    },
    {
      "epoch": 0.0016698672455539784,
      "grad_norm": 2.644784450531006,
      "learning_rate": 2.2241573033707866e-06,
      "loss": 1.3507,
      "step": 20
    },
    {
      "epoch": 0.0025048008683309676,
      "grad_norm": 1.4334214925765991,
      "learning_rate": 3.342134831460675e-06,
      "loss": 1.2368,
      "step": 30
    },
    {
      "epoch": 0.003339734491107957,
      "grad_norm": 1.1798350811004639,
      "learning_rate": 4.4601123595505624e-06,
      "loss": 1.0269,
      "step": 40
    },
    {
      "epoch": 0.004174668113884946,
      "grad_norm": 0.9337137937545776,
      "learning_rate": 5.5780898876404495e-06,
      "loss": 0.835,
      "step": 50
    },
    {
      "epoch": 0.005009601736661935,
      "grad_norm": 0.5626512765884399,
      "learning_rate": 6.6960674157303374e-06,
      "loss": 0.8018,
      "step": 60
    },
    {
      "epoch": 0.0058445353594389245,
      "grad_norm": 0.5007433891296387,
      "learning_rate": 7.814044943820226e-06,
      "loss": 0.7772,
      "step": 70
    },
    {
      "epoch": 0.006679468982215914,
      "grad_norm": 0.4939042329788208,
      "learning_rate": 8.932022471910112e-06,
      "loss": 0.7392,
      "step": 80
    },
    {
      "epoch": 0.007514402604992903,
      "grad_norm": 0.5686920285224915,
      "learning_rate": 1.005e-05,
      "loss": 0.7669,
      "step": 90
    },
    {
      "epoch": 0.008349336227769892,
      "grad_norm": 0.5208467245101929,
      "learning_rate": 1.1167977528089888e-05,
      "loss": 0.7672,
      "step": 100
    },
    {
      "epoch": 0.009184269850546881,
      "grad_norm": 0.4716687500476837,
      "learning_rate": 1.2285955056179776e-05,
      "loss": 0.7388,
      "step": 110
    },
    {
      "epoch": 0.01001920347332387,
      "grad_norm": 0.4500362277030945,
      "learning_rate": 1.3403932584269664e-05,
      "loss": 0.7456,
      "step": 120
    },
    {
      "epoch": 0.01085413709610086,
      "grad_norm": 0.4907788336277008,
      "learning_rate": 1.4521910112359552e-05,
      "loss": 0.7577,
      "step": 130
    },
    {
      "epoch": 0.011689070718877849,
      "grad_norm": 0.46692827343940735,
      "learning_rate": 1.563988764044944e-05,
      "loss": 0.7618,
      "step": 140
    },
    {
      "epoch": 0.012524004341654838,
      "grad_norm": 0.5163733959197998,
      "learning_rate": 1.6757865168539328e-05,
      "loss": 0.7743,
      "step": 150
    },
    {
      "epoch": 0.013358937964431827,
      "grad_norm": 0.459434449672699,
      "learning_rate": 1.7875842696629214e-05,
      "loss": 0.8005,
      "step": 160
    },
    {
      "epoch": 0.014193871587208817,
      "grad_norm": 0.48260498046875,
      "learning_rate": 1.89938202247191e-05,
      "loss": 0.7283,
      "step": 170
    },
    {
      "epoch": 0.015028805209985806,
      "grad_norm": 0.6386128067970276,
      "learning_rate": 2e-05,
      "loss": 0.7521,
      "step": 180
    },
    {
      "epoch": 0.015863738832762795,
      "grad_norm": 0.450025737285614,
      "learning_rate": 2e-05,
      "loss": 0.6999,
      "step": 190
    },
    {
      "epoch": 0.016698672455539784,
      "grad_norm": 0.4777669310569763,
      "learning_rate": 2e-05,
      "loss": 0.7388,
      "step": 200
    },
    {
      "epoch": 0.017533606078316773,
      "grad_norm": 0.44185250997543335,
      "learning_rate": 2e-05,
      "loss": 0.7528,
      "step": 210
    },
    {
      "epoch": 0.018368539701093763,
      "grad_norm": 0.48510533571243286,
      "learning_rate": 2e-05,
      "loss": 0.7419,
      "step": 220
    },
    {
      "epoch": 0.019203473323870752,
      "grad_norm": 0.4384409487247467,
      "learning_rate": 2e-05,
      "loss": 0.76,
      "step": 230
    },
    {
      "epoch": 0.02003840694664774,
      "grad_norm": 0.4512603282928467,
      "learning_rate": 2e-05,
      "loss": 0.7694,
      "step": 240
    },
    {
      "epoch": 0.02087334056942473,
      "grad_norm": 0.4747536778450012,
      "learning_rate": 2e-05,
      "loss": 0.756,
      "step": 250
    },
    {
      "epoch": 0.02170827419220172,
      "grad_norm": 0.4111518859863281,
      "learning_rate": 2e-05,
      "loss": 0.7286,
      "step": 260
    },
    {
      "epoch": 0.02254320781497871,
      "grad_norm": 0.39627984166145325,
      "learning_rate": 2e-05,
      "loss": 0.7802,
      "step": 270
    },
    {
      "epoch": 0.023378141437755698,
      "grad_norm": 0.45961877703666687,
      "learning_rate": 2e-05,
      "loss": 0.7671,
      "step": 280
    },
    {
      "epoch": 0.024213075060532687,
      "grad_norm": 0.4554557502269745,
      "learning_rate": 2e-05,
      "loss": 0.7404,
      "step": 290
    },
    {
      "epoch": 0.025048008683309676,
      "grad_norm": 0.4171745479106903,
      "learning_rate": 2e-05,
      "loss": 0.7177,
      "step": 300
    },
    {
      "epoch": 0.025882942306086665,
      "grad_norm": 0.42183369398117065,
      "learning_rate": 2e-05,
      "loss": 0.7447,
      "step": 310
    },
    {
      "epoch": 0.026717875928863655,
      "grad_norm": 0.9629115462303162,
      "learning_rate": 2e-05,
      "loss": 0.7802,
      "step": 320
    },
    {
      "epoch": 0.027552809551640644,
      "grad_norm": 0.3929850161075592,
      "learning_rate": 2e-05,
      "loss": 0.7022,
      "step": 330
    },
    {
      "epoch": 0.028387743174417633,
      "grad_norm": 0.4160802364349365,
      "learning_rate": 2e-05,
      "loss": 0.7431,
      "step": 340
    },
    {
      "epoch": 0.029222676797194622,
      "grad_norm": 0.43566712737083435,
      "learning_rate": 2e-05,
      "loss": 0.7409,
      "step": 350
    },
    {
      "epoch": 0.03005761041997161,
      "grad_norm": 0.47777512669563293,
      "learning_rate": 2e-05,
      "loss": 0.7661,
      "step": 360
    },
    {
      "epoch": 0.0308925440427486,
      "grad_norm": 0.45962175726890564,
      "learning_rate": 2e-05,
      "loss": 0.7364,
      "step": 370
    },
    {
      "epoch": 0.03172747766552559,
      "grad_norm": 0.4599931240081787,
      "learning_rate": 2e-05,
      "loss": 0.7255,
      "step": 380
    },
    {
      "epoch": 0.03256241128830258,
      "grad_norm": 0.42078301310539246,
      "learning_rate": 2e-05,
      "loss": 0.7568,
      "step": 390
    },
    {
      "epoch": 0.03339734491107957,
      "grad_norm": 0.41342827677726746,
      "learning_rate": 2e-05,
      "loss": 0.746,
      "step": 400
    },
    {
      "epoch": 0.03423227853385656,
      "grad_norm": 0.4753192663192749,
      "learning_rate": 2e-05,
      "loss": 0.7098,
      "step": 410
    },
    {
      "epoch": 0.03506721215663355,
      "grad_norm": 0.42074984312057495,
      "learning_rate": 2e-05,
      "loss": 0.7348,
      "step": 420
    },
    {
      "epoch": 0.03590214577941054,
      "grad_norm": 0.3717345893383026,
      "learning_rate": 2e-05,
      "loss": 0.7215,
      "step": 430
    },
    {
      "epoch": 0.036737079402187525,
      "grad_norm": 0.4751165509223938,
      "learning_rate": 2e-05,
      "loss": 0.7147,
      "step": 440
    },
    {
      "epoch": 0.03757201302496452,
      "grad_norm": 0.44281241297721863,
      "learning_rate": 2e-05,
      "loss": 0.7324,
      "step": 450
    },
    {
      "epoch": 0.038406946647741504,
      "grad_norm": 0.42347100377082825,
      "learning_rate": 2e-05,
      "loss": 0.7286,
      "step": 460
    },
    {
      "epoch": 0.039241880270518496,
      "grad_norm": 0.46966537833213806,
      "learning_rate": 2e-05,
      "loss": 0.7474,
      "step": 470
    },
    {
      "epoch": 0.04007681389329548,
      "grad_norm": 0.4589572548866272,
      "learning_rate": 2e-05,
      "loss": 0.7644,
      "step": 480
    },
    {
      "epoch": 0.040911747516072475,
      "grad_norm": 0.39805424213409424,
      "learning_rate": 2e-05,
      "loss": 0.6949,
      "step": 490
    },
    {
      "epoch": 0.04174668113884946,
      "grad_norm": 0.36458325386047363,
      "learning_rate": 2e-05,
      "loss": 0.7205,
      "step": 500
    },
    {
      "epoch": 0.04258161476162645,
      "grad_norm": 0.457324355840683,
      "learning_rate": 2e-05,
      "loss": 0.7049,
      "step": 510
    },
    {
      "epoch": 0.04341654838440344,
      "grad_norm": 0.39649268984794617,
      "learning_rate": 2e-05,
      "loss": 0.7298,
      "step": 520
    },
    {
      "epoch": 0.04425148200718043,
      "grad_norm": 0.4664701819419861,
      "learning_rate": 2e-05,
      "loss": 0.748,
      "step": 530
    },
    {
      "epoch": 0.04508641562995742,
      "grad_norm": 0.41996926069259644,
      "learning_rate": 2e-05,
      "loss": 0.7382,
      "step": 540
    },
    {
      "epoch": 0.04592134925273441,
      "grad_norm": 0.4385315179824829,
      "learning_rate": 2e-05,
      "loss": 0.7303,
      "step": 550
    },
    {
      "epoch": 0.046756282875511396,
      "grad_norm": 0.3684557378292084,
      "learning_rate": 2e-05,
      "loss": 0.723,
      "step": 560
    },
    {
      "epoch": 0.04759121649828839,
      "grad_norm": 0.4108535051345825,
      "learning_rate": 2e-05,
      "loss": 0.7505,
      "step": 570
    },
    {
      "epoch": 0.048426150121065374,
      "grad_norm": 0.4387914538383484,
      "learning_rate": 2e-05,
      "loss": 0.7649,
      "step": 580
    },
    {
      "epoch": 0.04926108374384237,
      "grad_norm": 0.9331011772155762,
      "learning_rate": 2e-05,
      "loss": 0.7209,
      "step": 590
    },
    {
      "epoch": 0.05009601736661935,
      "grad_norm": 0.4378844201564789,
      "learning_rate": 2e-05,
      "loss": 0.7596,
      "step": 600
    },
    {
      "epoch": 0.050930950989396345,
      "grad_norm": 0.41000136733055115,
      "learning_rate": 2e-05,
      "loss": 0.7652,
      "step": 610
    },
    {
      "epoch": 0.05176588461217333,
      "grad_norm": 0.3994877338409424,
      "learning_rate": 2e-05,
      "loss": 0.7096,
      "step": 620
    },
    {
      "epoch": 0.052600818234950324,
      "grad_norm": 0.4212743937969208,
      "learning_rate": 2e-05,
      "loss": 0.7607,
      "step": 630
    },
    {
      "epoch": 0.05343575185772731,
      "grad_norm": 0.4719401001930237,
      "learning_rate": 2e-05,
      "loss": 0.6994,
      "step": 640
    },
    {
      "epoch": 0.0542706854805043,
      "grad_norm": 0.45361122488975525,
      "learning_rate": 2e-05,
      "loss": 0.7521,
      "step": 650
    },
    {
      "epoch": 0.05510561910328129,
      "grad_norm": 0.3882407546043396,
      "learning_rate": 2e-05,
      "loss": 0.73,
      "step": 660
    },
    {
      "epoch": 0.05594055272605828,
      "grad_norm": 0.3987034559249878,
      "learning_rate": 2e-05,
      "loss": 0.7175,
      "step": 670
    },
    {
      "epoch": 0.056775486348835266,
      "grad_norm": 0.4764518141746521,
      "learning_rate": 2e-05,
      "loss": 0.7383,
      "step": 680
    },
    {
      "epoch": 0.05761041997161226,
      "grad_norm": 0.38849174976348877,
      "learning_rate": 2e-05,
      "loss": 0.7296,
      "step": 690
    },
    {
      "epoch": 0.058445353594389245,
      "grad_norm": 0.40971478819847107,
      "learning_rate": 2e-05,
      "loss": 0.7273,
      "step": 700
    },
    {
      "epoch": 0.05928028721716624,
      "grad_norm": 0.4441114068031311,
