{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.34963181148748157,
  "eval_steps": 500,
  "global_step": 5935,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000589101620029455,
      "grad_norm": 1.1187952756881714,
      "learning_rate": 2.910112359550562e-06,
      "loss": 0.7831,
      "step": 10
    },
    {
      "epoch": 0.00117820324005891,
      "grad_norm": 1.0689250230789185,
      "learning_rate": 3.921348314606742e-06,
      "loss": 0.8085,
      "step": 20
    },
    {
      "epoch": 0.0017673048600883653,
      "grad_norm": 0.741685688495636,
      "learning_rate": 4.932584269662922e-06,
      "loss": 0.7525,
      "step": 30
    },
    {
      "epoch": 0.00235640648011782,
      "grad_norm": 0.8915156722068787,
      "learning_rate": 5.943820224719102e-06,
      "loss": 0.7265,
      "step": 40
    },
    {
      "epoch": 0.0029455081001472753,
      "grad_norm": 0.6924229264259338,
      "learning_rate": 6.955056179775282e-06,
      "loss": 0.7388,
      "step": 50
    },
    {
      "epoch": 0.0035346097201767305,
      "grad_norm": 0.7231901288032532,
      "learning_rate": 7.966292134831462e-06,
      "loss": 0.7317,
      "step": 60
    },
    {
      "epoch": 0.004123711340206186,
      "grad_norm": 0.7959139943122864,
      "learning_rate": 8.977528089887641e-06,
      "loss": 0.7457,
      "step": 70
    },
    {
      "epoch": 0.00471281296023564,
      "grad_norm": 0.8100572824478149,
      "learning_rate": 9.988764044943822e-06,
      "loss": 0.7577,
      "step": 80
    },
    {
      "epoch": 0.005301914580265096,
      "grad_norm": 0.7159477472305298,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 0.7537,
      "step": 90
    },
    {
      "epoch": 0.005891016200294551,
      "grad_norm": 0.8149115443229675,
      "learning_rate": 1.2011235955056182e-05,
      "loss": 0.7762,
      "step": 100
    },
    {
      "epoch": 0.006480117820324006,
      "grad_norm": 0.8290235996246338,
      "learning_rate": 1.3022471910112362e-05,
      "loss": 0.7777,
      "step": 110
    },
    {
      "epoch": 0.007069219440353461,
      "grad_norm": 0.7166168093681335,
      "learning_rate": 1.403370786516854e-05,
      "loss": 0.7597,
      "step": 120
    },
    {
      "epoch": 0.007658321060382916,
      "grad_norm": 0.8164111375808716,
      "learning_rate": 1.504494382022472e-05,
      "loss": 0.7249,
      "step": 130
    },
    {
      "epoch": 0.008247422680412371,
      "grad_norm": 0.7679075598716736,
      "learning_rate": 1.60561797752809e-05,
      "loss": 0.7427,
      "step": 140
    },
    {
      "epoch": 0.008836524300441826,
      "grad_norm": 0.7330767512321472,
      "learning_rate": 1.706741573033708e-05,
      "loss": 0.7465,
      "step": 150
    },
    {
      "epoch": 0.00942562592047128,
      "grad_norm": 0.6833717823028564,
      "learning_rate": 1.8078651685393256e-05,
      "loss": 0.7656,
      "step": 160
    },
    {
      "epoch": 0.010014727540500737,
      "grad_norm": 0.7289340496063232,
      "learning_rate": 1.908988764044944e-05,
      "loss": 0.7791,
      "step": 170
    },
    {
      "epoch": 0.010603829160530192,
      "grad_norm": 0.6722760200500488,
      "learning_rate": 2e-05,
      "loss": 0.7861,
      "step": 180
    },
    {
      "epoch": 0.011192930780559647,
      "grad_norm": 0.6587191820144653,
      "learning_rate": 2e-05,
      "loss": 0.7593,
      "step": 190
    },
    {
      "epoch": 0.011782032400589101,
      "grad_norm": 1.342365026473999,
      "learning_rate": 2e-05,
      "loss": 0.7532,
      "step": 200
    },
    {
      "epoch": 0.012371134020618556,
      "grad_norm": 0.7734410166740417,
      "learning_rate": 2e-05,
      "loss": 0.7475,
      "step": 210
    },
    {
      "epoch": 0.012960235640648013,
      "grad_norm": 0.6449353098869324,
      "learning_rate": 2e-05,
      "loss": 0.7669,
      "step": 220
    },
    {
      "epoch": 0.013549337260677467,
      "grad_norm": 0.6131564378738403,
      "learning_rate": 2e-05,
      "loss": 0.7472,
      "step": 230
    },
    {
      "epoch": 0.014138438880706922,
      "grad_norm": 0.7450922131538391,
      "learning_rate": 2e-05,
      "loss": 0.7685,
      "step": 240
    },
    {
      "epoch": 0.014727540500736377,
      "grad_norm": 0.5752319693565369,
      "learning_rate": 2e-05,
      "loss": 0.7504,
      "step": 250
    },
    {
      "epoch": 0.015316642120765831,
      "grad_norm": 0.5238183736801147,
      "learning_rate": 2e-05,
      "loss": 0.7503,
      "step": 260
    },
    {
      "epoch": 0.015905743740795286,
      "grad_norm": 0.6273600459098816,
      "learning_rate": 2e-05,
      "loss": 0.7511,
      "step": 270
    },
    {
      "epoch": 0.016494845360824743,
      "grad_norm": 0.4939355254173279,
      "learning_rate": 2e-05,
      "loss": 0.7819,
      "step": 280
    },
    {
      "epoch": 0.017083946980854196,
      "grad_norm": 0.7005998492240906,
      "learning_rate": 2e-05,
      "loss": 0.7411,
      "step": 290
    },
    {
      "epoch": 0.017673048600883652,
      "grad_norm": 0.5144055485725403,
      "learning_rate": 2e-05,
      "loss": 0.7529,
      "step": 300
    },
    {
      "epoch": 0.01826215022091311,
      "grad_norm": 0.4719701111316681,
      "learning_rate": 2e-05,
      "loss": 0.755,
      "step": 310
    },
    {
      "epoch": 0.01885125184094256,
      "grad_norm": 0.5226520299911499,
      "learning_rate": 2e-05,
      "loss": 0.7586,
      "step": 320
    },
    {
      "epoch": 0.019440353460972018,
      "grad_norm": 0.5080249309539795,
      "learning_rate": 2e-05,
      "loss": 0.753,
      "step": 330
    },
    {
      "epoch": 0.020029455081001474,
      "grad_norm": 0.46481338143348694,
      "learning_rate": 2e-05,
      "loss": 0.7675,
      "step": 340
    },
    {
      "epoch": 0.020618556701030927,
      "grad_norm": 0.46027958393096924,
      "learning_rate": 2e-05,
      "loss": 0.7389,
      "step": 350
    },
    {
      "epoch": 0.021207658321060384,
      "grad_norm": 0.49125781655311584,
      "learning_rate": 2e-05,
      "loss": 0.758,
      "step": 360
    },
    {
      "epoch": 0.021796759941089837,
      "grad_norm": 0.4505169987678528,
      "learning_rate": 2e-05,
      "loss": 0.7586,
      "step": 370
    },
    {
      "epoch": 0.022385861561119293,
      "grad_norm": 0.5151411890983582,
      "learning_rate": 2e-05,
      "loss": 0.7431,
      "step": 380
    },
    {
      "epoch": 0.02297496318114875,
      "grad_norm": 0.4783925712108612,
      "learning_rate": 2e-05,
      "loss": 0.7412,
      "step": 390
    },
    {
      "epoch": 0.023564064801178203,
      "grad_norm": 0.4741362929344177,
      "learning_rate": 2e-05,
      "loss": 0.7413,
      "step": 400
    },
    {
      "epoch": 0.02415316642120766,
      "grad_norm": 0.46201640367507935,
      "learning_rate": 2e-05,
      "loss": 0.7445,
      "step": 410
    },
    {
      "epoch": 0.024742268041237112,
      "grad_norm": 0.4987025558948517,
      "learning_rate": 2e-05,
      "loss": 0.7517,
      "step": 420
    },
    {
      "epoch": 0.02533136966126657,
      "grad_norm": 0.4851631820201874,
      "learning_rate": 2e-05,
      "loss": 0.7594,
      "step": 430
    },
    {
      "epoch": 0.025920471281296025,
      "grad_norm": 0.47743213176727295,
      "learning_rate": 2e-05,
      "loss": 0.7565,
      "step": 440
    },
    {
      "epoch": 0.026509572901325478,
      "grad_norm": 0.5171688199043274,
      "learning_rate": 2e-05,
      "loss": 0.7569,
      "step": 450
    },
    {
      "epoch": 0.027098674521354935,
      "grad_norm": 0.5514922142028809,
      "learning_rate": 2e-05,
      "loss": 0.7244,
      "step": 460
    },
    {
      "epoch": 0.027687776141384388,
      "grad_norm": 0.4684111177921295,
      "learning_rate": 2e-05,
      "loss": 0.7346,
      "step": 470
    },
    {
      "epoch": 0.028276877761413844,
      "grad_norm": 0.49709105491638184,
      "learning_rate": 2e-05,
      "loss": 0.7644,
      "step": 480
    },
    {
      "epoch": 0.0288659793814433,
      "grad_norm": 0.4689137935638428,
      "learning_rate": 2e-05,
      "loss": 0.7314,
      "step": 490
    },
    {
      "epoch": 0.029455081001472753,
      "grad_norm": 0.48819467425346375,
      "learning_rate": 2e-05,
      "loss": 0.7665,
      "step": 500
    },
    {
      "epoch": 0.03004418262150221,
      "grad_norm": 0.46228793263435364,
      "learning_rate": 2e-05,
      "loss": 0.7693,
      "step": 510
    },
    {
      "epoch": 0.030633284241531663,
      "grad_norm": 0.4367271661758423,
      "learning_rate": 2e-05,
      "loss": 0.739,
      "step": 520
    },
    {
      "epoch": 0.03122238586156112,
      "grad_norm": 0.4632743000984192,
      "learning_rate": 2e-05,
      "loss": 0.7376,
      "step": 530
    },
    {
      "epoch": 0.03181148748159057,
      "grad_norm": 0.47965553402900696,
      "learning_rate": 2e-05,
      "loss": 0.7633,
      "step": 540
    },
    {
      "epoch": 0.03240058910162003,
      "grad_norm": 0.48231041431427,
      "learning_rate": 2e-05,
      "loss": 0.7412,
      "step": 550
    },
    {
      "epoch": 0.032989690721649485,
      "grad_norm": 0.48584654927253723,
      "learning_rate": 2e-05,
      "loss": 0.7347,
      "step": 560
    },
    {
      "epoch": 0.03357879234167894,
      "grad_norm": 0.45019200444221497,
      "learning_rate": 2e-05,
      "loss": 0.755,
      "step": 570
    },
    {
      "epoch": 0.03416789396170839,
      "grad_norm": 0.44921109080314636,
      "learning_rate": 2e-05,
      "loss": 0.7483,
      "step": 580
    },
    {
      "epoch": 0.03475699558173785,
      "grad_norm": 0.45112189650535583,
      "learning_rate": 2e-05,
      "loss": 0.7656,
      "step": 590
    },
    {
      "epoch": 0.035346097201767304,
      "grad_norm": 0.4544597268104553,
      "learning_rate": 2e-05,
      "loss": 0.7557,
      "step": 600
    },
    {
      "epoch": 0.03593519882179676,
      "grad_norm": 0.5381441712379456,
      "learning_rate": 2e-05,
      "loss": 0.7488,
      "step": 610
    },
    {
      "epoch": 0.03652430044182622,
      "grad_norm": 0.42246851325035095,
      "learning_rate": 2e-05,
      "loss": 0.7706,
      "step": 620
    },
    {
      "epoch": 0.03711340206185567,
      "grad_norm": 0.4562056362628937,
      "learning_rate": 2e-05,
      "loss": 0.7692,
      "step": 630
    },
    {
      "epoch": 0.03770250368188512,
      "grad_norm": 0.45282721519470215,
      "learning_rate": 2e-05,
      "loss": 0.7365,
      "step": 640
    },
    {
      "epoch": 0.03829160530191458,
      "grad_norm": 0.4427643120288849,
      "learning_rate": 2e-05,
      "loss": 0.7497,
      "step": 650
    },
    {
      "epoch": 0.038880706921944036,
      "grad_norm": 0.4979972839355469,
      "learning_rate": 2e-05,
      "loss": 0.7486,
      "step": 660
    },
    {
      "epoch": 0.03946980854197349,
      "grad_norm": 0.42347976565361023,
      "learning_rate": 2e-05,
      "loss": 0.7343,
      "step": 670
    },
    {
      "epoch": 0.04005891016200295,
      "grad_norm": 0.4554508924484253,
      "learning_rate": 2e-05,
      "loss": 0.7442,
      "step": 680
    },
    {
      "epoch": 0.0406480117820324,
      "grad_norm": 0.46228188276290894,
      "learning_rate": 2e-05,
      "loss": 0.7706,
      "step": 690
    },
    {
      "epoch": 0.041237113402061855,
      "grad_norm": 0.44366255402565,
      "learning_rate": 2e-05,
      "loss": 0.7382,
      "step": 700
    },
    {
      "epoch": 0.04182621502209131,
      "grad_norm": 0.418224573135376,
      "learning_rate": 2e-05,
      "loss": 0.7221,
      "step": 710
    },
    {
      "epoch": 0.04241531664212077,
      "grad_norm": 0.4644136428833008,
      "learning_rate": 2e-05,
      "loss": 0.7238,
      "step": 720
    },
    {
      "epoch": 0.04300441826215022,
      "grad_norm": 0.4094725251197815,
      "learning_rate": 2e-05,
      "loss": 0.7274,
