{
  "best_metric": 0.9470136761665344,
  "best_model_checkpoint": "./trained-race/checkpoint-9000",
  "epoch": 1.8443378827001107,
  "eval_steps": 1000,
  "global_step": 10000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 0.000199907783105865,
      "loss": 7.8876,
      "step": 10
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.00019981556621172998,
      "loss": 6.9577,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.000199723349317595,
      "loss": 6.1368,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019963113242345997,
      "loss": 5.5562,
      "step": 40
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019953891552932499,
      "loss": 4.6469,
      "step": 50
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019944669863518997,
      "loss": 3.999,
      "step": 60
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019935448174105498,
      "loss": 3.5516,
      "step": 70
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00019926226484691997,
      "loss": 3.2196,
      "step": 80
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019917004795278495,
      "loss": 3.0321,
      "step": 90
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019907783105864993,
      "loss": 2.7039,
      "step": 100
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019898561416451495,
      "loss": 2.6115,
      "step": 110
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019889339727037993,
      "loss": 2.4644,
      "step": 120
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00019880118037624494,
      "loss": 2.3589,
      "step": 130
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019870896348210993,
      "loss": 2.3657,
      "step": 140
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019861674658797494,
      "loss": 2.1013,
      "step": 150
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019852452969383992,
      "loss": 2.0926,
      "step": 160
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0001984323127997049,
      "loss": 2.1673,
      "step": 170
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0001983400959055699,
      "loss": 1.9943,
      "step": 180
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001982478790114349,
      "loss": 1.9654,
      "step": 190
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001981556621172999,
      "loss": 2.0219,
      "step": 200
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001980634452231649,
      "loss": 2.2223,
      "step": 210
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019797122832902988,
      "loss": 1.9974,
      "step": 220
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001978790114348949,
      "loss": 1.9526,
      "step": 230
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019778679454075988,
      "loss": 2.0874,
      "step": 240
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019769457764662486,
      "loss": 2.0008,
      "step": 250
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019760236075248985,
      "loss": 1.9077,
      "step": 260
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019751014385835486,
      "loss": 1.9373,
      "step": 270
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019741792696421984,
      "loss": 1.9091,
      "step": 280
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019732571007008486,
      "loss": 1.9555,
      "step": 290
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019723349317594984,
      "loss": 1.8459,
      "step": 300
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019714127628181485,
      "loss": 1.9188,
      "step": 310
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019704905938767984,
      "loss": 1.8289,
      "step": 320
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019695684249354482,
      "loss": 1.8993,
      "step": 330
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001968646255994098,
      "loss": 1.8261,
      "step": 340
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019677240870527482,
      "loss": 1.8536,
      "step": 350
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001966801918111398,
      "loss": 1.8515,
      "step": 360
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001965879749170048,
      "loss": 1.9419,
      "step": 370
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001964957580228698,
      "loss": 1.8086,
      "step": 380
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001964035411287348,
      "loss": 1.7413,
      "step": 390
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001963113242345998,
      "loss": 1.7866,
      "step": 400
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019621910734046478,
      "loss": 1.8727,
      "step": 410
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019612689044632976,
      "loss": 1.8292,
      "step": 420
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019603467355219477,
      "loss": 1.7826,
      "step": 430
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019594245665805976,
      "loss": 1.8574,
      "step": 440
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019585023976392477,
      "loss": 1.7725,
      "step": 450
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019575802286978975,
      "loss": 1.7879,
      "step": 460
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019566580597565476,
      "loss": 1.6836,
      "step": 470
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019557358908151975,
      "loss": 1.7871,
      "step": 480
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019548137218738473,
      "loss": 1.7409,
      "step": 490
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019538915529324972,
      "loss": 1.6154,
      "step": 500
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019529693839911473,
      "loss": 1.6699,
      "step": 510
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019520472150497971,
      "loss": 1.7106,
      "step": 520
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019511250461084473,
      "loss": 1.5808,
      "step": 530
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001950202877167097,
      "loss": 1.7415,
      "step": 540
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019492807082257472,
      "loss": 1.7023,
      "step": 550
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001948358539284397,
      "loss": 1.6189,
      "step": 560
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001947436370343047,
      "loss": 1.7952,
      "step": 570
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019465142014016967,
      "loss": 1.7501,
      "step": 580
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019455920324603469,
      "loss": 1.7323,
      "step": 590
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019446698635189967,
      "loss": 1.6671,
      "step": 600
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019437476945776468,
      "loss": 1.7774,
      "step": 610
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019428255256362967,
      "loss": 1.8237,
      "step": 620
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019419033566949468,
      "loss": 1.6837,
      "step": 630
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019409811877535966,
      "loss": 1.7524,
      "step": 640
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019400590188122465,
      "loss": 1.617,
      "step": 650
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019391368498708963,
      "loss": 1.6385,
      "step": 660
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019382146809295462,
      "loss": 1.7298,
      "step": 670
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019372925119881963,
      "loss": 1.8203,
      "step": 680
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019363703430468464,
      "loss": 1.6564,
      "step": 690
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019354481741054962,
      "loss": 1.6086,
      "step": 700
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001934526005164146,
      "loss": 1.7182,
      "step": 710
