{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9346246973365617,
  "eval_steps": 500,
  "global_step": 11194,
  "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": 0.35748395323753357,
      "learning_rate": 2e-05,
      "loss": 0.6729,
      "step": 5090
    },
    {
      "epoch": 0.4258161476162645,
      "grad_norm": 0.38644883036613464,
      "learning_rate": 2e-05,
      "loss": 0.6319,
      "step": 5100
    },
    {
      "epoch": 0.4266510812390415,
      "grad_norm": 0.34421461820602417,
      "learning_rate": 2e-05,
      "loss": 0.6147,
      "step": 5110
    },
    {
      "epoch": 0.4274860148618185,
      "grad_norm": 0.3712674677371979,
      "learning_rate": 2e-05,
      "loss": 0.6513,
      "step": 5120
    },
    {
      "epoch": 0.42832094848459545,
      "grad_norm": 0.37810656428337097,
      "learning_rate": 2e-05,
      "loss": 0.6782,
      "step": 5130
    },
    {
      "epoch": 0.4291558821073725,
      "grad_norm": 0.4167480170726776,
      "learning_rate": 2e-05,
      "loss": 0.6328,
      "step": 5140
    },
    {
      "epoch": 0.42999081573014947,
      "grad_norm": 0.61541348695755,
      "learning_rate": 2e-05,
      "loss": 0.64,
      "step": 5150
    },
    {
      "epoch": 0.43082574935292645,
      "grad_norm": 0.3594754636287689,
      "learning_rate": 2e-05,
      "loss": 0.6716,
      "step": 5160
    },
    {
      "epoch": 0.4316606829757034,
      "grad_norm": 0.35676905512809753,
      "learning_rate": 2e-05,
      "loss": 0.617,
      "step": 5170
    },
    {
      "epoch": 0.4324956165984804,
      "grad_norm": 0.34889644384384155,
      "learning_rate": 2e-05,
      "loss": 0.6308,
      "step": 5180
    },
    {
      "epoch": 0.4333305502212574,
      "grad_norm": 0.3722809851169586,
      "learning_rate": 2e-05,
      "loss": 0.6467,
      "step": 5190
    },
    {
      "epoch": 0.4341654838440344,
      "grad_norm": 0.43137869238853455,
      "learning_rate": 2e-05,
      "loss": 0.6738,
      "step": 5200
    },
    {
      "epoch": 0.4350004174668114,
      "grad_norm": 0.3877946138381958,
      "learning_rate": 2e-05,
      "loss": 0.6668,
      "step": 5210
    },
    {
      "epoch": 0.4358353510895884,
      "grad_norm": 0.39086174964904785,
      "learning_rate": 2e-05,
      "loss": 0.6408,
      "step": 5220
    },
    {
      "epoch": 0.43667028471236535,
      "grad_norm": 0.37674352526664734,
      "learning_rate": 2e-05,
      "loss": 0.6478,
      "step": 5230
    },
    {
      "epoch": 0.43750521833514233,
      "grad_norm": 0.3666381239891052,
      "learning_rate": 2e-05,
      "loss": 0.6393,
      "step": 5240
    },
    {
      "epoch": 0.43834015195791937,
      "grad_norm": 0.4082038700580597,
      "learning_rate": 2e-05,
      "loss": 0.6501,
      "step": 5250
    },
    {
      "epoch": 0.43917508558069634,
      "grad_norm": 0.34695321321487427,
      "learning_rate": 2e-05,
      "loss": 0.6444,
      "step": 5260
    },
    {
      "epoch": 0.4400100192034733,
      "grad_norm": 0.3737017512321472,
      "learning_rate": 2e-05,
      "loss": 0.64,
      "step": 5270
    },
    {
      "epoch": 0.4408449528262503,
      "grad_norm": 0.3662821054458618,
      "learning_rate": 2e-05,
      "loss": 0.6509,
      "step": 5280
    },
    {
      "epoch": 0.4416798864490273,
      "grad_norm": 0.4254063069820404,
      "learning_rate": 2e-05,
      "loss": 0.6288,
      "step": 5290
    },
    {
      "epoch": 0.4425148200718043,
      "grad_norm": 0.3648417294025421,
      "learning_rate": 2e-05,
      "loss": 0.6581,
      "step": 5300
    },
    {
      "epoch": 0.4433497536945813,
      "grad_norm": 0.3542555272579193,
      "learning_rate": 2e-05,
      "loss": 0.6315,
      "step": 5310
    },
    {
      "epoch": 0.4441846873173583,
      "grad_norm": 0.3174424171447754,
      "learning_rate": 2e-05,
      "loss": 0.5985,
      "step": 5320
    },
    {
      "epoch": 0.44501962094013525,
      "grad_norm": 0.35006508231163025,
      "learning_rate": 2e-05,
      "loss": 0.6407,
      "step": 5330
    },
    {
      "epoch": 0.44585455456291223,
      "grad_norm": 0.3580966591835022,
      "learning_rate": 2e-05,
      "loss": 0.6485,
      "step": 5340
    },
    {
      "epoch": 0.44668948818568927,
      "grad_norm": 0.3129090666770935,
      "learning_rate": 2e-05,
      "loss": 0.6582,
      "step": 5350
    },
    {
      "epoch": 0.44752442180846624,
      "grad_norm": 0.34629911184310913,
      "learning_rate": 2e-05,
      "loss": 0.6438,
      "step": 5360
    },
    {
      "epoch": 0.4483593554312432,
      "grad_norm": 0.33315667510032654,
      "learning_rate": 2e-05,
      "loss": 0.6656,
      "step": 5370
    },
    {
      "epoch": 0.4491942890540202,
      "grad_norm": 0.36381271481513977,
      "learning_rate": 2e-05,
      "loss": 0.6308,
      "step": 5380
    },
    {
      "epoch": 0.4500292226767972,
      "grad_norm": 0.3453449308872223,
      "learning_rate": 2e-05,
      "loss": 0.6335,
      "step": 5390
    },
    {
      "epoch": 0.45086415629957416,
      "grad_norm": 0.38269010186195374,
      "learning_rate": 2e-05,
      "loss": 0.6527,
      "step": 5400
    },
    {
      "epoch": 0.4516990899223512,
      "grad_norm": 0.3619810938835144,
      "learning_rate": 2e-05,
      "loss": 0.6169,
      "step": 5410
    },
    {
      "epoch": 0.4525340235451282,
      "grad_norm": 0.38358864188194275,
      "learning_rate": 2e-05,
      "loss": 0.6429,
      "step": 5420
    },
    {
      "epoch": 0.45336895716790515,
      "grad_norm": 0.35633301734924316,
      "learning_rate": 2e-05,
      "loss": 0.6437,
      "step": 5430
    },
    {
      "epoch": 0.45420389079068213,
      "grad_norm": 0.3583010733127594,
      "learning_rate": 2e-05,
      "loss": 0.6177,
      "step": 5440
    },
    {
      "epoch": 0.4550388244134591,
      "grad_norm": 0.371682345867157,
      "learning_rate": 2e-05,
      "loss": 0.6271,
      "step": 5450
    },
    {
      "epoch": 0.45587375803623614,
      "grad_norm": 0.3403146266937256,
      "learning_rate": 2e-05,
      "loss": 0.6653,
      "step": 5460
    },
    {
      "epoch": 0.4567086916590131,
      "grad_norm": 0.3783678710460663,
      "learning_rate": 2e-05,
      "loss": 0.6391,
      "step": 5470
    },
    {
      "epoch": 0.4575436252817901,
      "grad_norm": 0.33547651767730713,
      "learning_rate": 2e-05,
      "loss": 0.6832,
      "step": 5480
    },
    {
      "epoch": 0.4583785589045671,
      "grad_norm": 0.35089755058288574,
      "learning_rate": 2e-05,
      "loss": 0.6776,
      "step": 5490
    },
    {
      "epoch": 0.45921349252734406,
      "grad_norm": 0.4087474048137665,
      "learning_rate": 2e-05,
      "loss": 0.666,
      "step": 5500
    },
    {
      "epoch": 0.4600484261501211,
      "grad_norm": 0.3589717745780945,
      "learning_rate": 2e-05,
      "loss": 0.6732,
      "step": 5510
    },
    {
      "epoch": 0.46088335977289807,
      "grad_norm": 0.37691444158554077,
      "learning_rate": 2e-05,
      "loss": 0.6668,
      "step": 5520
    },
    {
      "epoch": 0.46171829339567505,
      "grad_norm": 0.38058099150657654,
      "learning_rate": 2e-05,
      "loss": 0.6719,
      "step": 5530
    },
    {
      "epoch": 0.46255322701845203,
      "grad_norm": 0.34354525804519653,
      "learning_rate": 2e-05,
      "loss": 0.6855,
      "step": 5540
    },
    {
      "epoch": 0.463388160641229,
      "grad_norm": 0.3705137372016907,
      "learning_rate": 2e-05,
      "loss": 0.671,
      "step": 5550
    },
    {
      "epoch": 0.464223094264006,
      "grad_norm": 0.5769681334495544,
      "learning_rate": 2e-05,
      "loss": 0.6378,
      "step": 5560
    },
    {
      "epoch": 0.465058027886783,
      "grad_norm": 0.38412007689476013,
      "learning_rate": 2e-05,
      "loss": 0.6661,
      "step": 5570
    },
    {
      "epoch": 0.46589296150956,
      "grad_norm": 0.4786997139453888,
      "learning_rate": 2e-05,
      "loss": 0.6327,
      "step": 5580
    },
    {
      "epoch": 0.466727895132337,
      "grad_norm": 0.34899458289146423,
      "learning_rate": 2e-05,
      "loss": 0.6253,
      "step": 5590
    },
    {
      "epoch": 0.46756282875511396,
      "grad_norm": 0.39586830139160156,
      "learning_rate": 2e-05,
      "loss": 0.642,
      "step": 5600
    },
    {
      "epoch": 0.46839776237789094,
      "grad_norm": 0.34808653593063354,
      "learning_rate": 2e-05,
      "loss": 0.6281,
      "step": 5610
    },
    {
      "epoch": 0.46923269600066797,
      "grad_norm": 0.36236539483070374,
      "learning_rate": 2e-05,
      "loss": 0.6356,
      "step": 5620
    },
    {
      "epoch": 0.47006762962344495,
      "grad_norm": 0.3778802156448364,
      "learning_rate": 2e-05,
      "loss": 0.6664,
      "step": 5630
    },
    {
      "epoch": 0.47090256324622193,
      "grad_norm": 0.34362590312957764,
      "learning_rate": 2e-05,
      "loss": 0.6604,
      "step": 5640
    },
    {
      "epoch": 0.4717374968689989,
      "grad_norm": 0.35422155261039734,
      "learning_rate": 2e-05,
      "loss": 0.6345,
      "step": 5650
    },
    {
      "epoch": 0.4725724304917759,
      "grad_norm": 0.3783043622970581,
      "learning_rate": 2e-05,
      "loss": 0.6519,
      "step": 5660
    },
    {
      "epoch": 0.4734073641145529,
      "grad_norm": 0.3652976155281067,
      "learning_rate": 2e-05,
      "loss": 0.664,
      "step": 5670
    },
    {
      "epoch": 0.4742422977373299,
      "grad_norm": 0.3395410478115082,
      "learning_rate": 2e-05,
      "loss": 0.632,
      "step": 5680
    },
    {
      "epoch": 0.4750772313601069,
      "grad_norm": 0.34737828373908997,
      "learning_rate": 2e-05,
      "loss": 0.6407,
      "step": 5690
    },
    {
      "epoch": 0.47591216498288386,
      "grad_norm": 0.3878612518310547,
      "learning_rate": 2e-05,
      "loss": 0.6394,
      "step": 5700
    },
    {
      "epoch": 0.47674709860566084,
      "grad_norm": 0.4202122092247009,
      "learning_rate": 2e-05,
      "loss": 0.6814,
      "step": 5710
    },
    {
      "epoch": 0.4775820322284378,
      "grad_norm": 0.3601955771446228,
      "learning_rate": 2e-05,
      "loss": 0.6276,
      "step": 5720
    },
    {
      "epoch": 0.47841696585121485,
      "grad_norm": 0.3742481768131256,
      "learning_rate": 2e-05,
      "loss": 0.6565,
      "step": 5730
    },
    {
      "epoch": 0.4792518994739918,
      "grad_norm": 0.367070734500885,
      "learning_rate": 2e-05,
      "loss": 0.634,
      "step": 5740
    },
    {
      "epoch": 0.4800868330967688,
      "grad_norm": 0.3631555736064911,
      "learning_rate": 2e-05,
      "loss": 0.6361,
      "step": 5750
    },
    {
      "epoch": 0.4809217667195458,
      "grad_norm": 0.3798542618751526,
      "learning_rate": 2e-05,
      "loss": 0.655,
      "step": 5760
    },
    {
      "epoch": 0.48175670034232276,
      "grad_norm": 0.3490215241909027,
      "learning_rate": 2e-05,
      "loss": 0.6568,
      "step": 5770
    },
    {
      "epoch": 0.4825916339650998,
      "grad_norm": 0.3756038248538971,
      "learning_rate": 2e-05,
      "loss": 0.646,
      "step": 5780
    },
    {
      "epoch": 0.4834265675878768,
      "grad_norm": 0.3717809319496155,
      "learning_rate": 2e-05,
      "loss": 0.6153,
      "step": 5790
    },
    {
      "epoch": 0.48426150121065376,
      "grad_norm": 0.360704630613327,
      "learning_rate": 2e-05,
      "loss": 0.636,
      "step": 5800
    },
    {
      "epoch": 0.48509643483343073,
      "grad_norm": 0.37858933210372925,
      "learning_rate": 2e-05,
      "loss": 0.6508,
      "step": 5810
    },
    {
      "epoch": 0.4859313684562077,
      "grad_norm": 0.37836575508117676,
      "learning_rate": 2e-05,
      "loss": 0.6774,
      "step": 5820
    },
    {
      "epoch": 0.48676630207898475,
      "grad_norm": 0.37883585691452026,
      "learning_rate": 2e-05,
      "loss": 0.6876,
      "step": 5830
    },
    {
      "epoch": 0.4876012357017617,
      "grad_norm": 0.3802921175956726,
      "learning_rate": 2e-05,
      "loss": 0.6103,
      "step": 5840
    },
    {
      "epoch": 0.4884361693245387,
      "grad_norm": 0.35831770300865173,
      "learning_rate": 2e-05,
      "loss": 0.6348,
      "step": 5850
    },
    {
      "epoch": 0.4892711029473157,