      "step": 730
    },
    {
      "epoch": 0.043593519882179674,
      "grad_norm": 0.431404173374176,
      "learning_rate": 2e-05,
      "loss": 0.7346,
      "step": 740
    },
    {
      "epoch": 0.044182621502209134,
      "grad_norm": 0.45599696040153503,
      "learning_rate": 2e-05,
      "loss": 0.7257,
      "step": 750
    },
    {
      "epoch": 0.04477172312223859,
      "grad_norm": 0.4296494424343109,
      "learning_rate": 2e-05,
      "loss": 0.7408,
      "step": 760
    },
    {
      "epoch": 0.04536082474226804,
      "grad_norm": 0.44864070415496826,
      "learning_rate": 2e-05,
      "loss": 0.7327,
      "step": 770
    },
    {
      "epoch": 0.0459499263622975,
      "grad_norm": 0.4230384826660156,
      "learning_rate": 2e-05,
      "loss": 0.7599,
      "step": 780
    },
    {
      "epoch": 0.04653902798232695,
      "grad_norm": 0.4637373685836792,
      "learning_rate": 2e-05,
      "loss": 0.7239,
      "step": 790
    },
    {
      "epoch": 0.047128129602356406,
      "grad_norm": 0.5369905233383179,
      "learning_rate": 2e-05,
      "loss": 0.7488,
      "step": 800
    },
    {
      "epoch": 0.04771723122238586,
      "grad_norm": 0.492144912481308,
      "learning_rate": 2e-05,
      "loss": 0.752,
      "step": 810
    },
    {
      "epoch": 0.04830633284241532,
      "grad_norm": 0.4071194529533386,
      "learning_rate": 2e-05,
      "loss": 0.7192,
      "step": 820
    },
    {
      "epoch": 0.04889543446244477,
      "grad_norm": 0.44176292419433594,
      "learning_rate": 2e-05,
      "loss": 0.7558,
      "step": 830
    },
    {
      "epoch": 0.049484536082474224,
      "grad_norm": 0.4265081286430359,
      "learning_rate": 2e-05,
      "loss": 0.7669,
      "step": 840
    },
    {
      "epoch": 0.050073637702503684,
      "grad_norm": 0.4696004092693329,
      "learning_rate": 2e-05,
      "loss": 0.7304,
      "step": 850
    },
    {
      "epoch": 0.05066273932253314,
      "grad_norm": 0.5208477973937988,
      "learning_rate": 2e-05,
      "loss": 0.7265,
      "step": 860
    },
    {
      "epoch": 0.05125184094256259,
      "grad_norm": 0.5095232725143433,
      "learning_rate": 2e-05,
      "loss": 0.7576,
      "step": 870
    },
    {
      "epoch": 0.05184094256259205,
      "grad_norm": 0.4431394934654236,
      "learning_rate": 2e-05,
      "loss": 0.7343,
      "step": 880
    },
    {
      "epoch": 0.0524300441826215,
      "grad_norm": 0.4195699989795685,
      "learning_rate": 2e-05,
      "loss": 0.738,
      "step": 890
    },
    {
      "epoch": 0.053019145802650956,
      "grad_norm": 0.4056459069252014,
      "learning_rate": 2e-05,
      "loss": 0.7368,
      "step": 900
    },
    {
      "epoch": 0.05360824742268041,
      "grad_norm": 0.4221668243408203,
      "learning_rate": 2e-05,
      "loss": 0.762,
      "step": 910
    },
    {
      "epoch": 0.05419734904270987,
      "grad_norm": 0.4228168725967407,
      "learning_rate": 2e-05,
      "loss": 0.7391,
      "step": 920
    },
    {
      "epoch": 0.05478645066273932,
      "grad_norm": 0.4960448741912842,
      "learning_rate": 2e-05,
      "loss": 0.7296,
      "step": 930
    },
    {
      "epoch": 0.055375552282768775,
      "grad_norm": 0.4253543019294739,
      "learning_rate": 2e-05,
      "loss": 0.727,
      "step": 940
    },
    {
      "epoch": 0.055964653902798235,
      "grad_norm": 0.46824753284454346,
      "learning_rate": 2e-05,
      "loss": 0.7394,
      "step": 950
    },
    {
      "epoch": 0.05655375552282769,
      "grad_norm": 0.5169887542724609,
      "learning_rate": 2e-05,
      "loss": 0.7303,
      "step": 960
    },
    {
      "epoch": 0.05714285714285714,
      "grad_norm": 0.43343982100486755,
      "learning_rate": 2e-05,
      "loss": 0.735,
      "step": 970
    },
    {
      "epoch": 0.0577319587628866,
      "grad_norm": 1.695380687713623,
      "learning_rate": 2e-05,
      "loss": 0.7665,
      "step": 980
    },
    {
      "epoch": 0.058321060382916054,
      "grad_norm": 0.41318148374557495,
      "learning_rate": 2e-05,
      "loss": 0.7429,
      "step": 990
    },
    {
      "epoch": 0.05891016200294551,
      "grad_norm": 0.47614580392837524,
      "learning_rate": 2e-05,
      "loss": 0.7343,
      "step": 1000
    },
    {
      "epoch": 0.05949926362297496,
      "grad_norm": 0.4229415953159332,
      "learning_rate": 2e-05,
      "loss": 0.7456,
      "step": 1010
    },
    {
      "epoch": 0.06008836524300442,
      "grad_norm": 0.47622278332710266,
      "learning_rate": 2e-05,
      "loss": 0.7293,
      "step": 1020
    },
    {
      "epoch": 0.06067746686303387,
      "grad_norm": 0.4421570897102356,
      "learning_rate": 2e-05,
      "loss": 0.7518,
      "step": 1030
    },
    {
      "epoch": 0.061266568483063326,
      "grad_norm": 0.4493580162525177,
      "learning_rate": 2e-05,
      "loss": 0.7131,
      "step": 1040
    },
    {
      "epoch": 0.061855670103092786,
      "grad_norm": 0.42369574308395386,
      "learning_rate": 2e-05,
      "loss": 0.7445,
      "step": 1050
    },
    {
      "epoch": 0.06244477172312224,
      "grad_norm": 0.5011873841285706,
      "learning_rate": 2e-05,
      "loss": 0.7186,
      "step": 1060
    },
    {
      "epoch": 0.06303387334315169,
      "grad_norm": 0.4701177477836609,
      "learning_rate": 2e-05,
      "loss": 0.7367,
      "step": 1070
    },
    {
      "epoch": 0.06362297496318114,
      "grad_norm": 0.5048543214797974,
      "learning_rate": 2e-05,
      "loss": 0.7686,
      "step": 1080
    },
    {
      "epoch": 0.0642120765832106,
      "grad_norm": 0.43808069825172424,
      "learning_rate": 2e-05,
      "loss": 0.7496,
      "step": 1090
    },
    {
      "epoch": 0.06480117820324006,
      "grad_norm": 0.4087897539138794,
      "learning_rate": 2e-05,
      "loss": 0.7428,
      "step": 1100
    },
    {
      "epoch": 0.06539027982326952,
      "grad_norm": 0.94446861743927,
      "learning_rate": 2e-05,
      "loss": 0.7528,
      "step": 1110
    },
    {
      "epoch": 0.06597938144329897,
      "grad_norm": 0.5377583503723145,
      "learning_rate": 2e-05,
      "loss": 0.7481,
      "step": 1120
    },
    {
      "epoch": 0.06656848306332842,
      "grad_norm": 0.48349031805992126,
      "learning_rate": 2e-05,
      "loss": 0.7193,
      "step": 1130
    },
    {
      "epoch": 0.06715758468335788,
      "grad_norm": 0.42557501792907715,
      "learning_rate": 2e-05,
      "loss": 0.7669,
      "step": 1140
    },
    {
      "epoch": 0.06774668630338733,
      "grad_norm": 0.5073610544204712,
      "learning_rate": 2e-05,
      "loss": 0.7277,
      "step": 1150
    },
    {
      "epoch": 0.06833578792341678,
      "grad_norm": 0.41803184151649475,
      "learning_rate": 2e-05,
      "loss": 0.7231,
      "step": 1160
    },
    {
      "epoch": 0.06892488954344625,
      "grad_norm": 0.4486277401447296,
      "learning_rate": 2e-05,
      "loss": 0.7348,
      "step": 1170
    },
    {
      "epoch": 0.0695139911634757,
      "grad_norm": 0.415258526802063,
      "learning_rate": 2e-05,
      "loss": 0.7541,
      "step": 1180
    },
    {
      "epoch": 0.07010309278350516,
      "grad_norm": 0.4587970972061157,
      "learning_rate": 2e-05,
      "loss": 0.7237,
      "step": 1190
    },
    {
      "epoch": 0.07069219440353461,
      "grad_norm": 0.5214656591415405,
      "learning_rate": 2e-05,
      "loss": 0.734,
      "step": 1200
    },
    {
      "epoch": 0.07128129602356406,
      "grad_norm": 0.4164435863494873,
      "learning_rate": 2e-05,
      "loss": 0.7187,
      "step": 1210
    },
    {
      "epoch": 0.07187039764359351,
      "grad_norm": 0.4546908140182495,
      "learning_rate": 2e-05,
      "loss": 0.7283,
      "step": 1220
    },
    {
      "epoch": 0.07245949926362298,
      "grad_norm": 0.4424133896827698,
      "learning_rate": 2e-05,
      "loss": 0.752,
      "step": 1230
    },
    {
      "epoch": 0.07304860088365243,
      "grad_norm": 0.443491667509079,
      "learning_rate": 2e-05,
      "loss": 0.7748,
      "step": 1240
    },
    {
      "epoch": 0.07363770250368189,
      "grad_norm": 0.44335657358169556,
      "learning_rate": 2e-05,
      "loss": 0.7453,
      "step": 1250
    },
    {
      "epoch": 0.07422680412371134,
      "grad_norm": 1.5585325956344604,
      "learning_rate": 2e-05,
      "loss": 0.7408,
      "step": 1260
    },
    {
      "epoch": 0.0748159057437408,
      "grad_norm": 0.41405925154685974,
      "learning_rate": 2e-05,
      "loss": 0.7703,
      "step": 1270
    },
    {
      "epoch": 0.07540500736377025,
      "grad_norm": 0.42153358459472656,
      "learning_rate": 2e-05,
      "loss": 0.724,
      "step": 1280
    },
    {
      "epoch": 0.0759941089837997,
      "grad_norm": 0.4355378746986389,
      "learning_rate": 2e-05,
      "loss": 0.7554,
      "step": 1290
    },
    {
      "epoch": 0.07658321060382917,
      "grad_norm": 0.41889065504074097,
      "learning_rate": 2e-05,
      "loss": 0.7185,
      "step": 1300
    },
    {
      "epoch": 0.07717231222385862,
      "grad_norm": 0.508222222328186,
      "learning_rate": 2e-05,
      "loss": 0.7345,
      "step": 1310
    },
    {
      "epoch": 0.07776141384388807,
      "grad_norm": 0.46808117628097534,
      "learning_rate": 2e-05,
      "loss": 0.7343,
      "step": 1320
    },
    {
      "epoch": 0.07835051546391752,
      "grad_norm": 0.4918811321258545,
      "learning_rate": 2e-05,
      "loss": 0.7327,
      "step": 1330
    },
    {
      "epoch": 0.07893961708394698,
      "grad_norm": 0.4409327208995819,
      "learning_rate": 2e-05,
      "loss": 0.7378,
      "step": 1340
    },
    {
      "epoch": 0.07952871870397643,
      "grad_norm": 0.4449012875556946,
      "learning_rate": 2e-05,
      "loss": 0.7319,
      "step": 1350
    },
    {
      "epoch": 0.0801178203240059,
      "grad_norm": 0.46577584743499756,
      "learning_rate": 2e-05,
      "loss": 0.7294,
      "step": 1360
    },
    {
      "epoch": 0.08070692194403535,
      "grad_norm": 0.6492505073547363,
      "learning_rate": 2e-05,
      "loss": 0.7226,
      "step": 1370
    },
    {
      "epoch": 0.0812960235640648,
      "grad_norm": 0.42449140548706055,
      "learning_rate": 2e-05,
      "loss": 0.7532,
      "step": 1380
    },
    {
      "epoch": 0.08188512518409426,
      "grad_norm": 0.4328237771987915,
      "learning_rate": 2e-05,
      "loss": 0.7396,
      "step": 1390
    },
    {
      "epoch": 0.08247422680412371,
      "grad_norm": 0.4322971701622009,
      "learning_rate": 2e-05,
      "loss": 0.7283,
      "step": 1400
    },
    {
      "epoch": 0.08306332842415316,
      "grad_norm": 0.4679484963417053,
      "learning_rate": 2e-05,
      "loss": 0.7279,
      "step": 1410
    },
    {
      "epoch": 0.08365243004418262,
      "grad_norm": 0.42679035663604736,
      "learning_rate": 2e-05,
      "loss": 0.7292,
      "step": 1420
    },
    {
      "epoch": 0.08424153166421208,
      "grad_norm": 0.41338619589805603,
      "learning_rate": 2e-05,
      "loss": 0.7474,
      "step": 1430
    },
    {
      "epoch": 0.08483063328424154,
      "grad_norm": 0.46909600496292114,
      "learning_rate": 2e-05,
      "loss": 0.7116,
      "step": 1440
    },
    {
      "epoch": 0.08541973490427099,
      "grad_norm": 0.7349217534065247,
      "learning_rate": 2e-05,
      "loss": 0.7249,
      "step": 1450
    },
    {
      "epoch": 0.08600883652430044,
      "grad_norm": 0.4277472198009491,
      "learning_rate": 2e-05,
      "loss": 0.7343,
      "step": 1460
    },
    {
      "epoch": 0.0865979381443299,
      "grad_norm": 0.5415593981742859,
      "learning_rate": 2e-05,
      "loss": 0.7304,
      "step": 1470
    },
    {
      "epoch": 0.08718703976435935,
      "grad_norm": 0.417657732963562,
      "learning_rate": 2e-05,
      "loss": 0.7356,
      "step": 1480
    },
    {
      "epoch": 0.0877761413843888,
