{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.3750394944707741,
  "eval_steps": 500,
  "global_step": 5935,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.000631911532385466,
      "grad_norm": 1.3719526529312134,
      "learning_rate": 2.910112359550562e-06,
      "loss": 1.1034,
      "step": 10
    },
    {
      "epoch": 0.001263823064770932,
      "grad_norm": 1.3293606042861938,
      "learning_rate": 3.921348314606742e-06,
      "loss": 1.0409,
      "step": 20
    },
    {
      "epoch": 0.0018957345971563982,
      "grad_norm": 1.001535177230835,
      "learning_rate": 4.932584269662922e-06,
      "loss": 1.0116,
      "step": 30
    },
    {
      "epoch": 0.002527646129541864,
      "grad_norm": 0.9550002813339233,
      "learning_rate": 5.943820224719102e-06,
      "loss": 0.9846,
      "step": 40
    },
    {
      "epoch": 0.00315955766192733,
      "grad_norm": 0.9018183350563049,
      "learning_rate": 6.955056179775282e-06,
      "loss": 0.964,
      "step": 50
    },
    {
      "epoch": 0.0037914691943127963,
      "grad_norm": 0.8208801746368408,
      "learning_rate": 7.966292134831462e-06,
      "loss": 0.9519,
      "step": 60
    },
    {
      "epoch": 0.0044233807266982625,
      "grad_norm": 0.8281100392341614,
      "learning_rate": 8.977528089887641e-06,
      "loss": 0.9736,
      "step": 70
    },
    {
      "epoch": 0.005055292259083728,
      "grad_norm": 0.8811385035514832,
      "learning_rate": 9.988764044943822e-06,
      "loss": 0.9287,
      "step": 80
    },
    {
      "epoch": 0.005687203791469194,
      "grad_norm": 0.8633086085319519,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 0.9478,
      "step": 90
    },
    {
      "epoch": 0.00631911532385466,
      "grad_norm": 0.826318085193634,
      "learning_rate": 1.2011235955056182e-05,
      "loss": 0.9045,
      "step": 100
    },
    {
      "epoch": 0.0069510268562401265,
      "grad_norm": 0.928963840007782,
      "learning_rate": 1.3022471910112362e-05,
      "loss": 0.9512,
      "step": 110
    },
    {
      "epoch": 0.007582938388625593,
      "grad_norm": 0.9523959755897522,
      "learning_rate": 1.403370786516854e-05,
      "loss": 0.9181,
      "step": 120
    },
    {
      "epoch": 0.008214849921011059,
      "grad_norm": 0.9578841924667358,
      "learning_rate": 1.504494382022472e-05,
      "loss": 0.928,
      "step": 130
    },
    {
      "epoch": 0.008846761453396525,
      "grad_norm": 0.9352409243583679,
      "learning_rate": 1.60561797752809e-05,
      "loss": 0.9325,
      "step": 140
    },
    {
      "epoch": 0.009478672985781991,
      "grad_norm": 0.8745258450508118,
      "learning_rate": 1.706741573033708e-05,
      "loss": 0.9502,
      "step": 150
    },
    {
      "epoch": 0.010110584518167456,
      "grad_norm": 0.8288115859031677,
      "learning_rate": 1.8078651685393256e-05,
      "loss": 0.9507,
      "step": 160
    },
    {
      "epoch": 0.010742496050552922,
      "grad_norm": 1.211285948753357,
      "learning_rate": 1.908988764044944e-05,
      "loss": 0.9347,
      "step": 170
    },
    {
      "epoch": 0.011374407582938388,
      "grad_norm": 0.8345915079116821,
      "learning_rate": 2e-05,
      "loss": 0.9589,
      "step": 180
    },
    {
      "epoch": 0.012006319115323854,
      "grad_norm": 0.8273001313209534,
      "learning_rate": 2e-05,
      "loss": 0.9263,
      "step": 190
    },
    {
      "epoch": 0.01263823064770932,
      "grad_norm": 0.7863247394561768,
      "learning_rate": 2e-05,
      "loss": 0.9252,
      "step": 200
    },
    {
      "epoch": 0.013270142180094787,
      "grad_norm": 0.7312272191047668,
      "learning_rate": 2e-05,
      "loss": 0.9206,
      "step": 210
    },
    {
      "epoch": 0.013902053712480253,
      "grad_norm": 0.8003473281860352,
      "learning_rate": 2e-05,
      "loss": 0.9291,
      "step": 220
    },
    {
      "epoch": 0.014533965244865719,
      "grad_norm": 0.7968817353248596,
      "learning_rate": 2e-05,
      "loss": 0.9268,
      "step": 230
    },
    {
      "epoch": 0.015165876777251185,
      "grad_norm": 0.787013053894043,
      "learning_rate": 2e-05,
      "loss": 0.9385,
      "step": 240
    },
    {
      "epoch": 0.01579778830963665,
      "grad_norm": 0.5825610756874084,
      "learning_rate": 2e-05,
      "loss": 0.9255,
      "step": 250
    },
    {
      "epoch": 0.016429699842022118,
      "grad_norm": 0.8487895131111145,
      "learning_rate": 2e-05,
      "loss": 0.9173,
      "step": 260
    },
    {
      "epoch": 0.017061611374407582,
      "grad_norm": 0.6732227802276611,
      "learning_rate": 2e-05,
      "loss": 0.9124,
      "step": 270
    },
    {
      "epoch": 0.01769352290679305,
      "grad_norm": 0.6557146310806274,
      "learning_rate": 2e-05,
      "loss": 0.9181,
      "step": 280
    },
    {
      "epoch": 0.018325434439178515,
      "grad_norm": 0.605126142501831,
      "learning_rate": 2e-05,
      "loss": 0.9163,
      "step": 290
    },
    {
      "epoch": 0.018957345971563982,
      "grad_norm": 0.610867977142334,
      "learning_rate": 2e-05,
      "loss": 0.9137,
      "step": 300
    },
    {
      "epoch": 0.019589257503949447,
      "grad_norm": 0.6153805255889893,
      "learning_rate": 2e-05,
      "loss": 0.8954,
      "step": 310
    },
    {
      "epoch": 0.02022116903633491,
      "grad_norm": 0.5910293459892273,
      "learning_rate": 2e-05,
      "loss": 0.9168,
      "step": 320
    },
    {
      "epoch": 0.02085308056872038,
      "grad_norm": 0.5992543697357178,
      "learning_rate": 2e-05,
      "loss": 0.9145,
      "step": 330
    },
    {
      "epoch": 0.021484992101105844,
      "grad_norm": 0.6959894299507141,
      "learning_rate": 2e-05,
      "loss": 0.9143,
      "step": 340
    },
    {
      "epoch": 0.022116903633491312,
      "grad_norm": 0.6059086322784424,
      "learning_rate": 2e-05,
      "loss": 0.9236,
      "step": 350
    },
    {
      "epoch": 0.022748815165876776,
      "grad_norm": 0.6001873016357422,
      "learning_rate": 2e-05,
      "loss": 0.9162,
      "step": 360
    },
    {
      "epoch": 0.023380726698262244,
      "grad_norm": 0.6104479432106018,
      "learning_rate": 2e-05,
      "loss": 0.8826,
      "step": 370
    },
    {
      "epoch": 0.02401263823064771,
      "grad_norm": 0.5393518209457397,
      "learning_rate": 2e-05,
      "loss": 0.9106,
      "step": 380
    },
    {
      "epoch": 0.024644549763033177,
      "grad_norm": 0.646965742111206,
      "learning_rate": 2e-05,
      "loss": 0.898,
      "step": 390
    },
    {
      "epoch": 0.02527646129541864,
      "grad_norm": 0.5867894291877747,
      "learning_rate": 2e-05,
      "loss": 0.9604,
      "step": 400
    },
    {
      "epoch": 0.02590837282780411,
      "grad_norm": 0.5493832230567932,
      "learning_rate": 2e-05,
      "loss": 0.9134,
      "step": 410
    },
    {
      "epoch": 0.026540284360189573,
      "grad_norm": 0.6109013557434082,
      "learning_rate": 2e-05,
      "loss": 0.9255,
      "step": 420
    },
    {
      "epoch": 0.027172195892575038,
      "grad_norm": 0.5416685938835144,
      "learning_rate": 2e-05,
      "loss": 0.9158,
      "step": 430
    },
    {
      "epoch": 0.027804107424960506,
      "grad_norm": 0.5394052863121033,
      "learning_rate": 2e-05,
      "loss": 0.9085,
      "step": 440
    },
    {
      "epoch": 0.02843601895734597,
      "grad_norm": 0.5504438281059265,
      "learning_rate": 2e-05,
      "loss": 0.9021,
      "step": 450
    },
    {
      "epoch": 0.029067930489731438,
      "grad_norm": 0.6121982932090759,
      "learning_rate": 2e-05,
      "loss": 0.9103,
      "step": 460
    },
    {
      "epoch": 0.029699842022116903,
      "grad_norm": 0.5534759163856506,
      "learning_rate": 2e-05,
      "loss": 0.913,
      "step": 470
    },
    {
      "epoch": 0.03033175355450237,
      "grad_norm": 0.5392954349517822,
      "learning_rate": 2e-05,
      "loss": 0.9033,
      "step": 480
    },
    {
      "epoch": 0.030963665086887835,
      "grad_norm": 0.5346894264221191,
      "learning_rate": 2e-05,
      "loss": 0.9401,
      "step": 490
    },
    {
      "epoch": 0.0315955766192733,
      "grad_norm": 0.580543577671051,
      "learning_rate": 2e-05,
      "loss": 0.9254,
      "step": 500
    },
    {
      "epoch": 0.03222748815165877,
      "grad_norm": 0.5434547066688538,
      "learning_rate": 2e-05,
      "loss": 0.899,
      "step": 510
    },
    {
      "epoch": 0.032859399684044235,
      "grad_norm": 0.5517114400863647,
      "learning_rate": 2e-05,
      "loss": 0.9203,
      "step": 520
    },
    {
      "epoch": 0.033491311216429696,
      "grad_norm": 0.5170676708221436,
      "learning_rate": 2e-05,
      "loss": 0.918,
      "step": 530
    },
    {
      "epoch": 0.034123222748815164,
      "grad_norm": 0.5187759399414062,
      "learning_rate": 2e-05,
      "loss": 0.9194,
      "step": 540
    },
    {
      "epoch": 0.03475513428120063,
      "grad_norm": 0.5202495455741882,
      "learning_rate": 2e-05,
      "loss": 0.8841,
      "step": 550
    },
    {
      "epoch": 0.0353870458135861,
      "grad_norm": 0.5262540578842163,
      "learning_rate": 2e-05,
      "loss": 0.9196,
      "step": 560
    },
    {
      "epoch": 0.03601895734597156,
      "grad_norm": 0.6277403831481934,
      "learning_rate": 2e-05,
      "loss": 0.8895,
      "step": 570
    },
    {
      "epoch": 0.03665086887835703,
      "grad_norm": 0.6800224184989929,
      "learning_rate": 2e-05,
      "loss": 0.8951,
      "step": 580
    },
    {
      "epoch": 0.0372827804107425,
      "grad_norm": 0.5502716898918152,
      "learning_rate": 2e-05,
      "loss": 0.9111,
      "step": 590
    },
    {
      "epoch": 0.037914691943127965,
      "grad_norm": 0.6231828927993774,
      "learning_rate": 2e-05,
      "loss": 0.9059,
      "step": 600
    },
    {
      "epoch": 0.038546603475513426,
      "grad_norm": 0.6243113875389099,
      "learning_rate": 2e-05,
      "loss": 0.8931,
      "step": 610
    },
    {
      "epoch": 0.039178515007898894,
      "grad_norm": 0.6058781147003174,
      "learning_rate": 2e-05,
      "loss": 0.915,
      "step": 620
    },
    {
      "epoch": 0.03981042654028436,
      "grad_norm": 0.579230010509491,
      "learning_rate": 2e-05,
      "loss": 0.9175,
      "step": 630
    },
    {
      "epoch": 0.04044233807266982,
      "grad_norm": 0.6037149429321289,
      "learning_rate": 2e-05,
      "loss": 0.9114,
      "step": 640
    },
    {
      "epoch": 0.04107424960505529,
      "grad_norm": 0.4986877739429474,
      "learning_rate": 2e-05,
      "loss": 0.901,
      "step": 650
    },
    {
      "epoch": 0.04170616113744076,
      "grad_norm": 0.6166608929634094,
      "learning_rate": 2e-05,
      "loss": 0.8761,
      "step": 660
    },
    {
      "epoch": 0.04233807266982623,
      "grad_norm": 0.552759051322937,
      "learning_rate": 2e-05,
      "loss": 0.9095,
      "step": 670
    },
    {
      "epoch": 0.04296998420221169,
      "grad_norm": 0.7433795928955078,
      "learning_rate": 2e-05,
      "loss": 0.9076,
      "step": 680
    },
    {
      "epoch": 0.043601895734597156,
      "grad_norm": 1.715183973312378,
      "learning_rate": 2e-05,
      "loss": 0.8737,
      "step": 690
    },
    {
      "epoch": 0.044233807266982623,
      "grad_norm": 0.9267755150794983,
      "learning_rate": 2e-05,
      "loss": 0.933,
      "step": 700
    },
    {
      "epoch": 0.04486571879936809,
      "grad_norm": 0.5714437365531921,
      "learning_rate": 2e-05,
      "loss": 0.8952,
      "step": 710
    },
    {
      "epoch": 0.04549763033175355,
      "grad_norm": 0.5523400902748108,
      "learning_rate": 2e-05,
      "loss": 0.9058,
      "step": 720
    },
    {
      "epoch": 0.04612954186413902,
      "grad_norm": 0.4988592267036438,
      "learning_rate": 2e-05,
      "loss": 0.8994,