      "grad_norm": 0.3879816234111786,
      "learning_rate": 2e-05,
      "loss": 0.6447,
      "step": 5860
    },
    {
      "epoch": 0.49010603657009266,
      "grad_norm": 0.4372434914112091,
      "learning_rate": 2e-05,
      "loss": 0.6083,
      "step": 5870
    },
    {
      "epoch": 0.49094097019286964,
      "grad_norm": 0.37333860993385315,
      "learning_rate": 2e-05,
      "loss": 0.6404,
      "step": 5880
    },
    {
      "epoch": 0.4917759038156467,
      "grad_norm": 0.4594089686870575,
      "learning_rate": 2e-05,
      "loss": 0.6459,
      "step": 5890
    },
    {
      "epoch": 0.49261083743842365,
      "grad_norm": 0.3620820939540863,
      "learning_rate": 2e-05,
      "loss": 0.6139,
      "step": 5900
    },
    {
      "epoch": 0.49344577106120063,
      "grad_norm": 0.38887137174606323,
      "learning_rate": 2e-05,
      "loss": 0.6506,
      "step": 5910
    },
    {
      "epoch": 0.4942807046839776,
      "grad_norm": 0.3226620852947235,
      "learning_rate": 2e-05,
      "loss": 0.6273,
      "step": 5920
    },
    {
      "epoch": 0.4951156383067546,
      "grad_norm": 0.36273685097694397,
      "learning_rate": 2e-05,
      "loss": 0.6426,
      "step": 5930
    },
    {
      "epoch": 0.4959505719295316,
      "grad_norm": 0.34463173151016235,
      "learning_rate": 2e-05,
      "loss": 0.5976,
      "step": 5940
    },
    {
      "epoch": 0.4967855055523086,
      "grad_norm": 0.3716309368610382,
      "learning_rate": 2e-05,
      "loss": 0.6388,
      "step": 5950
    },
    {
      "epoch": 0.4976204391750856,
      "grad_norm": 0.34396132826805115,
      "learning_rate": 2e-05,
      "loss": 0.6276,
      "step": 5960
    },
    {
      "epoch": 0.49845537279786256,
      "grad_norm": 0.3970491588115692,
      "learning_rate": 2e-05,
      "loss": 0.6451,
      "step": 5970
    },
    {
      "epoch": 0.49929030642063954,
      "grad_norm": 0.36600780487060547,
      "learning_rate": 2e-05,
      "loss": 0.6161,
      "step": 5980
    },
    {
      "epoch": 0.5001252400434165,
      "grad_norm": 0.355255663394928,
      "learning_rate": 2e-05,
      "loss": 0.655,
      "step": 5990
    },
    {
      "epoch": 0.5009601736661935,
      "grad_norm": 0.33628931641578674,
      "learning_rate": 2e-05,
      "loss": 0.6386,
      "step": 6000
    },
    {
      "epoch": 0.5017951072889705,
      "grad_norm": 0.4485655128955841,
      "learning_rate": 2e-05,
      "loss": 0.6308,
      "step": 6010
    },
    {
      "epoch": 0.5026300409117476,
      "grad_norm": 0.39465659856796265,
      "learning_rate": 2e-05,
      "loss": 0.6393,
      "step": 6020
    },
    {
      "epoch": 0.5034649745345245,
      "grad_norm": 0.34330031275749207,
      "learning_rate": 2e-05,
      "loss": 0.6105,
      "step": 6030
    },
    {
      "epoch": 0.5042999081573015,
      "grad_norm": 0.344465970993042,
      "learning_rate": 2e-05,
      "loss": 0.6074,
      "step": 6040
    },
    {
      "epoch": 0.5051348417800785,
      "grad_norm": 0.47026526927948,
      "learning_rate": 2e-05,
      "loss": 0.6466,
      "step": 6050
    },
    {
      "epoch": 0.5059697754028555,
      "grad_norm": 0.3524168133735657,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 6060
    },
    {
      "epoch": 0.5068047090256325,
      "grad_norm": 0.37515389919281006,
      "learning_rate": 2e-05,
      "loss": 0.6206,
      "step": 6070
    },
    {
      "epoch": 0.5076396426484094,
      "grad_norm": 0.3767631947994232,
      "learning_rate": 2e-05,
      "loss": 0.6335,
      "step": 6080
    },
    {
      "epoch": 0.5084745762711864,
      "grad_norm": 0.40053462982177734,
      "learning_rate": 2e-05,
      "loss": 0.6706,
      "step": 6090
    },
    {
      "epoch": 0.5093095098939634,
      "grad_norm": 0.3491964638233185,
      "learning_rate": 2e-05,
      "loss": 0.6291,
      "step": 6100
    },
    {
      "epoch": 0.5101444435167404,
      "grad_norm": 0.33106136322021484,
      "learning_rate": 2e-05,
      "loss": 0.6503,
      "step": 6110
    },
    {
      "epoch": 0.5109793771395174,
      "grad_norm": 0.3545686602592468,
      "learning_rate": 2e-05,
      "loss": 0.6446,
      "step": 6120
    },
    {
      "epoch": 0.5118143107622944,
      "grad_norm": 0.35029667615890503,
      "learning_rate": 2e-05,
      "loss": 0.6501,
      "step": 6130
    },
    {
      "epoch": 0.5126492443850714,
      "grad_norm": 0.3516990840435028,
      "learning_rate": 2e-05,
      "loss": 0.5994,
      "step": 6140
    },
    {
      "epoch": 0.5134841780078484,
      "grad_norm": 0.3215792179107666,
      "learning_rate": 2e-05,
      "loss": 0.6585,
      "step": 6150
    },
    {
      "epoch": 0.5143191116306254,
      "grad_norm": 0.3781889081001282,
      "learning_rate": 2e-05,
      "loss": 0.6274,
      "step": 6160
    },
    {
      "epoch": 0.5151540452534024,
      "grad_norm": 0.34870731830596924,
      "learning_rate": 2e-05,
      "loss": 0.6424,
      "step": 6170
    },
    {
      "epoch": 0.5159889788761793,
      "grad_norm": 0.34317317605018616,
      "learning_rate": 2e-05,
      "loss": 0.6071,
      "step": 6180
    },
    {
      "epoch": 0.5168239124989563,
      "grad_norm": 0.35119879245758057,
      "learning_rate": 2e-05,
      "loss": 0.6334,
      "step": 6190
    },
    {
      "epoch": 0.5176588461217333,
      "grad_norm": 0.3915646970272064,
      "learning_rate": 2e-05,
      "loss": 0.6148,
      "step": 6200
    },
    {
      "epoch": 0.5184937797445103,
      "grad_norm": 0.3533666133880615,
      "learning_rate": 2e-05,
      "loss": 0.63,
      "step": 6210
    },
    {
      "epoch": 0.5193287133672873,
      "grad_norm": 0.4127843677997589,
      "learning_rate": 2e-05,
      "loss": 0.6741,
      "step": 6220
    },
    {
      "epoch": 0.5201636469900643,
      "grad_norm": 0.3514946699142456,
      "learning_rate": 2e-05,
      "loss": 0.6031,
      "step": 6230
    },
    {
      "epoch": 0.5209985806128413,
      "grad_norm": 0.3975648581981659,
      "learning_rate": 2e-05,
      "loss": 0.6474,
      "step": 6240
    },
    {
      "epoch": 0.5218335142356183,
      "grad_norm": 0.367906391620636,
      "learning_rate": 2e-05,
      "loss": 0.6582,
      "step": 6250
    },
    {
      "epoch": 0.5226684478583953,
      "grad_norm": 0.35561466217041016,
      "learning_rate": 2e-05,
      "loss": 0.6386,
      "step": 6260
    },
    {
      "epoch": 0.5235033814811723,
      "grad_norm": 0.35916227102279663,
      "learning_rate": 2e-05,
      "loss": 0.6607,
      "step": 6270
    },
    {
      "epoch": 0.5243383151039492,
      "grad_norm": 0.3263804614543915,
      "learning_rate": 2e-05,
      "loss": 0.6387,
      "step": 6280
    },
    {
      "epoch": 0.5251732487267262,
      "grad_norm": 0.36467182636260986,
      "learning_rate": 2e-05,
      "loss": 0.6535,
      "step": 6290
    },
    {
      "epoch": 0.5260081823495032,
      "grad_norm": 0.3616567552089691,
      "learning_rate": 2e-05,
      "loss": 0.6263,
      "step": 6300
    },
    {
      "epoch": 0.5268431159722802,
      "grad_norm": 0.4070850908756256,
      "learning_rate": 2e-05,
      "loss": 0.6475,
      "step": 6310
    },
    {
      "epoch": 0.5276780495950572,
      "grad_norm": 0.3647271990776062,
      "learning_rate": 2e-05,
      "loss": 0.6301,
      "step": 6320
    },
    {
      "epoch": 0.5285129832178341,
      "grad_norm": 0.35368776321411133,
      "learning_rate": 2e-05,
      "loss": 0.6311,
      "step": 6330
    },
    {
      "epoch": 0.5293479168406112,
      "grad_norm": 0.41903790831565857,
      "learning_rate": 2e-05,
      "loss": 0.6749,
      "step": 6340
    },
    {
      "epoch": 0.5301828504633882,
      "grad_norm": 0.37777870893478394,
      "learning_rate": 2e-05,
      "loss": 0.6466,
      "step": 6350
    },
    {
      "epoch": 0.5310177840861652,
      "grad_norm": 0.3963848650455475,
      "learning_rate": 2e-05,
      "loss": 0.6328,
      "step": 6360
    },
    {
      "epoch": 0.5318527177089422,
      "grad_norm": 0.35825666785240173,
      "learning_rate": 2e-05,
      "loss": 0.6571,
      "step": 6370
    },
    {
      "epoch": 0.5326876513317191,
      "grad_norm": 0.39799344539642334,
      "learning_rate": 2e-05,
      "loss": 0.6703,
      "step": 6380
    },
    {
      "epoch": 0.5335225849544961,
      "grad_norm": 0.37488022446632385,
      "learning_rate": 2e-05,
      "loss": 0.6536,
      "step": 6390
    },
    {
      "epoch": 0.5343575185772731,
      "grad_norm": 0.35889244079589844,
      "learning_rate": 2e-05,
      "loss": 0.6278,
      "step": 6400
    },
    {
      "epoch": 0.5351924522000501,
      "grad_norm": 0.3627670705318451,
      "learning_rate": 2e-05,
      "loss": 0.6449,
      "step": 6410
    },
    {
      "epoch": 0.536027385822827,
      "grad_norm": 0.35776829719543457,
      "learning_rate": 2e-05,
      "loss": 0.6619,
      "step": 6420
    },
    {
      "epoch": 0.536862319445604,
      "grad_norm": 0.3869832158088684,
      "learning_rate": 2e-05,
      "loss": 0.621,
      "step": 6430
    },
    {
      "epoch": 0.537697253068381,
      "grad_norm": 0.3861088156700134,
      "learning_rate": 2e-05,
      "loss": 0.6265,
      "step": 6440
    },
    {
      "epoch": 0.5385321866911581,
      "grad_norm": 0.36469510197639465,
      "learning_rate": 2e-05,
      "loss": 0.6161,
      "step": 6450
    },
    {
      "epoch": 0.5393671203139351,
      "grad_norm": 0.35532933473587036,
      "learning_rate": 2e-05,
      "loss": 0.614,
      "step": 6460
    },
    {
      "epoch": 0.5402020539367121,
      "grad_norm": 0.3811328113079071,
      "learning_rate": 2e-05,
      "loss": 0.6311,
      "step": 6470
    },
    {
      "epoch": 0.541036987559489,
      "grad_norm": 0.38329440355300903,
      "learning_rate": 2e-05,
      "loss": 0.6287,
      "step": 6480
    },
    {
      "epoch": 0.541871921182266,
      "grad_norm": 0.3419809639453888,
      "learning_rate": 2e-05,
      "loss": 0.6374,
      "step": 6490
    },
    {
      "epoch": 0.542706854805043,
      "grad_norm": 0.3757476806640625,
      "learning_rate": 2e-05,
      "loss": 0.6464,
      "step": 6500
    },
    {
      "epoch": 0.54354178842782,
      "grad_norm": 0.3522132337093353,
      "learning_rate": 2e-05,
      "loss": 0.6135,
      "step": 6510
    },
    {
      "epoch": 0.544376722050597,
      "grad_norm": 0.3850357234477997,
      "learning_rate": 2e-05,
      "loss": 0.6264,
      "step": 6520
    },
    {
      "epoch": 0.5452116556733739,
      "grad_norm": 0.37768569588661194,
      "learning_rate": 2e-05,
      "loss": 0.6255,
      "step": 6530
    },
    {
      "epoch": 0.5460465892961509,
      "grad_norm": 0.396787166595459,
      "learning_rate": 2e-05,
      "loss": 0.6504,
      "step": 6540
    },
    {
      "epoch": 0.546881522918928,
      "grad_norm": 0.3368368148803711,
      "learning_rate": 2e-05,
      "loss": 0.6394,
      "step": 6550
    },
    {
      "epoch": 0.547716456541705,
      "grad_norm": 0.33096176385879517,
      "learning_rate": 2e-05,
      "loss": 0.6119,
      "step": 6560
    },
    {
      "epoch": 0.548551390164482,
      "grad_norm": 0.3885151445865631,
      "learning_rate": 2e-05,
      "loss": 0.6226,
      "step": 6570
    },
    {
      "epoch": 0.5493863237872589,
      "grad_norm": 0.3361867070198059,
      "learning_rate": 2e-05,
      "loss": 0.6181,
      "step": 6580
    },
    {
      "epoch": 0.5502212574100359,
      "grad_norm": 0.35794273018836975,
      "learning_rate": 2e-05,
      "loss": 0.5978,
      "step": 6590
    },
    {
      "epoch": 0.5510561910328129,
      "grad_norm": 0.33447200059890747,
      "learning_rate": 2e-05,
      "loss": 0.6326,
      "step": 6600
    },
    {
      "epoch": 0.5518911246555899,
      "grad_norm": 0.5452051758766174,
      "learning_rate": 2e-05,
      "loss": 0.6316,
      "step": 6610
    },
    {
      "epoch": 0.5527260582783668,
      "grad_norm": 0.36688533425331116,
      "learning_rate": 2e-05,
      "loss": 0.651,
      "step": 6620
    },
    {
      "epoch": 0.5535609919011438,
      "grad_norm": 2.3416237831115723,
      "learning_rate": 2e-05,
      "loss": 0.606,
      "step": 6630
    },
    {
      "epoch": 0.5543959255239208,
      "grad_norm": 0.690862238407135,
      "learning_rate": 2e-05,