      "grad_norm": 0.4391482174396515,
      "learning_rate": 2e-05,
      "loss": 0.7235,
      "step": 1490
    },
    {
      "epoch": 0.08836524300441827,
      "grad_norm": 0.46071121096611023,
      "learning_rate": 2e-05,
      "loss": 0.7277,
      "step": 1500
    },
    {
      "epoch": 0.08895434462444772,
      "grad_norm": 0.521892786026001,
      "learning_rate": 2e-05,
      "loss": 0.7776,
      "step": 1510
    },
    {
      "epoch": 0.08954344624447717,
      "grad_norm": 0.43941742181777954,
      "learning_rate": 2e-05,
      "loss": 0.7179,
      "step": 1520
    },
    {
      "epoch": 0.09013254786450663,
      "grad_norm": 0.4313391149044037,
      "learning_rate": 2e-05,
      "loss": 0.74,
      "step": 1530
    },
    {
      "epoch": 0.09072164948453608,
      "grad_norm": 0.560111403465271,
      "learning_rate": 2e-05,
      "loss": 0.7364,
      "step": 1540
    },
    {
      "epoch": 0.09131075110456553,
      "grad_norm": 0.4297552704811096,
      "learning_rate": 2e-05,
      "loss": 0.7237,
      "step": 1550
    },
    {
      "epoch": 0.091899852724595,
      "grad_norm": 0.4254915416240692,
      "learning_rate": 2e-05,
      "loss": 0.7141,
      "step": 1560
    },
    {
      "epoch": 0.09248895434462445,
      "grad_norm": 0.5323898196220398,
      "learning_rate": 2e-05,
      "loss": 0.7166,
      "step": 1570
    },
    {
      "epoch": 0.0930780559646539,
      "grad_norm": 0.4295822083950043,
      "learning_rate": 2e-05,
      "loss": 0.7103,
      "step": 1580
    },
    {
      "epoch": 0.09366715758468336,
      "grad_norm": 0.45835748314857483,
      "learning_rate": 2e-05,
      "loss": 0.722,
      "step": 1590
    },
    {
      "epoch": 0.09425625920471281,
      "grad_norm": 0.4352903664112091,
      "learning_rate": 2e-05,
      "loss": 0.7396,
      "step": 1600
    },
    {
      "epoch": 0.09484536082474226,
      "grad_norm": 0.42194947600364685,
      "learning_rate": 2e-05,
      "loss": 0.7174,
      "step": 1610
    },
    {
      "epoch": 0.09543446244477172,
      "grad_norm": 0.41238200664520264,
      "learning_rate": 2e-05,
      "loss": 0.7161,
      "step": 1620
    },
    {
      "epoch": 0.09602356406480118,
      "grad_norm": 0.5528646111488342,
      "learning_rate": 2e-05,
      "loss": 0.7377,
      "step": 1630
    },
    {
      "epoch": 0.09661266568483064,
      "grad_norm": 0.4552900195121765,
      "learning_rate": 2e-05,
      "loss": 0.7126,
      "step": 1640
    },
    {
      "epoch": 0.09720176730486009,
      "grad_norm": 0.5200653672218323,
      "learning_rate": 2e-05,
      "loss": 0.7421,
      "step": 1650
    },
    {
      "epoch": 0.09779086892488954,
      "grad_norm": 0.4877001643180847,
      "learning_rate": 2e-05,
      "loss": 0.7637,
      "step": 1660
    },
    {
      "epoch": 0.098379970544919,
      "grad_norm": 0.4906533658504486,
      "learning_rate": 2e-05,
      "loss": 0.7365,
      "step": 1670
    },
    {
      "epoch": 0.09896907216494845,
      "grad_norm": 0.47109371423721313,
      "learning_rate": 2e-05,
      "loss": 0.7152,
      "step": 1680
    },
    {
      "epoch": 0.0995581737849779,
      "grad_norm": 0.4154045283794403,
      "learning_rate": 2e-05,
      "loss": 0.7114,
      "step": 1690
    },
    {
      "epoch": 0.10014727540500737,
      "grad_norm": 0.4646206498146057,
      "learning_rate": 2e-05,
      "loss": 0.7276,
      "step": 1700
    },
    {
      "epoch": 0.10073637702503682,
      "grad_norm": 0.46414250135421753,
      "learning_rate": 2e-05,
      "loss": 0.7419,
      "step": 1710
    },
    {
      "epoch": 0.10132547864506627,
      "grad_norm": 0.5728378295898438,
      "learning_rate": 2e-05,
      "loss": 0.7351,
      "step": 1720
    },
    {
      "epoch": 0.10191458026509573,
      "grad_norm": 0.49045124650001526,
      "learning_rate": 2e-05,
      "loss": 0.7357,
      "step": 1730
    },
    {
      "epoch": 0.10250368188512518,
      "grad_norm": 0.40880975127220154,
      "learning_rate": 2e-05,
      "loss": 0.7404,
      "step": 1740
    },
    {
      "epoch": 0.10309278350515463,
      "grad_norm": 0.41513314843177795,
      "learning_rate": 2e-05,
      "loss": 0.7142,
      "step": 1750
    },
    {
      "epoch": 0.1036818851251841,
      "grad_norm": 0.4373551607131958,
      "learning_rate": 2e-05,
      "loss": 0.733,
      "step": 1760
    },
    {
      "epoch": 0.10427098674521355,
      "grad_norm": 0.4935678541660309,
      "learning_rate": 2e-05,
      "loss": 0.723,
      "step": 1770
    },
    {
      "epoch": 0.104860088365243,
      "grad_norm": 0.5708635449409485,
      "learning_rate": 2e-05,
      "loss": 0.7434,
      "step": 1780
    },
    {
      "epoch": 0.10544918998527246,
      "grad_norm": 0.44515126943588257,
      "learning_rate": 2e-05,
      "loss": 0.7481,
      "step": 1790
    },
    {
      "epoch": 0.10603829160530191,
      "grad_norm": 0.44421878457069397,
      "learning_rate": 2e-05,
      "loss": 0.7142,
      "step": 1800
    },
    {
      "epoch": 0.10662739322533137,
      "grad_norm": 0.4646596610546112,
      "learning_rate": 2e-05,
      "loss": 0.7169,
      "step": 1810
    },
    {
      "epoch": 0.10721649484536082,
      "grad_norm": 0.4383988380432129,
      "learning_rate": 2e-05,
      "loss": 0.7304,
      "step": 1820
    },
    {
      "epoch": 0.10780559646539029,
      "grad_norm": 0.4752575755119324,
      "learning_rate": 2e-05,
      "loss": 0.729,
      "step": 1830
    },
    {
      "epoch": 0.10839469808541974,
      "grad_norm": 0.5204085111618042,
      "learning_rate": 2e-05,
      "loss": 0.7303,
      "step": 1840
    },
    {
      "epoch": 0.10898379970544919,
      "grad_norm": 0.40991365909576416,
      "learning_rate": 2e-05,
      "loss": 0.6899,
      "step": 1850
    },
    {
      "epoch": 0.10957290132547864,
      "grad_norm": 0.511846661567688,
      "learning_rate": 2e-05,
      "loss": 0.7096,
      "step": 1860
    },
    {
      "epoch": 0.1101620029455081,
      "grad_norm": 0.3975253701210022,
      "learning_rate": 2e-05,
      "loss": 0.7464,
      "step": 1870
    },
    {
      "epoch": 0.11075110456553755,
      "grad_norm": 0.4428749680519104,
      "learning_rate": 2e-05,
      "loss": 0.7214,
      "step": 1880
    },
    {
      "epoch": 0.11134020618556702,
      "grad_norm": 0.6127219796180725,
      "learning_rate": 2e-05,
      "loss": 0.7163,
      "step": 1890
    },
    {
      "epoch": 0.11192930780559647,
      "grad_norm": 0.7660998106002808,
      "learning_rate": 2e-05,
      "loss": 0.7328,
      "step": 1900
    },
    {
      "epoch": 0.11251840942562592,
      "grad_norm": 0.4295995235443115,
      "learning_rate": 2e-05,
      "loss": 0.7419,
      "step": 1910
    },
    {
      "epoch": 0.11310751104565538,
      "grad_norm": 0.5128087997436523,
      "learning_rate": 2e-05,
      "loss": 0.7466,
      "step": 1920
    },
    {
      "epoch": 0.11369661266568483,
      "grad_norm": 0.4301770329475403,
      "learning_rate": 2e-05,
      "loss": 0.7115,
      "step": 1930
    },
    {
      "epoch": 0.11428571428571428,
      "grad_norm": 0.5042638182640076,
      "learning_rate": 2e-05,
      "loss": 0.7535,
      "step": 1940
    },
    {
      "epoch": 0.11487481590574374,
      "grad_norm": 0.39604488015174866,
      "learning_rate": 2e-05,
      "loss": 0.7089,
      "step": 1950
    },
    {
      "epoch": 0.1154639175257732,
      "grad_norm": 0.4408802092075348,
      "learning_rate": 2e-05,
      "loss": 0.7264,
      "step": 1960
    },
    {
      "epoch": 0.11605301914580265,
      "grad_norm": 0.44488540291786194,
      "learning_rate": 2e-05,
      "loss": 0.7287,
      "step": 1970
    },
    {
      "epoch": 0.11664212076583211,
      "grad_norm": 0.4116921126842499,
      "learning_rate": 2e-05,
      "loss": 0.7409,
      "step": 1980
    },
    {
      "epoch": 0.11723122238586156,
      "grad_norm": 0.47106829285621643,
      "learning_rate": 2e-05,
      "loss": 0.7338,
      "step": 1990
    },
    {
      "epoch": 0.11782032400589101,
      "grad_norm": 0.41873085498809814,
      "learning_rate": 2e-05,
      "loss": 0.7142,
      "step": 2000
    },
    {
      "epoch": 0.11840942562592047,
      "grad_norm": 0.606903612613678,
      "learning_rate": 2e-05,
      "loss": 0.7108,
      "step": 2010
    },
    {
      "epoch": 0.11899852724594992,
      "grad_norm": 0.42234426736831665,
      "learning_rate": 2e-05,
      "loss": 0.7445,
      "step": 2020
    },
    {
      "epoch": 0.11958762886597939,
      "grad_norm": 0.5257266759872437,
      "learning_rate": 2e-05,
      "loss": 0.725,
      "step": 2030
    },
    {
      "epoch": 0.12017673048600884,
      "grad_norm": 0.6711904406547546,
      "learning_rate": 2e-05,
      "loss": 0.7123,
      "step": 2040
    },
    {
      "epoch": 0.12076583210603829,
      "grad_norm": 0.427338182926178,
      "learning_rate": 2e-05,
      "loss": 0.7496,
      "step": 2050
    },
    {
      "epoch": 0.12135493372606775,
      "grad_norm": 0.5107380151748657,
      "learning_rate": 2e-05,
      "loss": 0.7162,
      "step": 2060
    },
    {
      "epoch": 0.1219440353460972,
      "grad_norm": 0.4388013482093811,
      "learning_rate": 2e-05,
      "loss": 0.7119,
      "step": 2070
    },
    {
      "epoch": 0.12253313696612665,
      "grad_norm": 0.4046862721443176,
      "learning_rate": 2e-05,
      "loss": 0.7115,
      "step": 2080
    },
    {
      "epoch": 0.12312223858615612,
      "grad_norm": 0.44036686420440674,
      "learning_rate": 2e-05,
      "loss": 0.7157,
      "step": 2090
    },
    {
      "epoch": 0.12371134020618557,
      "grad_norm": 0.4342750608921051,
      "learning_rate": 2e-05,
      "loss": 0.7165,
      "step": 2100
    },
    {
      "epoch": 0.12430044182621502,
      "grad_norm": 0.4218961000442505,
      "learning_rate": 2e-05,
      "loss": 0.7259,
      "step": 2110
    },
    {
      "epoch": 0.12488954344624448,
      "grad_norm": 0.39011839032173157,
      "learning_rate": 2e-05,
      "loss": 0.7316,
      "step": 2120
    },
    {
      "epoch": 0.12547864506627393,
      "grad_norm": 0.886987030506134,
      "learning_rate": 2e-05,
      "loss": 0.7082,
      "step": 2130
    },
    {
      "epoch": 0.12606774668630338,
      "grad_norm": 0.40313050150871277,
      "learning_rate": 2e-05,
      "loss": 0.7246,
      "step": 2140
    },
    {
      "epoch": 0.12665684830633284,
      "grad_norm": 0.4915687143802643,
      "learning_rate": 2e-05,
      "loss": 0.7297,
      "step": 2150
    },
    {
      "epoch": 0.1272459499263623,
      "grad_norm": 0.4565123915672302,
      "learning_rate": 2e-05,
      "loss": 0.7224,
      "step": 2160
    },
    {
      "epoch": 0.12783505154639174,
      "grad_norm": 0.43001580238342285,
      "learning_rate": 2e-05,
      "loss": 0.7029,
      "step": 2170
    },
    {
      "epoch": 0.1284241531664212,
      "grad_norm": 0.5716996788978577,
      "learning_rate": 2e-05,
      "loss": 0.7458,
      "step": 2180
    },
    {
      "epoch": 0.12901325478645068,
      "grad_norm": 0.425799161195755,
      "learning_rate": 2e-05,
      "loss": 0.7415,
      "step": 2190
    },
    {
      "epoch": 0.12960235640648013,
      "grad_norm": 0.40971073508262634,
      "learning_rate": 2e-05,
      "loss": 0.7146,
      "step": 2200
    },
    {
      "epoch": 0.13019145802650958,
      "grad_norm": 0.42143917083740234,
      "learning_rate": 2e-05,
      "loss": 0.7161,
      "step": 2210
    },
    {
      "epoch": 0.13078055964653904,
      "grad_norm": 0.41456344723701477,
      "learning_rate": 2e-05,
      "loss": 0.7242,
      "step": 2220
    },
    {
      "epoch": 0.1313696612665685,
      "grad_norm": 0.430918425321579,
      "learning_rate": 2e-05,
      "loss": 0.7073,
      "step": 2230
    },
    {
      "epoch": 0.13195876288659794,
      "grad_norm": 0.6099043488502502,