      "step": 730
    },
    {
      "epoch": 0.04676145339652449,
      "grad_norm": 0.545521080493927,
      "learning_rate": 2e-05,
      "loss": 0.9049,
      "step": 740
    },
    {
      "epoch": 0.04739336492890995,
      "grad_norm": 0.5209084749221802,
      "learning_rate": 2e-05,
      "loss": 0.922,
      "step": 750
    },
    {
      "epoch": 0.04802527646129542,
      "grad_norm": 0.5220515131950378,
      "learning_rate": 2e-05,
      "loss": 0.9045,
      "step": 760
    },
    {
      "epoch": 0.048657187993680885,
      "grad_norm": 0.5615672469139099,
      "learning_rate": 2e-05,
      "loss": 0.9276,
      "step": 770
    },
    {
      "epoch": 0.04928909952606635,
      "grad_norm": 0.5151423215866089,
      "learning_rate": 2e-05,
      "loss": 0.8802,
      "step": 780
    },
    {
      "epoch": 0.049921011058451814,
      "grad_norm": 0.5320910811424255,
      "learning_rate": 2e-05,
      "loss": 0.8965,
      "step": 790
    },
    {
      "epoch": 0.05055292259083728,
      "grad_norm": 0.6256475448608398,
      "learning_rate": 2e-05,
      "loss": 0.8897,
      "step": 800
    },
    {
      "epoch": 0.05118483412322275,
      "grad_norm": 0.5061290264129639,
      "learning_rate": 2e-05,
      "loss": 0.8909,
      "step": 810
    },
    {
      "epoch": 0.05181674565560822,
      "grad_norm": 0.5187636017799377,
      "learning_rate": 2e-05,
      "loss": 0.9063,
      "step": 820
    },
    {
      "epoch": 0.05244865718799368,
      "grad_norm": 0.5689096450805664,
      "learning_rate": 2e-05,
      "loss": 0.9125,
      "step": 830
    },
    {
      "epoch": 0.05308056872037915,
      "grad_norm": 0.560541033744812,
      "learning_rate": 2e-05,
      "loss": 0.9129,
      "step": 840
    },
    {
      "epoch": 0.053712480252764615,
      "grad_norm": 0.5733119249343872,
      "learning_rate": 2e-05,
      "loss": 0.8772,
      "step": 850
    },
    {
      "epoch": 0.054344391785150076,
      "grad_norm": 0.5279077887535095,
      "learning_rate": 2e-05,
      "loss": 0.8883,
      "step": 860
    },
    {
      "epoch": 0.054976303317535544,
      "grad_norm": 0.5335824489593506,
      "learning_rate": 2e-05,
      "loss": 0.9002,
      "step": 870
    },
    {
      "epoch": 0.05560821484992101,
      "grad_norm": 0.5062702894210815,
      "learning_rate": 2e-05,
      "loss": 0.8792,
      "step": 880
    },
    {
      "epoch": 0.05624012638230648,
      "grad_norm": 0.5492991209030151,
      "learning_rate": 2e-05,
      "loss": 0.8784,
      "step": 890
    },
    {
      "epoch": 0.05687203791469194,
      "grad_norm": 0.6579878330230713,
      "learning_rate": 2e-05,
      "loss": 0.9241,
      "step": 900
    },
    {
      "epoch": 0.05750394944707741,
      "grad_norm": 0.551071286201477,
      "learning_rate": 2e-05,
      "loss": 0.9067,
      "step": 910
    },
    {
      "epoch": 0.058135860979462876,
      "grad_norm": 0.5038785934448242,
      "learning_rate": 2e-05,
      "loss": 0.9083,
      "step": 920
    },
    {
      "epoch": 0.058767772511848344,
      "grad_norm": 0.5862524509429932,
      "learning_rate": 2e-05,
      "loss": 0.902,
      "step": 930
    },
    {
      "epoch": 0.059399684044233805,
      "grad_norm": 0.5894129872322083,
      "learning_rate": 2e-05,
      "loss": 0.8934,
      "step": 940
    },
    {
      "epoch": 0.06003159557661927,
      "grad_norm": 0.9521624445915222,
      "learning_rate": 2e-05,
      "loss": 0.8954,
      "step": 950
    },
    {
      "epoch": 0.06066350710900474,
      "grad_norm": 0.5209932327270508,
      "learning_rate": 2e-05,
      "loss": 0.8926,
      "step": 960
    },
    {
      "epoch": 0.0612954186413902,
      "grad_norm": 0.5649769902229309,
      "learning_rate": 2e-05,
      "loss": 0.8958,
      "step": 970
    },
    {
      "epoch": 0.06192733017377567,
      "grad_norm": 0.5092157125473022,
      "learning_rate": 2e-05,
      "loss": 0.8748,
      "step": 980
    },
    {
      "epoch": 0.06255924170616113,
      "grad_norm": 0.5467744469642639,
      "learning_rate": 2e-05,
      "loss": 0.9092,
      "step": 990
    },
    {
      "epoch": 0.0631911532385466,
      "grad_norm": 0.6695379614830017,
      "learning_rate": 2e-05,
      "loss": 0.8781,
      "step": 1000
    },
    {
      "epoch": 0.06382306477093207,
      "grad_norm": 0.5359766483306885,
      "learning_rate": 2e-05,
      "loss": 0.8728,
      "step": 1010
    },
    {
      "epoch": 0.06445497630331753,
      "grad_norm": 0.6037816405296326,
      "learning_rate": 2e-05,
      "loss": 0.8961,
      "step": 1020
    },
    {
      "epoch": 0.065086887835703,
      "grad_norm": 0.5401719212532043,
      "learning_rate": 2e-05,
      "loss": 0.9079,
      "step": 1030
    },
    {
      "epoch": 0.06571879936808847,
      "grad_norm": 0.4804134964942932,
      "learning_rate": 2e-05,
      "loss": 0.9085,
      "step": 1040
    },
    {
      "epoch": 0.06635071090047394,
      "grad_norm": 0.5073590278625488,
      "learning_rate": 2e-05,
      "loss": 0.8795,
      "step": 1050
    },
    {
      "epoch": 0.06698262243285939,
      "grad_norm": 0.5325782895088196,
      "learning_rate": 2e-05,
      "loss": 0.8943,
      "step": 1060
    },
    {
      "epoch": 0.06761453396524486,
      "grad_norm": 0.5039036273956299,
      "learning_rate": 2e-05,
      "loss": 0.9027,
      "step": 1070
    },
    {
      "epoch": 0.06824644549763033,
      "grad_norm": 0.4937596321105957,
      "learning_rate": 2e-05,
      "loss": 0.8922,
      "step": 1080
    },
    {
      "epoch": 0.0688783570300158,
      "grad_norm": 0.5820568203926086,
      "learning_rate": 2e-05,
      "loss": 0.9117,
      "step": 1090
    },
    {
      "epoch": 0.06951026856240126,
      "grad_norm": 0.46752944588661194,
      "learning_rate": 2e-05,
      "loss": 0.8919,
      "step": 1100
    },
    {
      "epoch": 0.07014218009478673,
      "grad_norm": 0.5803933143615723,
      "learning_rate": 2e-05,
      "loss": 0.899,
      "step": 1110
    },
    {
      "epoch": 0.0707740916271722,
      "grad_norm": 0.5495235323905945,
      "learning_rate": 2e-05,
      "loss": 0.8801,
      "step": 1120
    },
    {
      "epoch": 0.07140600315955767,
      "grad_norm": 0.782866895198822,
      "learning_rate": 2e-05,
      "loss": 0.8784,
      "step": 1130
    },
    {
      "epoch": 0.07203791469194312,
      "grad_norm": 0.5584379434585571,
      "learning_rate": 2e-05,
      "loss": 0.881,
      "step": 1140
    },
    {
      "epoch": 0.07266982622432859,
      "grad_norm": 0.5009581446647644,
      "learning_rate": 2e-05,
      "loss": 0.9107,
      "step": 1150
    },
    {
      "epoch": 0.07330173775671406,
      "grad_norm": 0.4949404299259186,
      "learning_rate": 2e-05,
      "loss": 0.8749,
      "step": 1160
    },
    {
      "epoch": 0.07393364928909953,
      "grad_norm": 0.5240024924278259,
      "learning_rate": 2e-05,
      "loss": 0.8747,
      "step": 1170
    },
    {
      "epoch": 0.074565560821485,
      "grad_norm": 0.512729823589325,
      "learning_rate": 2e-05,
      "loss": 0.8802,
      "step": 1180
    },
    {
      "epoch": 0.07519747235387046,
      "grad_norm": 0.5106031894683838,
      "learning_rate": 2e-05,
      "loss": 0.9088,
      "step": 1190
    },
    {
      "epoch": 0.07582938388625593,
      "grad_norm": 0.4788902699947357,
      "learning_rate": 2e-05,
      "loss": 0.8883,
      "step": 1200
    },
    {
      "epoch": 0.07646129541864138,
      "grad_norm": 0.5151655077934265,
      "learning_rate": 2e-05,
      "loss": 0.8886,
      "step": 1210
    },
    {
      "epoch": 0.07709320695102685,
      "grad_norm": 0.4810257852077484,
      "learning_rate": 2e-05,
      "loss": 0.8771,
      "step": 1220
    },
    {
      "epoch": 0.07772511848341232,
      "grad_norm": 0.48933422565460205,
      "learning_rate": 2e-05,
      "loss": 0.8821,
      "step": 1230
    },
    {
      "epoch": 0.07835703001579779,
      "grad_norm": 0.490460067987442,
      "learning_rate": 2e-05,
      "loss": 0.8701,
      "step": 1240
    },
    {
      "epoch": 0.07898894154818326,
      "grad_norm": 0.5137897729873657,
      "learning_rate": 2e-05,
      "loss": 0.8711,
      "step": 1250
    },
    {
      "epoch": 0.07962085308056872,
      "grad_norm": 0.4710884094238281,
      "learning_rate": 2e-05,
      "loss": 0.879,
      "step": 1260
    },
    {
      "epoch": 0.08025276461295419,
      "grad_norm": 0.5222970843315125,
      "learning_rate": 2e-05,
      "loss": 0.8558,
      "step": 1270
    },
    {
      "epoch": 0.08088467614533965,
      "grad_norm": 0.4963923990726471,
      "learning_rate": 2e-05,
      "loss": 0.8669,
      "step": 1280
    },
    {
      "epoch": 0.08151658767772511,
      "grad_norm": 0.4823462665081024,
      "learning_rate": 2e-05,
      "loss": 0.8631,
      "step": 1290
    },
    {
      "epoch": 0.08214849921011058,
      "grad_norm": 0.5219413042068481,
      "learning_rate": 2e-05,
      "loss": 0.895,
      "step": 1300
    },
    {
      "epoch": 0.08278041074249605,
      "grad_norm": 0.4815572500228882,
      "learning_rate": 2e-05,
      "loss": 0.8885,
      "step": 1310
    },
    {
      "epoch": 0.08341232227488152,
      "grad_norm": 0.4959852397441864,
      "learning_rate": 2e-05,
      "loss": 0.8634,
      "step": 1320
    },
    {
      "epoch": 0.08404423380726699,
      "grad_norm": 0.544163703918457,
      "learning_rate": 2e-05,
      "loss": 0.8764,
      "step": 1330
    },
    {
      "epoch": 0.08467614533965245,
      "grad_norm": 0.5251185297966003,
      "learning_rate": 2e-05,
      "loss": 0.8929,
      "step": 1340
    },
    {
      "epoch": 0.08530805687203792,
      "grad_norm": 0.5277668237686157,
      "learning_rate": 2e-05,
      "loss": 0.8775,
      "step": 1350
    },
    {
      "epoch": 0.08593996840442338,
      "grad_norm": 0.6873008012771606,
      "learning_rate": 2e-05,
      "loss": 0.8739,
      "step": 1360
    },
    {
      "epoch": 0.08657187993680884,
      "grad_norm": 0.4932924211025238,
      "learning_rate": 2e-05,
      "loss": 0.8667,
      "step": 1370
    },
    {
      "epoch": 0.08720379146919431,
      "grad_norm": 0.5071529746055603,
      "learning_rate": 2e-05,
      "loss": 0.888,
      "step": 1380
    },
    {
      "epoch": 0.08783570300157978,
      "grad_norm": 0.5023147463798523,
      "learning_rate": 2e-05,
      "loss": 0.8588,
      "step": 1390
    },
    {
      "epoch": 0.08846761453396525,
      "grad_norm": 0.4720936119556427,
      "learning_rate": 2e-05,
      "loss": 0.8854,
      "step": 1400
    },
    {
      "epoch": 0.08909952606635071,
      "grad_norm": 0.4973188042640686,
      "learning_rate": 2e-05,
      "loss": 0.886,
      "step": 1410
    },
    {
      "epoch": 0.08973143759873618,
      "grad_norm": 0.5486796498298645,
      "learning_rate": 2e-05,
      "loss": 0.8873,
      "step": 1420
    },
    {
      "epoch": 0.09036334913112164,
      "grad_norm": 0.49072250723838806,
      "learning_rate": 2e-05,
      "loss": 0.8701,
      "step": 1430
    },
    {
      "epoch": 0.0909952606635071,
      "grad_norm": 1.431967854499817,
      "learning_rate": 2e-05,
      "loss": 0.8583,
      "step": 1440
    },
    {
      "epoch": 0.09162717219589257,
      "grad_norm": 0.5038327574729919,
      "learning_rate": 2e-05,
      "loss": 0.8692,
      "step": 1450
    },
    {
      "epoch": 0.09225908372827804,
      "grad_norm": 0.4960634112358093,
      "learning_rate": 2e-05,
      "loss": 0.8557,
      "step": 1460
    },
    {
      "epoch": 0.09289099526066351,
      "grad_norm": 0.4850906729698181,
      "learning_rate": 2e-05,
      "loss": 0.8622,
      "step": 1470
    },
    {
      "epoch": 0.09352290679304898,
      "grad_norm": 0.5151680111885071,
      "learning_rate": 2e-05,
      "loss": 0.8857,
      "step": 1480
    },
    {