      "learning_rate": 2e-05,
      "loss": 0.7249,
      "step": 710
    },
    {
      "epoch": 0.06011522083994322,
      "grad_norm": 0.3833942413330078,
      "learning_rate": 2e-05,
      "loss": 0.7136,
      "step": 720
    },
    {
      "epoch": 0.060950154462720216,
      "grad_norm": 0.3802347779273987,
      "learning_rate": 2e-05,
      "loss": 0.7393,
      "step": 730
    },
    {
      "epoch": 0.0617850880854972,
      "grad_norm": 0.4086751639842987,
      "learning_rate": 2e-05,
      "loss": 0.7152,
      "step": 740
    },
    {
      "epoch": 0.06262002170827419,
      "grad_norm": 0.39086148142814636,
      "learning_rate": 2e-05,
      "loss": 0.7204,
      "step": 750
    },
    {
      "epoch": 0.06345495533105118,
      "grad_norm": 0.4143637418746948,
      "learning_rate": 2e-05,
      "loss": 0.7363,
      "step": 760
    },
    {
      "epoch": 0.06428988895382817,
      "grad_norm": 0.40541625022888184,
      "learning_rate": 2e-05,
      "loss": 0.7047,
      "step": 770
    },
    {
      "epoch": 0.06512482257660517,
      "grad_norm": 0.43499815464019775,
      "learning_rate": 2e-05,
      "loss": 0.7313,
      "step": 780
    },
    {
      "epoch": 0.06595975619938214,
      "grad_norm": 0.4995516836643219,
      "learning_rate": 2e-05,
      "loss": 0.7217,
      "step": 790
    },
    {
      "epoch": 0.06679468982215914,
      "grad_norm": 0.39953678846359253,
      "learning_rate": 2e-05,
      "loss": 0.7396,
      "step": 800
    },
    {
      "epoch": 0.06762962344493613,
      "grad_norm": 0.38989365100860596,
      "learning_rate": 2e-05,
      "loss": 0.7378,
      "step": 810
    },
    {
      "epoch": 0.06846455706771312,
      "grad_norm": 0.5271377563476562,
      "learning_rate": 2e-05,
      "loss": 0.7208,
      "step": 820
    },
    {
      "epoch": 0.0692994906904901,
      "grad_norm": 0.41988226771354675,
      "learning_rate": 2e-05,
      "loss": 0.7384,
      "step": 830
    },
    {
      "epoch": 0.0701344243132671,
      "grad_norm": 0.6820985078811646,
      "learning_rate": 2e-05,
      "loss": 0.721,
      "step": 840
    },
    {
      "epoch": 0.07096935793604409,
      "grad_norm": 0.45111504197120667,
      "learning_rate": 2e-05,
      "loss": 0.7292,
      "step": 850
    },
    {
      "epoch": 0.07180429155882108,
      "grad_norm": 0.42547836899757385,
      "learning_rate": 2e-05,
      "loss": 0.7111,
      "step": 860
    },
    {
      "epoch": 0.07263922518159806,
      "grad_norm": 0.4102250635623932,
      "learning_rate": 2e-05,
      "loss": 0.7237,
      "step": 870
    },
    {
      "epoch": 0.07347415880437505,
      "grad_norm": 0.39390629529953003,
      "learning_rate": 2e-05,
      "loss": 0.7549,
      "step": 880
    },
    {
      "epoch": 0.07430909242715204,
      "grad_norm": 0.45642760396003723,
      "learning_rate": 2e-05,
      "loss": 0.7633,
      "step": 890
    },
    {
      "epoch": 0.07514402604992904,
      "grad_norm": 0.4338949918746948,
      "learning_rate": 2e-05,
      "loss": 0.7234,
      "step": 900
    },
    {
      "epoch": 0.07597895967270601,
      "grad_norm": 0.4457014501094818,
      "learning_rate": 2e-05,
      "loss": 0.697,
      "step": 910
    },
    {
      "epoch": 0.07681389329548301,
      "grad_norm": 0.3864661157131195,
      "learning_rate": 2e-05,
      "loss": 0.7454,
      "step": 920
    },
    {
      "epoch": 0.07764882691826,
      "grad_norm": 0.4558520019054413,
      "learning_rate": 2e-05,
      "loss": 0.7176,
      "step": 930
    },
    {
      "epoch": 0.07848376054103699,
      "grad_norm": 0.40667447447776794,
      "learning_rate": 2e-05,
      "loss": 0.7354,
      "step": 940
    },
    {
      "epoch": 0.07931869416381397,
      "grad_norm": 0.4572715759277344,
      "learning_rate": 2e-05,
      "loss": 0.7536,
      "step": 950
    },
    {
      "epoch": 0.08015362778659096,
      "grad_norm": 0.3982183337211609,
      "learning_rate": 2e-05,
      "loss": 0.7087,
      "step": 960
    },
    {
      "epoch": 0.08098856140936796,
      "grad_norm": 0.3888983428478241,
      "learning_rate": 2e-05,
      "loss": 0.7578,
      "step": 970
    },
    {
      "epoch": 0.08182349503214495,
      "grad_norm": 0.40478619933128357,
      "learning_rate": 2e-05,
      "loss": 0.7126,
      "step": 980
    },
    {
      "epoch": 0.08265842865492193,
      "grad_norm": 0.48754480481147766,
      "learning_rate": 2e-05,
      "loss": 0.7314,
      "step": 990
    },
    {
      "epoch": 0.08349336227769892,
      "grad_norm": 0.38438543677330017,
      "learning_rate": 2e-05,
      "loss": 0.7413,
      "step": 1000
    },
    {
      "epoch": 0.08432829590047591,
      "grad_norm": 0.42465469241142273,
      "learning_rate": 2e-05,
      "loss": 0.7318,
      "step": 1010
    },
    {
      "epoch": 0.0851632295232529,
      "grad_norm": 0.5940709114074707,
      "learning_rate": 2e-05,
      "loss": 0.6863,
      "step": 1020
    },
    {
      "epoch": 0.08599816314602989,
      "grad_norm": 0.374142587184906,
      "learning_rate": 2e-05,
      "loss": 0.7137,
      "step": 1030
    },
    {
      "epoch": 0.08683309676880688,
      "grad_norm": 0.36702439188957214,
      "learning_rate": 2e-05,
      "loss": 0.6892,
      "step": 1040
    },
    {
      "epoch": 0.08766803039158387,
      "grad_norm": 0.363119900226593,
      "learning_rate": 2e-05,
      "loss": 0.7154,
      "step": 1050
    },
    {
      "epoch": 0.08850296401436086,
      "grad_norm": 0.4007730782032013,
      "learning_rate": 2e-05,
      "loss": 0.7001,
      "step": 1060
    },
    {
      "epoch": 0.08933789763713784,
      "grad_norm": 0.3801772892475128,
      "learning_rate": 2e-05,
      "loss": 0.713,
      "step": 1070
    },
    {
      "epoch": 0.09017283125991483,
      "grad_norm": 0.37953799962997437,
      "learning_rate": 2e-05,
      "loss": 0.6904,
      "step": 1080
    },
    {
      "epoch": 0.09100776488269183,
      "grad_norm": 0.3767666518688202,
      "learning_rate": 2e-05,
      "loss": 0.6867,
      "step": 1090
    },
    {
      "epoch": 0.09184269850546882,
      "grad_norm": 0.3752429187297821,
      "learning_rate": 2e-05,
      "loss": 0.6739,
      "step": 1100
    },
    {
      "epoch": 0.0926776321282458,
      "grad_norm": 0.4051717221736908,
      "learning_rate": 2e-05,
      "loss": 0.7001,
      "step": 1110
    },
    {
      "epoch": 0.09351256575102279,
      "grad_norm": 0.389317125082016,
      "learning_rate": 2e-05,
      "loss": 0.7089,
      "step": 1120
    },
    {
      "epoch": 0.09434749937379978,
      "grad_norm": 0.39081355929374695,
      "learning_rate": 2e-05,
      "loss": 0.7491,
      "step": 1130
    },
    {
      "epoch": 0.09518243299657678,
      "grad_norm": 0.9384667873382568,
      "learning_rate": 2e-05,
      "loss": 0.7606,
      "step": 1140
    },
    {
      "epoch": 0.09601736661935376,
      "grad_norm": 0.4086635708808899,
      "learning_rate": 2e-05,
      "loss": 0.6873,
      "step": 1150
    },
    {
      "epoch": 0.09685230024213075,
      "grad_norm": 0.5528045892715454,
      "learning_rate": 2e-05,
      "loss": 0.7105,
      "step": 1160
    },
    {
      "epoch": 0.09768723386490774,
      "grad_norm": 0.4137111306190491,
      "learning_rate": 2e-05,
      "loss": 0.7286,
      "step": 1170
    },
    {
      "epoch": 0.09852216748768473,
      "grad_norm": 0.3765595555305481,
      "learning_rate": 2e-05,
      "loss": 0.7176,
      "step": 1180
    },
    {
      "epoch": 0.09935710111046171,
      "grad_norm": 0.363838791847229,
      "learning_rate": 2e-05,
      "loss": 0.7176,
      "step": 1190
    },
    {
      "epoch": 0.1001920347332387,
      "grad_norm": 0.37655600905418396,
      "learning_rate": 2e-05,
      "loss": 0.72,
      "step": 1200
    },
    {
      "epoch": 0.1010269683560157,
      "grad_norm": 0.4103720784187317,
      "learning_rate": 2e-05,
      "loss": 0.6947,
      "step": 1210
    },
    {
      "epoch": 0.10186190197879269,
      "grad_norm": 0.4416528046131134,
      "learning_rate": 2e-05,
      "loss": 0.6984,
      "step": 1220
    },
    {
      "epoch": 0.10269683560156967,
      "grad_norm": 0.4061998724937439,
      "learning_rate": 2e-05,
      "loss": 0.705,
      "step": 1230
    },
    {
      "epoch": 0.10353176922434666,
      "grad_norm": 0.3562486171722412,
      "learning_rate": 2e-05,
      "loss": 0.6997,
      "step": 1240
    },
    {
      "epoch": 0.10436670284712365,
      "grad_norm": 0.3291058838367462,
      "learning_rate": 2e-05,
      "loss": 0.699,
      "step": 1250
    },
    {
      "epoch": 0.10520163646990065,
      "grad_norm": 0.36545422673225403,
      "learning_rate": 2e-05,
      "loss": 0.7427,
      "step": 1260
    },
    {
      "epoch": 0.10603657009267763,
      "grad_norm": 0.3796237111091614,
      "learning_rate": 2e-05,
      "loss": 0.7175,
      "step": 1270
    },
    {
      "epoch": 0.10687150371545462,
      "grad_norm": 0.3584843873977661,
      "learning_rate": 2e-05,
      "loss": 0.6931,
      "step": 1280
    },
    {
      "epoch": 0.10770643733823161,
      "grad_norm": 0.4039587676525116,
      "learning_rate": 2e-05,
      "loss": 0.709,
      "step": 1290
    },
    {
      "epoch": 0.1085413709610086,
      "grad_norm": 0.5920600295066833,
      "learning_rate": 2e-05,
      "loss": 0.7284,
      "step": 1300
    },
    {
      "epoch": 0.10937630458378558,
      "grad_norm": 0.41448742151260376,
      "learning_rate": 2e-05,
      "loss": 0.6784,
      "step": 1310
    },
    {
      "epoch": 0.11021123820656258,
      "grad_norm": 0.385707288980484,
      "learning_rate": 2e-05,
      "loss": 0.6937,
      "step": 1320
    },
    {
      "epoch": 0.11104617182933957,
      "grad_norm": 0.4448997974395752,
      "learning_rate": 2e-05,
      "loss": 0.7125,
      "step": 1330
    },
    {
      "epoch": 0.11188110545211656,
      "grad_norm": 0.46152302622795105,
      "learning_rate": 2e-05,
      "loss": 0.7389,
      "step": 1340
    },
    {
      "epoch": 0.11271603907489354,
      "grad_norm": 0.3662891089916229,
      "learning_rate": 2e-05,
      "loss": 0.7253,
      "step": 1350
    },
    {
      "epoch": 0.11355097269767053,
      "grad_norm": 0.40756434202194214,
      "learning_rate": 2e-05,
      "loss": 0.6773,
      "step": 1360
    },
    {
      "epoch": 0.11438590632044753,
      "grad_norm": 0.3913949728012085,
      "learning_rate": 2e-05,
      "loss": 0.7047,
      "step": 1370
    },
    {
      "epoch": 0.11522083994322452,
      "grad_norm": 0.3753698170185089,
      "learning_rate": 2e-05,
      "loss": 0.7021,
      "step": 1380
    },
    {
      "epoch": 0.1160557735660015,
      "grad_norm": 0.3978191614151001,
      "learning_rate": 2e-05,
      "loss": 0.7258,
      "step": 1390
    },
    {
      "epoch": 0.11689070718877849,
      "grad_norm": 0.3952433168888092,
      "learning_rate": 2e-05,
      "loss": 0.7125,
      "step": 1400
    },
    {
      "epoch": 0.11772564081155548,
      "grad_norm": 0.37945103645324707,
      "learning_rate": 2e-05,
      "loss": 0.6866,
      "step": 1410
    },
    {
      "epoch": 0.11856057443433247,
      "grad_norm": 0.37956511974334717,