      "learning_rate": 2e-05,
      "loss": 0.7474,
      "step": 2240
    },
    {
      "epoch": 0.1325478645066274,
      "grad_norm": 0.43032971024513245,
      "learning_rate": 2e-05,
      "loss": 0.7333,
      "step": 2250
    },
    {
      "epoch": 0.13313696612665685,
      "grad_norm": 0.4625254273414612,
      "learning_rate": 2e-05,
      "loss": 0.6958,
      "step": 2260
    },
    {
      "epoch": 0.1337260677466863,
      "grad_norm": 0.3948584198951721,
      "learning_rate": 2e-05,
      "loss": 0.7081,
      "step": 2270
    },
    {
      "epoch": 0.13431516936671575,
      "grad_norm": 0.4327255189418793,
      "learning_rate": 2e-05,
      "loss": 0.7492,
      "step": 2280
    },
    {
      "epoch": 0.1349042709867452,
      "grad_norm": 0.4343191683292389,
      "learning_rate": 2e-05,
      "loss": 0.7195,
      "step": 2290
    },
    {
      "epoch": 0.13549337260677466,
      "grad_norm": 0.4207633435726166,
      "learning_rate": 2e-05,
      "loss": 0.7205,
      "step": 2300
    },
    {
      "epoch": 0.1360824742268041,
      "grad_norm": 0.44352003931999207,
      "learning_rate": 2e-05,
      "loss": 0.7075,
      "step": 2310
    },
    {
      "epoch": 0.13667157584683357,
      "grad_norm": 0.3955673277378082,
      "learning_rate": 2e-05,
      "loss": 0.7309,
      "step": 2320
    },
    {
      "epoch": 0.13726067746686305,
      "grad_norm": 0.38486120104789734,
      "learning_rate": 2e-05,
      "loss": 0.7224,
      "step": 2330
    },
    {
      "epoch": 0.1378497790868925,
      "grad_norm": 0.3926263749599457,
      "learning_rate": 2e-05,
      "loss": 0.7101,
      "step": 2340
    },
    {
      "epoch": 0.13843888070692195,
      "grad_norm": 0.3783673644065857,
      "learning_rate": 2e-05,
      "loss": 0.6916,
      "step": 2350
    },
    {
      "epoch": 0.1390279823269514,
      "grad_norm": 0.40818560123443604,
      "learning_rate": 2e-05,
      "loss": 0.6931,
      "step": 2360
    },
    {
      "epoch": 0.13961708394698086,
      "grad_norm": 0.3975651264190674,
      "learning_rate": 2e-05,
      "loss": 0.7361,
      "step": 2370
    },
    {
      "epoch": 0.1402061855670103,
      "grad_norm": 0.5217942595481873,
      "learning_rate": 2e-05,
      "loss": 0.7366,
      "step": 2380
    },
    {
      "epoch": 0.14079528718703976,
      "grad_norm": 0.4319808781147003,
      "learning_rate": 2e-05,
      "loss": 0.7147,
      "step": 2390
    },
    {
      "epoch": 0.14138438880706922,
      "grad_norm": 0.3996039628982544,
      "learning_rate": 2e-05,
      "loss": 0.7347,
      "step": 2400
    },
    {
      "epoch": 0.14197349042709867,
      "grad_norm": 0.6555231213569641,
      "learning_rate": 2e-05,
      "loss": 0.748,
      "step": 2410
    },
    {
      "epoch": 0.14256259204712812,
      "grad_norm": 0.45214080810546875,
      "learning_rate": 2e-05,
      "loss": 0.6833,
      "step": 2420
    },
    {
      "epoch": 0.14315169366715758,
      "grad_norm": 0.6138016581535339,
      "learning_rate": 2e-05,
      "loss": 0.7095,
      "step": 2430
    },
    {
      "epoch": 0.14374079528718703,
      "grad_norm": 0.3984246253967285,
      "learning_rate": 2e-05,
      "loss": 0.7016,
      "step": 2440
    },
    {
      "epoch": 0.14432989690721648,
      "grad_norm": 0.4183751940727234,
      "learning_rate": 2e-05,
      "loss": 0.7206,
      "step": 2450
    },
    {
      "epoch": 0.14491899852724596,
      "grad_norm": 0.662306547164917,
      "learning_rate": 2e-05,
      "loss": 0.718,
      "step": 2460
    },
    {
      "epoch": 0.14550810014727542,
      "grad_norm": 0.3855864107608795,
      "learning_rate": 2e-05,
      "loss": 0.7275,
      "step": 2470
    },
    {
      "epoch": 0.14609720176730487,
      "grad_norm": 0.43764403462409973,
      "learning_rate": 2e-05,
      "loss": 0.7114,
      "step": 2480
    },
    {
      "epoch": 0.14668630338733432,
      "grad_norm": 0.41515758633613586,
      "learning_rate": 2e-05,
      "loss": 0.7471,
      "step": 2490
    },
    {
      "epoch": 0.14727540500736377,
      "grad_norm": 0.4112182855606079,
      "learning_rate": 2e-05,
      "loss": 0.6898,
      "step": 2500
    },
    {
      "epoch": 0.14786450662739323,
      "grad_norm": 0.4436074495315552,
      "learning_rate": 2e-05,
      "loss": 0.7029,
      "step": 2510
    },
    {
      "epoch": 0.14845360824742268,
      "grad_norm": 0.40902867913246155,
      "learning_rate": 2e-05,
      "loss": 0.7037,
      "step": 2520
    },
    {
      "epoch": 0.14904270986745213,
      "grad_norm": 0.4918319880962372,
      "learning_rate": 2e-05,
      "loss": 0.7065,
      "step": 2530
    },
    {
      "epoch": 0.1496318114874816,
      "grad_norm": 0.40850844979286194,
      "learning_rate": 2e-05,
      "loss": 0.7093,
      "step": 2540
    },
    {
      "epoch": 0.15022091310751104,
      "grad_norm": 0.44000327587127686,
      "learning_rate": 2e-05,
      "loss": 0.7346,
      "step": 2550
    },
    {
      "epoch": 0.1508100147275405,
      "grad_norm": 0.5778651833534241,
      "learning_rate": 2e-05,
      "loss": 0.7455,
      "step": 2560
    },
    {
      "epoch": 0.15139911634756995,
      "grad_norm": 0.4158194661140442,
      "learning_rate": 2e-05,
      "loss": 0.7203,
      "step": 2570
    },
    {
      "epoch": 0.1519882179675994,
      "grad_norm": 0.42827850580215454,
      "learning_rate": 2e-05,
      "loss": 0.7078,
      "step": 2580
    },
    {
      "epoch": 0.15257731958762888,
      "grad_norm": 0.4320778250694275,
      "learning_rate": 2e-05,
      "loss": 0.7256,
      "step": 2590
    },
    {
      "epoch": 0.15316642120765833,
      "grad_norm": 0.6728928089141846,
      "learning_rate": 2e-05,
      "loss": 0.7021,
      "step": 2600
    },
    {
      "epoch": 0.15375552282768779,
      "grad_norm": 0.40638744831085205,
      "learning_rate": 2e-05,
      "loss": 0.7156,
      "step": 2610
    },
    {
      "epoch": 0.15434462444771724,
      "grad_norm": 0.44968730211257935,
      "learning_rate": 2e-05,
      "loss": 0.7275,
      "step": 2620
    },
    {
      "epoch": 0.1549337260677467,
      "grad_norm": 0.44856545329093933,
      "learning_rate": 2e-05,
      "loss": 0.6964,
      "step": 2630
    },
    {
      "epoch": 0.15552282768777614,
      "grad_norm": 0.41701412200927734,
      "learning_rate": 2e-05,
      "loss": 0.7055,
      "step": 2640
    },
    {
      "epoch": 0.1561119293078056,
      "grad_norm": 0.4159757196903229,
      "learning_rate": 2e-05,
      "loss": 0.7117,
      "step": 2650
    },
    {
      "epoch": 0.15670103092783505,
      "grad_norm": 0.40794211626052856,
      "learning_rate": 2e-05,
      "loss": 0.7106,
      "step": 2660
    },
    {
      "epoch": 0.1572901325478645,
      "grad_norm": 0.4129374027252197,
      "learning_rate": 2e-05,
      "loss": 0.7171,
      "step": 2670
    },
    {
      "epoch": 0.15787923416789396,
      "grad_norm": 0.42779678106307983,
      "learning_rate": 2e-05,
      "loss": 0.7304,
      "step": 2680
    },
    {
      "epoch": 0.1584683357879234,
      "grad_norm": 0.4096926748752594,
      "learning_rate": 2e-05,
      "loss": 0.7238,
      "step": 2690
    },
    {
      "epoch": 0.15905743740795286,
      "grad_norm": 0.4360990524291992,
      "learning_rate": 2e-05,
      "loss": 0.7102,
      "step": 2700
    },
    {
      "epoch": 0.15964653902798231,
      "grad_norm": 0.41352882981300354,
      "learning_rate": 2e-05,
      "loss": 0.7252,
      "step": 2710
    },
    {
      "epoch": 0.1602356406480118,
      "grad_norm": 0.4187283217906952,
      "learning_rate": 2e-05,
      "loss": 0.7211,
      "step": 2720
    },
    {
      "epoch": 0.16082474226804125,
      "grad_norm": 0.3870251774787903,
      "learning_rate": 2e-05,
      "loss": 0.716,
      "step": 2730
    },
    {
      "epoch": 0.1614138438880707,
      "grad_norm": 0.4262825846672058,
      "learning_rate": 2e-05,
      "loss": 0.7122,
      "step": 2740
    },
    {
      "epoch": 0.16200294550810015,
      "grad_norm": 0.4767245352268219,
      "learning_rate": 2e-05,
      "loss": 0.699,
      "step": 2750
    },
    {
      "epoch": 0.1625920471281296,
      "grad_norm": 0.4373011589050293,
      "learning_rate": 2e-05,
      "loss": 0.6929,
      "step": 2760
    },
    {
      "epoch": 0.16318114874815906,
      "grad_norm": 0.38110506534576416,
      "learning_rate": 2e-05,
      "loss": 0.7317,
      "step": 2770
    },
    {
      "epoch": 0.1637702503681885,
      "grad_norm": 0.6112042665481567,
      "learning_rate": 2e-05,
      "loss": 0.6903,
      "step": 2780
    },
    {
      "epoch": 0.16435935198821797,
      "grad_norm": 0.4287341833114624,
      "learning_rate": 2e-05,
      "loss": 0.7106,
      "step": 2790
    },
    {
      "epoch": 0.16494845360824742,
      "grad_norm": 0.4338441491127014,
      "learning_rate": 2e-05,
      "loss": 0.7339,
      "step": 2800
    },
    {
      "epoch": 0.16553755522827687,
      "grad_norm": 0.4295300543308258,
      "learning_rate": 2e-05,
      "loss": 0.6971,
      "step": 2810
    },
    {
      "epoch": 0.16612665684830633,
      "grad_norm": 0.4455210268497467,
      "learning_rate": 2e-05,
      "loss": 0.7145,
      "step": 2820
    },
    {
      "epoch": 0.16671575846833578,
      "grad_norm": 0.6973451375961304,
      "learning_rate": 2e-05,
      "loss": 0.7059,
      "step": 2830
    },
    {
      "epoch": 0.16730486008836523,
      "grad_norm": 0.5928052663803101,
      "learning_rate": 2e-05,
      "loss": 0.7147,
      "step": 2840
    },
    {
      "epoch": 0.16789396170839468,
      "grad_norm": 0.44190120697021484,
      "learning_rate": 2e-05,
      "loss": 0.7111,
      "step": 2850
    },
    {
      "epoch": 0.16848306332842417,
      "grad_norm": 0.44275492429733276,
      "learning_rate": 2e-05,
      "loss": 0.7054,
      "step": 2860
    },
    {
      "epoch": 0.16907216494845362,
      "grad_norm": 0.410471111536026,
      "learning_rate": 2e-05,
      "loss": 0.7122,
      "step": 2870
    },
    {
      "epoch": 0.16966126656848307,
      "grad_norm": 0.43369925022125244,
      "learning_rate": 2e-05,
      "loss": 0.7019,
      "step": 2880
    },
    {
      "epoch": 0.17025036818851252,
      "grad_norm": 0.41688773036003113,
      "learning_rate": 2e-05,
      "loss": 0.7086,
      "step": 2890
    },
    {
      "epoch": 0.17083946980854198,
      "grad_norm": 0.43613725900650024,
      "learning_rate": 2e-05,
      "loss": 0.7068,
      "step": 2900
    },
    {
      "epoch": 0.17142857142857143,
      "grad_norm": 0.44705459475517273,
      "learning_rate": 2e-05,
      "loss": 0.6917,
      "step": 2910
    },
    {
      "epoch": 0.17201767304860088,
      "grad_norm": 0.43106403946876526,
      "learning_rate": 2e-05,
      "loss": 0.7187,
      "step": 2920
    },
    {
      "epoch": 0.17260677466863034,
      "grad_norm": 0.4107123613357544,
      "learning_rate": 2e-05,
      "loss": 0.7176,
      "step": 2930
    },
    {
      "epoch": 0.1731958762886598,
      "grad_norm": 0.470010370016098,
      "learning_rate": 2e-05,
      "loss": 0.7146,
      "step": 2940
    },
    {
      "epoch": 0.17378497790868924,
      "grad_norm": 0.371026873588562,
      "learning_rate": 2e-05,
      "loss": 0.7068,
      "step": 2950
    },
    {
      "epoch": 0.1743740795287187,
      "grad_norm": 0.4153003394603729,
      "learning_rate": 2e-05,
      "loss": 0.7249,
      "step": 2960
    },
    {
      "epoch": 0.17496318114874815,
      "grad_norm": 0.4121924638748169,
      "learning_rate": 2e-05,
      "loss": 0.697,
      "step": 2970
    },
    {
      "epoch": 0.1755522827687776,
      "grad_norm": 0.41051289439201355,
      "learning_rate": 2e-05,
      "loss": 0.7178,
      "step": 2980
    },
    {
      "epoch": 0.17614138438880708,
      "grad_norm": 0.4057338535785675,
      "learning_rate": 2e-05,
      "loss": 0.7245,