      "epoch": 0.09415481832543444,
      "grad_norm": 0.7864753603935242,
      "learning_rate": 2e-05,
      "loss": 0.8856,
      "step": 1490
    },
    {
      "epoch": 0.0947867298578199,
      "grad_norm": 0.495161235332489,
      "learning_rate": 2e-05,
      "loss": 0.9062,
      "step": 1500
    },
    {
      "epoch": 0.09541864139020537,
      "grad_norm": 0.5123103857040405,
      "learning_rate": 2e-05,
      "loss": 0.9329,
      "step": 1510
    },
    {
      "epoch": 0.09605055292259083,
      "grad_norm": 0.5555689334869385,
      "learning_rate": 2e-05,
      "loss": 0.8734,
      "step": 1520
    },
    {
      "epoch": 0.0966824644549763,
      "grad_norm": 0.4570164382457733,
      "learning_rate": 2e-05,
      "loss": 0.873,
      "step": 1530
    },
    {
      "epoch": 0.09731437598736177,
      "grad_norm": 0.4710372984409332,
      "learning_rate": 2e-05,
      "loss": 0.8746,
      "step": 1540
    },
    {
      "epoch": 0.09794628751974724,
      "grad_norm": 0.4700137972831726,
      "learning_rate": 2e-05,
      "loss": 0.8844,
      "step": 1550
    },
    {
      "epoch": 0.0985781990521327,
      "grad_norm": 0.5475497245788574,
      "learning_rate": 2e-05,
      "loss": 0.8901,
      "step": 1560
    },
    {
      "epoch": 0.09921011058451817,
      "grad_norm": 0.5088175535202026,
      "learning_rate": 2e-05,
      "loss": 0.8972,
      "step": 1570
    },
    {
      "epoch": 0.09984202211690363,
      "grad_norm": 0.48973360657691956,
      "learning_rate": 2e-05,
      "loss": 0.8805,
      "step": 1580
    },
    {
      "epoch": 0.1004739336492891,
      "grad_norm": 0.48626893758773804,
      "learning_rate": 2e-05,
      "loss": 0.9094,
      "step": 1590
    },
    {
      "epoch": 0.10110584518167456,
      "grad_norm": 0.4822232127189636,
      "learning_rate": 2e-05,
      "loss": 0.8793,
      "step": 1600
    },
    {
      "epoch": 0.10173775671406003,
      "grad_norm": 0.4699007570743561,
      "learning_rate": 2e-05,
      "loss": 0.8973,
      "step": 1610
    },
    {
      "epoch": 0.1023696682464455,
      "grad_norm": 0.5640358328819275,
      "learning_rate": 2e-05,
      "loss": 0.8687,
      "step": 1620
    },
    {
      "epoch": 0.10300157977883097,
      "grad_norm": 0.5482319593429565,
      "learning_rate": 2e-05,
      "loss": 0.8788,
      "step": 1630
    },
    {
      "epoch": 0.10363349131121644,
      "grad_norm": 0.48495763540267944,
      "learning_rate": 2e-05,
      "loss": 0.8902,
      "step": 1640
    },
    {
      "epoch": 0.10426540284360189,
      "grad_norm": 0.4935813248157501,
      "learning_rate": 2e-05,
      "loss": 0.8666,
      "step": 1650
    },
    {
      "epoch": 0.10489731437598736,
      "grad_norm": 0.5016494989395142,
      "learning_rate": 2e-05,
      "loss": 0.8922,
      "step": 1660
    },
    {
      "epoch": 0.10552922590837283,
      "grad_norm": 0.5238736867904663,
      "learning_rate": 2e-05,
      "loss": 0.881,
      "step": 1670
    },
    {
      "epoch": 0.1061611374407583,
      "grad_norm": 0.48621994256973267,
      "learning_rate": 2e-05,
      "loss": 0.8633,
      "step": 1680
    },
    {
      "epoch": 0.10679304897314376,
      "grad_norm": 0.49206703901290894,
      "learning_rate": 2e-05,
      "loss": 0.88,
      "step": 1690
    },
    {
      "epoch": 0.10742496050552923,
      "grad_norm": 0.4722195565700531,
      "learning_rate": 2e-05,
      "loss": 0.8716,
      "step": 1700
    },
    {
      "epoch": 0.1080568720379147,
      "grad_norm": 0.481782466173172,
      "learning_rate": 2e-05,
      "loss": 0.8767,
      "step": 1710
    },
    {
      "epoch": 0.10868878357030015,
      "grad_norm": 0.4795779883861542,
      "learning_rate": 2e-05,
      "loss": 0.8807,
      "step": 1720
    },
    {
      "epoch": 0.10932069510268562,
      "grad_norm": 0.5058030486106873,
      "learning_rate": 2e-05,
      "loss": 0.8759,
      "step": 1730
    },
    {
      "epoch": 0.10995260663507109,
      "grad_norm": 0.46553274989128113,
      "learning_rate": 2e-05,
      "loss": 0.8767,
      "step": 1740
    },
    {
      "epoch": 0.11058451816745656,
      "grad_norm": 0.5184619426727295,
      "learning_rate": 2e-05,
      "loss": 0.8625,
      "step": 1750
    },
    {
      "epoch": 0.11121642969984202,
      "grad_norm": 0.48755067586898804,
      "learning_rate": 2e-05,
      "loss": 0.8697,
      "step": 1760
    },
    {
      "epoch": 0.11184834123222749,
      "grad_norm": 0.4707954525947571,
      "learning_rate": 2e-05,
      "loss": 0.8679,
      "step": 1770
    },
    {
      "epoch": 0.11248025276461296,
      "grad_norm": 0.5027987957000732,
      "learning_rate": 2e-05,
      "loss": 0.8993,
      "step": 1780
    },
    {
      "epoch": 0.11311216429699843,
      "grad_norm": 1.081098198890686,
      "learning_rate": 2e-05,
      "loss": 0.8824,
      "step": 1790
    },
    {
      "epoch": 0.11374407582938388,
      "grad_norm": 0.4826985001564026,
      "learning_rate": 2e-05,
      "loss": 0.8703,
      "step": 1800
    },
    {
      "epoch": 0.11437598736176935,
      "grad_norm": 0.45513176918029785,
      "learning_rate": 2e-05,
      "loss": 0.8467,
      "step": 1810
    },
    {
      "epoch": 0.11500789889415482,
      "grad_norm": 0.7291151881217957,
      "learning_rate": 2e-05,
      "loss": 0.8607,
      "step": 1820
    },
    {
      "epoch": 0.11563981042654028,
      "grad_norm": 0.48444896936416626,
      "learning_rate": 2e-05,
      "loss": 0.8744,
      "step": 1830
    },
    {
      "epoch": 0.11627172195892575,
      "grad_norm": 0.46754005551338196,
      "learning_rate": 2e-05,
      "loss": 0.8714,
      "step": 1840
    },
    {
      "epoch": 0.11690363349131122,
      "grad_norm": 0.47295942902565,
      "learning_rate": 2e-05,
      "loss": 0.8741,
      "step": 1850
    },
    {
      "epoch": 0.11753554502369669,
      "grad_norm": 0.5298379063606262,
      "learning_rate": 2e-05,
      "loss": 0.8871,
      "step": 1860
    },
    {
      "epoch": 0.11816745655608214,
      "grad_norm": 0.49541521072387695,
      "learning_rate": 2e-05,
      "loss": 0.8696,
      "step": 1870
    },
    {
      "epoch": 0.11879936808846761,
      "grad_norm": 0.5147435665130615,
      "learning_rate": 2e-05,
      "loss": 0.8838,
      "step": 1880
    },
    {
      "epoch": 0.11943127962085308,
      "grad_norm": 0.46245381236076355,
      "learning_rate": 2e-05,
      "loss": 0.8781,
      "step": 1890
    },
    {
      "epoch": 0.12006319115323855,
      "grad_norm": 0.49722954630851746,
      "learning_rate": 2e-05,
      "loss": 0.875,
      "step": 1900
    },
    {
      "epoch": 0.12069510268562401,
      "grad_norm": 0.49780702590942383,
      "learning_rate": 2e-05,
      "loss": 0.8673,
      "step": 1910
    },
    {
      "epoch": 0.12132701421800948,
      "grad_norm": 0.4821648895740509,
      "learning_rate": 2e-05,
      "loss": 0.8983,
      "step": 1920
    },
    {
      "epoch": 0.12195892575039495,
      "grad_norm": 0.4544130563735962,
      "learning_rate": 2e-05,
      "loss": 0.8435,
      "step": 1930
    },
    {
      "epoch": 0.1225908372827804,
      "grad_norm": 0.5020303726196289,
      "learning_rate": 2e-05,
      "loss": 0.8582,
      "step": 1940
    },
    {
      "epoch": 0.12322274881516587,
      "grad_norm": 0.5067219734191895,
      "learning_rate": 2e-05,
      "loss": 0.8597,
      "step": 1950
    },
    {
      "epoch": 0.12385466034755134,
      "grad_norm": 0.46906647086143494,
      "learning_rate": 2e-05,
      "loss": 0.8571,
      "step": 1960
    },
    {
      "epoch": 0.12448657187993681,
      "grad_norm": 0.46474966406822205,
      "learning_rate": 2e-05,
      "loss": 0.875,
      "step": 1970
    },
    {
      "epoch": 0.12511848341232226,
      "grad_norm": 0.4918360114097595,
      "learning_rate": 2e-05,
      "loss": 0.8388,
      "step": 1980
    },
    {
      "epoch": 0.12575039494470774,
      "grad_norm": 0.4889756441116333,
      "learning_rate": 2e-05,
      "loss": 0.8686,
      "step": 1990
    },
    {
      "epoch": 0.1263823064770932,
      "grad_norm": 0.4613496661186218,
      "learning_rate": 2e-05,
      "loss": 0.8783,
      "step": 2000
    },
    {
      "epoch": 0.12701421800947868,
      "grad_norm": 0.45228028297424316,
      "learning_rate": 2e-05,
      "loss": 0.8963,
      "step": 2010
    },
    {
      "epoch": 0.12764612954186413,
      "grad_norm": 0.5558086037635803,
      "learning_rate": 2e-05,
      "loss": 0.8867,
      "step": 2020
    },
    {
      "epoch": 0.12827804107424962,
      "grad_norm": 0.4394373893737793,
      "learning_rate": 2e-05,
      "loss": 0.8813,
      "step": 2030
    },
    {
      "epoch": 0.12890995260663507,
      "grad_norm": 0.45812109112739563,
      "learning_rate": 2e-05,
      "loss": 0.8437,
      "step": 2040
    },
    {
      "epoch": 0.12954186413902052,
      "grad_norm": 0.5066508650779724,
      "learning_rate": 2e-05,
      "loss": 0.8666,
      "step": 2050
    },
    {
      "epoch": 0.130173775671406,
      "grad_norm": 0.9053667783737183,
      "learning_rate": 2e-05,
      "loss": 0.8664,
      "step": 2060
    },
    {
      "epoch": 0.13080568720379146,
      "grad_norm": 0.46928325295448303,
      "learning_rate": 2e-05,
      "loss": 0.8862,
      "step": 2070
    },
    {
      "epoch": 0.13143759873617694,
      "grad_norm": 0.4615163803100586,
      "learning_rate": 2e-05,
      "loss": 0.8596,
      "step": 2080
    },
    {
      "epoch": 0.1320695102685624,
      "grad_norm": 0.4925185739994049,
      "learning_rate": 2e-05,
      "loss": 0.8894,
      "step": 2090
    },
    {
      "epoch": 0.13270142180094788,
      "grad_norm": 0.4613633453845978,
      "learning_rate": 2e-05,
      "loss": 0.8629,
      "step": 2100
    },
    {
      "epoch": 0.13333333333333333,
      "grad_norm": 0.5064405202865601,
      "learning_rate": 2e-05,
      "loss": 0.8629,
      "step": 2110
    },
    {
      "epoch": 0.13396524486571879,
      "grad_norm": 0.498585045337677,
      "learning_rate": 2e-05,
      "loss": 0.8872,
      "step": 2120
    },
    {
      "epoch": 0.13459715639810427,
      "grad_norm": 0.4647330343723297,
      "learning_rate": 2e-05,
      "loss": 0.8606,
      "step": 2130
    },
    {
      "epoch": 0.13522906793048972,
      "grad_norm": 0.48744726181030273,
      "learning_rate": 2e-05,
      "loss": 0.8506,
      "step": 2140
    },
    {
      "epoch": 0.1358609794628752,
      "grad_norm": 0.4956341087818146,
      "learning_rate": 2e-05,
      "loss": 0.8545,
      "step": 2150
    },
    {
      "epoch": 0.13649289099526066,
      "grad_norm": 0.6944047808647156,
      "learning_rate": 2e-05,
      "loss": 0.8808,
      "step": 2160
    },
    {
      "epoch": 0.13712480252764614,
      "grad_norm": 0.5086840987205505,
      "learning_rate": 2e-05,
      "loss": 0.8589,
      "step": 2170
    },
    {
      "epoch": 0.1377567140600316,
      "grad_norm": 0.4839428961277008,
      "learning_rate": 2e-05,
      "loss": 0.8821,
      "step": 2180
    },
    {
      "epoch": 0.13838862559241707,
      "grad_norm": 0.5878944396972656,
      "learning_rate": 2e-05,
      "loss": 0.8815,
      "step": 2190
    },
    {
      "epoch": 0.13902053712480253,
      "grad_norm": 0.6157777905464172,
      "learning_rate": 2e-05,
      "loss": 0.8584,
      "step": 2200
    },
    {
      "epoch": 0.13965244865718798,
      "grad_norm": 0.48504042625427246,
      "learning_rate": 2e-05,
      "loss": 0.8624,
      "step": 2210
    },
    {
      "epoch": 0.14028436018957346,
      "grad_norm": 0.4809598922729492,
      "learning_rate": 2e-05,
      "loss": 0.8813,
      "step": 2220
    },
    {
      "epoch": 0.14091627172195892,
      "grad_norm": 0.4918782711029053,
      "learning_rate": 2e-05,
      "loss": 0.8752,
      "step": 2230
    },
    {
      "epoch": 0.1415481832543444,