      "learning_rate": 2e-05,
      "loss": 0.7207,
      "step": 1420
    },
    {
      "epoch": 0.11939550805710945,
      "grad_norm": 0.46238476037979126,
      "learning_rate": 2e-05,
      "loss": 0.7116,
      "step": 1430
    },
    {
      "epoch": 0.12023044167988645,
      "grad_norm": 0.5139086842536926,
      "learning_rate": 2e-05,
      "loss": 0.7079,
      "step": 1440
    },
    {
      "epoch": 0.12106537530266344,
      "grad_norm": 0.39234450459480286,
      "learning_rate": 2e-05,
      "loss": 0.7373,
      "step": 1450
    },
    {
      "epoch": 0.12190030892544043,
      "grad_norm": 0.35747602581977844,
      "learning_rate": 2e-05,
      "loss": 0.7352,
      "step": 1460
    },
    {
      "epoch": 0.12273524254821741,
      "grad_norm": 0.3996635675430298,
      "learning_rate": 2e-05,
      "loss": 0.6851,
      "step": 1470
    },
    {
      "epoch": 0.1235701761709944,
      "grad_norm": 0.4240017235279083,
      "learning_rate": 2e-05,
      "loss": 0.6853,
      "step": 1480
    },
    {
      "epoch": 0.1244051097937714,
      "grad_norm": 0.37402406334877014,
      "learning_rate": 2e-05,
      "loss": 0.7145,
      "step": 1490
    },
    {
      "epoch": 0.12524004341654837,
      "grad_norm": 0.3979440927505493,
      "learning_rate": 2e-05,
      "loss": 0.6867,
      "step": 1500
    },
    {
      "epoch": 0.12607497703932538,
      "grad_norm": 0.4218880534172058,
      "learning_rate": 2e-05,
      "loss": 0.7233,
      "step": 1510
    },
    {
      "epoch": 0.12690991066210236,
      "grad_norm": 0.4031672477722168,
      "learning_rate": 2e-05,
      "loss": 0.6906,
      "step": 1520
    },
    {
      "epoch": 0.12774484428487934,
      "grad_norm": 0.38655754923820496,
      "learning_rate": 2e-05,
      "loss": 0.6809,
      "step": 1530
    },
    {
      "epoch": 0.12857977790765635,
      "grad_norm": 0.3993000090122223,
      "learning_rate": 2e-05,
      "loss": 0.6899,
      "step": 1540
    },
    {
      "epoch": 0.12941471153043332,
      "grad_norm": 0.3644014596939087,
      "learning_rate": 2e-05,
      "loss": 0.6774,
      "step": 1550
    },
    {
      "epoch": 0.13024964515321033,
      "grad_norm": 0.38217946887016296,
      "learning_rate": 2e-05,
      "loss": 0.7211,
      "step": 1560
    },
    {
      "epoch": 0.1310845787759873,
      "grad_norm": 0.41936999559402466,
      "learning_rate": 2e-05,
      "loss": 0.7166,
      "step": 1570
    },
    {
      "epoch": 0.1319195123987643,
      "grad_norm": 0.3735484480857849,
      "learning_rate": 2e-05,
      "loss": 0.7103,
      "step": 1580
    },
    {
      "epoch": 0.1327544460215413,
      "grad_norm": 0.3685668706893921,
      "learning_rate": 2e-05,
      "loss": 0.6766,
      "step": 1590
    },
    {
      "epoch": 0.13358937964431827,
      "grad_norm": 0.4097385108470917,
      "learning_rate": 2e-05,
      "loss": 0.6958,
      "step": 1600
    },
    {
      "epoch": 0.13442431326709525,
      "grad_norm": 0.3247581422328949,
      "learning_rate": 2e-05,
      "loss": 0.7003,
      "step": 1610
    },
    {
      "epoch": 0.13525924688987226,
      "grad_norm": 0.3488894999027252,
      "learning_rate": 2e-05,
      "loss": 0.6967,
      "step": 1620
    },
    {
      "epoch": 0.13609418051264924,
      "grad_norm": 0.3828565180301666,
      "learning_rate": 2e-05,
      "loss": 0.7062,
      "step": 1630
    },
    {
      "epoch": 0.13692911413542624,
      "grad_norm": 0.39991557598114014,
      "learning_rate": 2e-05,
      "loss": 0.7122,
      "step": 1640
    },
    {
      "epoch": 0.13776404775820322,
      "grad_norm": 0.38803574442863464,
      "learning_rate": 2e-05,
      "loss": 0.7119,
      "step": 1650
    },
    {
      "epoch": 0.1385989813809802,
      "grad_norm": 0.39794668555259705,
      "learning_rate": 2e-05,
      "loss": 0.7072,
      "step": 1660
    },
    {
      "epoch": 0.1394339150037572,
      "grad_norm": 0.3571108877658844,
      "learning_rate": 2e-05,
      "loss": 0.6655,
      "step": 1670
    },
    {
      "epoch": 0.1402688486265342,
      "grad_norm": 0.3803035020828247,
      "learning_rate": 2e-05,
      "loss": 0.6978,
      "step": 1680
    },
    {
      "epoch": 0.14110378224931117,
      "grad_norm": 0.3964841663837433,
      "learning_rate": 2e-05,
      "loss": 0.7039,
      "step": 1690
    },
    {
      "epoch": 0.14193871587208817,
      "grad_norm": 0.3538469970226288,
      "learning_rate": 2e-05,
      "loss": 0.6992,
      "step": 1700
    },
    {
      "epoch": 0.14277364949486515,
      "grad_norm": 0.33465176820755005,
      "learning_rate": 2e-05,
      "loss": 0.6412,
      "step": 1710
    },
    {
      "epoch": 0.14360858311764216,
      "grad_norm": 0.39754611253738403,
      "learning_rate": 2e-05,
      "loss": 0.6957,
      "step": 1720
    },
    {
      "epoch": 0.14444351674041914,
      "grad_norm": 0.35193371772766113,
      "learning_rate": 2e-05,
      "loss": 0.6739,
      "step": 1730
    },
    {
      "epoch": 0.14527845036319612,
      "grad_norm": 0.36387869715690613,
      "learning_rate": 2e-05,
      "loss": 0.6811,
      "step": 1740
    },
    {
      "epoch": 0.14611338398597312,
      "grad_norm": 0.381991446018219,
      "learning_rate": 2e-05,
      "loss": 0.7192,
      "step": 1750
    },
    {
      "epoch": 0.1469483176087501,
      "grad_norm": 0.3615923523902893,
      "learning_rate": 2e-05,
      "loss": 0.677,
      "step": 1760
    },
    {
      "epoch": 0.1477832512315271,
      "grad_norm": 0.40365681052207947,
      "learning_rate": 2e-05,
      "loss": 0.6913,
      "step": 1770
    },
    {
      "epoch": 0.1486181848543041,
      "grad_norm": 0.44782131910324097,
      "learning_rate": 2e-05,
      "loss": 0.6924,
      "step": 1780
    },
    {
      "epoch": 0.14945311847708107,
      "grad_norm": 0.3924279808998108,
      "learning_rate": 2e-05,
      "loss": 0.7324,
      "step": 1790
    },
    {
      "epoch": 0.15028805209985807,
      "grad_norm": 0.35536661744117737,
      "learning_rate": 2e-05,
      "loss": 0.7101,
      "step": 1800
    },
    {
      "epoch": 0.15112298572263505,
      "grad_norm": 0.3691920340061188,
      "learning_rate": 2e-05,
      "loss": 0.6661,
      "step": 1810
    },
    {
      "epoch": 0.15195791934541203,
      "grad_norm": 0.37451237440109253,
      "learning_rate": 2e-05,
      "loss": 0.6778,
      "step": 1820
    },
    {
      "epoch": 0.15279285296818904,
      "grad_norm": 0.3587948977947235,
      "learning_rate": 2e-05,
      "loss": 0.6813,
      "step": 1830
    },
    {
      "epoch": 0.15362778659096601,
      "grad_norm": 0.46211594343185425,
      "learning_rate": 2e-05,
      "loss": 0.7048,
      "step": 1840
    },
    {
      "epoch": 0.15446272021374302,
      "grad_norm": 0.40670573711395264,
      "learning_rate": 2e-05,
      "loss": 0.7348,
      "step": 1850
    },
    {
      "epoch": 0.15529765383652,
      "grad_norm": 0.3951846659183502,
      "learning_rate": 2e-05,
      "loss": 0.7042,
      "step": 1860
    },
    {
      "epoch": 0.15613258745929698,
      "grad_norm": 0.3904159665107727,
      "learning_rate": 2e-05,
      "loss": 0.6851,
      "step": 1870
    },
    {
      "epoch": 0.15696752108207399,
      "grad_norm": 0.3530999720096588,
      "learning_rate": 2e-05,
      "loss": 0.7255,
      "step": 1880
    },
    {
      "epoch": 0.15780245470485096,
      "grad_norm": 0.37747418880462646,
      "learning_rate": 2e-05,
      "loss": 0.6937,
      "step": 1890
    },
    {
      "epoch": 0.15863738832762794,
      "grad_norm": 0.4633552134037018,
      "learning_rate": 2e-05,
      "loss": 0.6803,
      "step": 1900
    },
    {
      "epoch": 0.15947232195040495,
      "grad_norm": 0.3634198307991028,
      "learning_rate": 2e-05,
      "loss": 0.6702,
      "step": 1910
    },
    {
      "epoch": 0.16030725557318193,
      "grad_norm": 0.41056016087532043,
      "learning_rate": 2e-05,
      "loss": 0.7026,
      "step": 1920
    },
    {
      "epoch": 0.16114218919595893,
      "grad_norm": 0.3515470027923584,
      "learning_rate": 2e-05,
      "loss": 0.6895,
      "step": 1930
    },
    {
      "epoch": 0.1619771228187359,
      "grad_norm": 0.4402310848236084,
      "learning_rate": 2e-05,
      "loss": 0.7225,
      "step": 1940
    },
    {
      "epoch": 0.1628120564415129,
      "grad_norm": 0.3933682441711426,
      "learning_rate": 2e-05,
      "loss": 0.6916,
      "step": 1950
    },
    {
      "epoch": 0.1636469900642899,
      "grad_norm": 0.3949870467185974,
      "learning_rate": 2e-05,
      "loss": 0.7104,
      "step": 1960
    },
    {
      "epoch": 0.16448192368706688,
      "grad_norm": 0.3670644164085388,
      "learning_rate": 2e-05,
      "loss": 0.6781,
      "step": 1970
    },
    {
      "epoch": 0.16531685730984386,
      "grad_norm": 0.5817918181419373,
      "learning_rate": 2e-05,
      "loss": 0.717,
      "step": 1980
    },
    {
      "epoch": 0.16615179093262086,
      "grad_norm": 0.37324586510658264,
      "learning_rate": 2e-05,
      "loss": 0.7014,
      "step": 1990
    },
    {
      "epoch": 0.16698672455539784,
      "grad_norm": 0.3925936222076416,
      "learning_rate": 2e-05,
      "loss": 0.7129,
      "step": 2000
    },
    {
      "epoch": 0.16782165817817485,
      "grad_norm": 0.3621346950531006,
      "learning_rate": 2e-05,
      "loss": 0.6613,
      "step": 2010
    },
    {
      "epoch": 0.16865659180095183,
      "grad_norm": 0.412337988615036,
      "learning_rate": 2e-05,
      "loss": 0.7007,
      "step": 2020
    },
    {
      "epoch": 0.1694915254237288,
      "grad_norm": 0.3621259331703186,
      "learning_rate": 2e-05,
      "loss": 0.6655,
      "step": 2030
    },
    {
      "epoch": 0.1703264590465058,
      "grad_norm": 0.37553828954696655,
      "learning_rate": 2e-05,
      "loss": 0.666,
      "step": 2040
    },
    {
      "epoch": 0.1711613926692828,
      "grad_norm": 0.724480390548706,
      "learning_rate": 2e-05,
      "loss": 0.706,
      "step": 2050
    },
    {
      "epoch": 0.17199632629205977,
      "grad_norm": 0.4094829559326172,
      "learning_rate": 2e-05,
      "loss": 0.6642,
      "step": 2060
    },
    {
      "epoch": 0.17283125991483678,
      "grad_norm": 0.36326175928115845,
      "learning_rate": 2e-05,
      "loss": 0.6942,
      "step": 2070
    },
    {
      "epoch": 0.17366619353761376,
      "grad_norm": 0.3996007740497589,
      "learning_rate": 2e-05,
      "loss": 0.7191,
      "step": 2080
    },
    {
      "epoch": 0.17450112716039076,
      "grad_norm": 0.3675718903541565,
      "learning_rate": 2e-05,
      "loss": 0.6668,
      "step": 2090
    },
    {
      "epoch": 0.17533606078316774,
      "grad_norm": 0.4024317264556885,
      "learning_rate": 2e-05,
      "loss": 0.6919,
      "step": 2100
    },
    {
      "epoch": 0.17617099440594472,
      "grad_norm": 0.3986123204231262,
      "learning_rate": 2e-05,
      "loss": 0.716,
      "step": 2110
    },
    {
      "epoch": 0.17700592802872173,
      "grad_norm": 0.42046478390693665,
      "learning_rate": 2e-05,
      "loss": 0.717,
      "step": 2120
    },
    {
      "epoch": 0.1778408616514987,