      "step": 2990
    },
    {
      "epoch": 0.17673048600883653,
      "grad_norm": 0.39876967668533325,
      "learning_rate": 2e-05,
      "loss": 0.7038,
      "step": 3000
    },
    {
      "epoch": 0.177319587628866,
      "grad_norm": 0.46926066279411316,
      "learning_rate": 2e-05,
      "loss": 0.7105,
      "step": 3010
    },
    {
      "epoch": 0.17790868924889544,
      "grad_norm": 0.41574588418006897,
      "learning_rate": 2e-05,
      "loss": 0.7222,
      "step": 3020
    },
    {
      "epoch": 0.1784977908689249,
      "grad_norm": 0.4020771086215973,
      "learning_rate": 2e-05,
      "loss": 0.7279,
      "step": 3030
    },
    {
      "epoch": 0.17908689248895435,
      "grad_norm": 0.4329489767551422,
      "learning_rate": 2e-05,
      "loss": 0.7374,
      "step": 3040
    },
    {
      "epoch": 0.1796759941089838,
      "grad_norm": 0.44369325041770935,
      "learning_rate": 2e-05,
      "loss": 0.7038,
      "step": 3050
    },
    {
      "epoch": 0.18026509572901325,
      "grad_norm": 0.5440769195556641,
      "learning_rate": 2e-05,
      "loss": 0.7189,
      "step": 3060
    },
    {
      "epoch": 0.1808541973490427,
      "grad_norm": 0.4297233819961548,
      "learning_rate": 2e-05,
      "loss": 0.7355,
      "step": 3070
    },
    {
      "epoch": 0.18144329896907216,
      "grad_norm": 0.39824557304382324,
      "learning_rate": 2e-05,
      "loss": 0.706,
      "step": 3080
    },
    {
      "epoch": 0.1820324005891016,
      "grad_norm": 0.40815210342407227,
      "learning_rate": 2e-05,
      "loss": 0.7021,
      "step": 3090
    },
    {
      "epoch": 0.18262150220913106,
      "grad_norm": 0.5076918005943298,
      "learning_rate": 2e-05,
      "loss": 0.7269,
      "step": 3100
    },
    {
      "epoch": 0.18321060382916052,
      "grad_norm": 0.46184083819389343,
      "learning_rate": 2e-05,
      "loss": 0.6933,
      "step": 3110
    },
    {
      "epoch": 0.18379970544919,
      "grad_norm": 0.913251519203186,
      "learning_rate": 2e-05,
      "loss": 0.693,
      "step": 3120
    },
    {
      "epoch": 0.18438880706921945,
      "grad_norm": 0.4487643241882324,
      "learning_rate": 2e-05,
      "loss": 0.7147,
      "step": 3130
    },
    {
      "epoch": 0.1849779086892489,
      "grad_norm": 0.4097018539905548,
      "learning_rate": 2e-05,
      "loss": 0.6866,
      "step": 3140
    },
    {
      "epoch": 0.18556701030927836,
      "grad_norm": 0.42946818470954895,
      "learning_rate": 2e-05,
      "loss": 0.7057,
      "step": 3150
    },
    {
      "epoch": 0.1861561119293078,
      "grad_norm": 0.4244742691516876,
      "learning_rate": 2e-05,
      "loss": 0.7089,
      "step": 3160
    },
    {
      "epoch": 0.18674521354933726,
      "grad_norm": 0.4119102656841278,
      "learning_rate": 2e-05,
      "loss": 0.711,
      "step": 3170
    },
    {
      "epoch": 0.18733431516936672,
      "grad_norm": 0.4227426052093506,
      "learning_rate": 2e-05,
      "loss": 0.702,
      "step": 3180
    },
    {
      "epoch": 0.18792341678939617,
      "grad_norm": 0.44902554154396057,
      "learning_rate": 2e-05,
      "loss": 0.6886,
      "step": 3190
    },
    {
      "epoch": 0.18851251840942562,
      "grad_norm": 0.3865416646003723,
      "learning_rate": 2e-05,
      "loss": 0.7267,
      "step": 3200
    },
    {
      "epoch": 0.18910162002945508,
      "grad_norm": 0.49310725927352905,
      "learning_rate": 2e-05,
      "loss": 0.6939,
      "step": 3210
    },
    {
      "epoch": 0.18969072164948453,
      "grad_norm": 0.429551362991333,
      "learning_rate": 2e-05,
      "loss": 0.7103,
      "step": 3220
    },
    {
      "epoch": 0.19027982326951398,
      "grad_norm": 0.5094394683837891,
      "learning_rate": 2e-05,
      "loss": 0.7319,
      "step": 3230
    },
    {
      "epoch": 0.19086892488954343,
      "grad_norm": 0.43107378482818604,
      "learning_rate": 2e-05,
      "loss": 0.7044,
      "step": 3240
    },
    {
      "epoch": 0.19145802650957292,
      "grad_norm": 0.921887993812561,
      "learning_rate": 2e-05,
      "loss": 0.7616,
      "step": 3250
    },
    {
      "epoch": 0.19204712812960237,
      "grad_norm": 0.42933928966522217,
      "learning_rate": 2e-05,
      "loss": 0.6983,
      "step": 3260
    },
    {
      "epoch": 0.19263622974963182,
      "grad_norm": 0.3960442841053009,
      "learning_rate": 2e-05,
      "loss": 0.702,
      "step": 3270
    },
    {
      "epoch": 0.19322533136966127,
      "grad_norm": 0.4460879862308502,
      "learning_rate": 2e-05,
      "loss": 0.6944,
      "step": 3280
    },
    {
      "epoch": 0.19381443298969073,
      "grad_norm": 0.4470248222351074,
      "learning_rate": 2e-05,
      "loss": 0.7186,
      "step": 3290
    },
    {
      "epoch": 0.19440353460972018,
      "grad_norm": 0.44315096735954285,
      "learning_rate": 2e-05,
      "loss": 0.7667,
      "step": 3300
    },
    {
      "epoch": 0.19499263622974963,
      "grad_norm": 0.4374779760837555,
      "learning_rate": 2e-05,
      "loss": 0.7168,
      "step": 3310
    },
    {
      "epoch": 0.19558173784977909,
      "grad_norm": 0.4289309084415436,
      "learning_rate": 2e-05,
      "loss": 0.7155,
      "step": 3320
    },
    {
      "epoch": 0.19617083946980854,
      "grad_norm": 0.39460673928260803,
      "learning_rate": 2e-05,
      "loss": 0.7227,
      "step": 3330
    },
    {
      "epoch": 0.196759941089838,
      "grad_norm": 0.45456817746162415,
      "learning_rate": 2e-05,
      "loss": 0.7159,
      "step": 3340
    },
    {
      "epoch": 0.19734904270986744,
      "grad_norm": 0.4173303246498108,
      "learning_rate": 2e-05,
      "loss": 0.7252,
      "step": 3350
    },
    {
      "epoch": 0.1979381443298969,
      "grad_norm": 0.42761164903640747,
      "learning_rate": 2e-05,
      "loss": 0.7245,
      "step": 3360
    },
    {
      "epoch": 0.19852724594992635,
      "grad_norm": 0.3923100531101227,
      "learning_rate": 2e-05,
      "loss": 0.6949,
      "step": 3370
    },
    {
      "epoch": 0.1991163475699558,
      "grad_norm": 0.4013349115848541,
      "learning_rate": 2e-05,
      "loss": 0.7166,
      "step": 3380
    },
    {
      "epoch": 0.19970544918998528,
      "grad_norm": 0.40177324414253235,
      "learning_rate": 2e-05,
      "loss": 0.6968,
      "step": 3390
    },
    {
      "epoch": 0.20029455081001474,
      "grad_norm": 0.40452343225479126,
      "learning_rate": 2e-05,
      "loss": 0.7275,
      "step": 3400
    },
    {
      "epoch": 0.2008836524300442,
      "grad_norm": 0.4861152470111847,
      "learning_rate": 2e-05,
      "loss": 0.7023,
      "step": 3410
    },
    {
      "epoch": 0.20147275405007364,
      "grad_norm": 0.4972041845321655,
      "learning_rate": 2e-05,
      "loss": 0.6883,
      "step": 3420
    },
    {
      "epoch": 0.2020618556701031,
      "grad_norm": 0.4638412594795227,
      "learning_rate": 2e-05,
      "loss": 0.687,
      "step": 3430
    },
    {
      "epoch": 0.20265095729013255,
      "grad_norm": 0.4460333287715912,
      "learning_rate": 2e-05,
      "loss": 0.7041,
      "step": 3440
    },
    {
      "epoch": 0.203240058910162,
      "grad_norm": 0.42660534381866455,
      "learning_rate": 2e-05,
      "loss": 0.7156,
      "step": 3450
    },
    {
      "epoch": 0.20382916053019146,
      "grad_norm": 0.44786620140075684,
      "learning_rate": 2e-05,
      "loss": 0.6868,
      "step": 3460
    },
    {
      "epoch": 0.2044182621502209,
      "grad_norm": 0.4005437195301056,
      "learning_rate": 2e-05,
      "loss": 0.7026,
      "step": 3470
    },
    {
      "epoch": 0.20500736377025036,
      "grad_norm": 0.4299420118331909,
      "learning_rate": 2e-05,
      "loss": 0.7197,
      "step": 3480
    },
    {
      "epoch": 0.20559646539027981,
      "grad_norm": 0.44099679589271545,
      "learning_rate": 2e-05,
      "loss": 0.696,
      "step": 3490
    },
    {
      "epoch": 0.20618556701030927,
      "grad_norm": 0.6412930488586426,
      "learning_rate": 2e-05,
      "loss": 0.714,
      "step": 3500
    },
    {
      "epoch": 0.20677466863033872,
      "grad_norm": 0.3928799033164978,
      "learning_rate": 2e-05,
      "loss": 0.6904,
      "step": 3510
    },
    {
      "epoch": 0.2073637702503682,
      "grad_norm": 0.4050648808479309,
      "learning_rate": 2e-05,
      "loss": 0.689,
      "step": 3520
    },
    {
      "epoch": 0.20795287187039765,
      "grad_norm": 0.3692716360092163,
      "learning_rate": 2e-05,
      "loss": 0.6728,
      "step": 3530
    },
    {
      "epoch": 0.2085419734904271,
      "grad_norm": 0.4676406681537628,
      "learning_rate": 2e-05,
      "loss": 0.6776,
      "step": 3540
    },
    {
      "epoch": 0.20913107511045656,
      "grad_norm": 0.4601678252220154,
      "learning_rate": 2e-05,
      "loss": 0.6964,
      "step": 3550
    },
    {
      "epoch": 0.209720176730486,
      "grad_norm": 0.43913260102272034,
      "learning_rate": 2e-05,
      "loss": 0.6962,
      "step": 3560
    },
    {
      "epoch": 0.21030927835051547,
      "grad_norm": 0.40677475929260254,
      "learning_rate": 2e-05,
      "loss": 0.6906,
      "step": 3570
    },
    {
      "epoch": 0.21089837997054492,
      "grad_norm": 0.41579195857048035,
      "learning_rate": 2e-05,
      "loss": 0.6928,
      "step": 3580
    },
    {
      "epoch": 0.21148748159057437,
      "grad_norm": 0.39886176586151123,
      "learning_rate": 2e-05,
      "loss": 0.7066,
      "step": 3590
    },
    {
      "epoch": 0.21207658321060383,
      "grad_norm": 0.4432888627052307,
      "learning_rate": 2e-05,
      "loss": 0.6835,
      "step": 3600
    },
    {
      "epoch": 0.21266568483063328,
      "grad_norm": 0.4384315311908722,
      "learning_rate": 2e-05,
      "loss": 0.7247,
      "step": 3610
    },
    {
      "epoch": 0.21325478645066273,
      "grad_norm": 0.43290889263153076,
      "learning_rate": 2e-05,
      "loss": 0.7232,
      "step": 3620
    },
    {
      "epoch": 0.21384388807069218,
      "grad_norm": 0.4648188054561615,
      "learning_rate": 2e-05,
      "loss": 0.7113,
      "step": 3630
    },
    {
      "epoch": 0.21443298969072164,
      "grad_norm": 0.4099072813987732,
      "learning_rate": 2e-05,
      "loss": 0.7077,
      "step": 3640
    },
    {
      "epoch": 0.21502209131075112,
      "grad_norm": 0.4543364942073822,
      "learning_rate": 2e-05,
      "loss": 0.6944,
      "step": 3650
    },
    {
      "epoch": 0.21561119293078057,
      "grad_norm": 0.4100470542907715,
      "learning_rate": 2e-05,
      "loss": 0.7212,
      "step": 3660
    },
    {
      "epoch": 0.21620029455081002,
      "grad_norm": 0.5675593018531799,
      "learning_rate": 2e-05,
      "loss": 0.7104,
      "step": 3670
    },
    {
      "epoch": 0.21678939617083948,
      "grad_norm": 0.3948013186454773,
      "learning_rate": 2e-05,
      "loss": 0.7127,
      "step": 3680
    },
    {
      "epoch": 0.21737849779086893,
      "grad_norm": 0.4194417893886566,
      "learning_rate": 2e-05,
      "loss": 0.7087,
      "step": 3690
    },
    {
      "epoch": 0.21796759941089838,
      "grad_norm": 0.3860073387622833,
      "learning_rate": 2e-05,
      "loss": 0.7324,
      "step": 3700
    },
    {
      "epoch": 0.21855670103092784,
      "grad_norm": 0.42284122109413147,
      "learning_rate": 2e-05,
      "loss": 0.7062,
      "step": 3710
    },
    {
      "epoch": 0.2191458026509573,
      "grad_norm": 0.4071473181247711,
      "learning_rate": 2e-05,
      "loss": 0.7121,
      "step": 3720
    },
    {
      "epoch": 0.21973490427098674,
      "grad_norm": 0.40895453095436096,
      "learning_rate": 2e-05,
      "loss": 0.6595,
      "step": 3730
    },
    {
      "epoch": 0.2203240058910162,
      "grad_norm": 0.3910290002822876,
      "learning_rate": 2e-05,
      "loss": 0.7203,
      "step": 3740
    },
    {
      "epoch": 0.22091310751104565,