      "grad_norm": 0.4730665683746338,
      "learning_rate": 2e-05,
      "loss": 0.8701,
      "step": 2240
    },
    {
      "epoch": 0.14218009478672985,
      "grad_norm": 0.486869752407074,
      "learning_rate": 2e-05,
      "loss": 0.8854,
      "step": 2250
    },
    {
      "epoch": 0.14281200631911534,
      "grad_norm": 0.4495904743671417,
      "learning_rate": 2e-05,
      "loss": 0.8371,
      "step": 2260
    },
    {
      "epoch": 0.1434439178515008,
      "grad_norm": 0.4672189950942993,
      "learning_rate": 2e-05,
      "loss": 0.8603,
      "step": 2270
    },
    {
      "epoch": 0.14407582938388624,
      "grad_norm": 0.48257431387901306,
      "learning_rate": 2e-05,
      "loss": 0.8676,
      "step": 2280
    },
    {
      "epoch": 0.14470774091627173,
      "grad_norm": 0.4934057593345642,
      "learning_rate": 2e-05,
      "loss": 0.8736,
      "step": 2290
    },
    {
      "epoch": 0.14533965244865718,
      "grad_norm": 0.45726630091667175,
      "learning_rate": 2e-05,
      "loss": 0.8797,
      "step": 2300
    },
    {
      "epoch": 0.14597156398104266,
      "grad_norm": 0.4822031557559967,
      "learning_rate": 2e-05,
      "loss": 0.8741,
      "step": 2310
    },
    {
      "epoch": 0.14660347551342812,
      "grad_norm": 0.5092865824699402,
      "learning_rate": 2e-05,
      "loss": 0.8582,
      "step": 2320
    },
    {
      "epoch": 0.1472353870458136,
      "grad_norm": 1.198046326637268,
      "learning_rate": 2e-05,
      "loss": 0.8871,
      "step": 2330
    },
    {
      "epoch": 0.14786729857819905,
      "grad_norm": 0.6455650329589844,
      "learning_rate": 2e-05,
      "loss": 0.8568,
      "step": 2340
    },
    {
      "epoch": 0.1484992101105845,
      "grad_norm": 0.46702516078948975,
      "learning_rate": 2e-05,
      "loss": 0.8571,
      "step": 2350
    },
    {
      "epoch": 0.14913112164297,
      "grad_norm": 0.48596251010894775,
      "learning_rate": 2e-05,
      "loss": 0.8449,
      "step": 2360
    },
    {
      "epoch": 0.14976303317535544,
      "grad_norm": 0.4743686318397522,
      "learning_rate": 2e-05,
      "loss": 0.8634,
      "step": 2370
    },
    {
      "epoch": 0.15039494470774092,
      "grad_norm": 0.4790477752685547,
      "learning_rate": 2e-05,
      "loss": 0.8731,
      "step": 2380
    },
    {
      "epoch": 0.15102685624012638,
      "grad_norm": 0.494199275970459,
      "learning_rate": 2e-05,
      "loss": 0.8633,
      "step": 2390
    },
    {
      "epoch": 0.15165876777251186,
      "grad_norm": 0.4832230508327484,
      "learning_rate": 2e-05,
      "loss": 0.8718,
      "step": 2400
    },
    {
      "epoch": 0.15229067930489731,
      "grad_norm": 0.4704072177410126,
      "learning_rate": 2e-05,
      "loss": 0.8583,
      "step": 2410
    },
    {
      "epoch": 0.15292259083728277,
      "grad_norm": 0.5898045301437378,
      "learning_rate": 2e-05,
      "loss": 0.852,
      "step": 2420
    },
    {
      "epoch": 0.15355450236966825,
      "grad_norm": 0.5079308152198792,
      "learning_rate": 2e-05,
      "loss": 0.8638,
      "step": 2430
    },
    {
      "epoch": 0.1541864139020537,
      "grad_norm": 0.4585856795310974,
      "learning_rate": 2e-05,
      "loss": 0.8617,
      "step": 2440
    },
    {
      "epoch": 0.15481832543443919,
      "grad_norm": 0.5012873411178589,
      "learning_rate": 2e-05,
      "loss": 0.8591,
      "step": 2450
    },
    {
      "epoch": 0.15545023696682464,
      "grad_norm": 0.4680193066596985,
      "learning_rate": 2e-05,
      "loss": 0.8437,
      "step": 2460
    },
    {
      "epoch": 0.15608214849921012,
      "grad_norm": 0.5058935880661011,
      "learning_rate": 2e-05,
      "loss": 0.8758,
      "step": 2470
    },
    {
      "epoch": 0.15671406003159558,
      "grad_norm": 0.4877486228942871,
      "learning_rate": 2e-05,
      "loss": 0.8591,
      "step": 2480
    },
    {
      "epoch": 0.15734597156398103,
      "grad_norm": 0.5519242286682129,
      "learning_rate": 2e-05,
      "loss": 0.8508,
      "step": 2490
    },
    {
      "epoch": 0.1579778830963665,
      "grad_norm": 0.4628176689147949,
      "learning_rate": 2e-05,
      "loss": 0.8561,
      "step": 2500
    },
    {
      "epoch": 0.15860979462875197,
      "grad_norm": 0.5179667472839355,
      "learning_rate": 2e-05,
      "loss": 0.8499,
      "step": 2510
    },
    {
      "epoch": 0.15924170616113745,
      "grad_norm": 0.5802075266838074,
      "learning_rate": 2e-05,
      "loss": 0.8727,
      "step": 2520
    },
    {
      "epoch": 0.1598736176935229,
      "grad_norm": 0.7243200540542603,
      "learning_rate": 2e-05,
      "loss": 0.8497,
      "step": 2530
    },
    {
      "epoch": 0.16050552922590838,
      "grad_norm": 0.4918075203895569,
      "learning_rate": 2e-05,
      "loss": 0.8735,
      "step": 2540
    },
    {
      "epoch": 0.16113744075829384,
      "grad_norm": 0.46644020080566406,
      "learning_rate": 2e-05,
      "loss": 0.8321,
      "step": 2550
    },
    {
      "epoch": 0.1617693522906793,
      "grad_norm": 0.5539456605911255,
      "learning_rate": 2e-05,
      "loss": 0.8724,
      "step": 2560
    },
    {
      "epoch": 0.16240126382306477,
      "grad_norm": 0.47058072686195374,
      "learning_rate": 2e-05,
      "loss": 0.8514,
      "step": 2570
    },
    {
      "epoch": 0.16303317535545023,
      "grad_norm": 0.4589993953704834,
      "learning_rate": 2e-05,
      "loss": 0.839,
      "step": 2580
    },
    {
      "epoch": 0.1636650868878357,
      "grad_norm": 0.5339471697807312,
      "learning_rate": 2e-05,
      "loss": 0.8624,
      "step": 2590
    },
    {
      "epoch": 0.16429699842022116,
      "grad_norm": 0.4321252405643463,
      "learning_rate": 2e-05,
      "loss": 0.8558,
      "step": 2600
    },
    {
      "epoch": 0.16492890995260664,
      "grad_norm": 0.45498740673065186,
      "learning_rate": 2e-05,
      "loss": 0.8554,
      "step": 2610
    },
    {
      "epoch": 0.1655608214849921,
      "grad_norm": 0.48426494002342224,
      "learning_rate": 2e-05,
      "loss": 0.8652,
      "step": 2620
    },
    {
      "epoch": 0.16619273301737758,
      "grad_norm": 0.4380468428134918,
      "learning_rate": 2e-05,
      "loss": 0.8462,
      "step": 2630
    },
    {
      "epoch": 0.16682464454976303,
      "grad_norm": 0.48135820031166077,
      "learning_rate": 2e-05,
      "loss": 0.8663,
      "step": 2640
    },
    {
      "epoch": 0.1674565560821485,
      "grad_norm": 0.47653424739837646,
      "learning_rate": 2e-05,
      "loss": 0.8902,
      "step": 2650
    },
    {
      "epoch": 0.16808846761453397,
      "grad_norm": 0.500487208366394,
      "learning_rate": 2e-05,
      "loss": 0.8455,
      "step": 2660
    },
    {
      "epoch": 0.16872037914691942,
      "grad_norm": 0.45171165466308594,
      "learning_rate": 2e-05,
      "loss": 0.8536,
      "step": 2670
    },
    {
      "epoch": 0.1693522906793049,
      "grad_norm": 0.46864235401153564,
      "learning_rate": 2e-05,
      "loss": 0.8561,
      "step": 2680
    },
    {
      "epoch": 0.16998420221169036,
      "grad_norm": 0.43721145391464233,
      "learning_rate": 2e-05,
      "loss": 0.86,
      "step": 2690
    },
    {
      "epoch": 0.17061611374407584,
      "grad_norm": 0.5098836421966553,
      "learning_rate": 2e-05,
      "loss": 0.8558,
      "step": 2700
    },
    {
      "epoch": 0.1712480252764613,
      "grad_norm": 0.4822649657726288,
      "learning_rate": 2e-05,
      "loss": 0.8546,
      "step": 2710
    },
    {
      "epoch": 0.17187993680884675,
      "grad_norm": 0.5005358457565308,
      "learning_rate": 2e-05,
      "loss": 0.8506,
      "step": 2720
    },
    {
      "epoch": 0.17251184834123223,
      "grad_norm": 0.4592657685279846,
      "learning_rate": 2e-05,
      "loss": 0.8796,
      "step": 2730
    },
    {
      "epoch": 0.1731437598736177,
      "grad_norm": 0.47711560130119324,
      "learning_rate": 2e-05,
      "loss": 0.8481,
      "step": 2740
    },
    {
      "epoch": 0.17377567140600317,
      "grad_norm": 0.45625171065330505,
      "learning_rate": 2e-05,
      "loss": 0.8605,
      "step": 2750
    },
    {
      "epoch": 0.17440758293838862,
      "grad_norm": 0.4686576724052429,
      "learning_rate": 2e-05,
      "loss": 0.866,
      "step": 2760
    },
    {
      "epoch": 0.1750394944707741,
      "grad_norm": 0.46892648935317993,
      "learning_rate": 2e-05,
      "loss": 0.8714,
      "step": 2770
    },
    {
      "epoch": 0.17567140600315956,
      "grad_norm": 0.46695664525032043,
      "learning_rate": 2e-05,
      "loss": 0.8551,
      "step": 2780
    },
    {
      "epoch": 0.176303317535545,
      "grad_norm": 0.4646821618080139,
      "learning_rate": 2e-05,
      "loss": 0.8617,
      "step": 2790
    },
    {
      "epoch": 0.1769352290679305,
      "grad_norm": 0.45751631259918213,
      "learning_rate": 2e-05,
      "loss": 0.8472,
      "step": 2800
    },
    {
      "epoch": 0.17756714060031595,
      "grad_norm": 0.46256089210510254,
      "learning_rate": 2e-05,
      "loss": 0.859,
      "step": 2810
    },
    {
      "epoch": 0.17819905213270143,
      "grad_norm": 0.4825209975242615,
      "learning_rate": 2e-05,
      "loss": 0.8677,
      "step": 2820
    },
    {
      "epoch": 0.17883096366508688,
      "grad_norm": 0.5022207498550415,
      "learning_rate": 2e-05,
      "loss": 0.8619,
      "step": 2830
    },
    {
      "epoch": 0.17946287519747237,
      "grad_norm": 0.6094061732292175,
      "learning_rate": 2e-05,
      "loss": 0.8822,
      "step": 2840
    },
    {
      "epoch": 0.18009478672985782,
      "grad_norm": 0.47439998388290405,
      "learning_rate": 2e-05,
      "loss": 0.8874,
      "step": 2850
    },
    {
      "epoch": 0.18072669826224327,
      "grad_norm": 0.4702276885509491,
      "learning_rate": 2e-05,
      "loss": 0.873,
      "step": 2860
    },
    {
      "epoch": 0.18135860979462876,
      "grad_norm": 0.48905500769615173,
      "learning_rate": 2e-05,
      "loss": 0.8734,
      "step": 2870
    },
    {
      "epoch": 0.1819905213270142,
      "grad_norm": 0.4468541145324707,
      "learning_rate": 2e-05,
      "loss": 0.8928,
      "step": 2880
    },
    {
      "epoch": 0.1826224328593997,
      "grad_norm": 0.46830663084983826,
      "learning_rate": 2e-05,
      "loss": 0.8515,
      "step": 2890
    },
    {
      "epoch": 0.18325434439178515,
      "grad_norm": 0.48298606276512146,
      "learning_rate": 2e-05,
      "loss": 0.8377,
      "step": 2900
    },
    {
      "epoch": 0.18388625592417063,
      "grad_norm": 0.48953694105148315,
      "learning_rate": 2e-05,
      "loss": 0.8343,
      "step": 2910
    },
    {
      "epoch": 0.18451816745655608,
      "grad_norm": 0.47028350830078125,
      "learning_rate": 2e-05,
      "loss": 0.8276,
      "step": 2920
    },
    {
      "epoch": 0.18515007898894154,
      "grad_norm": 0.5009111166000366,
      "learning_rate": 2e-05,
      "loss": 0.8575,
      "step": 2930
    },
    {
      "epoch": 0.18578199052132702,
      "grad_norm": 0.4518514573574066,
      "learning_rate": 2e-05,
      "loss": 0.871,
      "step": 2940
    },
    {
      "epoch": 0.18641390205371247,
      "grad_norm": 0.4411482810974121,
      "learning_rate": 2e-05,
      "loss": 0.8767,
      "step": 2950
    },
    {
      "epoch": 0.18704581358609795,
      "grad_norm": 0.45347458124160767,
      "learning_rate": 2e-05,
      "loss": 0.845,
      "step": 2960
    },
    {
      "epoch": 0.1876777251184834,
      "grad_norm": 0.5517627596855164,
      "learning_rate": 2e-05,
      "loss": 0.8618,
      "step": 2970
    },
    {
      "epoch": 0.1883096366508689,
      "grad_norm": 0.4578409790992737,
      "learning_rate": 2e-05,
      "loss": 0.8577,
      "step": 2980
    },
    {
      "epoch": 0.18894154818325434,