      "grad_norm": 0.3795027732849121,
      "learning_rate": 2e-05,
      "loss": 0.7137,
      "step": 2130
    },
    {
      "epoch": 0.17867579527427568,
      "grad_norm": 0.39188075065612793,
      "learning_rate": 2e-05,
      "loss": 0.6503,
      "step": 2140
    },
    {
      "epoch": 0.1795107288970527,
      "grad_norm": 0.36276236176490784,
      "learning_rate": 2e-05,
      "loss": 0.6619,
      "step": 2150
    },
    {
      "epoch": 0.18034566251982967,
      "grad_norm": 0.3488128185272217,
      "learning_rate": 2e-05,
      "loss": 0.687,
      "step": 2160
    },
    {
      "epoch": 0.18118059614260668,
      "grad_norm": 0.44784921407699585,
      "learning_rate": 2e-05,
      "loss": 0.6882,
      "step": 2170
    },
    {
      "epoch": 0.18201552976538365,
      "grad_norm": 0.6095040440559387,
      "learning_rate": 2e-05,
      "loss": 0.7301,
      "step": 2180
    },
    {
      "epoch": 0.18285046338816063,
      "grad_norm": 0.3424732983112335,
      "learning_rate": 2e-05,
      "loss": 0.6673,
      "step": 2190
    },
    {
      "epoch": 0.18368539701093764,
      "grad_norm": 0.3422916531562805,
      "learning_rate": 2e-05,
      "loss": 0.7132,
      "step": 2200
    },
    {
      "epoch": 0.18452033063371462,
      "grad_norm": 0.35164374113082886,
      "learning_rate": 2e-05,
      "loss": 0.6696,
      "step": 2210
    },
    {
      "epoch": 0.1853552642564916,
      "grad_norm": 0.401510089635849,
      "learning_rate": 2e-05,
      "loss": 0.6992,
      "step": 2220
    },
    {
      "epoch": 0.1861901978792686,
      "grad_norm": 0.38584569096565247,
      "learning_rate": 2e-05,
      "loss": 0.7115,
      "step": 2230
    },
    {
      "epoch": 0.18702513150204558,
      "grad_norm": 0.3581119477748871,
      "learning_rate": 2e-05,
      "loss": 0.6684,
      "step": 2240
    },
    {
      "epoch": 0.1878600651248226,
      "grad_norm": 0.3552808463573456,
      "learning_rate": 2e-05,
      "loss": 0.6978,
      "step": 2250
    },
    {
      "epoch": 0.18869499874759957,
      "grad_norm": 0.3791872262954712,
      "learning_rate": 2e-05,
      "loss": 0.6649,
      "step": 2260
    },
    {
      "epoch": 0.18952993237037655,
      "grad_norm": 0.4128217399120331,
      "learning_rate": 2e-05,
      "loss": 0.7044,
      "step": 2270
    },
    {
      "epoch": 0.19036486599315355,
      "grad_norm": 0.6670289635658264,
      "learning_rate": 2e-05,
      "loss": 0.7054,
      "step": 2280
    },
    {
      "epoch": 0.19119979961593053,
      "grad_norm": 0.36319422721862793,
      "learning_rate": 2e-05,
      "loss": 0.7108,
      "step": 2290
    },
    {
      "epoch": 0.1920347332387075,
      "grad_norm": 0.37692052125930786,
      "learning_rate": 2e-05,
      "loss": 0.6655,
      "step": 2300
    },
    {
      "epoch": 0.19286966686148452,
      "grad_norm": 0.3617495894432068,
      "learning_rate": 2e-05,
      "loss": 0.6873,
      "step": 2310
    },
    {
      "epoch": 0.1937046004842615,
      "grad_norm": 0.4157564640045166,
      "learning_rate": 2e-05,
      "loss": 0.685,
      "step": 2320
    },
    {
      "epoch": 0.1945395341070385,
      "grad_norm": 0.38732683658599854,
      "learning_rate": 2e-05,
      "loss": 0.679,
      "step": 2330
    },
    {
      "epoch": 0.19537446772981548,
      "grad_norm": 0.4144394099712372,
      "learning_rate": 2e-05,
      "loss": 0.656,
      "step": 2340
    },
    {
      "epoch": 0.19620940135259246,
      "grad_norm": 0.48212432861328125,
      "learning_rate": 2e-05,
      "loss": 0.6674,
      "step": 2350
    },
    {
      "epoch": 0.19704433497536947,
      "grad_norm": 0.4362604022026062,
      "learning_rate": 2e-05,
      "loss": 0.667,
      "step": 2360
    },
    {
      "epoch": 0.19787926859814645,
      "grad_norm": 0.3660829961299896,
      "learning_rate": 2e-05,
      "loss": 0.6783,
      "step": 2370
    },
    {
      "epoch": 0.19871420222092342,
      "grad_norm": 0.4227180778980255,
      "learning_rate": 2e-05,
      "loss": 0.6965,
      "step": 2380
    },
    {
      "epoch": 0.19954913584370043,
      "grad_norm": 0.3671033978462219,
      "learning_rate": 2e-05,
      "loss": 0.7108,
      "step": 2390
    },
    {
      "epoch": 0.2003840694664774,
      "grad_norm": 0.35155659914016724,
      "learning_rate": 2e-05,
      "loss": 0.7087,
      "step": 2400
    },
    {
      "epoch": 0.20121900308925442,
      "grad_norm": 0.3577719032764435,
      "learning_rate": 2e-05,
      "loss": 0.6577,
      "step": 2410
    },
    {
      "epoch": 0.2020539367120314,
      "grad_norm": 0.43237510323524475,
      "learning_rate": 2e-05,
      "loss": 0.7113,
      "step": 2420
    },
    {
      "epoch": 0.20288887033480837,
      "grad_norm": 0.34467509388923645,
      "learning_rate": 2e-05,
      "loss": 0.6869,
      "step": 2430
    },
    {
      "epoch": 0.20372380395758538,
      "grad_norm": 0.41257065534591675,
      "learning_rate": 2e-05,
      "loss": 0.6879,
      "step": 2440
    },
    {
      "epoch": 0.20455873758036236,
      "grad_norm": 0.3533078730106354,
      "learning_rate": 2e-05,
      "loss": 0.6751,
      "step": 2450
    },
    {
      "epoch": 0.20539367120313934,
      "grad_norm": 0.36549070477485657,
      "learning_rate": 2e-05,
      "loss": 0.6968,
      "step": 2460
    },
    {
      "epoch": 0.20622860482591635,
      "grad_norm": 0.3747435212135315,
      "learning_rate": 2e-05,
      "loss": 0.6889,
      "step": 2470
    },
    {
      "epoch": 0.20706353844869332,
      "grad_norm": 0.3561590313911438,
      "learning_rate": 2e-05,
      "loss": 0.6331,
      "step": 2480
    },
    {
      "epoch": 0.20789847207147033,
      "grad_norm": 0.3813934326171875,
      "learning_rate": 2e-05,
      "loss": 0.6936,
      "step": 2490
    },
    {
      "epoch": 0.2087334056942473,
      "grad_norm": 0.3568154573440552,
      "learning_rate": 2e-05,
      "loss": 0.661,
      "step": 2500
    },
    {
      "epoch": 0.2095683393170243,
      "grad_norm": 0.39197221398353577,
      "learning_rate": 2e-05,
      "loss": 0.7107,
      "step": 2510
    },
    {
      "epoch": 0.2104032729398013,
      "grad_norm": 0.4126400351524353,
      "learning_rate": 2e-05,
      "loss": 0.6868,
      "step": 2520
    },
    {
      "epoch": 0.21123820656257827,
      "grad_norm": 0.3978210985660553,
      "learning_rate": 2e-05,
      "loss": 0.6903,
      "step": 2530
    },
    {
      "epoch": 0.21207314018535525,
      "grad_norm": 0.3680487871170044,
      "learning_rate": 2e-05,
      "loss": 0.6785,
      "step": 2540
    },
    {
      "epoch": 0.21290807380813226,
      "grad_norm": 0.35701045393943787,
      "learning_rate": 2e-05,
      "loss": 0.6436,
      "step": 2550
    },
    {
      "epoch": 0.21374300743090924,
      "grad_norm": 0.3677785396575928,
      "learning_rate": 2e-05,
      "loss": 0.666,
      "step": 2560
    },
    {
      "epoch": 0.21457794105368624,
      "grad_norm": 0.4512452185153961,
      "learning_rate": 2e-05,
      "loss": 0.6843,
      "step": 2570
    },
    {
      "epoch": 0.21541287467646322,
      "grad_norm": 0.41108766198158264,
      "learning_rate": 2e-05,
      "loss": 0.6778,
      "step": 2580
    },
    {
      "epoch": 0.2162478082992402,
      "grad_norm": 0.4080204665660858,
      "learning_rate": 2e-05,
      "loss": 0.6912,
      "step": 2590
    },
    {
      "epoch": 0.2170827419220172,
      "grad_norm": 0.3863216042518616,
      "learning_rate": 2e-05,
      "loss": 0.7123,
      "step": 2600
    },
    {
      "epoch": 0.2179176755447942,
      "grad_norm": 0.3424965441226959,
      "learning_rate": 2e-05,
      "loss": 0.7004,
      "step": 2610
    },
    {
      "epoch": 0.21875260916757117,
      "grad_norm": 0.3708108067512512,
      "learning_rate": 2e-05,
      "loss": 0.6705,
      "step": 2620
    },
    {
      "epoch": 0.21958754279034817,
      "grad_norm": 0.3766650855541229,
      "learning_rate": 2e-05,
      "loss": 0.678,
      "step": 2630
    },
    {
      "epoch": 0.22042247641312515,
      "grad_norm": 0.35487282276153564,
      "learning_rate": 2e-05,
      "loss": 0.6794,
      "step": 2640
    },
    {
      "epoch": 0.22125741003590216,
      "grad_norm": 0.37741872668266296,
      "learning_rate": 2e-05,
      "loss": 0.6515,
      "step": 2650
    },
    {
      "epoch": 0.22209234365867914,
      "grad_norm": 0.37090232968330383,
      "learning_rate": 2e-05,
      "loss": 0.7064,
      "step": 2660
    },
    {
      "epoch": 0.22292727728145612,
      "grad_norm": 0.3624257445335388,
      "learning_rate": 2e-05,
      "loss": 0.6698,
      "step": 2670
    },
    {
      "epoch": 0.22376221090423312,
      "grad_norm": 0.3474475145339966,
      "learning_rate": 2e-05,
      "loss": 0.7001,
      "step": 2680
    },
    {
      "epoch": 0.2245971445270101,
      "grad_norm": 0.3595317304134369,
      "learning_rate": 2e-05,
      "loss": 0.6748,
      "step": 2690
    },
    {
      "epoch": 0.22543207814978708,
      "grad_norm": 0.3778182566165924,
      "learning_rate": 2e-05,
      "loss": 0.6871,
      "step": 2700
    },
    {
      "epoch": 0.2262670117725641,
      "grad_norm": 0.3313213288784027,
      "learning_rate": 2e-05,
      "loss": 0.6974,
      "step": 2710
    },
    {
      "epoch": 0.22710194539534106,
      "grad_norm": 0.39664873480796814,
      "learning_rate": 2e-05,
      "loss": 0.681,
      "step": 2720
    },
    {
      "epoch": 0.22793687901811807,
      "grad_norm": 0.3566786050796509,
      "learning_rate": 2e-05,
      "loss": 0.6973,
      "step": 2730
    },
    {
      "epoch": 0.22877181264089505,
      "grad_norm": 0.3999096751213074,
      "learning_rate": 2e-05,
      "loss": 0.7009,
      "step": 2740
    },
    {
      "epoch": 0.22960674626367203,
      "grad_norm": 0.37713563442230225,
      "learning_rate": 2e-05,
      "loss": 0.7051,
      "step": 2750
    },
    {
      "epoch": 0.23044167988644904,
      "grad_norm": 0.34439313411712646,
      "learning_rate": 2e-05,
      "loss": 0.7153,
      "step": 2760
    },
    {
      "epoch": 0.23127661350922601,
      "grad_norm": 0.36476635932922363,
      "learning_rate": 2e-05,
      "loss": 0.697,
      "step": 2770
    },
    {
      "epoch": 0.232111547132003,
      "grad_norm": 0.3746335208415985,
      "learning_rate": 2e-05,
      "loss": 0.6677,
      "step": 2780
    },
    {
      "epoch": 0.23294648075478,
      "grad_norm": 0.39312872290611267,
      "learning_rate": 2e-05,
      "loss": 0.6655,
      "step": 2790
    },
    {
      "epoch": 0.23378141437755698,
      "grad_norm": 0.417848140001297,
      "learning_rate": 2e-05,
      "loss": 0.7016,
      "step": 2800
    },
    {
      "epoch": 0.23461634800033399,
      "grad_norm": 0.44341158866882324,
      "learning_rate": 2e-05,
      "loss": 0.6415,
      "step": 2810
    },
    {
      "epoch": 0.23545128162311096,
      "grad_norm": 0.3650837242603302,
      "learning_rate": 2e-05,
      "loss": 0.6421,
      "step": 2820
    },
    {
      "epoch": 0.23628621524588794,
      "grad_norm": 0.38273918628692627,
      "learning_rate": 2e-05,
      "loss": 0.7157,
      "step": 2830
    },
    {
      "epoch": 0.23712114886866495,