      "grad_norm": 0.4667680263519287,
      "learning_rate": 2e-05,
      "loss": 0.6912,
      "step": 3750
    },
    {
      "epoch": 0.2215022091310751,
      "grad_norm": 0.4072132110595703,
      "learning_rate": 2e-05,
      "loss": 0.6964,
      "step": 3760
    },
    {
      "epoch": 0.22209131075110455,
      "grad_norm": 0.45165884494781494,
      "learning_rate": 2e-05,
      "loss": 0.7058,
      "step": 3770
    },
    {
      "epoch": 0.22268041237113403,
      "grad_norm": 0.46648043394088745,
      "learning_rate": 2e-05,
      "loss": 0.7233,
      "step": 3780
    },
    {
      "epoch": 0.2232695139911635,
      "grad_norm": 0.42833346128463745,
      "learning_rate": 2e-05,
      "loss": 0.6908,
      "step": 3790
    },
    {
      "epoch": 0.22385861561119294,
      "grad_norm": 0.4137326776981354,
      "learning_rate": 2e-05,
      "loss": 0.7056,
      "step": 3800
    },
    {
      "epoch": 0.2244477172312224,
      "grad_norm": 0.40581566095352173,
      "learning_rate": 2e-05,
      "loss": 0.7188,
      "step": 3810
    },
    {
      "epoch": 0.22503681885125185,
      "grad_norm": 0.4089506268501282,
      "learning_rate": 2e-05,
      "loss": 0.7048,
      "step": 3820
    },
    {
      "epoch": 0.2256259204712813,
      "grad_norm": 0.4361477494239807,
      "learning_rate": 2e-05,
      "loss": 0.6911,
      "step": 3830
    },
    {
      "epoch": 0.22621502209131075,
      "grad_norm": 0.4451761841773987,
      "learning_rate": 2e-05,
      "loss": 0.711,
      "step": 3840
    },
    {
      "epoch": 0.2268041237113402,
      "grad_norm": 0.4079304039478302,
      "learning_rate": 2e-05,
      "loss": 0.7065,
      "step": 3850
    },
    {
      "epoch": 0.22739322533136966,
      "grad_norm": 0.39868396520614624,
      "learning_rate": 2e-05,
      "loss": 0.7068,
      "step": 3860
    },
    {
      "epoch": 0.2279823269513991,
      "grad_norm": 0.4223789572715759,
      "learning_rate": 2e-05,
      "loss": 0.6991,
      "step": 3870
    },
    {
      "epoch": 0.22857142857142856,
      "grad_norm": 0.44960352778434753,
      "learning_rate": 2e-05,
      "loss": 0.6957,
      "step": 3880
    },
    {
      "epoch": 0.22916053019145802,
      "grad_norm": 0.39986199140548706,
      "learning_rate": 2e-05,
      "loss": 0.697,
      "step": 3890
    },
    {
      "epoch": 0.22974963181148747,
      "grad_norm": 0.4080177843570709,
      "learning_rate": 2e-05,
      "loss": 0.7145,
      "step": 3900
    },
    {
      "epoch": 0.23033873343151692,
      "grad_norm": 0.3692325949668884,
      "learning_rate": 2e-05,
      "loss": 0.6936,
      "step": 3910
    },
    {
      "epoch": 0.2309278350515464,
      "grad_norm": 0.41063618659973145,
      "learning_rate": 2e-05,
      "loss": 0.7046,
      "step": 3920
    },
    {
      "epoch": 0.23151693667157586,
      "grad_norm": 0.4455985426902771,
      "learning_rate": 2e-05,
      "loss": 0.721,
      "step": 3930
    },
    {
      "epoch": 0.2321060382916053,
      "grad_norm": 0.42421069741249084,
      "learning_rate": 2e-05,
      "loss": 0.7094,
      "step": 3940
    },
    {
      "epoch": 0.23269513991163476,
      "grad_norm": 0.9296834468841553,
      "learning_rate": 2e-05,
      "loss": 0.7083,
      "step": 3950
    },
    {
      "epoch": 0.23328424153166422,
      "grad_norm": 0.4903710186481476,
      "learning_rate": 2e-05,
      "loss": 0.7304,
      "step": 3960
    },
    {
      "epoch": 0.23387334315169367,
      "grad_norm": 0.4305209517478943,
      "learning_rate": 2e-05,
      "loss": 0.6759,
      "step": 3970
    },
    {
      "epoch": 0.23446244477172312,
      "grad_norm": 0.4206168055534363,
      "learning_rate": 2e-05,
      "loss": 0.6879,
      "step": 3980
    },
    {
      "epoch": 0.23505154639175257,
      "grad_norm": 0.4069591760635376,
      "learning_rate": 2e-05,
      "loss": 0.7012,
      "step": 3990
    },
    {
      "epoch": 0.23564064801178203,
      "grad_norm": 0.6243274807929993,
      "learning_rate": 2e-05,
      "loss": 0.7281,
      "step": 4000
    },
    {
      "epoch": 0.23622974963181148,
      "grad_norm": 0.46947747468948364,
      "learning_rate": 2e-05,
      "loss": 0.709,
      "step": 4010
    },
    {
      "epoch": 0.23681885125184093,
      "grad_norm": 0.4082581400871277,
      "learning_rate": 2e-05,
      "loss": 0.7113,
      "step": 4020
    },
    {
      "epoch": 0.2374079528718704,
      "grad_norm": 0.48157668113708496,
      "learning_rate": 2e-05,
      "loss": 0.7188,
      "step": 4030
    },
    {
      "epoch": 0.23799705449189984,
      "grad_norm": 0.4539220929145813,
      "learning_rate": 2e-05,
      "loss": 0.6956,
      "step": 4040
    },
    {
      "epoch": 0.23858615611192932,
      "grad_norm": 0.39384758472442627,
      "learning_rate": 2e-05,
      "loss": 0.6888,
      "step": 4050
    },
    {
      "epoch": 0.23917525773195877,
      "grad_norm": 0.4415516257286072,
      "learning_rate": 2e-05,
      "loss": 0.685,
      "step": 4060
    },
    {
      "epoch": 0.23976435935198823,
      "grad_norm": 0.37274792790412903,
      "learning_rate": 2e-05,
      "loss": 0.7079,
      "step": 4070
    },
    {
      "epoch": 0.24035346097201768,
      "grad_norm": 0.42507970333099365,
      "learning_rate": 2e-05,
      "loss": 0.7007,
      "step": 4080
    },
    {
      "epoch": 0.24094256259204713,
      "grad_norm": 0.39337408542633057,
      "learning_rate": 2e-05,
      "loss": 0.6769,
      "step": 4090
    },
    {
      "epoch": 0.24153166421207659,
      "grad_norm": 0.41252169013023376,
      "learning_rate": 2e-05,
      "loss": 0.6966,
      "step": 4100
    },
    {
      "epoch": 0.24212076583210604,
      "grad_norm": 0.46128690242767334,
      "learning_rate": 2e-05,
      "loss": 0.6906,
      "step": 4110
    },
    {
      "epoch": 0.2427098674521355,
      "grad_norm": 0.4340006411075592,
      "learning_rate": 2e-05,
      "loss": 0.695,
      "step": 4120
    },
    {
      "epoch": 0.24329896907216494,
      "grad_norm": 0.41755911707878113,
      "learning_rate": 2e-05,
      "loss": 0.6973,
      "step": 4130
    },
    {
      "epoch": 0.2438880706921944,
      "grad_norm": 0.4026070237159729,
      "learning_rate": 2e-05,
      "loss": 0.6879,
      "step": 4140
    },
    {
      "epoch": 0.24447717231222385,
      "grad_norm": 0.41889482736587524,
      "learning_rate": 2e-05,
      "loss": 0.7024,
      "step": 4150
    },
    {
      "epoch": 0.2450662739322533,
      "grad_norm": 0.40596291422843933,
      "learning_rate": 2e-05,
      "loss": 0.7068,
      "step": 4160
    },
    {
      "epoch": 0.24565537555228276,
      "grad_norm": 0.4380647540092468,
      "learning_rate": 2e-05,
      "loss": 0.6721,
      "step": 4170
    },
    {
      "epoch": 0.24624447717231224,
      "grad_norm": 0.4007692039012909,
      "learning_rate": 2e-05,
      "loss": 0.6839,
      "step": 4180
    },
    {
      "epoch": 0.2468335787923417,
      "grad_norm": 0.4203566014766693,
      "learning_rate": 2e-05,
      "loss": 0.699,
      "step": 4190
    },
    {
      "epoch": 0.24742268041237114,
      "grad_norm": 0.4114213287830353,
      "learning_rate": 2e-05,
      "loss": 0.6935,
      "step": 4200
    },
    {
      "epoch": 0.2480117820324006,
      "grad_norm": 0.40831735730171204,
      "learning_rate": 2e-05,
      "loss": 0.6953,
      "step": 4210
    },
    {
      "epoch": 0.24860088365243005,
      "grad_norm": 0.39961013197898865,
      "learning_rate": 2e-05,
      "loss": 0.7179,
      "step": 4220
    },
    {
      "epoch": 0.2491899852724595,
      "grad_norm": 0.45752352476119995,
      "learning_rate": 2e-05,
      "loss": 0.6859,
      "step": 4230
    },
    {
      "epoch": 0.24977908689248896,
      "grad_norm": 0.39875659346580505,
      "learning_rate": 2e-05,
      "loss": 0.7112,
      "step": 4240
    },
    {
      "epoch": 0.2503681885125184,
      "grad_norm": 0.4168282449245453,
      "learning_rate": 2e-05,
      "loss": 0.7066,
      "step": 4250
    },
    {
      "epoch": 0.25095729013254786,
      "grad_norm": 0.415464848279953,
      "learning_rate": 2e-05,
      "loss": 0.6792,
      "step": 4260
    },
    {
      "epoch": 0.2515463917525773,
      "grad_norm": 0.44042834639549255,
      "learning_rate": 2e-05,
      "loss": 0.6951,
      "step": 4270
    },
    {
      "epoch": 0.25213549337260677,
      "grad_norm": 0.4377042055130005,
      "learning_rate": 2e-05,
      "loss": 0.6908,
      "step": 4280
    },
    {
      "epoch": 0.2527245949926362,
      "grad_norm": 0.4043514132499695,
      "learning_rate": 2e-05,
      "loss": 0.7071,
      "step": 4290
    },
    {
      "epoch": 0.2533136966126657,
      "grad_norm": 0.551868736743927,
      "learning_rate": 2e-05,
      "loss": 0.6837,
      "step": 4300
    },
    {
      "epoch": 0.2539027982326951,
      "grad_norm": 0.46551254391670227,
      "learning_rate": 2e-05,
      "loss": 0.6865,
      "step": 4310
    },
    {
      "epoch": 0.2544918998527246,
      "grad_norm": 0.41559070348739624,
      "learning_rate": 2e-05,
      "loss": 0.7011,
      "step": 4320
    },
    {
      "epoch": 0.25508100147275403,
      "grad_norm": 0.412230908870697,
      "learning_rate": 2e-05,
      "loss": 0.7133,
      "step": 4330
    },
    {
      "epoch": 0.2556701030927835,
      "grad_norm": 0.4170745313167572,
      "learning_rate": 2e-05,
      "loss": 0.6922,
      "step": 4340
    },
    {
      "epoch": 0.25625920471281294,
      "grad_norm": 0.5156211256980896,
      "learning_rate": 2e-05,
      "loss": 0.6829,
      "step": 4350
    },
    {
      "epoch": 0.2568483063328424,
      "grad_norm": 0.3799770474433899,
      "learning_rate": 2e-05,
      "loss": 0.6963,
      "step": 4360
    },
    {
      "epoch": 0.25743740795287184,
      "grad_norm": 0.5070363283157349,
      "learning_rate": 2e-05,
      "loss": 0.7203,
      "step": 4370
    },
    {
      "epoch": 0.25802650957290135,
      "grad_norm": 0.4407119154930115,
      "learning_rate": 2e-05,
      "loss": 0.7169,
      "step": 4380
    },
    {
      "epoch": 0.2586156111929308,
      "grad_norm": 0.42876699566841125,
      "learning_rate": 2e-05,
      "loss": 0.7141,
      "step": 4390
    },
    {
      "epoch": 0.25920471281296026,
      "grad_norm": 0.4825715720653534,
      "learning_rate": 2e-05,
      "loss": 0.6957,
      "step": 4400
    },
    {
      "epoch": 0.2597938144329897,
      "grad_norm": 0.39916175603866577,
      "learning_rate": 2e-05,
      "loss": 0.6981,
      "step": 4410
    },
    {
      "epoch": 0.26038291605301916,
      "grad_norm": 0.40878432989120483,
      "learning_rate": 2e-05,
      "loss": 0.6999,
      "step": 4420
    },
    {
      "epoch": 0.2609720176730486,
      "grad_norm": 0.5990281105041504,
      "learning_rate": 2e-05,
      "loss": 0.6778,
      "step": 4430
    },
    {
      "epoch": 0.26156111929307807,
      "grad_norm": 0.4314785599708557,
      "learning_rate": 2e-05,
      "loss": 0.6788,
      "step": 4440
    },
    {
      "epoch": 0.2621502209131075,
      "grad_norm": 0.45047637820243835,
      "learning_rate": 2e-05,
      "loss": 0.6946,
      "step": 4450
    },
    {
      "epoch": 0.262739322533137,
      "grad_norm": 0.43331000208854675,
      "learning_rate": 2e-05,
      "loss": 0.7431,
      "step": 4460
    },
    {
      "epoch": 0.26332842415316643,
      "grad_norm": 0.9719600677490234,
      "learning_rate": 2e-05,
      "loss": 0.6932,
      "step": 4470
    },
    {
      "epoch": 0.2639175257731959,
      "grad_norm": 0.415147066116333,
      "learning_rate": 2e-05,
      "loss": 0.6871,
      "step": 4480
    },
    {
      "epoch": 0.26450662739322534,
      "grad_norm": 0.4234153628349304,
      "learning_rate": 2e-05,
      "loss": 0.7103,
      "step": 4490
    },
    {
      "epoch": 0.2650957290132548,
      "grad_norm": 0.40550753474235535,