      "grad_norm": 0.5405149459838867,
      "learning_rate": 2e-05,
      "loss": 0.867,
      "step": 2990
    },
    {
      "epoch": 0.1895734597156398,
      "grad_norm": 0.4625633656978607,
      "learning_rate": 2e-05,
      "loss": 0.8738,
      "step": 3000
    },
    {
      "epoch": 0.19020537124802528,
      "grad_norm": 0.42636358737945557,
      "learning_rate": 2e-05,
      "loss": 0.8642,
      "step": 3010
    },
    {
      "epoch": 0.19083728278041073,
      "grad_norm": 0.5234770774841309,
      "learning_rate": 2e-05,
      "loss": 0.872,
      "step": 3020
    },
    {
      "epoch": 0.19146919431279621,
      "grad_norm": 0.4650782346725464,
      "learning_rate": 2e-05,
      "loss": 0.8335,
      "step": 3030
    },
    {
      "epoch": 0.19210110584518167,
      "grad_norm": 0.4408527612686157,
      "learning_rate": 2e-05,
      "loss": 0.8441,
      "step": 3040
    },
    {
      "epoch": 0.19273301737756715,
      "grad_norm": 0.4782986640930176,
      "learning_rate": 2e-05,
      "loss": 0.8862,
      "step": 3050
    },
    {
      "epoch": 0.1933649289099526,
      "grad_norm": 0.7020520567893982,
      "learning_rate": 2e-05,
      "loss": 0.8688,
      "step": 3060
    },
    {
      "epoch": 0.19399684044233806,
      "grad_norm": 0.45431479811668396,
      "learning_rate": 2e-05,
      "loss": 0.8571,
      "step": 3070
    },
    {
      "epoch": 0.19462875197472354,
      "grad_norm": 0.5147340297698975,
      "learning_rate": 2e-05,
      "loss": 0.8307,
      "step": 3080
    },
    {
      "epoch": 0.195260663507109,
      "grad_norm": 0.46793535351753235,
      "learning_rate": 2e-05,
      "loss": 0.8625,
      "step": 3090
    },
    {
      "epoch": 0.19589257503949448,
      "grad_norm": 0.4431324899196625,
      "learning_rate": 2e-05,
      "loss": 0.8632,
      "step": 3100
    },
    {
      "epoch": 0.19652448657187993,
      "grad_norm": 0.45821258425712585,
      "learning_rate": 2e-05,
      "loss": 0.8683,
      "step": 3110
    },
    {
      "epoch": 0.1971563981042654,
      "grad_norm": 0.4287582337856293,
      "learning_rate": 2e-05,
      "loss": 0.8222,
      "step": 3120
    },
    {
      "epoch": 0.19778830963665087,
      "grad_norm": 0.5040216445922852,
      "learning_rate": 2e-05,
      "loss": 0.8486,
      "step": 3130
    },
    {
      "epoch": 0.19842022116903635,
      "grad_norm": 0.46887287497520447,
      "learning_rate": 2e-05,
      "loss": 0.8762,
      "step": 3140
    },
    {
      "epoch": 0.1990521327014218,
      "grad_norm": 0.45551660656929016,
      "learning_rate": 2e-05,
      "loss": 0.8453,
      "step": 3150
    },
    {
      "epoch": 0.19968404423380726,
      "grad_norm": 0.4785887598991394,
      "learning_rate": 2e-05,
      "loss": 0.8553,
      "step": 3160
    },
    {
      "epoch": 0.20031595576619274,
      "grad_norm": 0.4773639142513275,
      "learning_rate": 2e-05,
      "loss": 0.8853,
      "step": 3170
    },
    {
      "epoch": 0.2009478672985782,
      "grad_norm": 0.48274144530296326,
      "learning_rate": 2e-05,
      "loss": 0.8519,
      "step": 3180
    },
    {
      "epoch": 0.20157977883096367,
      "grad_norm": 0.4556339681148529,
      "learning_rate": 2e-05,
      "loss": 0.8702,
      "step": 3190
    },
    {
      "epoch": 0.20221169036334913,
      "grad_norm": 0.5082745552062988,
      "learning_rate": 2e-05,
      "loss": 0.862,
      "step": 3200
    },
    {
      "epoch": 0.2028436018957346,
      "grad_norm": 0.4261406660079956,
      "learning_rate": 2e-05,
      "loss": 0.8469,
      "step": 3210
    },
    {
      "epoch": 0.20347551342812006,
      "grad_norm": 0.5073183178901672,
      "learning_rate": 2e-05,
      "loss": 0.8745,
      "step": 3220
    },
    {
      "epoch": 0.20410742496050552,
      "grad_norm": 0.4219917953014374,
      "learning_rate": 2e-05,
      "loss": 0.856,
      "step": 3230
    },
    {
      "epoch": 0.204739336492891,
      "grad_norm": 0.5122772455215454,
      "learning_rate": 2e-05,
      "loss": 0.879,
      "step": 3240
    },
    {
      "epoch": 0.20537124802527645,
      "grad_norm": 0.4538480043411255,
      "learning_rate": 2e-05,
      "loss": 0.871,
      "step": 3250
    },
    {
      "epoch": 0.20600315955766194,
      "grad_norm": 0.4952961802482605,
      "learning_rate": 2e-05,
      "loss": 0.8136,
      "step": 3260
    },
    {
      "epoch": 0.2066350710900474,
      "grad_norm": 0.4794948399066925,
      "learning_rate": 2e-05,
      "loss": 0.8632,
      "step": 3270
    },
    {
      "epoch": 0.20726698262243287,
      "grad_norm": 0.45920589566230774,
      "learning_rate": 2e-05,
      "loss": 0.8552,
      "step": 3280
    },
    {
      "epoch": 0.20789889415481833,
      "grad_norm": 0.5482389330863953,
      "learning_rate": 2e-05,
      "loss": 0.8537,
      "step": 3290
    },
    {
      "epoch": 0.20853080568720378,
      "grad_norm": 0.5436051487922668,
      "learning_rate": 2e-05,
      "loss": 0.8337,
      "step": 3300
    },
    {
      "epoch": 0.20916271721958926,
      "grad_norm": 0.4739297330379486,
      "learning_rate": 2e-05,
      "loss": 0.8586,
      "step": 3310
    },
    {
      "epoch": 0.20979462875197472,
      "grad_norm": 0.4466475546360016,
      "learning_rate": 2e-05,
      "loss": 0.847,
      "step": 3320
    },
    {
      "epoch": 0.2104265402843602,
      "grad_norm": 0.4614415466785431,
      "learning_rate": 2e-05,
      "loss": 0.8441,
      "step": 3330
    },
    {
      "epoch": 0.21105845181674565,
      "grad_norm": 0.46833336353302,
      "learning_rate": 2e-05,
      "loss": 0.8247,
      "step": 3340
    },
    {
      "epoch": 0.21169036334913113,
      "grad_norm": 0.48644742369651794,
      "learning_rate": 2e-05,
      "loss": 0.8416,
      "step": 3350
    },
    {
      "epoch": 0.2123222748815166,
      "grad_norm": 0.8044328093528748,
      "learning_rate": 2e-05,
      "loss": 0.8555,
      "step": 3360
    },
    {
      "epoch": 0.21295418641390204,
      "grad_norm": 0.5560572743415833,
      "learning_rate": 2e-05,
      "loss": 0.8594,
      "step": 3370
    },
    {
      "epoch": 0.21358609794628752,
      "grad_norm": 0.46371909976005554,
      "learning_rate": 2e-05,
      "loss": 0.843,
      "step": 3380
    },
    {
      "epoch": 0.21421800947867298,
      "grad_norm": 0.5273845791816711,
      "learning_rate": 2e-05,
      "loss": 0.8457,
      "step": 3390
    },
    {
      "epoch": 0.21484992101105846,
      "grad_norm": 0.4683104455471039,
      "learning_rate": 2e-05,
      "loss": 0.8649,
      "step": 3400
    },
    {
      "epoch": 0.2154818325434439,
      "grad_norm": 0.47392627596855164,
      "learning_rate": 2e-05,
      "loss": 0.8461,
      "step": 3410
    },
    {
      "epoch": 0.2161137440758294,
      "grad_norm": 0.4864743947982788,
      "learning_rate": 2e-05,
      "loss": 0.8626,
      "step": 3420
    },
    {
      "epoch": 0.21674565560821485,
      "grad_norm": 0.46061548590660095,
      "learning_rate": 2e-05,
      "loss": 0.847,
      "step": 3430
    },
    {
      "epoch": 0.2173775671406003,
      "grad_norm": 0.5065048933029175,
      "learning_rate": 2e-05,
      "loss": 0.8423,
      "step": 3440
    },
    {
      "epoch": 0.21800947867298578,
      "grad_norm": 0.5027605891227722,
      "learning_rate": 2e-05,
      "loss": 0.8625,
      "step": 3450
    },
    {
      "epoch": 0.21864139020537124,
      "grad_norm": 0.47832682728767395,
      "learning_rate": 2e-05,
      "loss": 0.8629,
      "step": 3460
    },
    {
      "epoch": 0.21927330173775672,
      "grad_norm": 0.5731452703475952,
      "learning_rate": 2e-05,
      "loss": 0.8488,
      "step": 3470
    },
    {
      "epoch": 0.21990521327014217,
      "grad_norm": 0.5229271650314331,
      "learning_rate": 2e-05,
      "loss": 0.8462,
      "step": 3480
    },
    {
      "epoch": 0.22053712480252766,
      "grad_norm": 0.49970465898513794,
      "learning_rate": 2e-05,
      "loss": 0.8571,
      "step": 3490
    },
    {
      "epoch": 0.2211690363349131,
      "grad_norm": 0.49529343843460083,
      "learning_rate": 2e-05,
      "loss": 0.8467,
      "step": 3500
    },
    {
      "epoch": 0.22180094786729856,
      "grad_norm": 0.5843809843063354,
      "learning_rate": 2e-05,
      "loss": 0.8217,
      "step": 3510
    },
    {
      "epoch": 0.22243285939968405,
      "grad_norm": 0.44670504331588745,
      "learning_rate": 2e-05,
      "loss": 0.8542,
      "step": 3520
    },
    {
      "epoch": 0.2230647709320695,
      "grad_norm": 0.4520668387413025,
      "learning_rate": 2e-05,
      "loss": 0.8322,
      "step": 3530
    },
    {
      "epoch": 0.22369668246445498,
      "grad_norm": 0.558677613735199,
      "learning_rate": 2e-05,
      "loss": 0.8486,
      "step": 3540
    },
    {
      "epoch": 0.22432859399684044,
      "grad_norm": 0.4501877427101135,
      "learning_rate": 2e-05,
      "loss": 0.8206,
      "step": 3550
    },
    {
      "epoch": 0.22496050552922592,
      "grad_norm": 0.4787459671497345,
      "learning_rate": 2e-05,
      "loss": 0.8376,
      "step": 3560
    },
    {
      "epoch": 0.22559241706161137,
      "grad_norm": 0.4680955111980438,
      "learning_rate": 2e-05,
      "loss": 0.8332,
      "step": 3570
    },
    {
      "epoch": 0.22622432859399685,
      "grad_norm": 0.46172869205474854,
      "learning_rate": 2e-05,
      "loss": 0.8347,
      "step": 3580
    },
    {
      "epoch": 0.2268562401263823,
      "grad_norm": 0.45753517746925354,
      "learning_rate": 2e-05,
      "loss": 0.8559,
      "step": 3590
    },
    {
      "epoch": 0.22748815165876776,
      "grad_norm": 0.49647700786590576,
      "learning_rate": 2e-05,
      "loss": 0.8697,
      "step": 3600
    },
    {
      "epoch": 0.22812006319115324,
      "grad_norm": 0.4633341431617737,
      "learning_rate": 2e-05,
      "loss": 0.8653,
      "step": 3610
    },
    {
      "epoch": 0.2287519747235387,
      "grad_norm": 0.48471298813819885,
      "learning_rate": 2e-05,
      "loss": 0.8559,
      "step": 3620
    },
    {
      "epoch": 0.22938388625592418,
      "grad_norm": 0.4376724362373352,
      "learning_rate": 2e-05,
      "loss": 0.8373,
      "step": 3630
    },
    {
      "epoch": 0.23001579778830963,
      "grad_norm": 0.47309640049934387,
      "learning_rate": 2e-05,
      "loss": 0.8421,
      "step": 3640
    },
    {
      "epoch": 0.23064770932069512,
      "grad_norm": 0.6322768330574036,
      "learning_rate": 2e-05,
      "loss": 0.846,
      "step": 3650
    },
    {
      "epoch": 0.23127962085308057,
      "grad_norm": 0.4769039452075958,
      "learning_rate": 2e-05,
      "loss": 0.8402,
      "step": 3660
    },
    {
      "epoch": 0.23191153238546602,
      "grad_norm": 0.5190998911857605,
      "learning_rate": 2e-05,
      "loss": 0.8558,
      "step": 3670
    },
    {
      "epoch": 0.2325434439178515,
      "grad_norm": 0.47469544410705566,
      "learning_rate": 2e-05,
      "loss": 0.8675,
      "step": 3680
    },
    {
      "epoch": 0.23317535545023696,
      "grad_norm": 0.43990078568458557,
      "learning_rate": 2e-05,
      "loss": 0.847,
      "step": 3690
    },
    {
      "epoch": 0.23380726698262244,
      "grad_norm": 0.45917657017707825,
      "learning_rate": 2e-05,
      "loss": 0.8499,
      "step": 3700
    },
    {
      "epoch": 0.2344391785150079,
      "grad_norm": 0.45578819513320923,
      "learning_rate": 2e-05,
      "loss": 0.8338,
      "step": 3710
    },
    {
      "epoch": 0.23507109004739338,
      "grad_norm": 0.47179514169692993,
      "learning_rate": 2e-05,
      "loss": 0.8641,
      "step": 3720
    },
    {
      "epoch": 0.23570300157977883,
      "grad_norm": 0.44691628217697144,
      "learning_rate": 2e-05,
      "loss": 0.829,
      "step": 3730
    },
    {
      "epoch": 0.23633491311216429,
      "grad_norm": 0.4435933828353882,