      "loss": 0.6089,
      "step": 6640
    },
    {
      "epoch": 0.5552308591466978,
      "grad_norm": 0.34501972794532776,
      "learning_rate": 2e-05,
      "loss": 0.6284,
      "step": 6650
    },
    {
      "epoch": 0.5560657927694749,
      "grad_norm": 0.3753596246242523,
      "learning_rate": 2e-05,
      "loss": 0.6314,
      "step": 6660
    },
    {
      "epoch": 0.5569007263922519,
      "grad_norm": 0.40744155645370483,
      "learning_rate": 2e-05,
      "loss": 0.6748,
      "step": 6670
    },
    {
      "epoch": 0.5577356600150288,
      "grad_norm": 0.3728608787059784,
      "learning_rate": 2e-05,
      "loss": 0.5853,
      "step": 6680
    },
    {
      "epoch": 0.5585705936378058,
      "grad_norm": 0.3556704521179199,
      "learning_rate": 2e-05,
      "loss": 0.655,
      "step": 6690
    },
    {
      "epoch": 0.5594055272605828,
      "grad_norm": 0.37264978885650635,
      "learning_rate": 2e-05,
      "loss": 0.5917,
      "step": 6700
    },
    {
      "epoch": 0.5602404608833598,
      "grad_norm": 0.33626896142959595,
      "learning_rate": 2e-05,
      "loss": 0.6145,
      "step": 6710
    },
    {
      "epoch": 0.5610753945061367,
      "grad_norm": 0.37021875381469727,
      "learning_rate": 2e-05,
      "loss": 0.6305,
      "step": 6720
    },
    {
      "epoch": 0.5619103281289137,
      "grad_norm": 0.35904595255851746,
      "learning_rate": 2e-05,
      "loss": 0.6374,
      "step": 6730
    },
    {
      "epoch": 0.5627452617516907,
      "grad_norm": 0.3719269633293152,
      "learning_rate": 2e-05,
      "loss": 0.633,
      "step": 6740
    },
    {
      "epoch": 0.5635801953744677,
      "grad_norm": 0.3872597813606262,
      "learning_rate": 2e-05,
      "loss": 0.6554,
      "step": 6750
    },
    {
      "epoch": 0.5644151289972447,
      "grad_norm": 0.3982311487197876,
      "learning_rate": 2e-05,
      "loss": 0.658,
      "step": 6760
    },
    {
      "epoch": 0.5652500626200218,
      "grad_norm": 0.36278238892555237,
      "learning_rate": 2e-05,
      "loss": 0.6453,
      "step": 6770
    },
    {
      "epoch": 0.5660849962427987,
      "grad_norm": 0.36902496218681335,
      "learning_rate": 2e-05,
      "loss": 0.6192,
      "step": 6780
    },
    {
      "epoch": 0.5669199298655757,
      "grad_norm": 0.37851348519325256,
      "learning_rate": 2e-05,
      "loss": 0.6212,
      "step": 6790
    },
    {
      "epoch": 0.5677548634883527,
      "grad_norm": 0.3578641414642334,
      "learning_rate": 2e-05,
      "loss": 0.6248,
      "step": 6800
    },
    {
      "epoch": 0.5685897971111297,
      "grad_norm": 0.3852931559085846,
      "learning_rate": 2e-05,
      "loss": 0.6381,
      "step": 6810
    },
    {
      "epoch": 0.5694247307339066,
      "grad_norm": 0.33246955275535583,
      "learning_rate": 2e-05,
      "loss": 0.6099,
      "step": 6820
    },
    {
      "epoch": 0.5702596643566836,
      "grad_norm": 0.3529711067676544,
      "learning_rate": 2e-05,
      "loss": 0.6018,
      "step": 6830
    },
    {
      "epoch": 0.5710945979794606,
      "grad_norm": 0.3788568377494812,
      "learning_rate": 2e-05,
      "loss": 0.6638,
      "step": 6840
    },
    {
      "epoch": 0.5719295316022376,
      "grad_norm": 0.35002124309539795,
      "learning_rate": 2e-05,
      "loss": 0.643,
      "step": 6850
    },
    {
      "epoch": 0.5727644652250146,
      "grad_norm": 0.4040302336215973,
      "learning_rate": 2e-05,
      "loss": 0.6548,
      "step": 6860
    },
    {
      "epoch": 0.5735993988477917,
      "grad_norm": 0.3852825164794922,
      "learning_rate": 2e-05,
      "loss": 0.642,
      "step": 6870
    },
    {
      "epoch": 0.5744343324705686,
      "grad_norm": 0.38133952021598816,
      "learning_rate": 2e-05,
      "loss": 0.6352,
      "step": 6880
    },
    {
      "epoch": 0.5752692660933456,
      "grad_norm": 0.33622294664382935,
      "learning_rate": 2e-05,
      "loss": 0.6014,
      "step": 6890
    },
    {
      "epoch": 0.5761041997161226,
      "grad_norm": 0.4087103009223938,
      "learning_rate": 2e-05,
      "loss": 0.64,
      "step": 6900
    },
    {
      "epoch": 0.5769391333388996,
      "grad_norm": 0.4508325457572937,
      "learning_rate": 2e-05,
      "loss": 0.6529,
      "step": 6910
    },
    {
      "epoch": 0.5777740669616765,
      "grad_norm": 0.36342644691467285,
      "learning_rate": 2e-05,
      "loss": 0.5946,
      "step": 6920
    },
    {
      "epoch": 0.5786090005844535,
      "grad_norm": 0.33947324752807617,
      "learning_rate": 2e-05,
      "loss": 0.6325,
      "step": 6930
    },
    {
      "epoch": 0.5794439342072305,
      "grad_norm": 0.3845723569393158,
      "learning_rate": 2e-05,
      "loss": 0.6503,
      "step": 6940
    },
    {
      "epoch": 0.5802788678300075,
      "grad_norm": 0.36079132556915283,
      "learning_rate": 2e-05,
      "loss": 0.6218,
      "step": 6950
    },
    {
      "epoch": 0.5811138014527845,
      "grad_norm": 0.34730052947998047,
      "learning_rate": 2e-05,
      "loss": 0.6143,
      "step": 6960
    },
    {
      "epoch": 0.5819487350755614,
      "grad_norm": 0.37332674860954285,
      "learning_rate": 2e-05,
      "loss": 0.6225,
      "step": 6970
    },
    {
      "epoch": 0.5827836686983385,
      "grad_norm": 0.3502192199230194,
      "learning_rate": 2e-05,
      "loss": 0.6196,
      "step": 6980
    },
    {
      "epoch": 0.5836186023211155,
      "grad_norm": 0.3406578302383423,
      "learning_rate": 2e-05,
      "loss": 0.6118,
      "step": 6990
    },
    {
      "epoch": 0.5844535359438925,
      "grad_norm": 0.3505347967147827,
      "learning_rate": 2e-05,
      "loss": 0.589,
      "step": 7000
    },
    {
      "epoch": 0.5852884695666695,
      "grad_norm": 0.3579750061035156,
      "learning_rate": 2e-05,
      "loss": 0.6102,
      "step": 7010
    },
    {
      "epoch": 0.5861234031894464,
      "grad_norm": 0.3577715754508972,
      "learning_rate": 2e-05,
      "loss": 0.6432,
      "step": 7020
    },
    {
      "epoch": 0.5869583368122234,
      "grad_norm": 0.360882431268692,
      "learning_rate": 2e-05,
      "loss": 0.6408,
      "step": 7030
    },
    {
      "epoch": 0.5877932704350004,
      "grad_norm": 0.342983603477478,
      "learning_rate": 2e-05,
      "loss": 0.6447,
      "step": 7040
    },
    {
      "epoch": 0.5886282040577774,
      "grad_norm": 0.39279457926750183,
      "learning_rate": 2e-05,
      "loss": 0.6588,
      "step": 7050
    },
    {
      "epoch": 0.5894631376805544,
      "grad_norm": 0.3919101655483246,
      "learning_rate": 2e-05,
      "loss": 0.6282,
      "step": 7060
    },
    {
      "epoch": 0.5902980713033313,
      "grad_norm": 0.4209004342556,
      "learning_rate": 2e-05,
      "loss": 0.6401,
      "step": 7070
    },
    {
      "epoch": 0.5911330049261084,
      "grad_norm": 0.38336417078971863,
      "learning_rate": 2e-05,
      "loss": 0.6086,
      "step": 7080
    },
    {
      "epoch": 0.5919679385488854,
      "grad_norm": 0.3805478513240814,
      "learning_rate": 2e-05,
      "loss": 0.6535,
      "step": 7090
    },
    {
      "epoch": 0.5928028721716624,
      "grad_norm": 0.4300934970378876,
      "learning_rate": 2e-05,
      "loss": 0.6478,
      "step": 7100
    },
    {
      "epoch": 0.5936378057944394,
      "grad_norm": 0.36743587255477905,
      "learning_rate": 2e-05,
      "loss": 0.6554,
      "step": 7110
    },
    {
      "epoch": 0.5944727394172163,
      "grad_norm": 0.38962113857269287,
      "learning_rate": 2e-05,
      "loss": 0.6363,
      "step": 7120
    },
    {
      "epoch": 0.5953076730399933,
      "grad_norm": 0.31849971413612366,
      "learning_rate": 2e-05,
      "loss": 0.6357,
      "step": 7130
    },
    {
      "epoch": 0.5961426066627703,
      "grad_norm": 0.3726116120815277,
      "learning_rate": 2e-05,
      "loss": 0.6449,
      "step": 7140
    },
    {
      "epoch": 0.5969775402855473,
      "grad_norm": 0.3691715598106384,
      "learning_rate": 2e-05,
      "loss": 0.626,
      "step": 7150
    },
    {
      "epoch": 0.5978124739083243,
      "grad_norm": 0.3862852454185486,
      "learning_rate": 2e-05,
      "loss": 0.6194,
      "step": 7160
    },
    {
      "epoch": 0.5986474075311012,
      "grad_norm": 0.3984801769256592,
      "learning_rate": 2e-05,
      "loss": 0.6401,
      "step": 7170
    },
    {
      "epoch": 0.5994823411538782,
      "grad_norm": 0.3971180021762848,
      "learning_rate": 2e-05,
      "loss": 0.6548,
      "step": 7180
    },
    {
      "epoch": 0.6003172747766553,
      "grad_norm": 0.3461320996284485,
      "learning_rate": 2e-05,
      "loss": 0.6323,
      "step": 7190
    },
    {
      "epoch": 0.6011522083994323,
      "grad_norm": 0.39283016324043274,
      "learning_rate": 2e-05,
      "loss": 0.6528,
      "step": 7200
    },
    {
      "epoch": 0.6019871420222093,
      "grad_norm": 0.38070911169052124,
      "learning_rate": 2e-05,
      "loss": 0.6323,
      "step": 7210
    },
    {
      "epoch": 0.6028220756449862,
      "grad_norm": 0.36780697107315063,
      "learning_rate": 2e-05,
      "loss": 0.6217,
      "step": 7220
    },
    {
      "epoch": 0.6036570092677632,
      "grad_norm": 0.4038870334625244,
      "learning_rate": 2e-05,
      "loss": 0.6224,
      "step": 7230
    },
    {
      "epoch": 0.6044919428905402,
      "grad_norm": 0.3862769901752472,
      "learning_rate": 2e-05,
      "loss": 0.6262,
      "step": 7240
    },
    {
      "epoch": 0.6053268765133172,
      "grad_norm": 0.35212066769599915,
      "learning_rate": 2e-05,
      "loss": 0.6277,
      "step": 7250
    },
    {
      "epoch": 0.6061618101360942,
      "grad_norm": 0.34551307559013367,
      "learning_rate": 2e-05,
      "loss": 0.6173,
      "step": 7260
    },
    {
      "epoch": 0.6069967437588711,
      "grad_norm": 0.3547622263431549,
      "learning_rate": 2e-05,
      "loss": 0.6172,
      "step": 7270
    },
    {
      "epoch": 0.6078316773816481,
      "grad_norm": 0.39236578345298767,
      "learning_rate": 2e-05,
      "loss": 0.6541,
      "step": 7280
    },
    {
      "epoch": 0.6086666110044251,
      "grad_norm": 0.554222583770752,
      "learning_rate": 2e-05,
      "loss": 0.6576,
      "step": 7290
    },
    {
      "epoch": 0.6095015446272022,
      "grad_norm": 0.3910732865333557,
      "learning_rate": 2e-05,
      "loss": 0.6019,
      "step": 7300
    },
    {
      "epoch": 0.6103364782499792,
      "grad_norm": 0.3806385397911072,
      "learning_rate": 2e-05,
      "loss": 0.6511,
      "step": 7310
    },
    {
      "epoch": 0.6111714118727561,
      "grad_norm": 0.35252174735069275,
      "learning_rate": 2e-05,
      "loss": 0.5998,
      "step": 7320
    },
    {
      "epoch": 0.6120063454955331,
      "grad_norm": 0.34954068064689636,
      "learning_rate": 2e-05,
      "loss": 0.6371,
      "step": 7330
    },
    {
      "epoch": 0.6128412791183101,
      "grad_norm": 0.38126328587532043,
      "learning_rate": 2e-05,
      "loss": 0.6106,
      "step": 7340
    },
    {
      "epoch": 0.6136762127410871,
      "grad_norm": 0.3403848111629486,
      "learning_rate": 2e-05,
      "loss": 0.6071,
      "step": 7350
    },
    {
      "epoch": 0.6145111463638641,
      "grad_norm": 0.37240487337112427,
      "learning_rate": 2e-05,
      "loss": 0.6016,
      "step": 7360
    },
    {
      "epoch": 0.615346079986641,
      "grad_norm": 0.3464624583721161,
      "learning_rate": 2e-05,
      "loss": 0.6058,
      "step": 7370
    },
    {
      "epoch": 0.616181013609418,
      "grad_norm": 0.4099377393722534,
      "learning_rate": 2e-05,
      "loss": 0.6427,
      "step": 7380
    },
    {
      "epoch": 0.617015947232195,
      "grad_norm": 0.3884275555610657,
      "learning_rate": 2e-05,
      "loss": 0.6338,
      "step": 7390
    },
    {
      "epoch": 0.6178508808549721,
      "grad_norm": 0.36154904961586,
      "learning_rate": 2e-05,
      "loss": 0.6118,
      "step": 7400
    },
    {
      "epoch": 0.6186858144777491,
      "grad_norm": 0.3710553050041199,
      "learning_rate": 2e-05,
      "loss": 0.6028,
      "step": 7410
    },
    {
      "epoch": 0.619520748100526,
      "grad_norm": 0.4147917330265045,
      "learning_rate": 2e-05,
      "loss": 0.6421,
      "step": 7420