      "grad_norm": 0.3887275755405426,
      "learning_rate": 2e-05,
      "loss": 0.6822,
      "step": 2840
    },
    {
      "epoch": 0.23795608249144193,
      "grad_norm": 0.3538888394832611,
      "learning_rate": 2e-05,
      "loss": 0.6842,
      "step": 2850
    },
    {
      "epoch": 0.2387910161142189,
      "grad_norm": 0.4602071940898895,
      "learning_rate": 2e-05,
      "loss": 0.6578,
      "step": 2860
    },
    {
      "epoch": 0.2396259497369959,
      "grad_norm": 0.38750094175338745,
      "learning_rate": 2e-05,
      "loss": 0.7375,
      "step": 2870
    },
    {
      "epoch": 0.2404608833597729,
      "grad_norm": 0.33201658725738525,
      "learning_rate": 2e-05,
      "loss": 0.71,
      "step": 2880
    },
    {
      "epoch": 0.2412958169825499,
      "grad_norm": 0.37085673213005066,
      "learning_rate": 2e-05,
      "loss": 0.6617,
      "step": 2890
    },
    {
      "epoch": 0.24213075060532688,
      "grad_norm": 0.3528018593788147,
      "learning_rate": 2e-05,
      "loss": 0.7013,
      "step": 2900
    },
    {
      "epoch": 0.24296568422810386,
      "grad_norm": 0.38753339648246765,
      "learning_rate": 2e-05,
      "loss": 0.6396,
      "step": 2910
    },
    {
      "epoch": 0.24380061785088086,
      "grad_norm": 0.3782426118850708,
      "learning_rate": 2e-05,
      "loss": 0.7163,
      "step": 2920
    },
    {
      "epoch": 0.24463555147365784,
      "grad_norm": 0.3527936339378357,
      "learning_rate": 2e-05,
      "loss": 0.6628,
      "step": 2930
    },
    {
      "epoch": 0.24547048509643482,
      "grad_norm": 0.3663792908191681,
      "learning_rate": 2e-05,
      "loss": 0.6884,
      "step": 2940
    },
    {
      "epoch": 0.24630541871921183,
      "grad_norm": 0.3781542479991913,
      "learning_rate": 2e-05,
      "loss": 0.7047,
      "step": 2950
    },
    {
      "epoch": 0.2471403523419888,
      "grad_norm": 0.3632984161376953,
      "learning_rate": 2e-05,
      "loss": 0.6909,
      "step": 2960
    },
    {
      "epoch": 0.2479752859647658,
      "grad_norm": 0.3737751841545105,
      "learning_rate": 2e-05,
      "loss": 0.6623,
      "step": 2970
    },
    {
      "epoch": 0.2488102195875428,
      "grad_norm": 0.33807748556137085,
      "learning_rate": 2e-05,
      "loss": 0.7159,
      "step": 2980
    },
    {
      "epoch": 0.24964515321031977,
      "grad_norm": 0.4013286828994751,
      "learning_rate": 2e-05,
      "loss": 0.7062,
      "step": 2990
    },
    {
      "epoch": 0.25048008683309675,
      "grad_norm": 0.3510397672653198,
      "learning_rate": 2e-05,
      "loss": 0.6639,
      "step": 3000
    },
    {
      "epoch": 0.2513150204558738,
      "grad_norm": 0.3942272961139679,
      "learning_rate": 2e-05,
      "loss": 0.6816,
      "step": 3010
    },
    {
      "epoch": 0.25214995407865076,
      "grad_norm": 0.3481244742870331,
      "learning_rate": 2e-05,
      "loss": 0.6693,
      "step": 3020
    },
    {
      "epoch": 0.25298488770142774,
      "grad_norm": 0.37340089678764343,
      "learning_rate": 2e-05,
      "loss": 0.6614,
      "step": 3030
    },
    {
      "epoch": 0.2538198213242047,
      "grad_norm": 0.389567494392395,
      "learning_rate": 2e-05,
      "loss": 0.697,
      "step": 3040
    },
    {
      "epoch": 0.2546547549469817,
      "grad_norm": 0.40338629484176636,
      "learning_rate": 2e-05,
      "loss": 0.6801,
      "step": 3050
    },
    {
      "epoch": 0.2554896885697587,
      "grad_norm": 1.6669360399246216,
      "learning_rate": 2e-05,
      "loss": 0.7063,
      "step": 3060
    },
    {
      "epoch": 0.2563246221925357,
      "grad_norm": 0.4049375355243683,
      "learning_rate": 2e-05,
      "loss": 0.7161,
      "step": 3070
    },
    {
      "epoch": 0.2571595558153127,
      "grad_norm": 0.3685857653617859,
      "learning_rate": 2e-05,
      "loss": 0.6955,
      "step": 3080
    },
    {
      "epoch": 0.25799448943808967,
      "grad_norm": 0.34307533502578735,
      "learning_rate": 2e-05,
      "loss": 0.6952,
      "step": 3090
    },
    {
      "epoch": 0.25882942306086665,
      "grad_norm": 0.36656075716018677,
      "learning_rate": 2e-05,
      "loss": 0.6742,
      "step": 3100
    },
    {
      "epoch": 0.2596643566836436,
      "grad_norm": 0.48263198137283325,
      "learning_rate": 2e-05,
      "loss": 0.7035,
      "step": 3110
    },
    {
      "epoch": 0.26049929030642066,
      "grad_norm": 0.3497573435306549,
      "learning_rate": 2e-05,
      "loss": 0.6729,
      "step": 3120
    },
    {
      "epoch": 0.26133422392919764,
      "grad_norm": 0.3446602523326874,
      "learning_rate": 2e-05,
      "loss": 0.6864,
      "step": 3130
    },
    {
      "epoch": 0.2621691575519746,
      "grad_norm": 0.4396670460700989,
      "learning_rate": 2e-05,
      "loss": 0.6651,
      "step": 3140
    },
    {
      "epoch": 0.2630040911747516,
      "grad_norm": 0.39971259236335754,
      "learning_rate": 2e-05,
      "loss": 0.6558,
      "step": 3150
    },
    {
      "epoch": 0.2638390247975286,
      "grad_norm": 0.3778079152107239,
      "learning_rate": 2e-05,
      "loss": 0.6926,
      "step": 3160
    },
    {
      "epoch": 0.2646739584203056,
      "grad_norm": 0.3553712069988251,
      "learning_rate": 2e-05,
      "loss": 0.6372,
      "step": 3170
    },
    {
      "epoch": 0.2655088920430826,
      "grad_norm": 0.3498518168926239,
      "learning_rate": 2e-05,
      "loss": 0.6775,
      "step": 3180
    },
    {
      "epoch": 0.26634382566585957,
      "grad_norm": 0.40803062915802,
      "learning_rate": 2e-05,
      "loss": 0.6999,
      "step": 3190
    },
    {
      "epoch": 0.26717875928863655,
      "grad_norm": 0.35266658663749695,
      "learning_rate": 2e-05,
      "loss": 0.6407,
      "step": 3200
    },
    {
      "epoch": 0.2680136929114135,
      "grad_norm": 0.3838484287261963,
      "learning_rate": 2e-05,
      "loss": 0.6994,
      "step": 3210
    },
    {
      "epoch": 0.2688486265341905,
      "grad_norm": 0.3680204153060913,
      "learning_rate": 2e-05,
      "loss": 0.6882,
      "step": 3220
    },
    {
      "epoch": 0.26968356015696754,
      "grad_norm": 0.36609959602355957,
      "learning_rate": 2e-05,
      "loss": 0.6918,
      "step": 3230
    },
    {
      "epoch": 0.2705184937797445,
      "grad_norm": 0.3467004597187042,
      "learning_rate": 2e-05,
      "loss": 0.668,
      "step": 3240
    },
    {
      "epoch": 0.2713534274025215,
      "grad_norm": 0.3607097864151001,
      "learning_rate": 2e-05,
      "loss": 0.6781,
      "step": 3250
    },
    {
      "epoch": 0.2721883610252985,
      "grad_norm": 0.34691914916038513,
      "learning_rate": 2e-05,
      "loss": 0.6742,
      "step": 3260
    },
    {
      "epoch": 0.27302329464807545,
      "grad_norm": 0.36507701873779297,
      "learning_rate": 2e-05,
      "loss": 0.6804,
      "step": 3270
    },
    {
      "epoch": 0.2738582282708525,
      "grad_norm": 0.36962178349494934,
      "learning_rate": 2e-05,
      "loss": 0.6913,
      "step": 3280
    },
    {
      "epoch": 0.27469316189362947,
      "grad_norm": 0.38439133763313293,
      "learning_rate": 2e-05,
      "loss": 0.6943,
      "step": 3290
    },
    {
      "epoch": 0.27552809551640645,
      "grad_norm": 0.3779873847961426,
      "learning_rate": 2e-05,
      "loss": 0.6749,
      "step": 3300
    },
    {
      "epoch": 0.2763630291391834,
      "grad_norm": 0.39938196539878845,
      "learning_rate": 2e-05,
      "loss": 0.692,
      "step": 3310
    },
    {
      "epoch": 0.2771979627619604,
      "grad_norm": 0.3708743155002594,
      "learning_rate": 2e-05,
      "loss": 0.679,
      "step": 3320
    },
    {
      "epoch": 0.27803289638473744,
      "grad_norm": 0.38334569334983826,
      "learning_rate": 2e-05,
      "loss": 0.6479,
      "step": 3330
    },
    {
      "epoch": 0.2788678300075144,
      "grad_norm": 0.4010925889015198,
      "learning_rate": 2e-05,
      "loss": 0.6434,
      "step": 3340
    },
    {
      "epoch": 0.2797027636302914,
      "grad_norm": 0.40315696597099304,
      "learning_rate": 2e-05,
      "loss": 0.6859,
      "step": 3350
    },
    {
      "epoch": 0.2805376972530684,
      "grad_norm": 0.34506651759147644,
      "learning_rate": 2e-05,
      "loss": 0.6633,
      "step": 3360
    },
    {
      "epoch": 0.28137263087584535,
      "grad_norm": 0.37208911776542664,
      "learning_rate": 2e-05,
      "loss": 0.6674,
      "step": 3370
    },
    {
      "epoch": 0.28220756449862233,
      "grad_norm": 0.48329827189445496,
      "learning_rate": 2e-05,
      "loss": 0.6636,
      "step": 3380
    },
    {
      "epoch": 0.28304249812139937,
      "grad_norm": 0.379374623298645,
      "learning_rate": 2e-05,
      "loss": 0.6569,
      "step": 3390
    },
    {
      "epoch": 0.28387743174417635,
      "grad_norm": 0.357758492231369,
      "learning_rate": 2e-05,
      "loss": 0.6542,
      "step": 3400
    },
    {
      "epoch": 0.2847123653669533,
      "grad_norm": 0.4022427499294281,
      "learning_rate": 2e-05,
      "loss": 0.6787,
      "step": 3410
    },
    {
      "epoch": 0.2855472989897303,
      "grad_norm": 0.35887256264686584,
      "learning_rate": 2e-05,
      "loss": 0.6665,
      "step": 3420
    },
    {
      "epoch": 0.2863822326125073,
      "grad_norm": 0.9388905763626099,
      "learning_rate": 2e-05,
      "loss": 0.7115,
      "step": 3430
    },
    {
      "epoch": 0.2872171662352843,
      "grad_norm": 0.35477712750434875,
      "learning_rate": 2e-05,
      "loss": 0.6839,
      "step": 3440
    },
    {
      "epoch": 0.2880520998580613,
      "grad_norm": 0.39069318771362305,
      "learning_rate": 2e-05,
      "loss": 0.7006,
      "step": 3450
    },
    {
      "epoch": 0.2888870334808383,
      "grad_norm": 0.37950125336647034,
      "learning_rate": 2e-05,
      "loss": 0.6832,
      "step": 3460
    },
    {
      "epoch": 0.28972196710361525,
      "grad_norm": 0.3914310336112976,
      "learning_rate": 2e-05,
      "loss": 0.6676,
      "step": 3470
    },
    {
      "epoch": 0.29055690072639223,
      "grad_norm": 0.3945288062095642,
      "learning_rate": 2e-05,
      "loss": 0.6996,
      "step": 3480
    },
    {
      "epoch": 0.29139183434916927,
      "grad_norm": 0.3539528250694275,
      "learning_rate": 2e-05,
      "loss": 0.6853,
      "step": 3490
    },
    {
      "epoch": 0.29222676797194624,
      "grad_norm": 0.3777148127555847,
      "learning_rate": 2e-05,
      "loss": 0.6013,
      "step": 3500
    },
    {
      "epoch": 0.2930617015947232,
      "grad_norm": 0.3967706859111786,
      "learning_rate": 2e-05,
      "loss": 0.6535,
      "step": 3510
    },
    {
      "epoch": 0.2938966352175002,
      "grad_norm": 0.35000601410865784,
      "learning_rate": 2e-05,
      "loss": 0.6617,
      "step": 3520
    },
    {
      "epoch": 0.2947315688402772,
      "grad_norm": 0.37971240282058716,
      "learning_rate": 2e-05,
      "loss": 0.6847,
      "step": 3530
    },
    {
      "epoch": 0.2955665024630542,
      "grad_norm": 0.38538622856140137,
      "learning_rate": 2e-05,
      "loss": 0.6789,
      "step": 3540
    },
    {
      "epoch": 0.2964014360858312,