      "learning_rate": 2e-05,
      "loss": 0.8296,
      "step": 3740
    },
    {
      "epoch": 0.23696682464454977,
      "grad_norm": 0.505792498588562,
      "learning_rate": 2e-05,
      "loss": 0.8355,
      "step": 3750
    },
    {
      "epoch": 0.23759873617693522,
      "grad_norm": 0.4689573645591736,
      "learning_rate": 2e-05,
      "loss": 0.8306,
      "step": 3760
    },
    {
      "epoch": 0.2382306477093207,
      "grad_norm": 0.513577938079834,
      "learning_rate": 2e-05,
      "loss": 0.847,
      "step": 3770
    },
    {
      "epoch": 0.23886255924170616,
      "grad_norm": 0.4486439824104309,
      "learning_rate": 2e-05,
      "loss": 0.8612,
      "step": 3780
    },
    {
      "epoch": 0.23949447077409164,
      "grad_norm": 0.4506240487098694,
      "learning_rate": 2e-05,
      "loss": 0.825,
      "step": 3790
    },
    {
      "epoch": 0.2401263823064771,
      "grad_norm": 0.4537910521030426,
      "learning_rate": 2e-05,
      "loss": 0.8609,
      "step": 3800
    },
    {
      "epoch": 0.24075829383886255,
      "grad_norm": 0.4724988639354706,
      "learning_rate": 2e-05,
      "loss": 0.838,
      "step": 3810
    },
    {
      "epoch": 0.24139020537124803,
      "grad_norm": 0.45192110538482666,
      "learning_rate": 2e-05,
      "loss": 0.8496,
      "step": 3820
    },
    {
      "epoch": 0.24202211690363348,
      "grad_norm": 0.4505990147590637,
      "learning_rate": 2e-05,
      "loss": 0.8461,
      "step": 3830
    },
    {
      "epoch": 0.24265402843601896,
      "grad_norm": 0.6071166396141052,
      "learning_rate": 2e-05,
      "loss": 0.8513,
      "step": 3840
    },
    {
      "epoch": 0.24328593996840442,
      "grad_norm": 0.4759773313999176,
      "learning_rate": 2e-05,
      "loss": 0.8604,
      "step": 3850
    },
    {
      "epoch": 0.2439178515007899,
      "grad_norm": 0.4953048527240753,
      "learning_rate": 2e-05,
      "loss": 0.8391,
      "step": 3860
    },
    {
      "epoch": 0.24454976303317535,
      "grad_norm": 0.46076062321662903,
      "learning_rate": 2e-05,
      "loss": 0.8597,
      "step": 3870
    },
    {
      "epoch": 0.2451816745655608,
      "grad_norm": 0.4764094948768616,
      "learning_rate": 2e-05,
      "loss": 0.8502,
      "step": 3880
    },
    {
      "epoch": 0.2458135860979463,
      "grad_norm": 0.45360779762268066,
      "learning_rate": 2e-05,
      "loss": 0.8505,
      "step": 3890
    },
    {
      "epoch": 0.24644549763033174,
      "grad_norm": 0.4578596353530884,
      "learning_rate": 2e-05,
      "loss": 0.8531,
      "step": 3900
    },
    {
      "epoch": 0.24707740916271723,
      "grad_norm": 0.4856429398059845,
      "learning_rate": 2e-05,
      "loss": 0.8456,
      "step": 3910
    },
    {
      "epoch": 0.24770932069510268,
      "grad_norm": 0.4444056451320648,
      "learning_rate": 2e-05,
      "loss": 0.8336,
      "step": 3920
    },
    {
      "epoch": 0.24834123222748816,
      "grad_norm": 0.453265517950058,
      "learning_rate": 2e-05,
      "loss": 0.8454,
      "step": 3930
    },
    {
      "epoch": 0.24897314375987362,
      "grad_norm": 0.44915109872817993,
      "learning_rate": 2e-05,
      "loss": 0.8627,
      "step": 3940
    },
    {
      "epoch": 0.24960505529225907,
      "grad_norm": 0.4355114698410034,
      "learning_rate": 2e-05,
      "loss": 0.8476,
      "step": 3950
    },
    {
      "epoch": 0.2502369668246445,
      "grad_norm": 0.4746149182319641,
      "learning_rate": 2e-05,
      "loss": 0.8525,
      "step": 3960
    },
    {
      "epoch": 0.25086887835703003,
      "grad_norm": 0.46624839305877686,
      "learning_rate": 2e-05,
      "loss": 0.8384,
      "step": 3970
    },
    {
      "epoch": 0.2515007898894155,
      "grad_norm": 0.4566735625267029,
      "learning_rate": 2e-05,
      "loss": 0.857,
      "step": 3980
    },
    {
      "epoch": 0.25213270142180094,
      "grad_norm": 0.4428395926952362,
      "learning_rate": 2e-05,
      "loss": 0.8419,
      "step": 3990
    },
    {
      "epoch": 0.2527646129541864,
      "grad_norm": 0.44233736395835876,
      "learning_rate": 2e-05,
      "loss": 0.8335,
      "step": 4000
    },
    {
      "epoch": 0.2533965244865719,
      "grad_norm": 0.4331214725971222,
      "learning_rate": 2e-05,
      "loss": 0.8123,
      "step": 4010
    },
    {
      "epoch": 0.25402843601895736,
      "grad_norm": 0.4452461004257202,
      "learning_rate": 2e-05,
      "loss": 0.8433,
      "step": 4020
    },
    {
      "epoch": 0.2546603475513428,
      "grad_norm": 0.5020315647125244,
      "learning_rate": 2e-05,
      "loss": 0.845,
      "step": 4030
    },
    {
      "epoch": 0.25529225908372827,
      "grad_norm": 0.46302923560142517,
      "learning_rate": 2e-05,
      "loss": 0.8525,
      "step": 4040
    },
    {
      "epoch": 0.2559241706161137,
      "grad_norm": 0.4996463656425476,
      "learning_rate": 2e-05,
      "loss": 0.8573,
      "step": 4050
    },
    {
      "epoch": 0.25655608214849923,
      "grad_norm": 0.44467058777809143,
      "learning_rate": 2e-05,
      "loss": 0.8414,
      "step": 4060
    },
    {
      "epoch": 0.2571879936808847,
      "grad_norm": 0.46636804938316345,
      "learning_rate": 2e-05,
      "loss": 0.8557,
      "step": 4070
    },
    {
      "epoch": 0.25781990521327014,
      "grad_norm": 0.4668079912662506,
      "learning_rate": 2e-05,
      "loss": 0.8772,
      "step": 4080
    },
    {
      "epoch": 0.2584518167456556,
      "grad_norm": 0.5015170574188232,
      "learning_rate": 2e-05,
      "loss": 0.8281,
      "step": 4090
    },
    {
      "epoch": 0.25908372827804105,
      "grad_norm": 0.44769486784935,
      "learning_rate": 2e-05,
      "loss": 0.8623,
      "step": 4100
    },
    {
      "epoch": 0.25971563981042656,
      "grad_norm": 0.44267430901527405,
      "learning_rate": 2e-05,
      "loss": 0.8314,
      "step": 4110
    },
    {
      "epoch": 0.260347551342812,
      "grad_norm": 0.5009731650352478,
      "learning_rate": 2e-05,
      "loss": 0.8353,
      "step": 4120
    },
    {
      "epoch": 0.26097946287519747,
      "grad_norm": 0.47171640396118164,
      "learning_rate": 2e-05,
      "loss": 0.8584,
      "step": 4130
    },
    {
      "epoch": 0.2616113744075829,
      "grad_norm": 0.4540732502937317,
      "learning_rate": 2e-05,
      "loss": 0.8367,
      "step": 4140
    },
    {
      "epoch": 0.26224328593996843,
      "grad_norm": 0.533187747001648,
      "learning_rate": 2e-05,
      "loss": 0.8496,
      "step": 4150
    },
    {
      "epoch": 0.2628751974723539,
      "grad_norm": 0.46570831537246704,
      "learning_rate": 2e-05,
      "loss": 0.8367,
      "step": 4160
    },
    {
      "epoch": 0.26350710900473934,
      "grad_norm": 0.43925151228904724,
      "learning_rate": 2e-05,
      "loss": 0.8584,
      "step": 4170
    },
    {
      "epoch": 0.2641390205371248,
      "grad_norm": 0.44512003660202026,
      "learning_rate": 2e-05,
      "loss": 0.8582,
      "step": 4180
    },
    {
      "epoch": 0.26477093206951025,
      "grad_norm": 0.46296021342277527,
      "learning_rate": 2e-05,
      "loss": 0.8566,
      "step": 4190
    },
    {
      "epoch": 0.26540284360189575,
      "grad_norm": 0.4663885235786438,
      "learning_rate": 2e-05,
      "loss": 0.8505,
      "step": 4200
    },
    {
      "epoch": 0.2660347551342812,
      "grad_norm": 0.4963429570198059,
      "learning_rate": 2e-05,
      "loss": 0.831,
      "step": 4210
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 0.47315335273742676,
      "learning_rate": 2e-05,
      "loss": 0.8512,
      "step": 4220
    },
    {
      "epoch": 0.2672985781990521,
      "grad_norm": 0.48192426562309265,
      "learning_rate": 2e-05,
      "loss": 0.8326,
      "step": 4230
    },
    {
      "epoch": 0.26793048973143757,
      "grad_norm": 0.47754642367362976,
      "learning_rate": 2e-05,
      "loss": 0.8219,
      "step": 4240
    },
    {
      "epoch": 0.2685624012638231,
      "grad_norm": 0.45899492502212524,
      "learning_rate": 2e-05,
      "loss": 0.8371,
      "step": 4250
    },
    {
      "epoch": 0.26919431279620853,
      "grad_norm": 0.45363181829452515,
      "learning_rate": 2e-05,
      "loss": 0.8572,
      "step": 4260
    },
    {
      "epoch": 0.269826224328594,
      "grad_norm": 0.474301278591156,
      "learning_rate": 2e-05,
      "loss": 0.8459,
      "step": 4270
    },
    {
      "epoch": 0.27045813586097944,
      "grad_norm": 0.4889231026172638,
      "learning_rate": 2e-05,
      "loss": 0.8381,
      "step": 4280
    },
    {
      "epoch": 0.27109004739336495,
      "grad_norm": 0.4646555185317993,
      "learning_rate": 2e-05,
      "loss": 0.8719,
      "step": 4290
    },
    {
      "epoch": 0.2717219589257504,
      "grad_norm": 0.43774381279945374,
      "learning_rate": 2e-05,
      "loss": 0.8657,
      "step": 4300
    },
    {
      "epoch": 0.27235387045813586,
      "grad_norm": 0.6450157165527344,
      "learning_rate": 2e-05,
      "loss": 0.8319,
      "step": 4310
    },
    {
      "epoch": 0.2729857819905213,
      "grad_norm": 0.46757811307907104,
      "learning_rate": 2e-05,
      "loss": 0.8514,
      "step": 4320
    },
    {
      "epoch": 0.27361769352290677,
      "grad_norm": 0.6075416803359985,
      "learning_rate": 2e-05,
      "loss": 0.8381,
      "step": 4330
    },
    {
      "epoch": 0.2742496050552923,
      "grad_norm": 0.477455198764801,
      "learning_rate": 2e-05,
      "loss": 0.8228,
      "step": 4340
    },
    {
      "epoch": 0.27488151658767773,
      "grad_norm": 0.49315664172172546,
      "learning_rate": 2e-05,
      "loss": 0.8409,
      "step": 4350
    },
    {
      "epoch": 0.2755134281200632,
      "grad_norm": 0.5885541439056396,
      "learning_rate": 2e-05,
      "loss": 0.8484,
      "step": 4360
    },
    {
      "epoch": 0.27614533965244864,
      "grad_norm": 0.4655112624168396,
      "learning_rate": 2e-05,
      "loss": 0.8453,
      "step": 4370
    },
    {
      "epoch": 0.27677725118483415,
      "grad_norm": 0.4779876172542572,
      "learning_rate": 2e-05,
      "loss": 0.8293,
      "step": 4380
    },
    {
      "epoch": 0.2774091627172196,
      "grad_norm": 0.4869404733181,
      "learning_rate": 2e-05,
      "loss": 0.8295,
      "step": 4390
    },
    {
      "epoch": 0.27804107424960506,
      "grad_norm": 0.46487656235694885,
      "learning_rate": 2e-05,
      "loss": 0.8317,
      "step": 4400
    },
    {
      "epoch": 0.2786729857819905,
      "grad_norm": 0.4718816876411438,
      "learning_rate": 2e-05,
      "loss": 0.8208,
      "step": 4410
    },
    {
      "epoch": 0.27930489731437597,
      "grad_norm": 0.47306886315345764,
      "learning_rate": 2e-05,
      "loss": 0.8459,
      "step": 4420
    },
    {
      "epoch": 0.2799368088467615,
      "grad_norm": 0.5678571462631226,
      "learning_rate": 2e-05,
      "loss": 0.8371,
      "step": 4430
    },
    {
      "epoch": 0.28056872037914693,
      "grad_norm": 0.47235390543937683,
      "learning_rate": 2e-05,
      "loss": 0.8339,
      "step": 4440
    },
    {
      "epoch": 0.2812006319115324,
      "grad_norm": 0.46147507429122925,
      "learning_rate": 2e-05,
      "loss": 0.8369,
      "step": 4450
    },
    {
      "epoch": 0.28183254344391784,
      "grad_norm": 0.47175613045692444,
      "learning_rate": 2e-05,
      "loss": 0.8236,
      "step": 4460
    },
    {
      "epoch": 0.2824644549763033,
      "grad_norm": 0.45858657360076904,
      "learning_rate": 2e-05,
      "loss": 0.8302,
      "step": 4470
    },
    {
      "epoch": 0.2830963665086888,
      "grad_norm": 0.4359367787837982,
      "learning_rate": 2e-05,
      "loss": 0.8042,
      "step": 4480
    },
    {
      "epoch": 0.28372827804107426,
      "grad_norm": 0.4913497865200043,
      "learning_rate": 2e-05,