    },
    {
      "epoch": 0.620355681723303,
      "grad_norm": 0.3236616849899292,
      "learning_rate": 2e-05,
      "loss": 0.6212,
      "step": 7430
    },
    {
      "epoch": 0.62119061534608,
      "grad_norm": 0.37347686290740967,
      "learning_rate": 2e-05,
      "loss": 0.6469,
      "step": 7440
    },
    {
      "epoch": 0.622025548968857,
      "grad_norm": 0.35045894980430603,
      "learning_rate": 2e-05,
      "loss": 0.6424,
      "step": 7450
    },
    {
      "epoch": 0.622860482591634,
      "grad_norm": 0.36104148626327515,
      "learning_rate": 2e-05,
      "loss": 0.6513,
      "step": 7460
    },
    {
      "epoch": 0.6236954162144109,
      "grad_norm": 0.3592577576637268,
      "learning_rate": 2e-05,
      "loss": 0.5951,
      "step": 7470
    },
    {
      "epoch": 0.6245303498371879,
      "grad_norm": 0.3420862853527069,
      "learning_rate": 2e-05,
      "loss": 0.6146,
      "step": 7480
    },
    {
      "epoch": 0.6253652834599649,
      "grad_norm": 0.3220742344856262,
      "learning_rate": 2e-05,
      "loss": 0.6327,
      "step": 7490
    },
    {
      "epoch": 0.6262002170827419,
      "grad_norm": 0.3457016944885254,
      "learning_rate": 2e-05,
      "loss": 0.5998,
      "step": 7500
    },
    {
      "epoch": 0.627035150705519,
      "grad_norm": 0.42156949639320374,
      "learning_rate": 2e-05,
      "loss": 0.6461,
      "step": 7510
    },
    {
      "epoch": 0.6278700843282959,
      "grad_norm": 0.33992454409599304,
      "learning_rate": 2e-05,
      "loss": 0.6112,
      "step": 7520
    },
    {
      "epoch": 0.6287050179510729,
      "grad_norm": 0.3505455553531647,
      "learning_rate": 2e-05,
      "loss": 0.5914,
      "step": 7530
    },
    {
      "epoch": 0.6295399515738499,
      "grad_norm": 0.42522716522216797,
      "learning_rate": 2e-05,
      "loss": 0.5891,
      "step": 7540
    },
    {
      "epoch": 0.6303748851966269,
      "grad_norm": 0.38674265146255493,
      "learning_rate": 2e-05,
      "loss": 0.6394,
      "step": 7550
    },
    {
      "epoch": 0.6312098188194039,
      "grad_norm": 0.35257601737976074,
      "learning_rate": 2e-05,
      "loss": 0.6093,
      "step": 7560
    },
    {
      "epoch": 0.6320447524421808,
      "grad_norm": 0.39018532633781433,
      "learning_rate": 2e-05,
      "loss": 0.6183,
      "step": 7570
    },
    {
      "epoch": 0.6328796860649578,
      "grad_norm": 0.4343474805355072,
      "learning_rate": 2e-05,
      "loss": 0.6384,
      "step": 7580
    },
    {
      "epoch": 0.6337146196877348,
      "grad_norm": 0.33827558159828186,
      "learning_rate": 2e-05,
      "loss": 0.6108,
      "step": 7590
    },
    {
      "epoch": 0.6345495533105118,
      "grad_norm": 0.5442385077476501,
      "learning_rate": 2e-05,
      "loss": 0.5976,
      "step": 7600
    },
    {
      "epoch": 0.6353844869332888,
      "grad_norm": 0.35988712310791016,
      "learning_rate": 2e-05,
      "loss": 0.6701,
      "step": 7610
    },
    {
      "epoch": 0.6362194205560658,
      "grad_norm": 0.39446479082107544,
      "learning_rate": 2e-05,
      "loss": 0.6455,
      "step": 7620
    },
    {
      "epoch": 0.6370543541788428,
      "grad_norm": 0.3593360483646393,
      "learning_rate": 2e-05,
      "loss": 0.614,
      "step": 7630
    },
    {
      "epoch": 0.6378892878016198,
      "grad_norm": 0.405366986989975,
      "learning_rate": 2e-05,
      "loss": 0.652,
      "step": 7640
    },
    {
      "epoch": 0.6387242214243968,
      "grad_norm": 0.4077059030532837,
      "learning_rate": 2e-05,
      "loss": 0.6549,
      "step": 7650
    },
    {
      "epoch": 0.6395591550471738,
      "grad_norm": 0.3993997275829315,
      "learning_rate": 2e-05,
      "loss": 0.6185,
      "step": 7660
    },
    {
      "epoch": 0.6403940886699507,
      "grad_norm": 0.3736080229282379,
      "learning_rate": 2e-05,
      "loss": 0.651,
      "step": 7670
    },
    {
      "epoch": 0.6412290222927277,
      "grad_norm": 0.3890738785266876,
      "learning_rate": 2e-05,
      "loss": 0.6554,
      "step": 7680
    },
    {
      "epoch": 0.6420639559155047,
      "grad_norm": 0.3307528495788574,
      "learning_rate": 2e-05,
      "loss": 0.6343,
      "step": 7690
    },
    {
      "epoch": 0.6428988895382817,
      "grad_norm": 0.39791086316108704,
      "learning_rate": 2e-05,
      "loss": 0.6063,
      "step": 7700
    },
    {
      "epoch": 0.6437338231610586,
      "grad_norm": 0.36169809103012085,
      "learning_rate": 2e-05,
      "loss": 0.6292,
      "step": 7710
    },
    {
      "epoch": 0.6445687567838357,
      "grad_norm": 0.3543538749217987,
      "learning_rate": 2e-05,
      "loss": 0.605,
      "step": 7720
    },
    {
      "epoch": 0.6454036904066127,
      "grad_norm": 0.3725712299346924,
      "learning_rate": 2e-05,
      "loss": 0.6114,
      "step": 7730
    },
    {
      "epoch": 0.6462386240293897,
      "grad_norm": 0.38478484749794006,
      "learning_rate": 2e-05,
      "loss": 0.6279,
      "step": 7740
    },
    {
      "epoch": 0.6470735576521667,
      "grad_norm": 0.3763516843318939,
      "learning_rate": 2e-05,
      "loss": 0.6238,
      "step": 7750
    },
    {
      "epoch": 0.6479084912749437,
      "grad_norm": 0.39952021837234497,
      "learning_rate": 2e-05,
      "loss": 0.6143,
      "step": 7760
    },
    {
      "epoch": 0.6487434248977206,
      "grad_norm": 0.35652899742126465,
      "learning_rate": 2e-05,
      "loss": 0.6444,
      "step": 7770
    },
    {
      "epoch": 0.6495783585204976,
      "grad_norm": 0.40300849080085754,
      "learning_rate": 2e-05,
      "loss": 0.6328,
      "step": 7780
    },
    {
      "epoch": 0.6504132921432746,
      "grad_norm": 0.36284902691841125,
      "learning_rate": 2e-05,
      "loss": 0.6119,
      "step": 7790
    },
    {
      "epoch": 0.6512482257660516,
      "grad_norm": 0.3799496591091156,
      "learning_rate": 2e-05,
      "loss": 0.6022,
      "step": 7800
    },
    {
      "epoch": 0.6520831593888285,
      "grad_norm": 0.3699091970920563,
      "learning_rate": 2e-05,
      "loss": 0.6371,
      "step": 7810
    },
    {
      "epoch": 0.6529180930116055,
      "grad_norm": 0.343892365694046,
      "learning_rate": 2e-05,
      "loss": 0.6467,
      "step": 7820
    },
    {
      "epoch": 0.6537530266343826,
      "grad_norm": 0.4449147880077362,
      "learning_rate": 2e-05,
      "loss": 0.6568,
      "step": 7830
    },
    {
      "epoch": 0.6545879602571596,
      "grad_norm": 0.32980215549468994,
      "learning_rate": 2e-05,
      "loss": 0.5862,
      "step": 7840
    },
    {
      "epoch": 0.6554228938799366,
      "grad_norm": 0.3580218553543091,
      "learning_rate": 2e-05,
      "loss": 0.5985,
      "step": 7850
    },
    {
      "epoch": 0.6562578275027136,
      "grad_norm": 0.310857355594635,
      "learning_rate": 2e-05,
      "loss": 0.6406,
      "step": 7860
    },
    {
      "epoch": 0.6570927611254905,
      "grad_norm": 0.415312796831131,
      "learning_rate": 2e-05,
      "loss": 0.6119,
      "step": 7870
    },
    {
      "epoch": 0.6579276947482675,
      "grad_norm": 0.378924697637558,
      "learning_rate": 2e-05,
      "loss": 0.6179,
      "step": 7880
    },
    {
      "epoch": 0.6587626283710445,
      "grad_norm": 0.3965006172657013,
      "learning_rate": 2e-05,
      "loss": 0.6119,
      "step": 7890
    },
    {
      "epoch": 0.6595975619938215,
      "grad_norm": 0.32662075757980347,
      "learning_rate": 2e-05,
      "loss": 0.617,
      "step": 7900
    },
    {
      "epoch": 0.6604324956165984,
      "grad_norm": 0.35183387994766235,
      "learning_rate": 2e-05,
      "loss": 0.6131,
      "step": 7910
    },
    {
      "epoch": 0.6612674292393754,
      "grad_norm": 0.43516552448272705,
      "learning_rate": 2e-05,
      "loss": 0.6196,
      "step": 7920
    },
    {
      "epoch": 0.6621023628621524,
      "grad_norm": 0.34107694029808044,
      "learning_rate": 2e-05,
      "loss": 0.6184,
      "step": 7930
    },
    {
      "epoch": 0.6629372964849295,
      "grad_norm": 0.3809938132762909,
      "learning_rate": 2e-05,
      "loss": 0.6475,
      "step": 7940
    },
    {
      "epoch": 0.6637722301077065,
      "grad_norm": 0.368051141500473,
      "learning_rate": 2e-05,
      "loss": 0.6147,
      "step": 7950
    },
    {
      "epoch": 0.6646071637304835,
      "grad_norm": 0.36812081933021545,
      "learning_rate": 2e-05,
      "loss": 0.6119,
      "step": 7960
    },
    {
      "epoch": 0.6654420973532604,
      "grad_norm": 0.31720709800720215,
      "learning_rate": 2e-05,
      "loss": 0.6569,
      "step": 7970
    },
    {
      "epoch": 0.6662770309760374,
      "grad_norm": 0.38973110914230347,
      "learning_rate": 2e-05,
      "loss": 0.6409,
      "step": 7980
    },
    {
      "epoch": 0.6671119645988144,
      "grad_norm": 0.4488024413585663,
      "learning_rate": 2e-05,
      "loss": 0.6154,
      "step": 7990
    },
    {
      "epoch": 0.6679468982215914,
      "grad_norm": 0.3259122669696808,
      "learning_rate": 2e-05,
      "loss": 0.645,
      "step": 8000
    },
    {
      "epoch": 0.6687818318443683,
      "grad_norm": 0.34954050183296204,
      "learning_rate": 2e-05,
      "loss": 0.6255,
      "step": 8010
    },
    {
      "epoch": 0.6696167654671453,
      "grad_norm": 0.3668088912963867,
      "learning_rate": 2e-05,
      "loss": 0.6273,
      "step": 8020
    },
    {
      "epoch": 0.6704516990899223,
      "grad_norm": 0.362095445394516,
      "learning_rate": 2e-05,
      "loss": 0.6217,
      "step": 8030
    },
    {
      "epoch": 0.6712866327126994,
      "grad_norm": 0.3309608995914459,
      "learning_rate": 2e-05,
      "loss": 0.6068,
      "step": 8040
    },
    {
      "epoch": 0.6721215663354764,
      "grad_norm": 0.3655224144458771,
      "learning_rate": 2e-05,
      "loss": 0.576,
      "step": 8050
    },
    {
      "epoch": 0.6729564999582534,
      "grad_norm": 0.3824118971824646,
      "learning_rate": 2e-05,
      "loss": 0.6479,
      "step": 8060
    },
    {
      "epoch": 0.6737914335810303,
      "grad_norm": 0.35779502987861633,
      "learning_rate": 2e-05,
      "loss": 0.6198,
      "step": 8070
    },
    {
      "epoch": 0.6746263672038073,
      "grad_norm": 0.36893609166145325,
      "learning_rate": 2e-05,
      "loss": 0.577,
      "step": 8080
    },
    {
      "epoch": 0.6754613008265843,
      "grad_norm": 0.3613140881061554,
      "learning_rate": 2e-05,
      "loss": 0.5987,
      "step": 8090
    },
    {
      "epoch": 0.6762962344493613,
      "grad_norm": 0.40050366520881653,
      "learning_rate": 2e-05,
      "loss": 0.6641,
      "step": 8100
    },
    {
      "epoch": 0.6771311680721382,
      "grad_norm": 0.35525432229042053,
      "learning_rate": 2e-05,
      "loss": 0.6323,
      "step": 8110
    },
    {
      "epoch": 0.6779661016949152,
      "grad_norm": 0.3979458212852478,
      "learning_rate": 2e-05,
      "loss": 0.6504,
      "step": 8120
    },
    {
      "epoch": 0.6788010353176922,
      "grad_norm": 0.35750463604927063,
      "learning_rate": 2e-05,
      "loss": 0.6041,
      "step": 8130
    },
    {
      "epoch": 0.6796359689404692,
      "grad_norm": 0.37350085377693176,
      "learning_rate": 2e-05,
      "loss": 0.6388,
      "step": 8140
    },
    {
      "epoch": 0.6804709025632463,
      "grad_norm": 0.36661744117736816,
      "learning_rate": 2e-05,
      "loss": 0.619,
      "step": 8150
    },
    {
      "epoch": 0.6813058361860233,
      "grad_norm": 0.34453845024108887,
      "learning_rate": 2e-05,
      "loss": 0.6347,
      "step": 8160
    },
    {
      "epoch": 0.6821407698088002,
      "grad_norm": 0.3704468607902527,
      "learning_rate": 2e-05,
      "loss": 0.6499,
      "step": 8170
    },
    {
      "epoch": 0.6829757034315772,
      "grad_norm": 0.3720283508300781,
      "learning_rate": 2e-05,
      "loss": 0.6169,
      "step": 8180
    },
    {
      "epoch": 0.6838106370543542,
      "grad_norm": 0.36334046721458435,
      "learning_rate": 2e-05,
      "loss": 0.6393,
      "step": 8190
    },
    {
      "epoch": 0.6846455706771312,
      "grad_norm": 0.3491136431694031,
      "learning_rate": 2e-05,
      "loss": 0.5935,
      "step": 8200
    },
    {