      "grad_norm": 0.38987165689468384,
      "learning_rate": 2e-05,
      "loss": 0.6844,
      "step": 3550
    },
    {
      "epoch": 0.2972363697086082,
      "grad_norm": 0.3420151174068451,
      "learning_rate": 2e-05,
      "loss": 0.6583,
      "step": 3560
    },
    {
      "epoch": 0.29807130333138515,
      "grad_norm": 0.38039082288742065,
      "learning_rate": 2e-05,
      "loss": 0.6898,
      "step": 3570
    },
    {
      "epoch": 0.29890623695416213,
      "grad_norm": 0.35230252146720886,
      "learning_rate": 2e-05,
      "loss": 0.65,
      "step": 3580
    },
    {
      "epoch": 0.2997411705769391,
      "grad_norm": 0.3664659559726715,
      "learning_rate": 2e-05,
      "loss": 0.6795,
      "step": 3590
    },
    {
      "epoch": 0.30057610419971614,
      "grad_norm": 0.4027750492095947,
      "learning_rate": 2e-05,
      "loss": 0.657,
      "step": 3600
    },
    {
      "epoch": 0.3014110378224931,
      "grad_norm": 0.3652893900871277,
      "learning_rate": 2e-05,
      "loss": 0.6863,
      "step": 3610
    },
    {
      "epoch": 0.3022459714452701,
      "grad_norm": 0.32467252016067505,
      "learning_rate": 2e-05,
      "loss": 0.6623,
      "step": 3620
    },
    {
      "epoch": 0.3030809050680471,
      "grad_norm": 0.35403797030448914,
      "learning_rate": 2e-05,
      "loss": 0.6754,
      "step": 3630
    },
    {
      "epoch": 0.30391583869082406,
      "grad_norm": 0.37590375542640686,
      "learning_rate": 2e-05,
      "loss": 0.6737,
      "step": 3640
    },
    {
      "epoch": 0.3047507723136011,
      "grad_norm": 0.3438977599143982,
      "learning_rate": 2e-05,
      "loss": 0.6555,
      "step": 3650
    },
    {
      "epoch": 0.30558570593637807,
      "grad_norm": 0.3637774884700775,
      "learning_rate": 2e-05,
      "loss": 0.6644,
      "step": 3660
    },
    {
      "epoch": 0.30642063955915505,
      "grad_norm": 1.2078490257263184,
      "learning_rate": 2e-05,
      "loss": 0.7085,
      "step": 3670
    },
    {
      "epoch": 0.30725557318193203,
      "grad_norm": 0.4078238010406494,
      "learning_rate": 2e-05,
      "loss": 0.6441,
      "step": 3680
    },
    {
      "epoch": 0.308090506804709,
      "grad_norm": 0.36774277687072754,
      "learning_rate": 2e-05,
      "loss": 0.665,
      "step": 3690
    },
    {
      "epoch": 0.30892544042748604,
      "grad_norm": 0.34963279962539673,
      "learning_rate": 2e-05,
      "loss": 0.6971,
      "step": 3700
    },
    {
      "epoch": 0.309760374050263,
      "grad_norm": 0.36340829730033875,
      "learning_rate": 2e-05,
      "loss": 0.657,
      "step": 3710
    },
    {
      "epoch": 0.31059530767304,
      "grad_norm": 0.30774226784706116,
      "learning_rate": 2e-05,
      "loss": 0.6288,
      "step": 3720
    },
    {
      "epoch": 0.311430241295817,
      "grad_norm": 0.3752991259098053,
      "learning_rate": 2e-05,
      "loss": 0.7037,
      "step": 3730
    },
    {
      "epoch": 0.31226517491859396,
      "grad_norm": 0.3525054454803467,
      "learning_rate": 2e-05,
      "loss": 0.6352,
      "step": 3740
    },
    {
      "epoch": 0.31310010854137094,
      "grad_norm": 0.3878222107887268,
      "learning_rate": 2e-05,
      "loss": 0.6686,
      "step": 3750
    },
    {
      "epoch": 0.31393504216414797,
      "grad_norm": 0.4150567948818207,
      "learning_rate": 2e-05,
      "loss": 0.6826,
      "step": 3760
    },
    {
      "epoch": 0.31476997578692495,
      "grad_norm": 0.37096673250198364,
      "learning_rate": 2e-05,
      "loss": 0.6759,
      "step": 3770
    },
    {
      "epoch": 0.31560490940970193,
      "grad_norm": 0.34516632556915283,
      "learning_rate": 2e-05,
      "loss": 0.6703,
      "step": 3780
    },
    {
      "epoch": 0.3164398430324789,
      "grad_norm": 0.3539997339248657,
      "learning_rate": 2e-05,
      "loss": 0.6806,
      "step": 3790
    },
    {
      "epoch": 0.3172747766552559,
      "grad_norm": 0.39132675528526306,
      "learning_rate": 2e-05,
      "loss": 0.6852,
      "step": 3800
    },
    {
      "epoch": 0.3181097102780329,
      "grad_norm": 0.3675609529018402,
      "learning_rate": 2e-05,
      "loss": 0.6617,
      "step": 3810
    },
    {
      "epoch": 0.3189446439008099,
      "grad_norm": 0.3339812159538269,
      "learning_rate": 2e-05,
      "loss": 0.6715,
      "step": 3820
    },
    {
      "epoch": 0.3197795775235869,
      "grad_norm": 0.3698909282684326,
      "learning_rate": 2e-05,
      "loss": 0.661,
      "step": 3830
    },
    {
      "epoch": 0.32061451114636386,
      "grad_norm": 0.395443856716156,
      "learning_rate": 2e-05,
      "loss": 0.6382,
      "step": 3840
    },
    {
      "epoch": 0.32144944476914084,
      "grad_norm": 0.38289499282836914,
      "learning_rate": 2e-05,
      "loss": 0.644,
      "step": 3850
    },
    {
      "epoch": 0.32228437839191787,
      "grad_norm": 0.3726746737957001,
      "learning_rate": 2e-05,
      "loss": 0.7051,
      "step": 3860
    },
    {
      "epoch": 0.32311931201469485,
      "grad_norm": 0.38410791754722595,
      "learning_rate": 2e-05,
      "loss": 0.6573,
      "step": 3870
    },
    {
      "epoch": 0.3239542456374718,
      "grad_norm": 0.39256834983825684,
      "learning_rate": 2e-05,
      "loss": 0.6588,
      "step": 3880
    },
    {
      "epoch": 0.3247891792602488,
      "grad_norm": 0.3729894757270813,
      "learning_rate": 2e-05,
      "loss": 0.6542,
      "step": 3890
    },
    {
      "epoch": 0.3256241128830258,
      "grad_norm": 0.37289127707481384,
      "learning_rate": 2e-05,
      "loss": 0.7037,
      "step": 3900
    },
    {
      "epoch": 0.32645904650580276,
      "grad_norm": 0.3670271337032318,
      "learning_rate": 2e-05,
      "loss": 0.7013,
      "step": 3910
    },
    {
      "epoch": 0.3272939801285798,
      "grad_norm": 0.3461996912956238,
      "learning_rate": 2e-05,
      "loss": 0.6555,
      "step": 3920
    },
    {
      "epoch": 0.3281289137513568,
      "grad_norm": 0.37942495942115784,
      "learning_rate": 2e-05,
      "loss": 0.6631,
      "step": 3930
    },
    {
      "epoch": 0.32896384737413376,
      "grad_norm": 0.36742621660232544,
      "learning_rate": 2e-05,
      "loss": 0.653,
      "step": 3940
    },
    {
      "epoch": 0.32979878099691073,
      "grad_norm": 0.3810204863548279,
      "learning_rate": 2e-05,
      "loss": 0.6707,
      "step": 3950
    },
    {
      "epoch": 0.3306337146196877,
      "grad_norm": 0.3833245635032654,
      "learning_rate": 2e-05,
      "loss": 0.7017,
      "step": 3960
    },
    {
      "epoch": 0.33146864824246475,
      "grad_norm": 0.3555945158004761,
      "learning_rate": 2e-05,
      "loss": 0.704,
      "step": 3970
    },
    {
      "epoch": 0.3323035818652417,
      "grad_norm": 0.35338345170021057,
      "learning_rate": 2e-05,
      "loss": 0.6715,
      "step": 3980
    },
    {
      "epoch": 0.3331385154880187,
      "grad_norm": 0.37337103486061096,
      "learning_rate": 2e-05,
      "loss": 0.6746,
      "step": 3990
    },
    {
      "epoch": 0.3339734491107957,
      "grad_norm": 0.39499807357788086,
      "learning_rate": 2e-05,
      "loss": 0.6355,
      "step": 4000
    },
    {
      "epoch": 0.33480838273357266,
      "grad_norm": 0.3578856587409973,
      "learning_rate": 2e-05,
      "loss": 0.6468,
      "step": 4010
    },
    {
      "epoch": 0.3356433163563497,
      "grad_norm": 0.35872864723205566,
      "learning_rate": 2e-05,
      "loss": 0.6489,
      "step": 4020
    },
    {
      "epoch": 0.3364782499791267,
      "grad_norm": 0.36053839325904846,
      "learning_rate": 2e-05,
      "loss": 0.6837,
      "step": 4030
    },
    {
      "epoch": 0.33731318360190365,
      "grad_norm": 0.35407933592796326,
      "learning_rate": 2e-05,
      "loss": 0.6772,
      "step": 4040
    },
    {
      "epoch": 0.33814811722468063,
      "grad_norm": 0.3584800362586975,
      "learning_rate": 2e-05,
      "loss": 0.6652,
      "step": 4050
    },
    {
      "epoch": 0.3389830508474576,
      "grad_norm": 0.3831849992275238,
      "learning_rate": 2e-05,
      "loss": 0.6683,
      "step": 4060
    },
    {
      "epoch": 0.3398179844702346,
      "grad_norm": 0.3577152192592621,
      "learning_rate": 2e-05,
      "loss": 0.6799,
      "step": 4070
    },
    {
      "epoch": 0.3406529180930116,
      "grad_norm": 0.36730220913887024,
      "learning_rate": 2e-05,
      "loss": 0.7125,
      "step": 4080
    },
    {
      "epoch": 0.3414878517157886,
      "grad_norm": 0.37625598907470703,
      "learning_rate": 2e-05,
      "loss": 0.6706,
      "step": 4090
    },
    {
      "epoch": 0.3423227853385656,
      "grad_norm": 0.367351770401001,
      "learning_rate": 2e-05,
      "loss": 0.6499,
      "step": 4100
    },
    {
      "epoch": 0.34315771896134256,
      "grad_norm": 0.3668467104434967,
      "learning_rate": 2e-05,
      "loss": 0.6691,
      "step": 4110
    },
    {
      "epoch": 0.34399265258411954,
      "grad_norm": 0.3462507724761963,
      "learning_rate": 2e-05,
      "loss": 0.667,
      "step": 4120
    },
    {
      "epoch": 0.3448275862068966,
      "grad_norm": 0.3935559093952179,
      "learning_rate": 2e-05,
      "loss": 0.6847,
      "step": 4130
    },
    {
      "epoch": 0.34566251982967355,
      "grad_norm": 0.4254509508609772,
      "learning_rate": 2e-05,
      "loss": 0.6427,
      "step": 4140
    },
    {
      "epoch": 0.34649745345245053,
      "grad_norm": 0.41700831055641174,
      "learning_rate": 2e-05,
      "loss": 0.6954,
      "step": 4150
    },
    {
      "epoch": 0.3473323870752275,
      "grad_norm": 0.37758785486221313,
      "learning_rate": 2e-05,
      "loss": 0.6512,
      "step": 4160
    },
    {
      "epoch": 0.3481673206980045,
      "grad_norm": 0.4406028091907501,
      "learning_rate": 2e-05,
      "loss": 0.6759,
      "step": 4170
    },
    {
      "epoch": 0.3490022543207815,
      "grad_norm": 0.3895401060581207,
      "learning_rate": 2e-05,
      "loss": 0.6632,
      "step": 4180
    },
    {
      "epoch": 0.3498371879435585,
      "grad_norm": 0.36757349967956543,
      "learning_rate": 2e-05,
      "loss": 0.6507,
      "step": 4190
    },
    {
      "epoch": 0.3506721215663355,
      "grad_norm": 0.4160826802253723,
      "learning_rate": 2e-05,
      "loss": 0.6705,
      "step": 4200
    },
    {
      "epoch": 0.35150705518911246,
      "grad_norm": 0.37626156210899353,
      "learning_rate": 2e-05,
      "loss": 0.6607,
      "step": 4210
    },
    {
      "epoch": 0.35234198881188944,
      "grad_norm": 0.39758870005607605,
      "learning_rate": 2e-05,
      "loss": 0.6613,
      "step": 4220
    },
    {
      "epoch": 0.3531769224346664,
      "grad_norm": 0.3542587459087372,
      "learning_rate": 2e-05,
      "loss": 0.6404,
      "step": 4230
    },
    {
      "epoch": 0.35401185605744345,
      "grad_norm": 0.3708060085773468,
      "learning_rate": 2e-05,
      "loss": 0.6663,
      "step": 4240
    },
    {
      "epoch": 0.35484678968022043,
      "grad_norm": 0.3636868894100189,
      "learning_rate": 2e-05,
      "loss": 0.6475,
      "step": 4250
    },
    {
      "epoch": 0.3556817233029974,