      "learning_rate": 2e-05,
      "loss": 0.6989,
      "step": 4500
    },
    {
      "epoch": 0.26568483063328424,
      "grad_norm": 0.41490641236305237,
      "learning_rate": 2e-05,
      "loss": 0.7029,
      "step": 4510
    },
    {
      "epoch": 0.2662739322533137,
      "grad_norm": 0.4242832362651825,
      "learning_rate": 2e-05,
      "loss": 0.6975,
      "step": 4520
    },
    {
      "epoch": 0.26686303387334315,
      "grad_norm": 0.3949618339538574,
      "learning_rate": 2e-05,
      "loss": 0.6658,
      "step": 4530
    },
    {
      "epoch": 0.2674521354933726,
      "grad_norm": 0.49901503324508667,
      "learning_rate": 2e-05,
      "loss": 0.6739,
      "step": 4540
    },
    {
      "epoch": 0.26804123711340205,
      "grad_norm": 0.4465973675251007,
      "learning_rate": 2e-05,
      "loss": 0.6789,
      "step": 4550
    },
    {
      "epoch": 0.2686303387334315,
      "grad_norm": 0.4555221199989319,
      "learning_rate": 2e-05,
      "loss": 0.6703,
      "step": 4560
    },
    {
      "epoch": 0.26921944035346096,
      "grad_norm": 0.42599591612815857,
      "learning_rate": 2e-05,
      "loss": 0.6911,
      "step": 4570
    },
    {
      "epoch": 0.2698085419734904,
      "grad_norm": 0.422658771276474,
      "learning_rate": 2e-05,
      "loss": 0.6945,
      "step": 4580
    },
    {
      "epoch": 0.27039764359351987,
      "grad_norm": 0.4176231920719147,
      "learning_rate": 2e-05,
      "loss": 0.7343,
      "step": 4590
    },
    {
      "epoch": 0.2709867452135493,
      "grad_norm": 0.43195536732673645,
      "learning_rate": 2e-05,
      "loss": 0.7021,
      "step": 4600
    },
    {
      "epoch": 0.27157584683357877,
      "grad_norm": 0.4497572183609009,
      "learning_rate": 2e-05,
      "loss": 0.7074,
      "step": 4610
    },
    {
      "epoch": 0.2721649484536082,
      "grad_norm": 0.46830621361732483,
      "learning_rate": 2e-05,
      "loss": 0.7479,
      "step": 4620
    },
    {
      "epoch": 0.2727540500736377,
      "grad_norm": 0.4642024040222168,
      "learning_rate": 2e-05,
      "loss": 0.6958,
      "step": 4630
    },
    {
      "epoch": 0.27334315169366713,
      "grad_norm": 0.41414839029312134,
      "learning_rate": 2e-05,
      "loss": 0.699,
      "step": 4640
    },
    {
      "epoch": 0.27393225331369664,
      "grad_norm": 0.40390071272850037,
      "learning_rate": 2e-05,
      "loss": 0.699,
      "step": 4650
    },
    {
      "epoch": 0.2745213549337261,
      "grad_norm": 0.39850613474845886,
      "learning_rate": 2e-05,
      "loss": 0.6886,
      "step": 4660
    },
    {
      "epoch": 0.27511045655375554,
      "grad_norm": 0.41028398275375366,
      "learning_rate": 2e-05,
      "loss": 0.6899,
      "step": 4670
    },
    {
      "epoch": 0.275699558173785,
      "grad_norm": 0.6875986456871033,
      "learning_rate": 2e-05,
      "loss": 0.713,
      "step": 4680
    },
    {
      "epoch": 0.27628865979381445,
      "grad_norm": 0.413207471370697,
      "learning_rate": 2e-05,
      "loss": 0.677,
      "step": 4690
    },
    {
      "epoch": 0.2768777614138439,
      "grad_norm": 0.40974339842796326,
      "learning_rate": 2e-05,
      "loss": 0.7282,
      "step": 4700
    },
    {
      "epoch": 0.27746686303387336,
      "grad_norm": 0.3963793218135834,
      "learning_rate": 2e-05,
      "loss": 0.6918,
      "step": 4710
    },
    {
      "epoch": 0.2780559646539028,
      "grad_norm": 0.4228408932685852,
      "learning_rate": 2e-05,
      "loss": 0.6776,
      "step": 4720
    },
    {
      "epoch": 0.27864506627393226,
      "grad_norm": 0.5265048742294312,
      "learning_rate": 2e-05,
      "loss": 0.6734,
      "step": 4730
    },
    {
      "epoch": 0.2792341678939617,
      "grad_norm": 0.41529589891433716,
      "learning_rate": 2e-05,
      "loss": 0.6921,
      "step": 4740
    },
    {
      "epoch": 0.27982326951399117,
      "grad_norm": 0.5591365694999695,
      "learning_rate": 2e-05,
      "loss": 0.6866,
      "step": 4750
    },
    {
      "epoch": 0.2804123711340206,
      "grad_norm": 0.5308225750923157,
      "learning_rate": 2e-05,
      "loss": 0.7017,
      "step": 4760
    },
    {
      "epoch": 0.2810014727540501,
      "grad_norm": 0.3693162500858307,
      "learning_rate": 2e-05,
      "loss": 0.6678,
      "step": 4770
    },
    {
      "epoch": 0.2815905743740795,
      "grad_norm": 0.38088929653167725,
      "learning_rate": 2e-05,
      "loss": 0.6803,
      "step": 4780
    },
    {
      "epoch": 0.282179675994109,
      "grad_norm": 0.7079582810401917,
      "learning_rate": 2e-05,
      "loss": 0.6801,
      "step": 4790
    },
    {
      "epoch": 0.28276877761413843,
      "grad_norm": 0.4070088267326355,
      "learning_rate": 2e-05,
      "loss": 0.6771,
      "step": 4800
    },
    {
      "epoch": 0.2833578792341679,
      "grad_norm": 0.43372347950935364,
      "learning_rate": 2e-05,
      "loss": 0.6893,
      "step": 4810
    },
    {
      "epoch": 0.28394698085419734,
      "grad_norm": 0.45391398668289185,
      "learning_rate": 2e-05,
      "loss": 0.685,
      "step": 4820
    },
    {
      "epoch": 0.2845360824742268,
      "grad_norm": 0.47494325041770935,
      "learning_rate": 2e-05,
      "loss": 0.691,
      "step": 4830
    },
    {
      "epoch": 0.28512518409425625,
      "grad_norm": 0.4002147912979126,
      "learning_rate": 2e-05,
      "loss": 0.7105,
      "step": 4840
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 0.4951745271682739,
      "learning_rate": 2e-05,
      "loss": 0.6867,
      "step": 4850
    },
    {
      "epoch": 0.28630338733431515,
      "grad_norm": 0.4014371931552887,
      "learning_rate": 2e-05,
      "loss": 0.6824,
      "step": 4860
    },
    {
      "epoch": 0.2868924889543446,
      "grad_norm": 0.45175793766975403,
      "learning_rate": 2e-05,
      "loss": 0.7255,
      "step": 4870
    },
    {
      "epoch": 0.28748159057437406,
      "grad_norm": 0.4240052402019501,
      "learning_rate": 2e-05,
      "loss": 0.6849,
      "step": 4880
    },
    {
      "epoch": 0.2880706921944035,
      "grad_norm": 0.44377070665359497,
      "learning_rate": 2e-05,
      "loss": 0.7082,
      "step": 4890
    },
    {
      "epoch": 0.28865979381443296,
      "grad_norm": 0.4380945563316345,
      "learning_rate": 2e-05,
      "loss": 0.6844,
      "step": 4900
    },
    {
      "epoch": 0.28924889543446247,
      "grad_norm": 0.49835464358329773,
      "learning_rate": 2e-05,
      "loss": 0.683,
      "step": 4910
    },
    {
      "epoch": 0.2898379970544919,
      "grad_norm": 0.4041672348976135,
      "learning_rate": 2e-05,
      "loss": 0.6775,
      "step": 4920
    },
    {
      "epoch": 0.2904270986745214,
      "grad_norm": 0.3982612192630768,
      "learning_rate": 2e-05,
      "loss": 0.6855,
      "step": 4930
    },
    {
      "epoch": 0.29101620029455083,
      "grad_norm": 0.48156970739364624,
      "learning_rate": 2e-05,
      "loss": 0.682,
      "step": 4940
    },
    {
      "epoch": 0.2916053019145803,
      "grad_norm": 0.44620630145072937,
      "learning_rate": 2e-05,
      "loss": 0.7026,
      "step": 4950
    },
    {
      "epoch": 0.29219440353460974,
      "grad_norm": 0.38848385214805603,
      "learning_rate": 2e-05,
      "loss": 0.6782,
      "step": 4960
    },
    {
      "epoch": 0.2927835051546392,
      "grad_norm": 0.4821191132068634,
      "learning_rate": 2e-05,
      "loss": 0.6939,
      "step": 4970
    },
    {
      "epoch": 0.29337260677466864,
      "grad_norm": 0.4406975209712982,
      "learning_rate": 2e-05,
      "loss": 0.6749,
      "step": 4980
    },
    {
      "epoch": 0.2939617083946981,
      "grad_norm": 0.4304104447364807,
      "learning_rate": 2e-05,
      "loss": 0.6958,
      "step": 4990
    },
    {
      "epoch": 0.29455081001472755,
      "grad_norm": 0.4109821617603302,
      "learning_rate": 2e-05,
      "loss": 0.6719,
      "step": 5000
    },
    {
      "epoch": 0.295139911634757,
      "grad_norm": 0.43141236901283264,
      "learning_rate": 2e-05,
      "loss": 0.6724,
      "step": 5010
    },
    {
      "epoch": 0.29572901325478645,
      "grad_norm": 0.42312124371528625,
      "learning_rate": 2e-05,
      "loss": 0.7069,
      "step": 5020
    },
    {
      "epoch": 0.2963181148748159,
      "grad_norm": 0.43027567863464355,
      "learning_rate": 2e-05,
      "loss": 0.6996,
      "step": 5030
    },
    {
      "epoch": 0.29690721649484536,
      "grad_norm": 0.37733063101768494,
      "learning_rate": 2e-05,
      "loss": 0.6993,
      "step": 5040
    },
    {
      "epoch": 0.2974963181148748,
      "grad_norm": 0.4357035756111145,
      "learning_rate": 2e-05,
      "loss": 0.6982,
      "step": 5050
    },
    {
      "epoch": 0.29808541973490427,
      "grad_norm": 0.47058531641960144,
      "learning_rate": 2e-05,
      "loss": 0.6867,
      "step": 5060
    },
    {
      "epoch": 0.2986745213549337,
      "grad_norm": 0.44040125608444214,
      "learning_rate": 2e-05,
      "loss": 0.6904,
      "step": 5070
    },
    {
      "epoch": 0.2992636229749632,
      "grad_norm": 0.44315773248672485,
      "learning_rate": 2e-05,
      "loss": 0.649,
      "step": 5080
    },
    {
      "epoch": 0.2998527245949926,
      "grad_norm": 0.4275892972946167,
      "learning_rate": 2e-05,
      "loss": 0.6908,
      "step": 5090
    },
    {
      "epoch": 0.3004418262150221,
      "grad_norm": 0.428265780210495,
      "learning_rate": 2e-05,
      "loss": 0.6732,
      "step": 5100
    },
    {
      "epoch": 0.30103092783505153,
      "grad_norm": 0.4170130789279938,
      "learning_rate": 2e-05,
      "loss": 0.6891,
      "step": 5110
    },
    {
      "epoch": 0.301620029455081,
      "grad_norm": 0.41354796290397644,
      "learning_rate": 2e-05,
      "loss": 0.7121,
      "step": 5120
    },
    {
      "epoch": 0.30220913107511044,
      "grad_norm": 0.43516281247138977,
      "learning_rate": 2e-05,
      "loss": 0.6837,
      "step": 5130
    },
    {
      "epoch": 0.3027982326951399,
      "grad_norm": 0.9627923369407654,
      "learning_rate": 2e-05,
      "loss": 0.6952,
      "step": 5140
    },
    {
      "epoch": 0.30338733431516934,
      "grad_norm": 0.40709105134010315,
      "learning_rate": 2e-05,
      "loss": 0.7107,
      "step": 5150
    },
    {
      "epoch": 0.3039764359351988,
      "grad_norm": 0.4350499212741852,
      "learning_rate": 2e-05,
      "loss": 0.7038,
      "step": 5160
    },
    {
      "epoch": 0.30456553755522825,
      "grad_norm": 0.4102758467197418,
      "learning_rate": 2e-05,
      "loss": 0.6736,
      "step": 5170
    },
    {
      "epoch": 0.30515463917525776,
      "grad_norm": 0.4135771691799164,
      "learning_rate": 2e-05,
      "loss": 0.7002,
      "step": 5180
    },
    {
      "epoch": 0.3057437407952872,
      "grad_norm": 0.45237410068511963,
      "learning_rate": 2e-05,
      "loss": 0.687,
      "step": 5190
    },
    {
      "epoch": 0.30633284241531666,
      "grad_norm": 0.402154803276062,
      "learning_rate": 2e-05,
      "loss": 0.6694,
      "step": 5200
    },
    {
      "epoch": 0.3069219440353461,
      "grad_norm": 0.4044947922229767,
      "learning_rate": 2e-05,
      "loss": 0.7024,
      "step": 5210
    },
    {
      "epoch": 0.30751104565537557,
      "grad_norm": 0.4093049466609955,
      "learning_rate": 2e-05,
      "loss": 0.691,
      "step": 5220
    },
    {
      "epoch": 0.308100147275405,
      "grad_norm": 0.47435611486434937,
      "learning_rate": 2e-05,
      "loss": 0.6629,
      "step": 5230
    },
    {
      "epoch": 0.3086892488954345,
      "grad_norm": 0.40708109736442566,
      "learning_rate": 2e-05,
      "loss": 0.6688,
      "step": 5240
    },
    {
      "epoch": 0.30927835051546393,
      "grad_norm": 0.4180338978767395,
      "learning_rate": 2e-05,
      "loss": 0.6984,
      "step": 5250
    },