      "epoch": 0.6854805042999081,
      "grad_norm": 0.36256110668182373,
      "learning_rate": 2e-05,
      "loss": 0.6416,
      "step": 8210
    },
    {
      "epoch": 0.6863154379226851,
      "grad_norm": 0.3605171740055084,
      "learning_rate": 2e-05,
      "loss": 0.6269,
      "step": 8220
    },
    {
      "epoch": 0.6871503715454621,
      "grad_norm": 0.31848451495170593,
      "learning_rate": 2e-05,
      "loss": 0.5924,
      "step": 8230
    },
    {
      "epoch": 0.6879853051682391,
      "grad_norm": 0.38697347044944763,
      "learning_rate": 2e-05,
      "loss": 0.6126,
      "step": 8240
    },
    {
      "epoch": 0.6888202387910161,
      "grad_norm": 0.38452598452568054,
      "learning_rate": 2e-05,
      "loss": 0.593,
      "step": 8250
    },
    {
      "epoch": 0.6896551724137931,
      "grad_norm": 0.35967740416526794,
      "learning_rate": 2e-05,
      "loss": 0.6093,
      "step": 8260
    },
    {
      "epoch": 0.6904901060365701,
      "grad_norm": 0.3865363299846649,
      "learning_rate": 2e-05,
      "loss": 0.6258,
      "step": 8270
    },
    {
      "epoch": 0.6913250396593471,
      "grad_norm": 0.5585871934890747,
      "learning_rate": 2e-05,
      "loss": 0.6285,
      "step": 8280
    },
    {
      "epoch": 0.6921599732821241,
      "grad_norm": 0.3831752836704254,
      "learning_rate": 2e-05,
      "loss": 0.58,
      "step": 8290
    },
    {
      "epoch": 0.6929949069049011,
      "grad_norm": 0.37848177552223206,
      "learning_rate": 2e-05,
      "loss": 0.6246,
      "step": 8300
    },
    {
      "epoch": 0.693829840527678,
      "grad_norm": 0.5990774631500244,
      "learning_rate": 2e-05,
      "loss": 0.6366,
      "step": 8310
    },
    {
      "epoch": 0.694664774150455,
      "grad_norm": 0.37825098633766174,
      "learning_rate": 2e-05,
      "loss": 0.6429,
      "step": 8320
    },
    {
      "epoch": 0.695499707773232,
      "grad_norm": 0.3840556740760803,
      "learning_rate": 2e-05,
      "loss": 0.6154,
      "step": 8330
    },
    {
      "epoch": 0.696334641396009,
      "grad_norm": 0.37030690908432007,
      "learning_rate": 2e-05,
      "loss": 0.6349,
      "step": 8340
    },
    {
      "epoch": 0.697169575018786,
      "grad_norm": 0.39243191480636597,
      "learning_rate": 2e-05,
      "loss": 0.6052,
      "step": 8350
    },
    {
      "epoch": 0.698004508641563,
      "grad_norm": 0.3358215391635895,
      "learning_rate": 2e-05,
      "loss": 0.62,
      "step": 8360
    },
    {
      "epoch": 0.69883944226434,
      "grad_norm": 0.3688078820705414,
      "learning_rate": 2e-05,
      "loss": 0.6231,
      "step": 8370
    },
    {
      "epoch": 0.699674375887117,
      "grad_norm": 0.3490513861179352,
      "learning_rate": 2e-05,
      "loss": 0.632,
      "step": 8380
    },
    {
      "epoch": 0.700509309509894,
      "grad_norm": 0.47990018129348755,
      "learning_rate": 2e-05,
      "loss": 0.656,
      "step": 8390
    },
    {
      "epoch": 0.701344243132671,
      "grad_norm": 0.36011454463005066,
      "learning_rate": 2e-05,
      "loss": 0.6028,
      "step": 8400
    },
    {
      "epoch": 0.7021791767554479,
      "grad_norm": 0.38842275738716125,
      "learning_rate": 2e-05,
      "loss": 0.6351,
      "step": 8410
    },
    {
      "epoch": 0.7030141103782249,
      "grad_norm": 0.4088204801082611,
      "learning_rate": 2e-05,
      "loss": 0.6117,
      "step": 8420
    },
    {
      "epoch": 0.7038490440010019,
      "grad_norm": 0.36256250739097595,
      "learning_rate": 2e-05,
      "loss": 0.6262,
      "step": 8430
    },
    {
      "epoch": 0.7046839776237789,
      "grad_norm": 0.3916327655315399,
      "learning_rate": 2e-05,
      "loss": 0.6059,
      "step": 8440
    },
    {
      "epoch": 0.7055189112465559,
      "grad_norm": 0.334215372800827,
      "learning_rate": 2e-05,
      "loss": 0.629,
      "step": 8450
    },
    {
      "epoch": 0.7063538448693328,
      "grad_norm": 0.36916741728782654,
      "learning_rate": 2e-05,
      "loss": 0.6204,
      "step": 8460
    },
    {
      "epoch": 0.7071887784921099,
      "grad_norm": 0.3638446033000946,
      "learning_rate": 2e-05,
      "loss": 0.6145,
      "step": 8470
    },
    {
      "epoch": 0.7080237121148869,
      "grad_norm": 0.3380838632583618,
      "learning_rate": 2e-05,
      "loss": 0.6148,
      "step": 8480
    },
    {
      "epoch": 0.7088586457376639,
      "grad_norm": 0.37337175011634827,
      "learning_rate": 2e-05,
      "loss": 0.64,
      "step": 8490
    },
    {
      "epoch": 0.7096935793604409,
      "grad_norm": 0.3561973571777344,
      "learning_rate": 2e-05,
      "loss": 0.6195,
      "step": 8500
    },
    {
      "epoch": 0.7105285129832178,
      "grad_norm": 0.3795963525772095,
      "learning_rate": 2e-05,
      "loss": 0.5913,
      "step": 8510
    },
    {
      "epoch": 0.7113634466059948,
      "grad_norm": 0.3356948792934418,
      "learning_rate": 2e-05,
      "loss": 0.6171,
      "step": 8520
    },
    {
      "epoch": 0.7121983802287718,
      "grad_norm": 0.389492005109787,
      "learning_rate": 2e-05,
      "loss": 0.6171,
      "step": 8530
    },
    {
      "epoch": 0.7130333138515488,
      "grad_norm": 0.3794238865375519,
      "learning_rate": 2e-05,
      "loss": 0.6274,
      "step": 8540
    },
    {
      "epoch": 0.7138682474743258,
      "grad_norm": 0.38909241557121277,
      "learning_rate": 2e-05,
      "loss": 0.6289,
      "step": 8550
    },
    {
      "epoch": 0.7147031810971027,
      "grad_norm": 0.4157065451145172,
      "learning_rate": 2e-05,
      "loss": 0.6261,
      "step": 8560
    },
    {
      "epoch": 0.7155381147198797,
      "grad_norm": 0.3396424353122711,
      "learning_rate": 2e-05,
      "loss": 0.6242,
      "step": 8570
    },
    {
      "epoch": 0.7163730483426568,
      "grad_norm": 0.5149165391921997,
      "learning_rate": 2e-05,
      "loss": 0.6014,
      "step": 8580
    },
    {
      "epoch": 0.7172079819654338,
      "grad_norm": 0.41299954056739807,
      "learning_rate": 2e-05,
      "loss": 0.6088,
      "step": 8590
    },
    {
      "epoch": 0.7180429155882108,
      "grad_norm": 0.35286012291908264,
      "learning_rate": 2e-05,
      "loss": 0.5906,
      "step": 8600
    },
    {
      "epoch": 0.7188778492109877,
      "grad_norm": 0.3411950170993805,
      "learning_rate": 2e-05,
      "loss": 0.6147,
      "step": 8610
    },
    {
      "epoch": 0.7197127828337647,
      "grad_norm": 0.46006834506988525,
      "learning_rate": 2e-05,
      "loss": 0.6171,
      "step": 8620
    },
    {
      "epoch": 0.7205477164565417,
      "grad_norm": 0.3588429391384125,
      "learning_rate": 2e-05,
      "loss": 0.6137,
      "step": 8630
    },
    {
      "epoch": 0.7213826500793187,
      "grad_norm": 0.41396811604499817,
      "learning_rate": 2e-05,
      "loss": 0.6333,
      "step": 8640
    },
    {
      "epoch": 0.7222175837020957,
      "grad_norm": 0.40604695677757263,
      "learning_rate": 2e-05,
      "loss": 0.6036,
      "step": 8650
    },
    {
      "epoch": 0.7230525173248726,
      "grad_norm": 0.5695397257804871,
      "learning_rate": 2e-05,
      "loss": 0.6168,
      "step": 8660
    },
    {
      "epoch": 0.7238874509476496,
      "grad_norm": 0.36406224966049194,
      "learning_rate": 2e-05,
      "loss": 0.6358,
      "step": 8670
    },
    {
      "epoch": 0.7247223845704267,
      "grad_norm": 0.3366695046424866,
      "learning_rate": 2e-05,
      "loss": 0.6111,
      "step": 8680
    },
    {
      "epoch": 0.7255573181932037,
      "grad_norm": 0.43011474609375,
      "learning_rate": 2e-05,
      "loss": 0.6373,
      "step": 8690
    },
    {
      "epoch": 0.7263922518159807,
      "grad_norm": 0.4564869999885559,
      "learning_rate": 2e-05,
      "loss": 0.6299,
      "step": 8700
    },
    {
      "epoch": 0.7272271854387576,
      "grad_norm": 0.3484053313732147,
      "learning_rate": 2e-05,
      "loss": 0.6052,
      "step": 8710
    },
    {
      "epoch": 0.7280621190615346,
      "grad_norm": 0.3960491418838501,
      "learning_rate": 2e-05,
      "loss": 0.6276,
      "step": 8720
    },
    {
      "epoch": 0.7288970526843116,
      "grad_norm": 0.32983118295669556,
      "learning_rate": 2e-05,
      "loss": 0.6214,
      "step": 8730
    },
    {
      "epoch": 0.7297319863070886,
      "grad_norm": 0.3551155626773834,
      "learning_rate": 2e-05,
      "loss": 0.6575,
      "step": 8740
    },
    {
      "epoch": 0.7305669199298656,
      "grad_norm": 0.36505311727523804,
      "learning_rate": 2e-05,
      "loss": 0.6196,
      "step": 8750
    },
    {
      "epoch": 0.7314018535526425,
      "grad_norm": 0.36678943037986755,
      "learning_rate": 2e-05,
      "loss": 0.6323,
      "step": 8760
    },
    {
      "epoch": 0.7322367871754195,
      "grad_norm": 1.2073694467544556,
      "learning_rate": 2e-05,
      "loss": 0.6621,
      "step": 8770
    },
    {
      "epoch": 0.7330717207981965,
      "grad_norm": 0.33506250381469727,
      "learning_rate": 2e-05,
      "loss": 0.6318,
      "step": 8780
    },
    {
      "epoch": 0.7339066544209736,
      "grad_norm": 0.3531174659729004,
      "learning_rate": 2e-05,
      "loss": 0.6369,
      "step": 8790
    },
    {
      "epoch": 0.7347415880437506,
      "grad_norm": 0.4113178551197052,
      "learning_rate": 2e-05,
      "loss": 0.6465,
      "step": 8800
    },
    {
      "epoch": 0.7355765216665275,
      "grad_norm": 0.3323570191860199,
      "learning_rate": 2e-05,
      "loss": 0.5897,
      "step": 8810
    },
    {
      "epoch": 0.7364114552893045,
      "grad_norm": 0.3450741171836853,
      "learning_rate": 2e-05,
      "loss": 0.6281,
      "step": 8820
    },
    {
      "epoch": 0.7372463889120815,
      "grad_norm": 0.41418057680130005,
      "learning_rate": 2e-05,
      "loss": 0.6005,
      "step": 8830
    },
    {
      "epoch": 0.7380813225348585,
      "grad_norm": 0.3248048424720764,
      "learning_rate": 2e-05,
      "loss": 0.6077,
      "step": 8840
    },
    {
      "epoch": 0.7389162561576355,
      "grad_norm": 0.36467882990837097,
      "learning_rate": 2e-05,
      "loss": 0.6007,
      "step": 8850
    },
    {
      "epoch": 0.7397511897804124,
      "grad_norm": 0.36008143424987793,
      "learning_rate": 2e-05,
      "loss": 0.6178,
      "step": 8860
    },
    {
      "epoch": 0.7405861234031894,
      "grad_norm": 0.38576602935791016,
      "learning_rate": 2e-05,
      "loss": 0.6265,
      "step": 8870
    },
    {
      "epoch": 0.7414210570259664,
      "grad_norm": 0.3826843202114105,
      "learning_rate": 2e-05,
      "loss": 0.5833,
      "step": 8880
    },
    {
      "epoch": 0.7422559906487435,
      "grad_norm": 0.34177497029304504,
      "learning_rate": 2e-05,
      "loss": 0.5759,
      "step": 8890
    },
    {
      "epoch": 0.7430909242715205,
      "grad_norm": 0.4049939513206482,
      "learning_rate": 2e-05,
      "loss": 0.6114,
      "step": 8900
    },
    {
      "epoch": 0.7439258578942974,
      "grad_norm": 0.3932447135448456,
      "learning_rate": 2e-05,
      "loss": 0.6525,
      "step": 8910
    },
    {
      "epoch": 0.7447607915170744,
      "grad_norm": 0.3476998507976532,
      "learning_rate": 2e-05,
      "loss": 0.6291,
      "step": 8920
    },
    {
      "epoch": 0.7455957251398514,
      "grad_norm": 0.3328748345375061,
      "learning_rate": 2e-05,
      "loss": 0.58,
      "step": 8930
    },
    {
      "epoch": 0.7464306587626284,
      "grad_norm": 0.3468411862850189,
      "learning_rate": 2e-05,
      "loss": 0.6161,
      "step": 8940
    },
    {
      "epoch": 0.7472655923854054,
      "grad_norm": 0.37306737899780273,
      "learning_rate": 2e-05,
      "loss": 0.6561,
      "step": 8950
    },
    {
      "epoch": 0.7481005260081823,
      "grad_norm": 0.36130857467651367,
      "learning_rate": 2e-05,
      "loss": 0.6062,
      "step": 8960
    },
    {
      "epoch": 0.7489354596309593,
      "grad_norm": 0.37640470266342163,
      "learning_rate": 2e-05,
      "loss": 0.5987,
      "step": 8970
    },
    {
      "epoch": 0.7497703932537363,
      "grad_norm": 0.35466036200523376,
      "learning_rate": 2e-05,
      "loss": 0.6083,
      "step": 8980
    },
    {
      "epoch": 0.7506053268765133,
      "grad_norm": 0.3857974410057068,