      "grad_norm": 0.38279253244400024,
      "learning_rate": 2e-05,
      "loss": 0.6214,
      "step": 4260
    },
    {
      "epoch": 0.3565166569257744,
      "grad_norm": 0.41041916608810425,
      "learning_rate": 2e-05,
      "loss": 0.6664,
      "step": 4270
    },
    {
      "epoch": 0.35735159054855137,
      "grad_norm": 0.3527238070964813,
      "learning_rate": 2e-05,
      "loss": 0.6668,
      "step": 4280
    },
    {
      "epoch": 0.3581865241713284,
      "grad_norm": 0.3648574948310852,
      "learning_rate": 2e-05,
      "loss": 0.6612,
      "step": 4290
    },
    {
      "epoch": 0.3590214577941054,
      "grad_norm": 0.36532264947891235,
      "learning_rate": 2e-05,
      "loss": 0.6537,
      "step": 4300
    },
    {
      "epoch": 0.35985639141688236,
      "grad_norm": 0.34127771854400635,
      "learning_rate": 2e-05,
      "loss": 0.6766,
      "step": 4310
    },
    {
      "epoch": 0.36069132503965934,
      "grad_norm": 0.3816087543964386,
      "learning_rate": 2e-05,
      "loss": 0.6581,
      "step": 4320
    },
    {
      "epoch": 0.3615262586624363,
      "grad_norm": 0.35879552364349365,
      "learning_rate": 2e-05,
      "loss": 0.6359,
      "step": 4330
    },
    {
      "epoch": 0.36236119228521335,
      "grad_norm": 0.4297371804714203,
      "learning_rate": 2e-05,
      "loss": 0.6558,
      "step": 4340
    },
    {
      "epoch": 0.36319612590799033,
      "grad_norm": 0.3856416940689087,
      "learning_rate": 2e-05,
      "loss": 0.6707,
      "step": 4350
    },
    {
      "epoch": 0.3640310595307673,
      "grad_norm": 0.3447234630584717,
      "learning_rate": 2e-05,
      "loss": 0.6749,
      "step": 4360
    },
    {
      "epoch": 0.3648659931535443,
      "grad_norm": 0.39188873767852783,
      "learning_rate": 2e-05,
      "loss": 0.6658,
      "step": 4370
    },
    {
      "epoch": 0.36570092677632127,
      "grad_norm": 0.3640318810939789,
      "learning_rate": 2e-05,
      "loss": 0.662,
      "step": 4380
    },
    {
      "epoch": 0.36653586039909825,
      "grad_norm": 0.34400489926338196,
      "learning_rate": 2e-05,
      "loss": 0.644,
      "step": 4390
    },
    {
      "epoch": 0.3673707940218753,
      "grad_norm": 0.36888739466667175,
      "learning_rate": 2e-05,
      "loss": 0.6544,
      "step": 4400
    },
    {
      "epoch": 0.36820572764465226,
      "grad_norm": 0.42386606335639954,
      "learning_rate": 2e-05,
      "loss": 0.6903,
      "step": 4410
    },
    {
      "epoch": 0.36904066126742924,
      "grad_norm": 0.4172605872154236,
      "learning_rate": 2e-05,
      "loss": 0.6508,
      "step": 4420
    },
    {
      "epoch": 0.3698755948902062,
      "grad_norm": 0.36239171028137207,
      "learning_rate": 2e-05,
      "loss": 0.6839,
      "step": 4430
    },
    {
      "epoch": 0.3707105285129832,
      "grad_norm": 0.34572529792785645,
      "learning_rate": 2e-05,
      "loss": 0.6657,
      "step": 4440
    },
    {
      "epoch": 0.37154546213576023,
      "grad_norm": 0.3653551936149597,
      "learning_rate": 2e-05,
      "loss": 0.6344,
      "step": 4450
    },
    {
      "epoch": 0.3723803957585372,
      "grad_norm": 0.37788355350494385,
      "learning_rate": 2e-05,
      "loss": 0.6724,
      "step": 4460
    },
    {
      "epoch": 0.3732153293813142,
      "grad_norm": 0.3315238058567047,
      "learning_rate": 2e-05,
      "loss": 0.667,
      "step": 4470
    },
    {
      "epoch": 0.37405026300409117,
      "grad_norm": 0.33371374011039734,
      "learning_rate": 2e-05,
      "loss": 0.6657,
      "step": 4480
    },
    {
      "epoch": 0.37488519662686814,
      "grad_norm": 0.3910922408103943,
      "learning_rate": 2e-05,
      "loss": 0.6782,
      "step": 4490
    },
    {
      "epoch": 0.3757201302496452,
      "grad_norm": 0.33257752656936646,
      "learning_rate": 2e-05,
      "loss": 0.6445,
      "step": 4500
    },
    {
      "epoch": 0.37655506387242216,
      "grad_norm": 0.44457560777664185,
      "learning_rate": 2e-05,
      "loss": 0.6541,
      "step": 4510
    },
    {
      "epoch": 0.37738999749519914,
      "grad_norm": 0.4875108599662781,
      "learning_rate": 2e-05,
      "loss": 0.6326,
      "step": 4520
    },
    {
      "epoch": 0.3782249311179761,
      "grad_norm": 0.36632639169692993,
      "learning_rate": 2e-05,
      "loss": 0.6636,
      "step": 4530
    },
    {
      "epoch": 0.3790598647407531,
      "grad_norm": 0.36051562428474426,
      "learning_rate": 2e-05,
      "loss": 0.6346,
      "step": 4540
    },
    {
      "epoch": 0.3798947983635301,
      "grad_norm": 0.3570646643638611,
      "learning_rate": 2e-05,
      "loss": 0.6459,
      "step": 4550
    },
    {
      "epoch": 0.3807297319863071,
      "grad_norm": 0.31021231412887573,
      "learning_rate": 2e-05,
      "loss": 0.6728,
      "step": 4560
    },
    {
      "epoch": 0.3815646656090841,
      "grad_norm": 0.39026519656181335,
      "learning_rate": 2e-05,
      "loss": 0.6427,
      "step": 4570
    },
    {
      "epoch": 0.38239959923186106,
      "grad_norm": 0.37352192401885986,
      "learning_rate": 2e-05,
      "loss": 0.6572,
      "step": 4580
    },
    {
      "epoch": 0.38323453285463804,
      "grad_norm": 0.3814341723918915,
      "learning_rate": 2e-05,
      "loss": 0.6458,
      "step": 4590
    },
    {
      "epoch": 0.384069466477415,
      "grad_norm": 0.37026679515838623,
      "learning_rate": 2e-05,
      "loss": 0.6639,
      "step": 4600
    },
    {
      "epoch": 0.38490440010019206,
      "grad_norm": 0.43374890089035034,
      "learning_rate": 2e-05,
      "loss": 0.6623,
      "step": 4610
    },
    {
      "epoch": 0.38573933372296904,
      "grad_norm": 0.3498901426792145,
      "learning_rate": 2e-05,
      "loss": 0.6629,
      "step": 4620
    },
    {
      "epoch": 0.386574267345746,
      "grad_norm": 0.4091528654098511,
      "learning_rate": 2e-05,
      "loss": 0.6672,
      "step": 4630
    },
    {
      "epoch": 0.387409200968523,
      "grad_norm": 0.38661715388298035,
      "learning_rate": 2e-05,
      "loss": 0.6221,
      "step": 4640
    },
    {
      "epoch": 0.38824413459129997,
      "grad_norm": 0.3787270486354828,
      "learning_rate": 2e-05,
      "loss": 0.6667,
      "step": 4650
    },
    {
      "epoch": 0.389079068214077,
      "grad_norm": 0.3583727478981018,
      "learning_rate": 2e-05,
      "loss": 0.6621,
      "step": 4660
    },
    {
      "epoch": 0.389914001836854,
      "grad_norm": 0.3378801643848419,
      "learning_rate": 2e-05,
      "loss": 0.6592,
      "step": 4670
    },
    {
      "epoch": 0.39074893545963096,
      "grad_norm": 0.38119861483573914,
      "learning_rate": 2e-05,
      "loss": 0.6517,
      "step": 4680
    },
    {
      "epoch": 0.39158386908240794,
      "grad_norm": 0.3805156648159027,
      "learning_rate": 2e-05,
      "loss": 0.6861,
      "step": 4690
    },
    {
      "epoch": 0.3924188027051849,
      "grad_norm": 0.4050471782684326,
      "learning_rate": 2e-05,
      "loss": 0.6543,
      "step": 4700
    },
    {
      "epoch": 0.3932537363279619,
      "grad_norm": 0.3710922300815582,
      "learning_rate": 2e-05,
      "loss": 0.6617,
      "step": 4710
    },
    {
      "epoch": 0.39408866995073893,
      "grad_norm": 0.3692035973072052,
      "learning_rate": 2e-05,
      "loss": 0.6437,
      "step": 4720
    },
    {
      "epoch": 0.3949236035735159,
      "grad_norm": 0.424715131521225,
      "learning_rate": 2e-05,
      "loss": 0.6375,
      "step": 4730
    },
    {
      "epoch": 0.3957585371962929,
      "grad_norm": 0.35713499784469604,
      "learning_rate": 2e-05,
      "loss": 0.6504,
      "step": 4740
    },
    {
      "epoch": 0.39659347081906987,
      "grad_norm": 0.39688199758529663,
      "learning_rate": 2e-05,
      "loss": 0.6492,
      "step": 4750
    },
    {
      "epoch": 0.39742840444184685,
      "grad_norm": 0.38314980268478394,
      "learning_rate": 2e-05,
      "loss": 0.6541,
      "step": 4760
    },
    {
      "epoch": 0.3982633380646239,
      "grad_norm": 0.3715704679489136,
      "learning_rate": 2e-05,
      "loss": 0.6719,
      "step": 4770
    },
    {
      "epoch": 0.39909827168740086,
      "grad_norm": 0.35048532485961914,
      "learning_rate": 2e-05,
      "loss": 0.6714,
      "step": 4780
    },
    {
      "epoch": 0.39993320531017784,
      "grad_norm": 0.368452250957489,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 4790
    },
    {
      "epoch": 0.4007681389329548,
      "grad_norm": 0.3463035225868225,
      "learning_rate": 2e-05,
      "loss": 0.6398,
      "step": 4800
    },
    {
      "epoch": 0.4016030725557318,
      "grad_norm": 0.5245319604873657,
      "learning_rate": 2e-05,
      "loss": 0.6594,
      "step": 4810
    },
    {
      "epoch": 0.40243800617850883,
      "grad_norm": 0.3590163290500641,
      "learning_rate": 2e-05,
      "loss": 0.6846,
      "step": 4820
    },
    {
      "epoch": 0.4032729398012858,
      "grad_norm": 0.3746512234210968,
      "learning_rate": 2e-05,
      "loss": 0.682,
      "step": 4830
    },
    {
      "epoch": 0.4041078734240628,
      "grad_norm": 0.33228105306625366,
      "learning_rate": 2e-05,
      "loss": 0.6479,
      "step": 4840
    },
    {
      "epoch": 0.40494280704683977,
      "grad_norm": 0.3701796531677246,
      "learning_rate": 2e-05,
      "loss": 0.6837,
      "step": 4850
    },
    {
      "epoch": 0.40577774066961675,
      "grad_norm": 0.36138930916786194,
      "learning_rate": 2e-05,
      "loss": 0.6473,
      "step": 4860
    },
    {
      "epoch": 0.4066126742923937,
      "grad_norm": 0.35197946429252625,
      "learning_rate": 2e-05,
      "loss": 0.6515,
      "step": 4870
    },
    {
      "epoch": 0.40744760791517076,
      "grad_norm": 0.4873769283294678,
      "learning_rate": 2e-05,
      "loss": 0.6549,
      "step": 4880
    },
    {
      "epoch": 0.40828254153794774,
      "grad_norm": 0.44790661334991455,
      "learning_rate": 2e-05,
      "loss": 0.6394,
      "step": 4890
    },
    {
      "epoch": 0.4091174751607247,
      "grad_norm": 0.3719101548194885,
      "learning_rate": 2e-05,
      "loss": 0.6286,
      "step": 4900
    },
    {
      "epoch": 0.4099524087835017,
      "grad_norm": 0.4109218418598175,
      "learning_rate": 2e-05,
      "loss": 0.6603,
      "step": 4910
    },
    {
      "epoch": 0.4107873424062787,
      "grad_norm": 0.37327003479003906,
      "learning_rate": 2e-05,
      "loss": 0.6595,
      "step": 4920
    },
    {
      "epoch": 0.4116222760290557,
      "grad_norm": 0.3086226284503937,
      "learning_rate": 2e-05,
      "loss": 0.6398,
      "step": 4930
    },
    {
      "epoch": 0.4124572096518327,
      "grad_norm": 0.340065062046051,
      "learning_rate": 2e-05,
      "loss": 0.6327,
      "step": 4940
    },
    {
      "epoch": 0.41329214327460967,
      "grad_norm": 0.7166498303413391,
      "learning_rate": 2e-05,
      "loss": 0.667,
      "step": 4950
    },
    {
      "epoch": 0.41412707689738665,
      "grad_norm": 0.33677515387535095,
      "learning_rate": 2e-05,
      "loss": 0.6354,
      "step": 4960
    },
    {
      "epoch": 0.4149620105201636,
      "grad_norm": 0.3930906057357788,