      "loss": 0.838,
      "step": 4490
    },
    {
      "epoch": 0.2843601895734597,
      "grad_norm": 0.4531594514846802,
      "learning_rate": 2e-05,
      "loss": 0.8281,
      "step": 4500
    },
    {
      "epoch": 0.28499210110584516,
      "grad_norm": 0.504382848739624,
      "learning_rate": 2e-05,
      "loss": 0.8391,
      "step": 4510
    },
    {
      "epoch": 0.2856240126382307,
      "grad_norm": 0.6337704658508301,
      "learning_rate": 2e-05,
      "loss": 0.858,
      "step": 4520
    },
    {
      "epoch": 0.2862559241706161,
      "grad_norm": 0.4782826900482178,
      "learning_rate": 2e-05,
      "loss": 0.8332,
      "step": 4530
    },
    {
      "epoch": 0.2868878357030016,
      "grad_norm": 0.48696208000183105,
      "learning_rate": 2e-05,
      "loss": 0.8361,
      "step": 4540
    },
    {
      "epoch": 0.28751974723538704,
      "grad_norm": 0.43620461225509644,
      "learning_rate": 2e-05,
      "loss": 0.8264,
      "step": 4550
    },
    {
      "epoch": 0.2881516587677725,
      "grad_norm": 0.44903993606567383,
      "learning_rate": 2e-05,
      "loss": 0.8241,
      "step": 4560
    },
    {
      "epoch": 0.288783570300158,
      "grad_norm": 0.4964090883731842,
      "learning_rate": 2e-05,
      "loss": 0.8437,
      "step": 4570
    },
    {
      "epoch": 0.28941548183254345,
      "grad_norm": 0.47316983342170715,
      "learning_rate": 2e-05,
      "loss": 0.8561,
      "step": 4580
    },
    {
      "epoch": 0.2900473933649289,
      "grad_norm": 0.4800175130367279,
      "learning_rate": 2e-05,
      "loss": 0.8267,
      "step": 4590
    },
    {
      "epoch": 0.29067930489731436,
      "grad_norm": 0.46129193902015686,
      "learning_rate": 2e-05,
      "loss": 0.8495,
      "step": 4600
    },
    {
      "epoch": 0.2913112164296998,
      "grad_norm": 0.5344136953353882,
      "learning_rate": 2e-05,
      "loss": 0.8201,
      "step": 4610
    },
    {
      "epoch": 0.2919431279620853,
      "grad_norm": 0.4293859899044037,
      "learning_rate": 2e-05,
      "loss": 0.7982,
      "step": 4620
    },
    {
      "epoch": 0.2925750394944708,
      "grad_norm": 0.4427815079689026,
      "learning_rate": 2e-05,
      "loss": 0.8148,
      "step": 4630
    },
    {
      "epoch": 0.29320695102685623,
      "grad_norm": 0.46747544407844543,
      "learning_rate": 2e-05,
      "loss": 0.8592,
      "step": 4640
    },
    {
      "epoch": 0.2938388625592417,
      "grad_norm": 1.13631272315979,
      "learning_rate": 2e-05,
      "loss": 0.818,
      "step": 4650
    },
    {
      "epoch": 0.2944707740916272,
      "grad_norm": 0.44885706901550293,
      "learning_rate": 2e-05,
      "loss": 0.8044,
      "step": 4660
    },
    {
      "epoch": 0.29510268562401265,
      "grad_norm": 0.4977491796016693,
      "learning_rate": 2e-05,
      "loss": 0.8116,
      "step": 4670
    },
    {
      "epoch": 0.2957345971563981,
      "grad_norm": 0.7201970219612122,
      "learning_rate": 2e-05,
      "loss": 0.8399,
      "step": 4680
    },
    {
      "epoch": 0.29636650868878356,
      "grad_norm": 0.463399201631546,
      "learning_rate": 2e-05,
      "loss": 0.8505,
      "step": 4690
    },
    {
      "epoch": 0.296998420221169,
      "grad_norm": 0.4751105010509491,
      "learning_rate": 2e-05,
      "loss": 0.8557,
      "step": 4700
    },
    {
      "epoch": 0.2976303317535545,
      "grad_norm": 0.47626441717147827,
      "learning_rate": 2e-05,
      "loss": 0.826,
      "step": 4710
    },
    {
      "epoch": 0.29826224328594,
      "grad_norm": 0.4821193218231201,
      "learning_rate": 2e-05,
      "loss": 0.8358,
      "step": 4720
    },
    {
      "epoch": 0.29889415481832543,
      "grad_norm": 0.4608619511127472,
      "learning_rate": 2e-05,
      "loss": 0.8273,
      "step": 4730
    },
    {
      "epoch": 0.2995260663507109,
      "grad_norm": 0.46025869250297546,
      "learning_rate": 2e-05,
      "loss": 0.8176,
      "step": 4740
    },
    {
      "epoch": 0.3001579778830964,
      "grad_norm": 0.4340529143810272,
      "learning_rate": 2e-05,
      "loss": 0.8312,
      "step": 4750
    },
    {
      "epoch": 0.30078988941548185,
      "grad_norm": 0.5388309359550476,
      "learning_rate": 2e-05,
      "loss": 0.8303,
      "step": 4760
    },
    {
      "epoch": 0.3014218009478673,
      "grad_norm": 0.4544883370399475,
      "learning_rate": 2e-05,
      "loss": 0.8301,
      "step": 4770
    },
    {
      "epoch": 0.30205371248025276,
      "grad_norm": 0.5687497854232788,
      "learning_rate": 2e-05,
      "loss": 0.8265,
      "step": 4780
    },
    {
      "epoch": 0.3026856240126382,
      "grad_norm": 0.5265123844146729,
      "learning_rate": 2e-05,
      "loss": 0.8392,
      "step": 4790
    },
    {
      "epoch": 0.3033175355450237,
      "grad_norm": 0.46169963479042053,
      "learning_rate": 2e-05,
      "loss": 0.8441,
      "step": 4800
    },
    {
      "epoch": 0.3039494470774092,
      "grad_norm": 0.4648761749267578,
      "learning_rate": 2e-05,
      "loss": 0.8495,
      "step": 4810
    },
    {
      "epoch": 0.30458135860979463,
      "grad_norm": 0.4892057478427887,
      "learning_rate": 2e-05,
      "loss": 0.8244,
      "step": 4820
    },
    {
      "epoch": 0.3052132701421801,
      "grad_norm": 0.4654883146286011,
      "learning_rate": 2e-05,
      "loss": 0.8287,
      "step": 4830
    },
    {
      "epoch": 0.30584518167456554,
      "grad_norm": 0.4781225323677063,
      "learning_rate": 2e-05,
      "loss": 0.8236,
      "step": 4840
    },
    {
      "epoch": 0.30647709320695105,
      "grad_norm": 0.4732058346271515,
      "learning_rate": 2e-05,
      "loss": 0.8355,
      "step": 4850
    },
    {
      "epoch": 0.3071090047393365,
      "grad_norm": 0.4513225555419922,
      "learning_rate": 2e-05,
      "loss": 0.8037,
      "step": 4860
    },
    {
      "epoch": 0.30774091627172195,
      "grad_norm": 0.5412766933441162,
      "learning_rate": 2e-05,
      "loss": 0.8633,
      "step": 4870
    },
    {
      "epoch": 0.3083728278041074,
      "grad_norm": 0.46940046548843384,
      "learning_rate": 2e-05,
      "loss": 0.8307,
      "step": 4880
    },
    {
      "epoch": 0.3090047393364929,
      "grad_norm": 0.46186816692352295,
      "learning_rate": 2e-05,
      "loss": 0.8383,
      "step": 4890
    },
    {
      "epoch": 0.30963665086887837,
      "grad_norm": 0.48696672916412354,
      "learning_rate": 2e-05,
      "loss": 0.8294,
      "step": 4900
    },
    {
      "epoch": 0.3102685624012638,
      "grad_norm": 0.5054580569267273,
      "learning_rate": 2e-05,
      "loss": 0.8323,
      "step": 4910
    },
    {
      "epoch": 0.3109004739336493,
      "grad_norm": 0.5813882946968079,
      "learning_rate": 2e-05,
      "loss": 0.831,
      "step": 4920
    },
    {
      "epoch": 0.31153238546603473,
      "grad_norm": 0.4755232632160187,
      "learning_rate": 2e-05,
      "loss": 0.8333,
      "step": 4930
    },
    {
      "epoch": 0.31216429699842024,
      "grad_norm": 0.4346970319747925,
      "learning_rate": 2e-05,
      "loss": 0.8347,
      "step": 4940
    },
    {
      "epoch": 0.3127962085308057,
      "grad_norm": 0.4913887679576874,
      "learning_rate": 2e-05,
      "loss": 0.8282,
      "step": 4950
    },
    {
      "epoch": 0.31342812006319115,
      "grad_norm": 0.47380706667900085,
      "learning_rate": 2e-05,
      "loss": 0.8276,
      "step": 4960
    },
    {
      "epoch": 0.3140600315955766,
      "grad_norm": 0.47869932651519775,
      "learning_rate": 2e-05,
      "loss": 0.8352,
      "step": 4970
    },
    {
      "epoch": 0.31469194312796206,
      "grad_norm": 0.46003836393356323,
      "learning_rate": 2e-05,
      "loss": 0.8539,
      "step": 4980
    },
    {
      "epoch": 0.31532385466034757,
      "grad_norm": 0.4738747775554657,
      "learning_rate": 2e-05,
      "loss": 0.8123,
      "step": 4990
    },
    {
      "epoch": 0.315955766192733,
      "grad_norm": 0.48640719056129456,
      "learning_rate": 2e-05,
      "loss": 0.824,
      "step": 5000
    },
    {
      "epoch": 0.3165876777251185,
      "grad_norm": 0.4773779511451721,
      "learning_rate": 2e-05,
      "loss": 0.8207,
      "step": 5010
    },
    {
      "epoch": 0.31721958925750393,
      "grad_norm": 0.4434696435928345,
      "learning_rate": 2e-05,
      "loss": 0.8231,
      "step": 5020
    },
    {
      "epoch": 0.31785150078988944,
      "grad_norm": 0.5556390285491943,
      "learning_rate": 2e-05,
      "loss": 0.8181,
      "step": 5030
    },
    {
      "epoch": 0.3184834123222749,
      "grad_norm": 0.6391692757606506,
      "learning_rate": 2e-05,
      "loss": 0.8288,
      "step": 5040
    },
    {
      "epoch": 0.31911532385466035,
      "grad_norm": 0.48827236890792847,
      "learning_rate": 2e-05,
      "loss": 0.8215,
      "step": 5050
    },
    {
      "epoch": 0.3197472353870458,
      "grad_norm": 0.5423160195350647,
      "learning_rate": 2e-05,
      "loss": 0.8198,
      "step": 5060
    },
    {
      "epoch": 0.32037914691943126,
      "grad_norm": 0.45193660259246826,
      "learning_rate": 2e-05,
      "loss": 0.8406,
      "step": 5070
    },
    {
      "epoch": 0.32101105845181677,
      "grad_norm": 0.4477918744087219,
      "learning_rate": 2e-05,
      "loss": 0.82,
      "step": 5080
    },
    {
      "epoch": 0.3216429699842022,
      "grad_norm": 0.4462607502937317,
      "learning_rate": 2e-05,
      "loss": 0.8261,
      "step": 5090
    },
    {
      "epoch": 0.3222748815165877,
      "grad_norm": 0.5247912406921387,
      "learning_rate": 2e-05,
      "loss": 0.846,
      "step": 5100
    },
    {
      "epoch": 0.32290679304897313,
      "grad_norm": 0.45022520422935486,
      "learning_rate": 2e-05,
      "loss": 0.8287,
      "step": 5110
    },
    {
      "epoch": 0.3235387045813586,
      "grad_norm": 0.47860753536224365,
      "learning_rate": 2e-05,
      "loss": 0.8134,
      "step": 5120
    },
    {
      "epoch": 0.3241706161137441,
      "grad_norm": 0.49048587679862976,
      "learning_rate": 2e-05,
      "loss": 0.8371,
      "step": 5130
    },
    {
      "epoch": 0.32480252764612955,
      "grad_norm": 0.4868302345275879,
      "learning_rate": 2e-05,
      "loss": 0.8337,
      "step": 5140
    },
    {
      "epoch": 0.325434439178515,
      "grad_norm": 0.4755631387233734,
      "learning_rate": 2e-05,
      "loss": 0.835,
      "step": 5150
    },
    {
      "epoch": 0.32606635071090045,
      "grad_norm": 0.45582303404808044,
      "learning_rate": 2e-05,
      "loss": 0.833,
      "step": 5160
    },
    {
      "epoch": 0.32669826224328596,
      "grad_norm": 0.5564966201782227,
      "learning_rate": 2e-05,
      "loss": 0.8553,
      "step": 5170
    },
    {
      "epoch": 0.3273301737756714,
      "grad_norm": 0.46671995520591736,
      "learning_rate": 2e-05,
      "loss": 0.8386,
      "step": 5180
    },
    {
      "epoch": 0.32796208530805687,
      "grad_norm": 0.5555332899093628,
      "learning_rate": 2e-05,
      "loss": 0.8202,
      "step": 5190
    },
    {
      "epoch": 0.3285939968404423,
      "grad_norm": 0.4633268415927887,
      "learning_rate": 2e-05,
      "loss": 0.8344,
      "step": 5200
    },
    {
      "epoch": 0.3292259083728278,
      "grad_norm": 0.509212076663971,
      "learning_rate": 2e-05,
      "loss": 0.8356,
      "step": 5210
    },
    {
      "epoch": 0.3298578199052133,
      "grad_norm": 0.44110676646232605,
      "learning_rate": 2e-05,
      "loss": 0.8289,
      "step": 5220
    },
    {
      "epoch": 0.33048973143759874,
      "grad_norm": 0.4653122127056122,
      "learning_rate": 2e-05,
      "loss": 0.8414,
      "step": 5230
    },
    {
      "epoch": 0.3311216429699842,
      "grad_norm": 0.4613790512084961,
      "learning_rate": 2e-05,
      "loss": 0.829,
      "step": 5240
    },
    {