    {
      "epoch": 0.3098674521354934,
      "grad_norm": 0.43557310104370117,
      "learning_rate": 2e-05,
      "loss": 0.6648,
      "step": 5260
    },
    {
      "epoch": 0.31045655375552283,
      "grad_norm": 0.40062248706817627,
      "learning_rate": 2e-05,
      "loss": 0.6964,
      "step": 5270
    },
    {
      "epoch": 0.3110456553755523,
      "grad_norm": 0.439288467168808,
      "learning_rate": 2e-05,
      "loss": 0.7116,
      "step": 5280
    },
    {
      "epoch": 0.31163475699558174,
      "grad_norm": 0.39990952610969543,
      "learning_rate": 2e-05,
      "loss": 0.6676,
      "step": 5290
    },
    {
      "epoch": 0.3122238586156112,
      "grad_norm": 0.4019280970096588,
      "learning_rate": 2e-05,
      "loss": 0.6807,
      "step": 5300
    },
    {
      "epoch": 0.31281296023564065,
      "grad_norm": 0.41289186477661133,
      "learning_rate": 2e-05,
      "loss": 0.7097,
      "step": 5310
    },
    {
      "epoch": 0.3134020618556701,
      "grad_norm": 0.3916700482368469,
      "learning_rate": 2e-05,
      "loss": 0.7049,
      "step": 5320
    },
    {
      "epoch": 0.31399116347569955,
      "grad_norm": 0.39174118638038635,
      "learning_rate": 2e-05,
      "loss": 0.6711,
      "step": 5330
    },
    {
      "epoch": 0.314580265095729,
      "grad_norm": 0.48452794551849365,
      "learning_rate": 2e-05,
      "loss": 0.7047,
      "step": 5340
    },
    {
      "epoch": 0.31516936671575846,
      "grad_norm": 0.4124949276447296,
      "learning_rate": 1.999194520388485e-05,
      "loss": 0.719,
      "step": 5350
    },
    {
      "epoch": 0.3157584683357879,
      "grad_norm": 0.4392179548740387,
      "learning_rate": 1.9959247303157763e-05,
      "loss": 0.7171,
      "step": 5360
    },
    {
      "epoch": 0.31634756995581736,
      "grad_norm": 0.413244366645813,
      "learning_rate": 1.990149424459817e-05,
      "loss": 0.6834,
      "step": 5370
    },
    {
      "epoch": 0.3169366715758468,
      "grad_norm": 0.42354679107666016,
      "learning_rate": 1.9818847538648688e-05,
      "loss": 0.6786,
      "step": 5380
    },
    {
      "epoch": 0.31752577319587627,
      "grad_norm": 0.465719610452652,
      "learning_rate": 1.971153831256721e-05,
      "loss": 0.6345,
      "step": 5390
    },
    {
      "epoch": 0.3181148748159057,
      "grad_norm": 0.41384217143058777,
      "learning_rate": 1.9579866664063435e-05,
      "loss": 0.6979,
      "step": 5400
    },
    {
      "epoch": 0.3187039764359352,
      "grad_norm": 0.40369582176208496,
      "learning_rate": 1.9424200822054782e-05,
      "loss": 0.6693,
      "step": 5410
    },
    {
      "epoch": 0.31929307805596463,
      "grad_norm": 0.45909035205841064,
      "learning_rate": 1.9244976116888626e-05,
      "loss": 0.6729,
      "step": 5420
    },
    {
      "epoch": 0.3198821796759941,
      "grad_norm": 0.4269322454929352,
      "learning_rate": 1.904269376291071e-05,
      "loss": 0.6809,
      "step": 5430
    },
    {
      "epoch": 0.3204712812960236,
      "grad_norm": 0.44284433126449585,
      "learning_rate": 1.88179194567844e-05,
      "loss": 0.689,
      "step": 5440
    },
    {
      "epoch": 0.32106038291605304,
      "grad_norm": 0.623070478439331,
      "learning_rate": 1.8571281795480632e-05,
      "loss": 0.7005,
      "step": 5450
    },
    {
      "epoch": 0.3216494845360825,
      "grad_norm": 0.4040432870388031,
      "learning_rate": 1.8303470518362808e-05,
      "loss": 0.6729,
      "step": 5460
    },
    {
      "epoch": 0.32223858615611195,
      "grad_norm": 0.4163486659526825,
      "learning_rate": 1.801523457828271e-05,
      "loss": 0.7119,
      "step": 5470
    },
    {
      "epoch": 0.3228276877761414,
      "grad_norm": 0.4070464074611664,
      "learning_rate": 1.7707380047081796e-05,
      "loss": 0.6961,
      "step": 5480
    },
    {
      "epoch": 0.32341678939617086,
      "grad_norm": 0.41157838702201843,
      "learning_rate": 1.7380767861355252e-05,
      "loss": 0.6638,
      "step": 5490
    },
    {
      "epoch": 0.3240058910162003,
      "grad_norm": 0.4094664752483368,
      "learning_rate": 1.703631141478299e-05,
      "loss": 0.6776,
      "step": 5500
    },
    {
      "epoch": 0.32459499263622976,
      "grad_norm": 0.4159485995769501,
      "learning_rate": 1.6674974003760705e-05,
      "loss": 0.6716,
      "step": 5510
    },
    {
      "epoch": 0.3251840942562592,
      "grad_norm": 0.4784594774246216,
      "learning_rate": 1.6297766133474556e-05,
      "loss": 0.6719,
      "step": 5520
    },
    {
      "epoch": 0.32577319587628867,
      "grad_norm": 0.4779771864414215,
      "learning_rate": 1.5905742691953186e-05,
      "loss": 0.6657,
      "step": 5530
    },
    {
      "epoch": 0.3263622974963181,
      "grad_norm": 0.4335428476333618,
      "learning_rate": 1.55e-05,
      "loss": 0.6693,
      "step": 5540
    },
    {
      "epoch": 0.3269513991163476,
      "grad_norm": 0.4379546046257019,
      "learning_rate": 1.5081672745255862e-05,
      "loss": 0.6811,
      "step": 5550
    },
    {
      "epoch": 0.327540500736377,
      "grad_norm": 0.4272069036960602,
      "learning_rate": 1.4651930808966206e-05,
      "loss": 0.652,
      "step": 5560
    },
    {
      "epoch": 0.3281296023564065,
      "grad_norm": 0.508344292640686,
      "learning_rate": 1.4211975994326848e-05,
      "loss": 0.68,
      "step": 5570
    },
    {
      "epoch": 0.32871870397643593,
      "grad_norm": 0.4721388518810272,
      "learning_rate": 1.376303866555779e-05,
      "loss": 0.6836,
      "step": 5580
    },
    {
      "epoch": 0.3293078055964654,
      "grad_norm": 0.40316617488861084,
      "learning_rate": 1.3306374307104164e-05,
      "loss": 0.6854,
      "step": 5590
    },
    {
      "epoch": 0.32989690721649484,
      "grad_norm": 0.40988579392433167,
      "learning_rate": 1.2843260012586718e-05,
      "loss": 0.7224,
      "step": 5600
    },
    {
      "epoch": 0.3304860088365243,
      "grad_norm": 0.4444100558757782,
      "learning_rate": 1.2374990913320726e-05,
      "loss": 0.6682,
      "step": 5610
    },
    {
      "epoch": 0.33107511045655375,
      "grad_norm": 0.393057644367218,
      "learning_rate": 1.1902876556391247e-05,
      "loss": 0.6796,
      "step": 5620
    },
    {
      "epoch": 0.3316642120765832,
      "grad_norm": 0.4164564907550812,
      "learning_rate": 1.1428237242413685e-05,
      "loss": 0.6264,
      "step": 5630
    },
    {
      "epoch": 0.33225331369661265,
      "grad_norm": 0.4044070541858673,
      "learning_rate": 1.095240033322132e-05,
      "loss": 0.6844,
      "step": 5640
    },
    {
      "epoch": 0.3328424153166421,
      "grad_norm": 0.6211689710617065,
      "learning_rate": 1.047669653980572e-05,
      "loss": 0.6621,
      "step": 5650
    },
    {
      "epoch": 0.33343151693667156,
      "grad_norm": 0.3830164074897766,
      "learning_rate": 1.0002456200890902e-05,
      "loss": 0.687,
      "step": 5660
    },
    {
      "epoch": 0.334020618556701,
      "grad_norm": 0.4077165126800537,
      "learning_rate": 9.531005562548668e-06,
      "loss": 0.6979,
      "step": 5670
    },
    {
      "epoch": 0.33460972017673046,
      "grad_norm": 0.434323787689209,
      "learning_rate": 9.06366306925925e-06,
      "loss": 0.7037,
      "step": 5680
    },
    {
      "epoch": 0.3351988217967599,
      "grad_norm": 0.39463865756988525,
      "learning_rate": 8.601735676789686e-06,
      "loss": 0.6816,
      "step": 5690
    },
    {
      "epoch": 0.33578792341678937,
      "grad_norm": 0.42032647132873535,
      "learning_rate": 8.14651519720118e-06,
      "loss": 0.6744,
      "step": 5700
    },
    {
      "epoch": 0.3363770250368189,
      "grad_norm": 0.48433321714401245,
      "learning_rate": 7.699274686206866e-06,
      "loss": 0.7189,
      "step": 5710
    },
    {
      "epoch": 0.33696612665684833,
      "grad_norm": 0.4152541160583496,
      "learning_rate": 7.261264882983024e-06,
      "loss": 0.7099,
      "step": 5720
    },
    {
      "epoch": 0.3375552282768778,
      "grad_norm": 0.38766705989837646,
      "learning_rate": 6.8337107123900386e-06,
      "loss": 0.6746,
      "step": 5730
    },
    {
      "epoch": 0.33814432989690724,
      "grad_norm": 0.45199644565582275,
      "learning_rate": 6.417807859384881e-06,
      "loss": 0.6872,
      "step": 5740
    },
    {
      "epoch": 0.3387334315169367,
      "grad_norm": 0.4452700614929199,
      "learning_rate": 6.014719425205009e-06,
      "loss": 0.6918,
      "step": 5750
    },
    {
      "epoch": 0.33932253313696614,
      "grad_norm": 0.40021347999572754,
      "learning_rate": 5.6255726746749275e-06,
      "loss": 0.6639,
      "step": 5760
    },
    {
      "epoch": 0.3399116347569956,
      "grad_norm": 0.40448781847953796,
      "learning_rate": 5.251455883731728e-06,
      "loss": 0.6764,
      "step": 5770
    },
    {
      "epoch": 0.34050073637702505,
      "grad_norm": 0.4461590349674225,
      "learning_rate": 4.893415295985812e-06,
      "loss": 0.6851,
      "step": 5780
    },
    {
      "epoch": 0.3410898379970545,
      "grad_norm": 0.439993679523468,
      "learning_rate": 4.552452196827885e-06,
      "loss": 0.6635,
      "step": 5790
    },
    {
      "epoch": 0.34167893961708395,
      "grad_norm": 0.46094948053359985,
      "learning_rate": 4.229520113264785e-06,
      "loss": 0.6658,
      "step": 5800
    },
    {
      "epoch": 0.3422680412371134,
      "grad_norm": 0.4167616069316864,
      "learning_rate": 3.925522147314915e-06,
      "loss": 0.6832,
      "step": 5810
    },
    {
      "epoch": 0.34285714285714286,
      "grad_norm": 0.40989744663238525,
      "learning_rate": 3.641308450420675e-06,
      "loss": 0.6687,
      "step": 5820
    },
    {
      "epoch": 0.3434462444771723,
      "grad_norm": 0.45302531123161316,
      "learning_rate": 3.37767384594087e-06,
      "loss": 0.6713,
      "step": 5830
    },
    {
      "epoch": 0.34403534609720177,
      "grad_norm": 0.4904053211212158,
      "learning_rate": 3.1353556063719357e-06,
      "loss": 0.6732,
      "step": 5840
    },
    {
      "epoch": 0.3446244477172312,
      "grad_norm": 0.4581778347492218,
      "learning_rate": 2.9150313915141474e-06,
      "loss": 0.6651,
      "step": 5850
    },
    {
      "epoch": 0.3452135493372607,
      "grad_norm": 0.42740631103515625,
      "learning_rate": 2.717317353348891e-06,
      "loss": 0.6767,
      "step": 5860
    },
    {
      "epoch": 0.3458026509572901,
      "grad_norm": 0.4436577558517456,
      "learning_rate": 2.5427664129268253e-06,
      "loss": 0.6756,
      "step": 5870
    },
    {
      "epoch": 0.3463917525773196,
      "grad_norm": 0.4009878635406494,
      "learning_rate": 2.391866714085717e-06,
      "loss": 0.6951,
      "step": 5880
    },
    {
      "epoch": 0.34698085419734903,
      "grad_norm": 0.3839609622955322,
      "learning_rate": 2.2650402583222753e-06,
      "loss": 0.6591,
      "step": 5890
    },
    {
      "epoch": 0.3475699558173785,
      "grad_norm": 0.4072750210762024,
      "learning_rate": 2.1626417246356398e-06,
      "loss": 0.6473,
      "step": 5900
    },
    {
      "epoch": 0.34815905743740794,
      "grad_norm": 0.39682498574256897,
      "learning_rate": 2.084957477642894e-06,
      "loss": 0.6622,
      "step": 5910
    },
    {
      "epoch": 0.3487481590574374,
      "grad_norm": 0.4176507592201233,
      "learning_rate": 2.0322047667405147e-06,
      "loss": 0.6889,
      "step": 5920
    },
    {
      "epoch": 0.34933726067746684,
      "grad_norm": 0.4035165011882782,
      "learning_rate": 2.0045311185513342e-06,
      "loss": 0.6926,
      "step": 5930
    }
  ],
  "logging_steps": 10,
  "max_steps": 5935,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2035,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.6494000762504348e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}