      "learning_rate": 2e-05,
      "loss": 0.6712,
      "step": 4970
    },
    {
      "epoch": 0.41579694414294066,
      "grad_norm": 0.35926154255867004,
      "learning_rate": 2e-05,
      "loss": 0.67,
      "step": 4980
    },
    {
      "epoch": 0.41663187776571764,
      "grad_norm": 0.3671875298023224,
      "learning_rate": 2e-05,
      "loss": 0.6186,
      "step": 4990
    },
    {
      "epoch": 0.4174668113884946,
      "grad_norm": 0.382476806640625,
      "learning_rate": 2e-05,
      "loss": 0.6393,
      "step": 5000
    },
    {
      "epoch": 0.4183017450112716,
      "grad_norm": 0.3704940676689148,
      "learning_rate": 2e-05,
      "loss": 0.6723,
      "step": 5010
    },
    {
      "epoch": 0.4191366786340486,
      "grad_norm": 0.34426355361938477,
      "learning_rate": 2e-05,
      "loss": 0.608,
      "step": 5020
    },
    {
      "epoch": 0.41997161225682555,
      "grad_norm": 0.38724127411842346,
      "learning_rate": 2e-05,
      "loss": 0.6355,
      "step": 5030
    },
    {
      "epoch": 0.4208065458796026,
      "grad_norm": 0.3683294951915741,
      "learning_rate": 2e-05,
      "loss": 0.6349,
      "step": 5040
    },
    {
      "epoch": 0.42164147950237957,
      "grad_norm": 0.3408048152923584,
      "learning_rate": 2e-05,
      "loss": 0.6337,
      "step": 5050
    },
    {
      "epoch": 0.42247641312515655,
      "grad_norm": 0.4431440234184265,
      "learning_rate": 2e-05,
      "loss": 0.6526,
      "step": 5060
    },
    {
      "epoch": 0.4233113467479335,
      "grad_norm": 0.4292963445186615,
      "learning_rate": 2e-05,
      "loss": 0.6279,
      "step": 5070
    },
    {
      "epoch": 0.4241462803707105,
      "grad_norm": 0.3619665801525116,
      "learning_rate": 2e-05,
      "loss": 0.6357,
      "step": 5080
    },
    {
      "epoch": 0.42498121399348754,
      "grad_norm": 0.35609790682792664,
      "learning_rate": 1.9999810479633703e-05,
      "loss": 0.6729,
      "step": 5090
    },
    {
      "epoch": 0.4258161476162645,
      "grad_norm": 0.3645738661289215,
      "learning_rate": 1.997707677020786e-05,
      "loss": 0.6319,
      "step": 5100
    },
    {
      "epoch": 0.4266510812390415,
      "grad_norm": 0.3419838845729828,
      "learning_rate": 1.9916538196124637e-05,
      "loss": 0.6146,
      "step": 5110
    },
    {
      "epoch": 0.4274860148618185,
      "grad_norm": 0.37080222368240356,
      "learning_rate": 1.9818425303863402e-05,
      "loss": 0.6512,
      "step": 5120
    },
    {
      "epoch": 0.42832094848459545,
      "grad_norm": 0.3506437838077545,
      "learning_rate": 1.9683111732582887e-05,
      "loss": 0.678,
      "step": 5130
    },
    {
      "epoch": 0.4291558821073725,
      "grad_norm": 0.41475293040275574,
      "learning_rate": 1.951111279120706e-05,
      "loss": 0.6327,
      "step": 5140
    },
    {
      "epoch": 0.42999081573014947,
      "grad_norm": 0.37617531418800354,
      "learning_rate": 1.9303083495996028e-05,
      "loss": 0.6397,
      "step": 5150
    },
    {
      "epoch": 0.43082574935292645,
      "grad_norm": 0.3370538651943207,
      "learning_rate": 1.9059816076075482e-05,
      "loss": 0.6713,
      "step": 5160
    },
    {
      "epoch": 0.4316606829757034,
      "grad_norm": 0.3520298898220062,
      "learning_rate": 1.8782236956424142e-05,
      "loss": 0.6165,
      "step": 5170
    },
    {
      "epoch": 0.4324956165984804,
      "grad_norm": 0.34632763266563416,
      "learning_rate": 1.8471403229808897e-05,
      "loss": 0.6303,
      "step": 5180
    },
    {
      "epoch": 0.4333305502212574,
      "grad_norm": 0.39343270659446716,
      "learning_rate": 1.8128498631103333e-05,
      "loss": 0.646,
      "step": 5190
    },
    {
      "epoch": 0.4341654838440344,
      "grad_norm": 0.3651024401187897,
      "learning_rate": 1.7754829029320507e-05,
      "loss": 0.673,
      "step": 5200
    },
    {
      "epoch": 0.4350004174668114,
      "grad_norm": 0.37827959656715393,
      "learning_rate": 1.7351817454527473e-05,
      "loss": 0.6658,
      "step": 5210
    },
    {
      "epoch": 0.4358353510895884,
      "grad_norm": 0.37647971510887146,
      "learning_rate": 1.6920998678580345e-05,
      "loss": 0.6394,
      "step": 5220
    },
    {
      "epoch": 0.43667028471236535,
      "grad_norm": 0.37646958231925964,
      "learning_rate": 1.6464013370317854e-05,
      "loss": 0.6462,
      "step": 5230
    },
    {
      "epoch": 0.43750521833514233,
      "grad_norm": 0.3650120794773102,
      "learning_rate": 1.5982601847471915e-05,
      "loss": 0.637,
      "step": 5240
    },
    {
      "epoch": 0.43834015195791937,
      "grad_norm": 0.3861883282661438,
      "learning_rate": 1.5478597449089553e-05,
      "loss": 0.6476,
      "step": 5250
    },
    {
      "epoch": 0.43917508558069634,
      "grad_norm": 0.3427833914756775,
      "learning_rate": 1.4953919553705725e-05,
      "loss": 0.6417,
      "step": 5260
    },
    {
      "epoch": 0.4400100192034733,
      "grad_norm": 0.36627012491226196,
      "learning_rate": 1.441056626985559e-05,
      "loss": 0.6371,
      "step": 5270
    },
    {
      "epoch": 0.4408449528262503,
      "grad_norm": 0.36164045333862305,
      "learning_rate": 1.3850606826762668e-05,
      "loss": 0.6476,
      "step": 5280
    },
    {
      "epoch": 0.4416798864490273,
      "grad_norm": 0.3590315878391266,
      "learning_rate": 1.3276173694181128e-05,
      "loss": 0.6249,
      "step": 5290
    },
    {
      "epoch": 0.4425148200718043,
      "grad_norm": 0.34728389978408813,
      "learning_rate": 1.2689454461401822e-05,
      "loss": 0.6541,
      "step": 5300
    },
    {
      "epoch": 0.4433497536945813,
      "grad_norm": 0.3587271273136139,
      "learning_rate": 1.2092683506348939e-05,
      "loss": 0.6271,
      "step": 5310
    },
    {
      "epoch": 0.4441846873173583,
      "grad_norm": 0.3040362596511841,
      "learning_rate": 1.14881334864935e-05,
      "loss": 0.5943,
      "step": 5320
    },
    {
      "epoch": 0.44501962094013525,
      "grad_norm": 0.3750527799129486,
      "learning_rate": 1.0878106683988481e-05,
      "loss": 0.636,
      "step": 5330
    },
    {
      "epoch": 0.44585455456291223,
      "grad_norm": 0.3489191234111786,
      "learning_rate": 1.026492623798551e-05,
      "loss": 0.6419,
      "step": 5340
    },
    {
      "epoch": 0.44668948818568927,
      "grad_norm": 0.31228408217430115,
      "learning_rate": 9.650927297522743e-06,
      "loss": 0.6517,
      "step": 5350
    },
    {
      "epoch": 0.44752442180846624,
      "grad_norm": 0.3532041013240814,
      "learning_rate": 9.038448128676012e-06,
      "loss": 0.6376,
      "step": 5360
    },
    {
      "epoch": 0.4483593554312432,
      "grad_norm": 0.31956541538238525,
      "learning_rate": 8.42982120983949e-06,
      "loss": 0.6579,
      "step": 5370
    },
    {
      "epoch": 0.4491942890540202,
      "grad_norm": 0.33854755759239197,
      "learning_rate": 7.827364349047303e-06,
      "loss": 0.6238,
      "step": 5380
    },
    {
      "epoch": 0.4500292226767972,
      "grad_norm": 0.33272674679756165,
      "learning_rate": 7.233371857163706e-06,
      "loss": 0.6254,
      "step": 5390
    },
    {
      "epoch": 0.45086415629957416,
      "grad_norm": 0.5521824359893799,
      "learning_rate": 6.650105810556492e-06,
      "loss": 0.644,
      "step": 5400
    },
    {
      "epoch": 0.4516990899223512,
      "grad_norm": 0.35825058817863464,
      "learning_rate": 6.079787436527681e-06,
      "loss": 0.6096,
      "step": 5410
    },
    {
      "epoch": 0.4525340235451282,
      "grad_norm": 0.36263272166252136,
      "learning_rate": 5.524588654307961e-06,
      "loss": 0.6341,
      "step": 5420
    },
    {
      "epoch": 0.45336895716790515,
      "grad_norm": 0.3541775643825531,
      "learning_rate": 4.986623803828985e-06,
      "loss": 0.6337,
      "step": 5430
    },
    {
      "epoch": 0.45420389079068213,
      "grad_norm": 0.3440946042537689,
      "learning_rate": 4.467941593772423e-06,
      "loss": 0.6088,
      "step": 5440
    },
    {
      "epoch": 0.4550388244134591,
      "grad_norm": 0.3492544889450073,
      "learning_rate": 3.9705172995597606e-06,
      "loss": 0.618,
      "step": 5450
    },
    {
      "epoch": 0.45587375803623614,
      "grad_norm": 0.3273930847644806,
      "learning_rate": 3.4962452409948346e-06,
      "loss": 0.6556,
      "step": 5460
    },
    {
      "epoch": 0.4567086916590131,
      "grad_norm": 0.36364230513572693,
      "learning_rate": 3.046931568206211e-06,
      "loss": 0.6284,
      "step": 5470
    },
    {
      "epoch": 0.4575436252817901,
      "grad_norm": 0.32170334458351135,
      "learning_rate": 2.624287383362404e-06,
      "loss": 0.6715,
      "step": 5480
    },
    {
      "epoch": 0.4583785589045671,
      "grad_norm": 0.35907304286956787,
      "learning_rate": 2.2299222243541845e-06,
      "loss": 0.6671,
      "step": 5490
    },
    {
      "epoch": 0.45921349252734406,
      "grad_norm": 0.4123549163341522,
      "learning_rate": 1.8653379352598246e-06,
      "loss": 0.6547,
      "step": 5500
    },
    {
      "epoch": 0.4600484261501211,
      "grad_norm": 0.34175819158554077,
      "learning_rate": 1.5319229469361075e-06,
      "loss": 0.6614,
      "step": 5510
    },
    {
      "epoch": 0.46088335977289807,
      "grad_norm": 0.35448622703552246,
      "learning_rate": 1.2309469895160917e-06,
      "loss": 0.6563,
      "step": 5520
    },
    {
      "epoch": 0.46171829339567505,
      "grad_norm": 0.36138349771499634,
      "learning_rate": 9.635562569497974e-07,
      "loss": 0.6615,
      "step": 5530
    },
    {
      "epoch": 0.46255322701845203,
      "grad_norm": 0.33417871594429016,
      "learning_rate": 7.307690420024867e-07,
      "loss": 0.6735,
      "step": 5540
    },
    {
      "epoch": 0.463388160641229,
      "grad_norm": 0.34698307514190674,
      "learning_rate": 5.334718583335942e-07,
      "loss": 0.6595,
      "step": 5550
    },
    {
      "epoch": 0.464223094264006,
      "grad_norm": 0.3464798331260681,
      "learning_rate": 3.7241606442442763e-07,
      "loss": 0.6275,
      "step": 5560
    },
    {
      "epoch": 0.465058027886783,
      "grad_norm": 0.35537174344062805,
      "learning_rate": 2.482150022115879e-07,
      "loss": 0.6542,
      "step": 5570
    },
    {
      "epoch": 0.46589296150956,
      "grad_norm": 0.3661311864852905,
      "learning_rate": 1.6134166132293174e-07,
      "loss": 0.6211,
      "step": 5580
    },
    {
      "epoch": 0.466727895132337,
      "grad_norm": 0.33591344952583313,
      "learning_rate": 1.1212687781127357e-07,
      "loss": 0.616,
      "step": 5590
    },
    {
      "epoch": 0.46731234866828086,
      "step": 5597,
      "total_flos": 2.2737574718397743e+19,
      "train_loss": 0.02194896335577952,
      "train_runtime": 1801.4216,
      "train_samples_per_second": 37.284,
      "train_steps_per_second": 3.107
    }
  ],
  "logging_steps": 10,
  "max_steps": 5597,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 159,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.2737574718397743e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}