      "epoch": 0.33175355450236965,
      "grad_norm": 0.5667819976806641,
      "learning_rate": 2e-05,
      "loss": 0.833,
      "step": 5250
    },
    {
      "epoch": 0.33238546603475516,
      "grad_norm": 0.48280099034309387,
      "learning_rate": 2e-05,
      "loss": 0.8166,
      "step": 5260
    },
    {
      "epoch": 0.3330173775671406,
      "grad_norm": 0.4703671634197235,
      "learning_rate": 2e-05,
      "loss": 0.822,
      "step": 5270
    },
    {
      "epoch": 0.33364928909952607,
      "grad_norm": 0.4868001341819763,
      "learning_rate": 2e-05,
      "loss": 0.8286,
      "step": 5280
    },
    {
      "epoch": 0.3342812006319115,
      "grad_norm": 0.44161519408226013,
      "learning_rate": 2e-05,
      "loss": 0.8257,
      "step": 5290
    },
    {
      "epoch": 0.334913112164297,
      "grad_norm": 0.4368448555469513,
      "learning_rate": 2e-05,
      "loss": 0.8232,
      "step": 5300
    },
    {
      "epoch": 0.3355450236966825,
      "grad_norm": 0.45230305194854736,
      "learning_rate": 2e-05,
      "loss": 0.8323,
      "step": 5310
    },
    {
      "epoch": 0.33617693522906794,
      "grad_norm": 0.49776390194892883,
      "learning_rate": 2e-05,
      "loss": 0.8395,
      "step": 5320
    },
    {
      "epoch": 0.3368088467614534,
      "grad_norm": 0.46672677993774414,
      "learning_rate": 2e-05,
      "loss": 0.8263,
      "step": 5330
    },
    {
      "epoch": 0.33744075829383885,
      "grad_norm": 0.46882596611976624,
      "learning_rate": 2e-05,
      "loss": 0.8018,
      "step": 5340
    },
    {
      "epoch": 0.3380726698262243,
      "grad_norm": 0.44535496830940247,
      "learning_rate": 1.999194520388485e-05,
      "loss": 0.8088,
      "step": 5350
    },
    {
      "epoch": 0.3387045813586098,
      "grad_norm": 0.5008165240287781,
      "learning_rate": 1.9959247303157763e-05,
      "loss": 0.844,
      "step": 5360
    },
    {
      "epoch": 0.33933649289099527,
      "grad_norm": 0.4688977897167206,
      "learning_rate": 1.990149424459817e-05,
      "loss": 0.8532,
      "step": 5370
    },
    {
      "epoch": 0.3399684044233807,
      "grad_norm": 0.5142953395843506,
      "learning_rate": 1.9818847538648688e-05,
      "loss": 0.8176,
      "step": 5380
    },
    {
      "epoch": 0.3406003159557662,
      "grad_norm": 0.48431527614593506,
      "learning_rate": 1.971153831256721e-05,
      "loss": 0.8371,
      "step": 5390
    },
    {
      "epoch": 0.3412322274881517,
      "grad_norm": 0.4619187116622925,
      "learning_rate": 1.9579866664063435e-05,
      "loss": 0.8406,
      "step": 5400
    },
    {
      "epoch": 0.34186413902053714,
      "grad_norm": 0.8965077996253967,
      "learning_rate": 1.9424200822054782e-05,
      "loss": 0.8369,
      "step": 5410
    },
    {
      "epoch": 0.3424960505529226,
      "grad_norm": 0.44973745942115784,
      "learning_rate": 1.9244976116888626e-05,
      "loss": 0.8161,
      "step": 5420
    },
    {
      "epoch": 0.34312796208530805,
      "grad_norm": 0.4818698465824127,
      "learning_rate": 1.904269376291071e-05,
      "loss": 0.8426,
      "step": 5430
    },
    {
      "epoch": 0.3437598736176935,
      "grad_norm": 0.4372403621673584,
      "learning_rate": 1.88179194567844e-05,
      "loss": 0.83,
      "step": 5440
    },
    {
      "epoch": 0.344391785150079,
      "grad_norm": 0.45972108840942383,
      "learning_rate": 1.8571281795480632e-05,
      "loss": 0.8291,
      "step": 5450
    },
    {
      "epoch": 0.34502369668246446,
      "grad_norm": 0.48019173741340637,
      "learning_rate": 1.8303470518362808e-05,
      "loss": 0.8289,
      "step": 5460
    },
    {
      "epoch": 0.3456556082148499,
      "grad_norm": 0.4868035614490509,
      "learning_rate": 1.801523457828271e-05,
      "loss": 0.8165,
      "step": 5470
    },
    {
      "epoch": 0.3462875197472354,
      "grad_norm": 0.4691777229309082,
      "learning_rate": 1.7707380047081796e-05,
      "loss": 0.846,
      "step": 5480
    },
    {
      "epoch": 0.3469194312796208,
      "grad_norm": 0.465721070766449,
      "learning_rate": 1.7380767861355252e-05,
      "loss": 0.8259,
      "step": 5490
    },
    {
      "epoch": 0.34755134281200634,
      "grad_norm": 0.5202421545982361,
      "learning_rate": 1.703631141478299e-05,
      "loss": 0.8267,
      "step": 5500
    },
    {
      "epoch": 0.3481832543443918,
      "grad_norm": 0.4645007252693176,
      "learning_rate": 1.6674974003760705e-05,
      "loss": 0.8346,
      "step": 5510
    },
    {
      "epoch": 0.34881516587677724,
      "grad_norm": 0.47416257858276367,
      "learning_rate": 1.6297766133474556e-05,
      "loss": 0.8567,
      "step": 5520
    },
    {
      "epoch": 0.3494470774091627,
      "grad_norm": 0.42838436365127563,
      "learning_rate": 1.5905742691953186e-05,
      "loss": 0.8381,
      "step": 5530
    },
    {
      "epoch": 0.3500789889415482,
      "grad_norm": 0.470024049282074,
      "learning_rate": 1.55e-05,
      "loss": 0.8226,
      "step": 5540
    },
    {
      "epoch": 0.35071090047393366,
      "grad_norm": 0.4637489914894104,
      "learning_rate": 1.5081672745255862e-05,
      "loss": 0.8301,
      "step": 5550
    },
    {
      "epoch": 0.3513428120063191,
      "grad_norm": 0.47107744216918945,
      "learning_rate": 1.4651930808966206e-05,
      "loss": 0.848,
      "step": 5560
    },
    {
      "epoch": 0.35197472353870457,
      "grad_norm": 0.5376831889152527,
      "learning_rate": 1.4211975994326848e-05,
      "loss": 0.8566,
      "step": 5570
    },
    {
      "epoch": 0.35260663507109,
      "grad_norm": 0.46883130073547363,
      "learning_rate": 1.376303866555779e-05,
      "loss": 0.8033,
      "step": 5580
    },
    {
      "epoch": 0.35323854660347553,
      "grad_norm": 0.46029284596443176,
      "learning_rate": 1.3306374307104164e-05,
      "loss": 0.8203,
      "step": 5590
    },
    {
      "epoch": 0.353870458135861,
      "grad_norm": 0.49120739102363586,
      "learning_rate": 1.2843260012586718e-05,
      "loss": 0.8291,
      "step": 5600
    },
    {
      "epoch": 0.35450236966824644,
      "grad_norm": 0.49688807129859924,
      "learning_rate": 1.2374990913320726e-05,
      "loss": 0.8269,
      "step": 5610
    },
    {
      "epoch": 0.3551342812006319,
      "grad_norm": 0.46060746908187866,
      "learning_rate": 1.1902876556391247e-05,
      "loss": 0.816,
      "step": 5620
    },
    {
      "epoch": 0.35576619273301735,
      "grad_norm": 0.46067968010902405,
      "learning_rate": 1.1428237242413685e-05,
      "loss": 0.8277,
      "step": 5630
    },
    {
      "epoch": 0.35639810426540286,
      "grad_norm": 0.45033684372901917,
      "learning_rate": 1.095240033322132e-05,
      "loss": 0.8222,
      "step": 5640
    },
    {
      "epoch": 0.3570300157977883,
      "grad_norm": 0.4710417687892914,
      "learning_rate": 1.047669653980572e-05,
      "loss": 0.8326,
      "step": 5650
    },
    {
      "epoch": 0.35766192733017377,
      "grad_norm": 0.4336843192577362,
      "learning_rate": 1.0002456200890902e-05,
      "loss": 0.7873,
      "step": 5660
    },
    {
      "epoch": 0.3582938388625592,
      "grad_norm": 0.48323360085487366,
      "learning_rate": 9.531005562548668e-06,
      "loss": 0.8111,
      "step": 5670
    },
    {
      "epoch": 0.35892575039494473,
      "grad_norm": 0.4541269540786743,
      "learning_rate": 9.06366306925925e-06,
      "loss": 0.8134,
      "step": 5680
    },
    {
      "epoch": 0.3595576619273302,
      "grad_norm": 0.45144543051719666,
      "learning_rate": 8.601735676789686e-06,
      "loss": 0.8087,
      "step": 5690
    },
    {
      "epoch": 0.36018957345971564,
      "grad_norm": 0.4903709888458252,
      "learning_rate": 8.14651519720118e-06,
      "loss": 0.8202,
      "step": 5700
    },
    {
      "epoch": 0.3608214849921011,
      "grad_norm": 0.46367889642715454,
      "learning_rate": 7.699274686206866e-06,
      "loss": 0.8508,
      "step": 5710
    },
    {
      "epoch": 0.36145339652448655,
      "grad_norm": 0.46012982726097107,
      "learning_rate": 7.261264882983024e-06,
      "loss": 0.8243,
      "step": 5720
    },
    {
      "epoch": 0.36208530805687206,
      "grad_norm": 0.5042491555213928,
      "learning_rate": 6.8337107123900386e-06,
      "loss": 0.8237,
      "step": 5730
    },
    {
      "epoch": 0.3627172195892575,
      "grad_norm": 0.43738189339637756,
      "learning_rate": 6.417807859384881e-06,
      "loss": 0.8192,
      "step": 5740
    },
    {
      "epoch": 0.36334913112164297,
      "grad_norm": 0.5394167900085449,
      "learning_rate": 6.014719425205009e-06,
      "loss": 0.828,
      "step": 5750
    },
    {
      "epoch": 0.3639810426540284,
      "grad_norm": 0.4541049599647522,
      "learning_rate": 5.6255726746749275e-06,
      "loss": 0.7941,
      "step": 5760
    },
    {
      "epoch": 0.36461295418641393,
      "grad_norm": 0.4468422532081604,
      "learning_rate": 5.251455883731728e-06,
      "loss": 0.8007,
      "step": 5770
    },
    {
      "epoch": 0.3652448657187994,
      "grad_norm": 0.4709740877151489,
      "learning_rate": 4.893415295985812e-06,
      "loss": 0.8329,
      "step": 5780
    },
    {
      "epoch": 0.36587677725118484,
      "grad_norm": 0.4302249252796173,
      "learning_rate": 4.552452196827885e-06,
      "loss": 0.819,
      "step": 5790
    },
    {
      "epoch": 0.3665086887835703,
      "grad_norm": 0.5065787434577942,
      "learning_rate": 4.229520113264785e-06,
      "loss": 0.8055,
      "step": 5800
    },
    {
      "epoch": 0.36714060031595575,
      "grad_norm": 0.48239853978157043,
      "learning_rate": 3.925522147314915e-06,
      "loss": 0.816,
      "step": 5810
    },
    {
      "epoch": 0.36777251184834125,
      "grad_norm": 0.4948682188987732,
      "learning_rate": 3.641308450420675e-06,
      "loss": 0.8225,
      "step": 5820
    },
    {
      "epoch": 0.3684044233807267,
      "grad_norm": 0.7623666524887085,
      "learning_rate": 3.37767384594087e-06,
      "loss": 0.8017,
      "step": 5830
    },
    {
      "epoch": 0.36903633491311216,
      "grad_norm": 0.4984063506126404,
      "learning_rate": 3.1353556063719357e-06,
      "loss": 0.8154,
      "step": 5840
    },
    {
      "epoch": 0.3696682464454976,
      "grad_norm": 0.4427468478679657,
      "learning_rate": 2.9150313915141474e-06,
      "loss": 0.803,
      "step": 5850
    },
    {
      "epoch": 0.37030015797788307,
      "grad_norm": 0.44285380840301514,
      "learning_rate": 2.717317353348891e-06,
      "loss": 0.7852,
      "step": 5860
    },
    {
      "epoch": 0.3709320695102686,
      "grad_norm": 0.7217434644699097,
      "learning_rate": 2.5427664129268253e-06,
      "loss": 0.7992,
      "step": 5870
    },
    {
      "epoch": 0.37156398104265403,
      "grad_norm": 0.42715734243392944,
      "learning_rate": 2.391866714085717e-06,
      "loss": 0.7999,
      "step": 5880
    },
    {
      "epoch": 0.3721958925750395,
      "grad_norm": 0.4381905198097229,
      "learning_rate": 2.2650402583222753e-06,
      "loss": 0.8304,
      "step": 5890
    },
    {
      "epoch": 0.37282780410742494,
      "grad_norm": 0.4326131343841553,
      "learning_rate": 2.1626417246356398e-06,
      "loss": 0.8135,
      "step": 5900
    },
    {
      "epoch": 0.37345971563981045,
      "grad_norm": 0.433256059885025,
      "learning_rate": 2.084957477642894e-06,
      "loss": 0.8088,
      "step": 5910
    },
    {
      "epoch": 0.3740916271721959,
      "grad_norm": 0.4412059187889099,
      "learning_rate": 2.0322047667405147e-06,
      "loss": 0.8088,
      "step": 5920
    },
    {
      "epoch": 0.37472353870458136,
      "grad_norm": 0.4553843140602112,
      "learning_rate": 2.0045311185513342e-06,
      "loss": 0.8019,
      "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
}