      "learning_rate": 2e-05,
      "loss": 0.6088,
      "step": 8990
    },
    {
      "epoch": 0.7514402604992904,
      "grad_norm": 0.32704639434814453,
      "learning_rate": 2e-05,
      "loss": 0.5941,
      "step": 9000
    },
    {
      "epoch": 0.7522751941220673,
      "grad_norm": 0.34643012285232544,
      "learning_rate": 2e-05,
      "loss": 0.5978,
      "step": 9010
    },
    {
      "epoch": 0.7531101277448443,
      "grad_norm": 0.42115890979766846,
      "learning_rate": 2e-05,
      "loss": 0.62,
      "step": 9020
    },
    {
      "epoch": 0.7539450613676213,
      "grad_norm": 0.33378666639328003,
      "learning_rate": 2e-05,
      "loss": 0.5494,
      "step": 9030
    },
    {
      "epoch": 0.7547799949903983,
      "grad_norm": 0.3829306364059448,
      "learning_rate": 2e-05,
      "loss": 0.608,
      "step": 9040
    },
    {
      "epoch": 0.7556149286131753,
      "grad_norm": 0.37077975273132324,
      "learning_rate": 2e-05,
      "loss": 0.581,
      "step": 9050
    },
    {
      "epoch": 0.7564498622359522,
      "grad_norm": 0.35445499420166016,
      "learning_rate": 2e-05,
      "loss": 0.6136,
      "step": 9060
    },
    {
      "epoch": 0.7572847958587292,
      "grad_norm": 0.4109140634536743,
      "learning_rate": 2e-05,
      "loss": 0.6165,
      "step": 9070
    },
    {
      "epoch": 0.7581197294815062,
      "grad_norm": 0.40386489033699036,
      "learning_rate": 2e-05,
      "loss": 0.5862,
      "step": 9080
    },
    {
      "epoch": 0.7589546631042832,
      "grad_norm": 0.3835025429725647,
      "learning_rate": 2e-05,
      "loss": 0.633,
      "step": 9090
    },
    {
      "epoch": 0.7597895967270601,
      "grad_norm": 0.5909358263015747,
      "learning_rate": 2e-05,
      "loss": 0.6245,
      "step": 9100
    },
    {
      "epoch": 0.7606245303498372,
      "grad_norm": 0.33725836873054504,
      "learning_rate": 2e-05,
      "loss": 0.6025,
      "step": 9110
    },
    {
      "epoch": 0.7614594639726142,
      "grad_norm": 0.3663126826286316,
      "learning_rate": 2e-05,
      "loss": 0.6296,
      "step": 9120
    },
    {
      "epoch": 0.7622943975953912,
      "grad_norm": 0.36287397146224976,
      "learning_rate": 2e-05,
      "loss": 0.6051,
      "step": 9130
    },
    {
      "epoch": 0.7631293312181682,
      "grad_norm": 0.37856441736221313,
      "learning_rate": 2e-05,
      "loss": 0.6403,
      "step": 9140
    },
    {
      "epoch": 0.7639642648409452,
      "grad_norm": 0.42316287755966187,
      "learning_rate": 2e-05,
      "loss": 0.5968,
      "step": 9150
    },
    {
      "epoch": 0.7647991984637221,
      "grad_norm": 0.3926558196544647,
      "learning_rate": 2e-05,
      "loss": 0.6214,
      "step": 9160
    },
    {
      "epoch": 0.7656341320864991,
      "grad_norm": 0.4099116325378418,
      "learning_rate": 2e-05,
      "loss": 0.6129,
      "step": 9170
    },
    {
      "epoch": 0.7664690657092761,
      "grad_norm": 0.3692742586135864,
      "learning_rate": 2e-05,
      "loss": 0.6125,
      "step": 9180
    },
    {
      "epoch": 0.7673039993320531,
      "grad_norm": 0.4389698803424835,
      "learning_rate": 2e-05,
      "loss": 0.6315,
      "step": 9190
    },
    {
      "epoch": 0.76813893295483,
      "grad_norm": 0.40900108218193054,
      "learning_rate": 2e-05,
      "loss": 0.6274,
      "step": 9200
    },
    {
      "epoch": 0.7689738665776071,
      "grad_norm": 0.3634421229362488,
      "learning_rate": 2e-05,
      "loss": 0.6343,
      "step": 9210
    },
    {
      "epoch": 0.7698088002003841,
      "grad_norm": 0.36201366782188416,
      "learning_rate": 2e-05,
      "loss": 0.6145,
      "step": 9220
    },
    {
      "epoch": 0.7706437338231611,
      "grad_norm": 0.34093189239501953,
      "learning_rate": 2e-05,
      "loss": 0.6455,
      "step": 9230
    },
    {
      "epoch": 0.7714786674459381,
      "grad_norm": 0.38290971517562866,
      "learning_rate": 2e-05,
      "loss": 0.6184,
      "step": 9240
    },
    {
      "epoch": 0.772313601068715,
      "grad_norm": 0.3850368857383728,
      "learning_rate": 2e-05,
      "loss": 0.6345,
      "step": 9250
    },
    {
      "epoch": 0.773148534691492,
      "grad_norm": 0.37834233045578003,
      "learning_rate": 2e-05,
      "loss": 0.5896,
      "step": 9260
    },
    {
      "epoch": 0.773983468314269,
      "grad_norm": 0.33898690342903137,
      "learning_rate": 2e-05,
      "loss": 0.5952,
      "step": 9270
    },
    {
      "epoch": 0.774818401937046,
      "grad_norm": 0.40454182028770447,
      "learning_rate": 2e-05,
      "loss": 0.6107,
      "step": 9280
    },
    {
      "epoch": 0.775653335559823,
      "grad_norm": 0.34523066878318787,
      "learning_rate": 2e-05,
      "loss": 0.5784,
      "step": 9290
    },
    {
      "epoch": 0.7764882691825999,
      "grad_norm": 0.3787095248699188,
      "learning_rate": 2e-05,
      "loss": 0.6187,
      "step": 9300
    },
    {
      "epoch": 0.7773232028053769,
      "grad_norm": 0.35530802607536316,
      "learning_rate": 2e-05,
      "loss": 0.6205,
      "step": 9310
    },
    {
      "epoch": 0.778158136428154,
      "grad_norm": 0.3759779930114746,
      "learning_rate": 2e-05,
      "loss": 0.6266,
      "step": 9320
    },
    {
      "epoch": 0.778993070050931,
      "grad_norm": 0.37251630425453186,
      "learning_rate": 2e-05,
      "loss": 0.6323,
      "step": 9330
    },
    {
      "epoch": 0.779828003673708,
      "grad_norm": 0.38062775135040283,
      "learning_rate": 2e-05,
      "loss": 0.6133,
      "step": 9340
    },
    {
      "epoch": 0.780662937296485,
      "grad_norm": 0.3865993618965149,
      "learning_rate": 2e-05,
      "loss": 0.6268,
      "step": 9350
    },
    {
      "epoch": 0.7814978709192619,
      "grad_norm": 0.3553386926651001,
      "learning_rate": 2e-05,
      "loss": 0.657,
      "step": 9360
    },
    {
      "epoch": 0.7823328045420389,
      "grad_norm": 0.3627830743789673,
      "learning_rate": 2e-05,
      "loss": 0.5934,
      "step": 9370
    },
    {
      "epoch": 0.7831677381648159,
      "grad_norm": 0.3538872003555298,
      "learning_rate": 2e-05,
      "loss": 0.6276,
      "step": 9380
    },
    {
      "epoch": 0.7840026717875929,
      "grad_norm": 0.3829791247844696,
      "learning_rate": 2e-05,
      "loss": 0.6086,
      "step": 9390
    },
    {
      "epoch": 0.7848376054103698,
      "grad_norm": 0.3650110065937042,
      "learning_rate": 2e-05,
      "loss": 0.6196,
      "step": 9400
    },
    {
      "epoch": 0.7856725390331468,
      "grad_norm": 0.4135454595088959,
      "learning_rate": 2e-05,
      "loss": 0.6019,
      "step": 9410
    },
    {
      "epoch": 0.7865074726559238,
      "grad_norm": 0.39289310574531555,
      "learning_rate": 2e-05,
      "loss": 0.6332,
      "step": 9420
    },
    {
      "epoch": 0.7873424062787009,
      "grad_norm": 0.34821635484695435,
      "learning_rate": 2e-05,
      "loss": 0.6471,
      "step": 9430
    },
    {
      "epoch": 0.7881773399014779,
      "grad_norm": 0.340297669172287,
      "learning_rate": 2e-05,
      "loss": 0.6041,
      "step": 9440
    },
    {
      "epoch": 0.7890122735242548,
      "grad_norm": 0.33500832319259644,
      "learning_rate": 2e-05,
      "loss": 0.5999,
      "step": 9450
    },
    {
      "epoch": 0.7898472071470318,
      "grad_norm": 0.3863445222377777,
      "learning_rate": 2e-05,
      "loss": 0.5989,
      "step": 9460
    },
    {
      "epoch": 0.7906821407698088,
      "grad_norm": 0.38019225001335144,
      "learning_rate": 2e-05,
      "loss": 0.5756,
      "step": 9470
    },
    {
      "epoch": 0.7915170743925858,
      "grad_norm": 0.3725093603134155,
      "learning_rate": 2e-05,
      "loss": 0.61,
      "step": 9480
    },
    {
      "epoch": 0.7923520080153628,
      "grad_norm": 0.3705242872238159,
      "learning_rate": 2e-05,
      "loss": 0.6345,
      "step": 9490
    },
    {
      "epoch": 0.7931869416381397,
      "grad_norm": 1.1226922273635864,
      "learning_rate": 2e-05,
      "loss": 0.5983,
      "step": 9500
    },
    {
      "epoch": 0.7940218752609167,
      "grad_norm": 0.3787461817264557,
      "learning_rate": 2e-05,
      "loss": 0.5977,
      "step": 9510
    },
    {
      "epoch": 0.7948568088836937,
      "grad_norm": 0.3609832227230072,
      "learning_rate": 2e-05,
      "loss": 0.6109,
      "step": 9520
    },
    {
      "epoch": 0.7956917425064708,
      "grad_norm": 0.36195361614227295,
      "learning_rate": 2e-05,
      "loss": 0.609,
      "step": 9530
    },
    {
      "epoch": 0.7965266761292478,
      "grad_norm": 0.3786957263946533,
      "learning_rate": 2e-05,
      "loss": 0.5761,
      "step": 9540
    },
    {
      "epoch": 0.7973616097520247,
      "grad_norm": 0.5323551893234253,
      "learning_rate": 2e-05,
      "loss": 0.655,
      "step": 9550
    },
    {
      "epoch": 0.7981965433748017,
      "grad_norm": 0.4058271646499634,
      "learning_rate": 2e-05,
      "loss": 0.6345,
      "step": 9560
    },
    {
      "epoch": 0.7990314769975787,
      "grad_norm": 0.3330802023410797,
      "learning_rate": 2e-05,
      "loss": 0.6017,
      "step": 9570
    },
    {
      "epoch": 0.7998664106203557,
      "grad_norm": 0.35336217284202576,
      "learning_rate": 2e-05,
      "loss": 0.6297,
      "step": 9580
    },
    {
      "epoch": 0.8007013442431327,
      "grad_norm": 0.3388120234012604,
      "learning_rate": 2e-05,
      "loss": 0.6185,
      "step": 9590
    },
    {
      "epoch": 0.8015362778659096,
      "grad_norm": 0.3326929211616516,
      "learning_rate": 2e-05,
      "loss": 0.6322,
      "step": 9600
    },
    {
      "epoch": 0.8023712114886866,
      "grad_norm": 0.379774272441864,
      "learning_rate": 2e-05,
      "loss": 0.6067,
      "step": 9610
    },
    {
      "epoch": 0.8032061451114636,
      "grad_norm": 0.3607538938522339,
      "learning_rate": 2e-05,
      "loss": 0.6055,
      "step": 9620
    },
    {
      "epoch": 0.8040410787342406,
      "grad_norm": 0.4780648946762085,
      "learning_rate": 2e-05,
      "loss": 0.6147,
      "step": 9630
    },
    {
      "epoch": 0.8048760123570177,
      "grad_norm": 0.3941080868244171,
      "learning_rate": 2e-05,
      "loss": 0.5785,
      "step": 9640
    },
    {
      "epoch": 0.8057109459797946,
      "grad_norm": 0.4163796603679657,
      "learning_rate": 2e-05,
      "loss": 0.6173,
      "step": 9650
    },
    {
      "epoch": 0.8065458796025716,
      "grad_norm": 0.34561049938201904,
      "learning_rate": 2e-05,
      "loss": 0.5995,
      "step": 9660
    },
    {
      "epoch": 0.8073808132253486,
      "grad_norm": 0.35494714975357056,
      "learning_rate": 2e-05,
      "loss": 0.6843,
      "step": 9670
    },
    {
      "epoch": 0.8082157468481256,
      "grad_norm": 0.3530466854572296,
      "learning_rate": 2e-05,
      "loss": 0.6258,
      "step": 9680
    },
    {
      "epoch": 0.8090506804709026,
      "grad_norm": 0.3739253878593445,
      "learning_rate": 2e-05,
      "loss": 0.6236,
      "step": 9690
    },
    {
      "epoch": 0.8098856140936795,
      "grad_norm": 0.3684515058994293,
      "learning_rate": 2e-05,
      "loss": 0.5855,
      "step": 9700
    },
    {
      "epoch": 0.8107205477164565,
      "grad_norm": 0.36686304211616516,
      "learning_rate": 2e-05,
      "loss": 0.6048,
      "step": 9710
    },
    {
      "epoch": 0.8115554813392335,
      "grad_norm": 0.3791918158531189,
      "learning_rate": 2e-05,
      "loss": 0.6308,
      "step": 9720
    },
    {
      "epoch": 0.8123904149620105,
      "grad_norm": 0.36236298084259033,
      "learning_rate": 2e-05,
      "loss": 0.6018,
      "step": 9730
    },
    {
      "epoch": 0.8132253485847875,
      "grad_norm": 0.35410210490226746,
      "learning_rate": 2e-05,
      "loss": 0.5974,
      "step": 9740
    },
    {
      "epoch": 0.8140602822075645,
      "grad_norm": 0.3910984992980957,
      "learning_rate": 2e-05,
      "loss": 0.6297,
      "step": 9750
    },
    {
      "epoch": 0.8148952158303415,
      "grad_norm": 0.3511691689491272,
      "learning_rate": 2e-05,
      "loss": 0.5602,
      "step": 9760
    },
    {
      "epoch": 0.8157301494531185,
      "grad_norm": 0.37893590331077576,
      "learning_rate": 2e-05,
      "loss": 0.6076,
      "step": 9770
    },
    {
      "epoch": 0.8165650830758955,
      "grad_norm": 0.3633483350276947,
      "learning_rate": 2e-05,
      "loss": 0.5953,
      "step": 9780
    },
    {
      "epoch": 0.8174000166986725,
      "grad_norm": 0.410077303647995,
      "learning_rate": 2e-05,
      "loss": 0.5967,
      "step": 9790
    },
    {
      "epoch": 0.8182349503214494,
      "grad_norm": 0.38097113370895386,
      "learning_rate": 2e-05,
      "loss": 0.6468,
      "step": 9800
    },
    {
      "epoch": 0.8190698839442264,
      "grad_norm": 0.3786362111568451,
      "learning_rate": 2e-05,
      "loss": 0.5822,
      "step": 9810
    },
    {
      "epoch": 0.8199048175670034,
      "grad_norm": 0.413521409034729,
      "learning_rate": 2e-05,
      "loss": 0.6306,
      "step": 9820
    },
    {
      "epoch": 0.8207397511897804,
      "grad_norm": 0.3841961920261383,
      "learning_rate": 2e-05,
      "loss": 0.612,
      "step": 9830
    },
    {
      "epoch": 0.8215746848125574,
      "grad_norm": 0.3671470284461975,
      "learning_rate": 2e-05,
      "loss": 0.6074,
      "step": 9840
    },
    {
      "epoch": 0.8224096184353344,
      "grad_norm": 0.3668075501918793,
      "learning_rate": 2e-05,
      "loss": 0.5926,
      "step": 9850
    },
    {
      "epoch": 0.8232445520581114,
      "grad_norm": 0.3639094829559326,
      "learning_rate": 2e-05,
      "loss": 0.5692,
      "step": 9860
    },
    {
      "epoch": 0.8240794856808884,
      "grad_norm": 0.38630393147468567,
      "learning_rate": 2e-05,
      "loss": 0.5792,
      "step": 9870
    },
    {
      "epoch": 0.8249144193036654,
      "grad_norm": 0.38509032130241394,
      "learning_rate": 2e-05,
      "loss": 0.632,
      "step": 9880
    },
    {
      "epoch": 0.8257493529264424,
      "grad_norm": 0.39054349064826965,
      "learning_rate": 2e-05,
      "loss": 0.6194,
      "step": 9890
    },
    {
      "epoch": 0.8265842865492193,
      "grad_norm": 0.3866192400455475,
      "learning_rate": 2e-05,
      "loss": 0.5863,
      "step": 9900
    },
    {
      "epoch": 0.8274192201719963,
      "grad_norm": 0.3836771547794342,
      "learning_rate": 2e-05,
      "loss": 0.6374,
      "step": 9910
    },
    {
      "epoch": 0.8282541537947733,
      "grad_norm": 0.3885362446308136,
      "learning_rate": 2e-05,
      "loss": 0.5957,
      "step": 9920
    },
    {
      "epoch": 0.8290890874175503,
      "grad_norm": 0.390948623418808,
      "learning_rate": 2e-05,
      "loss": 0.6341,
      "step": 9930
    },
    {
      "epoch": 0.8299240210403273,
      "grad_norm": 0.43137142062187195,
      "learning_rate": 2e-05,
      "loss": 0.627,
      "step": 9940
    },
    {
      "epoch": 0.8307589546631042,
      "grad_norm": 0.3218252956867218,
      "learning_rate": 2e-05,
      "loss": 0.5806,
      "step": 9950
    },
    {
      "epoch": 0.8315938882858813,
      "grad_norm": 0.38680320978164673,
      "learning_rate": 2e-05,
      "loss": 0.5865,
      "step": 9960
    },
    {
      "epoch": 0.8324288219086583,
      "grad_norm": 0.3924795687198639,
      "learning_rate": 2e-05,
      "loss": 0.6205,
      "step": 9970
    },
    {
      "epoch": 0.8332637555314353,
      "grad_norm": 0.32647204399108887,
      "learning_rate": 2e-05,
      "loss": 0.5881,
      "step": 9980
    },
    {
      "epoch": 0.8340986891542123,
      "grad_norm": 0.3603630065917969,
      "learning_rate": 2e-05,
      "loss": 0.6249,
      "step": 9990
    },
    {
      "epoch": 0.8349336227769892,
      "grad_norm": 0.3946162462234497,
      "learning_rate": 2e-05,
      "loss": 0.6617,
      "step": 10000
    },
    {
      "epoch": 0.8357685563997662,
      "grad_norm": 0.3502800762653351,
      "learning_rate": 2e-05,
      "loss": 0.609,
      "step": 10010
    },
    {
      "epoch": 0.8366034900225432,
      "grad_norm": 0.35262271761894226,
      "learning_rate": 2e-05,
      "loss": 0.5936,
      "step": 10020
    },
    {
      "epoch": 0.8374384236453202,
      "grad_norm": 0.4137907922267914,
      "learning_rate": 2e-05,
      "loss": 0.6185,
      "step": 10030
    },
    {
      "epoch": 0.8382733572680972,
      "grad_norm": 0.35547366738319397,
      "learning_rate": 2e-05,
      "loss": 0.6108,
      "step": 10040
    },
    {
      "epoch": 0.8391082908908741,
      "grad_norm": 0.39250293374061584,
      "learning_rate": 2e-05,
      "loss": 0.5965,
      "step": 10050
    },
    {
      "epoch": 0.8399432245136511,
      "grad_norm": 0.34358125925064087,
      "learning_rate": 2e-05,
      "loss": 0.5836,
      "step": 10060
    },
    {
      "epoch": 0.8407781581364282,
      "grad_norm": 0.34757301211357117,
      "learning_rate": 2e-05,
      "loss": 0.6083,
      "step": 10070
    },
    {
      "epoch": 0.8416130917592052,
      "grad_norm": 0.35956668853759766,
      "learning_rate": 2e-05,
      "loss": 0.5883,
      "step": 10080
    },
    {
      "epoch": 0.8424480253819822,
      "grad_norm": 0.40053948760032654,
      "learning_rate": 2e-05,
      "loss": 0.6141,
      "step": 10090
    },
    {
      "epoch": 0.8432829590047591,
      "grad_norm": 0.3407866954803467,
      "learning_rate": 2e-05,
      "loss": 0.5909,
      "step": 10100
    },
    {
      "epoch": 0.8441178926275361,
      "grad_norm": 0.3778192102909088,
      "learning_rate": 2e-05,
      "loss": 0.6215,
      "step": 10110
    },
    {
      "epoch": 0.8449528262503131,
      "grad_norm": 0.3503149151802063,
      "learning_rate": 2e-05,
      "loss": 0.6339,
      "step": 10120
    },
    {
      "epoch": 0.8457877598730901,
      "grad_norm": 0.4496859312057495,
      "learning_rate": 2e-05,
      "loss": 0.5951,
      "step": 10130
    },
    {
      "epoch": 0.846622693495867,
      "grad_norm": 0.37010782957077026,
      "learning_rate": 2e-05,
      "loss": 0.6214,
      "step": 10140
    },
    {
      "epoch": 0.847457627118644,
      "grad_norm": 0.39074718952178955,
      "learning_rate": 2e-05,
      "loss": 0.6253,
      "step": 10150
    },
    {
      "epoch": 0.848292560741421,
      "grad_norm": 0.35743311047554016,
      "learning_rate": 2e-05,
      "loss": 0.5869,
      "step": 10160
    },
    {
      "epoch": 0.8491274943641981,
      "grad_norm": 0.39758819341659546,
      "learning_rate": 2e-05,
      "loss": 0.5771,
      "step": 10170
    },
    {
      "epoch": 0.8499624279869751,
      "grad_norm": 0.37007418274879456,
      "learning_rate": 1.999957358086795e-05,
      "loss": 0.6082,
      "step": 10180
    },
    {
      "epoch": 0.8507973616097521,
      "grad_norm": 0.39760851860046387,
      "learning_rate": 1.9991993813020483e-05,
      "loss": 0.6326,
      "step": 10190
    },
    {
      "epoch": 0.851632295232529,
      "grad_norm": 0.32918378710746765,
      "learning_rate": 1.9974946373028628e-05,
      "loss": 0.5971,
      "step": 10200
    },
    {
      "epoch": 0.852467228855306,
      "grad_norm": 0.36492687463760376,
      "learning_rate": 1.994844749501764e-05,
      "loss": 0.5636,
      "step": 10210
    },
    {
      "epoch": 0.853302162478083,
      "grad_norm": 0.35209789872169495,
      "learning_rate": 1.9912522413632988e-05,
      "loss": 0.6031,
      "step": 10220
    },
    {
      "epoch": 0.85413709610086,
      "grad_norm": 0.3639342188835144,
      "learning_rate": 1.9867205340009656e-05,
      "loss": 0.5787,
      "step": 10230
    },
    {
      "epoch": 0.854972029723637,
      "grad_norm": 0.3678273856639862,
      "learning_rate": 1.9812539429193164e-05,
      "loss": 0.6111,
      "step": 10240
    },
    {
      "epoch": 0.8558069633464139,
      "grad_norm": 0.38371023535728455,
      "learning_rate": 1.9748576739043423e-05,
      "loss": 0.6276,
      "step": 10250
    },
    {
      "epoch": 0.8566418969691909,
      "grad_norm": 0.3652210831642151,
      "learning_rate": 1.967537818066047e-05,
      "loss": 0.5668,
      "step": 10260
    },
    {
      "epoch": 0.8574768305919679,
      "grad_norm": 0.3719028830528259,
      "learning_rate": 1.959301346037941e-05,
      "loss": 0.5951,
      "step": 10270
    },
    {
      "epoch": 0.858311764214745,
      "grad_norm": 0.392001748085022,
      "learning_rate": 1.9501561013389646e-05,
      "loss": 0.6174,
      "step": 10280
    },
    {
      "epoch": 0.859146697837522,
      "grad_norm": 0.3708076477050781,
      "learning_rate": 1.9401107929041795e-05,
      "loss": 0.5796,
      "step": 10290
    },
    {
      "epoch": 0.8599816314602989,
      "grad_norm": 0.37791723012924194,
      "learning_rate": 1.9291749867913225e-05,
      "loss": 0.5762,
      "step": 10300
    },
    {
      "epoch": 0.8608165650830759,
      "grad_norm": 0.3666907250881195,
      "learning_rate": 1.91735909707114e-05,
      "loss": 0.5754,
      "step": 10310
    },
    {
      "epoch": 0.8616514987058529,
      "grad_norm": 0.3981567919254303,
      "learning_rate": 1.9046743759101545e-05,
      "loss": 0.6103,
      "step": 10320
    },
    {
      "epoch": 0.8624864323286299,
      "grad_norm": 0.35273322463035583,
      "learning_rate": 1.8911329028553324e-05,
      "loss": 0.6066,
      "step": 10330
    },
    {
      "epoch": 0.8633213659514068,
      "grad_norm": 0.32982000708580017,
      "learning_rate": 1.876747573330835e-05,
      "loss": 0.6241,
      "step": 10340
    },
    {
      "epoch": 0.8641562995741838,
      "grad_norm": 0.35347530245780945,
      "learning_rate": 1.8615320863578244e-05,
      "loss": 0.5879,
      "step": 10350
    },
    {
      "epoch": 0.8649912331969608,
      "grad_norm": 0.3892061710357666,
      "learning_rate": 1.8455009315090036e-05,
      "loss": 0.6041,
      "step": 10360
    },
    {
      "epoch": 0.8658261668197378,
      "grad_norm": 0.3646557033061981,
      "learning_rate": 1.8286693751103294e-05,
      "loss": 0.6092,
      "step": 10370
    },
    {
      "epoch": 0.8666611004425148,
      "grad_norm": 0.3309522569179535,
      "learning_rate": 1.811053445703024e-05,
      "loss": 0.6002,
      "step": 10380
    },
    {
      "epoch": 0.8674960340652919,
      "grad_norm": 0.37891459465026855,
      "learning_rate": 1.7926699187797375e-05,
      "loss": 0.6034,
      "step": 10390
    },
    {
      "epoch": 0.8683309676880688,
      "grad_norm": 0.40277040004730225,
      "learning_rate": 1.7735363008093995e-05,
      "loss": 0.5968,
      "step": 10400
    },
    {
      "epoch": 0.8691659013108458,
      "grad_norm": 0.3684222996234894,
      "learning_rate": 1.7536708125659577e-05,
      "loss": 0.6188,
      "step": 10410
    },
    {
      "epoch": 0.8700008349336228,
      "grad_norm": 0.34457340836524963,
      "learning_rate": 1.733092371776905e-05,
      "loss": 0.603,
      "step": 10420
    },
    {
      "epoch": 0.8708357685563998,
      "grad_norm": 0.3593447208404541,
      "learning_rate": 1.7118205751080915e-05,
      "loss": 0.5948,
      "step": 10430
    },
    {
      "epoch": 0.8716707021791767,
      "grad_norm": 0.358451247215271,
      "learning_rate": 1.6898756795019935e-05,
      "loss": 0.5895,
      "step": 10440
    },
    {
      "epoch": 0.8725056358019537,
      "grad_norm": 0.3518572747707367,
      "learning_rate": 1.667278582887211e-05,
      "loss": 0.6011,
      "step": 10450
    },
    {
      "epoch": 0.8733405694247307,
      "grad_norm": 0.3565864562988281,
      "learning_rate": 1.6440508042775545e-05,
      "loss": 0.6101,
      "step": 10460
    },
    {
      "epoch": 0.8741755030475077,
      "grad_norm": 0.3328714072704315,
      "learning_rate": 1.6202144632796796e-05,
      "loss": 0.5977,
      "step": 10470
    },
    {
      "epoch": 0.8750104366702847,
      "grad_norm": 0.4373946189880371,
      "learning_rate": 1.595792259028783e-05,
      "loss": 0.5964,
      "step": 10480
    },
    {
      "epoch": 0.8758453702930618,
      "grad_norm": 0.40729203820228577,
      "learning_rate": 1.570807448572419e-05,
      "loss": 0.5886,
      "step": 10490
    },
    {
      "epoch": 0.8766803039158387,
      "grad_norm": 0.3842219412326813,
      "learning_rate": 1.5452838247230197e-05,
      "loss": 0.6097,
      "step": 10500
    },
    {
      "epoch": 0.8775152375386157,
      "grad_norm": 0.4176235496997833,
      "learning_rate": 1.5192456934002106e-05,
      "loss": 0.6032,
      "step": 10510
    },
    {
      "epoch": 0.8783501711613927,
      "grad_norm": 0.3615528643131256,
      "learning_rate": 1.492717850484502e-05,
      "loss": 0.5786,
      "step": 10520
    },
    {
      "epoch": 0.8791851047841697,
      "grad_norm": 0.3478465676307678,
      "learning_rate": 1.4657255582043904e-05,
      "loss": 0.5685,
      "step": 10530
    },
    {
      "epoch": 0.8800200384069466,
      "grad_norm": 0.461124986410141,
      "learning_rate": 1.4382945210793663e-05,
      "loss": 0.5941,
      "step": 10540
    },
    {
      "epoch": 0.8808549720297236,
      "grad_norm": 0.38237428665161133,
      "learning_rate": 1.4104508614417267e-05,
      "loss": 0.6097,
      "step": 10550
    },
    {
      "epoch": 0.8816899056525006,
      "grad_norm": 0.339219868183136,
      "learning_rate": 1.382221094560512e-05,
      "loss": 0.5803,
      "step": 10560
    },
    {
      "epoch": 0.8825248392752776,
      "grad_norm": 0.3779853880405426,
      "learning_rate": 1.3536321033912506e-05,
      "loss": 0.6179,
      "step": 10570
    },
    {
      "epoch": 0.8833597728980546,
      "grad_norm": 0.3430643081665039,
      "learning_rate": 1.3247111129755609e-05,
      "loss": 0.6132,
      "step": 10580
    },
    {
      "epoch": 0.8841947065208315,
      "grad_norm": 0.3803752362728119,
      "learning_rate": 1.2954856645149868e-05,
      "loss": 0.5695,
      "step": 10590
    },
    {
      "epoch": 0.8850296401436086,
      "grad_norm": 0.3585555851459503,
      "learning_rate": 1.2659835891437553e-05,
      "loss": 0.6059,
      "step": 10600
    },
    {
      "epoch": 0.8858645737663856,
      "grad_norm": 0.3682415783405304,
      "learning_rate": 1.236232981425436e-05,
      "loss": 0.5948,
      "step": 10610
    },
    {
      "epoch": 0.8866995073891626,
      "grad_norm": 0.35322144627571106,
      "learning_rate": 1.2062621725987402e-05,
      "loss": 0.6181,
      "step": 10620
    },
    {
      "epoch": 0.8875344410119396,
      "grad_norm": 0.3899747431278229,
      "learning_rate": 1.1760997035979318e-05,
      "loss": 0.5956,
      "step": 10630
    },
    {
      "epoch": 0.8883693746347165,
      "grad_norm": 0.3507107198238373,
      "learning_rate": 1.1457742978735514e-05,
      "loss": 0.6072,
      "step": 10640
    },
    {
      "epoch": 0.8892043082574935,
      "grad_norm": 0.3397108018398285,
      "learning_rate": 1.1153148340393298e-05,
      "loss": 0.5847,
      "step": 10650
    },
    {
      "epoch": 0.8900392418802705,
      "grad_norm": 0.36083757877349854,
      "learning_rate": 1.0847503183713424e-05,
      "loss": 0.5935,
      "step": 10660
    },
    {
      "epoch": 0.8908741755030475,
      "grad_norm": 0.32317498326301575,
      "learning_rate": 1.0541098571855896e-05,
      "loss": 0.5598,
      "step": 10670
    },
    {
      "epoch": 0.8917091091258245,
      "grad_norm": 0.3648549020290375,
      "learning_rate": 1.0234226291203123e-05,
      "loss": 0.5757,
      "step": 10680
    },
    {
      "epoch": 0.8925440427486014,
      "grad_norm": 0.36010846495628357,
      "learning_rate": 9.927178573494333e-06,
      "loss": 0.6076,
      "step": 10690
    },
    {
      "epoch": 0.8933789763713785,
      "grad_norm": 0.35454869270324707,
      "learning_rate": 9.620247817535902e-06,
      "loss": 0.5499,
      "step": 10700
    },
    {
      "epoch": 0.8942139099941555,
      "grad_norm": 0.34091150760650635,
      "learning_rate": 9.313726310752581e-06,
      "loss": 0.5649,
      "step": 10710
    },
    {
      "epoch": 0.8950488436169325,
      "grad_norm": 0.4124930799007416,
      "learning_rate": 9.00790595084477e-06,
      "loss": 0.5775,
      "step": 10720
    },
    {
      "epoch": 0.8958837772397095,
      "grad_norm": 0.46319183707237244,
      "learning_rate": 8.703077967816944e-06,
      "loss": 0.5794,
      "step": 10730
    },
    {
      "epoch": 0.8967187108624864,
      "grad_norm": 0.34511756896972656,
      "learning_rate": 8.39953264664192e-06,
      "loss": 0.5889,
      "step": 10740
    },
    {
      "epoch": 0.8975536444852634,
      "grad_norm": 0.3821766972541809,
      "learning_rate": 8.097559050825058e-06,
      "loss": 0.5924,
      "step": 10750
    },
    {
      "epoch": 0.8983885781080404,
      "grad_norm": 0.3572334945201874,
      "learning_rate": 7.79744474713166e-06,
      "loss": 0.5908,
      "step": 10760
    },
    {
      "epoch": 0.8992235117308174,
      "grad_norm": 0.36048629879951477,
      "learning_rate": 7.499475531739696e-06,
      "loss": 0.5894,
      "step": 10770
    },
    {
      "epoch": 0.9000584453535944,
      "grad_norm": 0.35607296228408813,
      "learning_rate": 7.203935158078662e-06,
      "loss": 0.5969,
      "step": 10780
    },
    {
      "epoch": 0.9008933789763713,
      "grad_norm": 0.35151657462120056,
      "learning_rate": 6.911105066613709e-06,
      "loss": 0.5583,
      "step": 10790
    },
    {
      "epoch": 0.9017283125991483,
      "grad_norm": 0.36734962463378906,
      "learning_rate": 6.621264116832394e-06,
      "loss": 0.5913,
      "step": 10800
    },
    {
      "epoch": 0.9025632462219254,
      "grad_norm": 0.33584025502204895,
      "learning_rate": 6.334688321689274e-06,
      "loss": 0.5793,
      "step": 10810
    },
    {
      "epoch": 0.9033981798447024,
      "grad_norm": 0.3446446657180786,
      "learning_rate": 6.051650584761237e-06,
      "loss": 0.5634,
      "step": 10820
    },
    {
      "epoch": 0.9042331134674794,
      "grad_norm": 0.3458070158958435,
      "learning_rate": 5.772420440363841e-06,
      "loss": 0.5793,
      "step": 10830
    },
    {
      "epoch": 0.9050680470902563,
      "grad_norm": 0.37715059518814087,
      "learning_rate": 5.497263796876207e-06,
      "loss": 0.6141,
      "step": 10840
    },
    {
      "epoch": 0.9059029807130333,
      "grad_norm": 0.36355647444725037,
      "learning_rate": 5.226442683518813e-06,
      "loss": 0.5809,
      "step": 10850
    },
    {
      "epoch": 0.9067379143358103,
      "grad_norm": 0.3633350133895874,
      "learning_rate": 4.960215000825432e-06,
      "loss": 0.5773,
      "step": 10860
    },
    {
      "epoch": 0.9075728479585873,
      "grad_norm": 0.35102567076683044,
      "learning_rate": 4.698834275046711e-06,
      "loss": 0.5832,
      "step": 10870
    },
    {
      "epoch": 0.9084077815813643,
      "grad_norm": 0.3875003159046173,
      "learning_rate": 4.442549416719423e-06,
      "loss": 0.6144,
      "step": 10880
    },
    {
      "epoch": 0.9092427152041412,
      "grad_norm": 0.3612220883369446,
      "learning_rate": 4.1916044836311565e-06,
      "loss": 0.599,
      "step": 10890
    },
    {
      "epoch": 0.9100776488269182,
      "grad_norm": 0.3629154562950134,
      "learning_rate": 3.94623844840626e-06,
      "loss": 0.5804,
      "step": 10900
    },
    {
      "epoch": 0.9109125824496952,
      "grad_norm": 0.3650902509689331,
      "learning_rate": 3.7066849709343467e-06,
      "loss": 0.5725,
      "step": 10910
    },
    {
      "epoch": 0.9117475160724723,
      "grad_norm": 0.336750328540802,
      "learning_rate": 3.473172175858059e-06,
      "loss": 0.5957,
      "step": 10920
    },
    {
      "epoch": 0.9125824496952493,
      "grad_norm": 0.3504043221473694,
      "learning_rate": 3.2459224353320046e-06,
      "loss": 0.5703,
      "step": 10930
    },
    {
      "epoch": 0.9134173833180262,
      "grad_norm": 0.3550267219543457,
      "learning_rate": 3.025152157259735e-06,
      "loss": 0.5976,
      "step": 10940
    },
    {
      "epoch": 0.9142523169408032,
      "grad_norm": 0.3537559509277344,
      "learning_rate": 2.811071579210425e-06,
      "loss": 0.5907,
      "step": 10950
    },
    {
      "epoch": 0.9150872505635802,
      "grad_norm": 0.36417996883392334,
      "learning_rate": 2.603884568211513e-06,
      "loss": 0.6071,
      "step": 10960
    },
    {
      "epoch": 0.9159221841863572,
      "grad_norm": 0.36349087953567505,
      "learning_rate": 2.4037884266079434e-06,
      "loss": 0.5813,
      "step": 10970
    },
    {
      "epoch": 0.9167571178091342,
      "grad_norm": 0.3333248496055603,
      "learning_rate": 2.2109737041728955e-06,
      "loss": 0.5803,
      "step": 10980
    },
    {
      "epoch": 0.9175920514319111,
      "grad_norm": 0.36495158076286316,
      "learning_rate": 2.0256240166489454e-06,
      "loss": 0.6052,
      "step": 10990
    },
    {
      "epoch": 0.9184269850546881,
      "grad_norm": 0.33543699979782104,
      "learning_rate": 1.847915870892413e-06,
      "loss": 0.5754,
      "step": 11000
    },
    {
      "epoch": 0.9192619186774651,
      "grad_norm": 0.3315180540084839,
      "learning_rate": 1.6780184967874715e-06,
      "loss": 0.5366,
      "step": 11010
    },
    {
      "epoch": 0.9200968523002422,
      "grad_norm": 0.36481913924217224,
      "learning_rate": 1.5160936860900294e-06,
      "loss": 0.5975,
      "step": 11020
    },
    {
      "epoch": 0.9209317859230192,
      "grad_norm": 0.3755260705947876,
      "learning_rate": 1.3622956383548762e-06,
      "loss": 0.5952,
      "step": 11030
    },
    {
      "epoch": 0.9217667195457961,
      "grad_norm": 0.3339856266975403,
      "learning_rate": 1.216770814092834e-06,
      "loss": 0.6047,
      "step": 11040
    },
    {
      "epoch": 0.9226016531685731,
      "grad_norm": 0.3536217510700226,
      "learning_rate": 1.0796577952977013e-06,
      "loss": 0.6029,
      "step": 11050
    },
    {
      "epoch": 0.9234365867913501,
      "grad_norm": 0.3443988263607025,
      "learning_rate": 9.510871534758705e-07,
      "loss": 0.594,
      "step": 11060
    },
    {
      "epoch": 0.9242715204141271,
      "grad_norm": 0.3684104382991791,
      "learning_rate": 8.311813253042334e-07,
      "loss": 0.5881,
      "step": 11070
    },
    {
      "epoch": 0.9251064540369041,
      "grad_norm": 0.35818350315093994,
      "learning_rate": 7.200544960348308e-07,
      "loss": 0.5578,
      "step": 11080
    },
    {
      "epoch": 0.925941387659681,
      "grad_norm": 0.3459317088127136,
      "learning_rate": 6.178124907572545e-07,
      "loss": 0.5963,
      "step": 11090
    },
    {
      "epoch": 0.926776321282458,
      "grad_norm": 0.34455206990242004,
      "learning_rate": 5.245526736223528e-07,
      "loss": 0.5722,
      "step": 11100
    },
    {
      "epoch": 0.927611254905235,
      "grad_norm": 0.4079967439174652,
      "learning_rate": 4.403638551232205e-07,
      "loss": 0.5657,
      "step": 11110
    },
    {
      "epoch": 0.928446188528012,
      "grad_norm": 0.33032408356666565,
      "learning_rate": 3.6532620752175616e-07,
      "loss": 0.584,
      "step": 11120
    },
    {
      "epoch": 0.9292811221507891,
      "grad_norm": 0.3536728322505951,
      "learning_rate": 2.9951118850133247e-07,
      "loss": 0.5905,
      "step": 11130
    },
    {
      "epoch": 0.930116055773566,
      "grad_norm": 0.3770751953125,
      "learning_rate": 2.429814731182774e-07,
      "loss": 0.5859,
      "step": 11140
    },
    {
      "epoch": 0.930950989396343,
      "grad_norm": 0.33062511682510376,
      "learning_rate": 1.957908941169813e-07,
      "loss": 0.5852,
      "step": 11150
    },
    {
      "epoch": 0.93178592301912,
      "grad_norm": 0.3163359761238098,
      "learning_rate": 1.5798439066545254e-07,
      "loss": 0.5804,
      "step": 11160
    },
    {
      "epoch": 0.932620856641897,
      "grad_norm": 0.35043323040008545,
      "learning_rate": 1.2959796556014471e-07,
      "loss": 0.5582,
      "step": 11170
    },
    {
      "epoch": 0.933455790264674,
      "grad_norm": 0.3704030215740204,
      "learning_rate": 1.1065865094081884e-07,
      "loss": 0.5803,
      "step": 11180
    },
    {
      "epoch": 0.9342907238874509,
      "grad_norm": 0.3586052656173706,
      "learning_rate": 1.011844825480734e-07,
      "loss": 0.578,
      "step": 11190
    },
    {
      "epoch": 0.9346246973365617,
      "step": 11194,
      "total_flos": 4.547514943786923e+19,
      "train_loss": 0.003304087116779207,
      "train_runtime": 610.4281,
      "train_samples_per_second": 220.055,
      "train_steps_per_second": 18.338
    }
  ],
  "logging_steps": 10,
  "max_steps": 11194,
  "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": 4.547514943786923e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}