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019336038362227962,
      "loss": 1.8396,
      "step": 720
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0001932681667281446,
      "loss": 1.6539,
      "step": 730
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001931759498340096,
      "loss": 1.6745,
      "step": 740
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019308373293987457,
      "loss": 1.7868,
      "step": 750
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019299151604573958,
      "loss": 1.6292,
      "step": 760
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001928992991516046,
      "loss": 1.5977,
      "step": 770
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019280708225746958,
      "loss": 1.6717,
      "step": 780
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019271486536333456,
      "loss": 1.712,
      "step": 790
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019262264846919958,
      "loss": 1.6402,
      "step": 800
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019253043157506456,
      "loss": 1.6647,
      "step": 810
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019243821468092954,
      "loss": 1.6568,
      "step": 820
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019234599778679453,
      "loss": 1.5253,
      "step": 830
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019225378089265954,
      "loss": 1.646,
      "step": 840
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019216156399852455,
      "loss": 1.7057,
      "step": 850
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019206934710438954,
      "loss": 1.5224,
      "step": 860
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019197713021025452,
      "loss": 1.6126,
      "step": 870
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019188491331611953,
      "loss": 1.5545,
      "step": 880
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019179269642198452,
      "loss": 1.6877,
      "step": 890
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001917004795278495,
      "loss": 1.6015,
      "step": 900
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00019160826263371449,
      "loss": 1.6627,
      "step": 910
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001915160457395795,
      "loss": 1.5924,
      "step": 920
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001914238288454445,
      "loss": 1.5444,
      "step": 930
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001913316119513095,
      "loss": 1.5869,
      "step": 940
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019123939505717448,
      "loss": 1.5678,
      "step": 950
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001911471781630395,
      "loss": 1.5879,
      "step": 960
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019105496126890447,
      "loss": 1.5347,
      "step": 970
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019096274437476946,
      "loss": 1.6076,
      "step": 980
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019087052748063444,
      "loss": 1.5945,
      "step": 990
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019077831058649945,
      "loss": 1.4995,
      "step": 1000
    },
    {
      "epoch": 0.18,
      "eval_accuracy": 0.4135475625342341,
      "eval_loss": 1.513289451599121,
      "eval_runtime": 120.4122,
      "eval_samples_per_second": 90.971,
      "eval_steps_per_second": 11.378,
      "step": 1000
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019068609369236446,
      "loss": 1.513,
      "step": 1010
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019059387679822945,
      "loss": 1.6145,
      "step": 1020
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019050165990409443,
      "loss": 1.6611,
      "step": 1030
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019040944300995944,
      "loss": 1.5784,
      "step": 1040
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019031722611582443,
      "loss": 1.5153,
      "step": 1050
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019022500922168941,
      "loss": 1.5148,
      "step": 1060
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001901327923275544,
      "loss": 1.6189,
      "step": 1070
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001900405754334194,
      "loss": 1.6063,
      "step": 1080
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00018994835853928442,
      "loss": 1.5173,
      "step": 1090
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001898561416451494,
      "loss": 1.4855,
      "step": 1100
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001897639247510144,
      "loss": 1.6097,
      "step": 1110
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001896717078568794,
      "loss": 1.6251,
      "step": 1120
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00018957949096274439,
      "loss": 1.5885,
      "step": 1130
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00018948727406860937,
      "loss": 1.5966,
      "step": 1140
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00018939505717447435,
      "loss": 1.5552,
      "step": 1150
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00018930284028033937,
      "loss": 1.6476,
      "step": 1160
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00018921062338620438,
      "loss": 1.492,
      "step": 1170
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00018911840649206936,
      "loss": 1.4321,
      "step": 1180
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00018902618959793435,
      "loss": 1.5384,
      "step": 1190
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00018893397270379936,
      "loss": 1.5005,
      "step": 1200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00018884175580966434,
      "loss": 1.4933,
      "step": 1210
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00018874953891552933,
      "loss": 1.5114,
      "step": 1220
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001886573220213943,
      "loss": 1.5656,
      "step": 1230
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00018856510512725932,
      "loss": 1.4009,
      "step": 1240
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00018847288823312433,
      "loss": 1.4853,
      "step": 1250
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00018838067133898932,
      "loss": 1.5312,
      "step": 1260
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001882884544448543,
      "loss": 1.5566,
      "step": 1270
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001881962375507193,
      "loss": 1.4965,
      "step": 1280
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001881040206565843,
      "loss": 1.5982,
      "step": 1290
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00018801180376244928,
      "loss": 1.4838,
      "step": 1300
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00018791958686831427,
      "loss": 1.4434,
      "step": 1310
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00018782736997417928,
      "loss": 1.4311,
      "step": 1320
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001877351530800443,
      "loss": 1.4896,
      "step": 1330
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00018764293618590928,
      "loss": 1.4999,
      "step": 1340
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00018755071929177426,
      "loss": 1.3174,
      "step": 1350
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00018745850239763924,
      "loss": 1.3892,
      "step": 1360
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00018736628550350426,
      "loss": 1.4754,
      "step": 1370
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00018727406860936924,
      "loss": 1.4074,
      "step": 1380
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00018718185171523422,
      "loss": 1.526,
      "step": 1390
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00018708963482109924,
      "loss": 1.3958,
      "step": 1400
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00018699741792696425,
      "loss": 1.473,
      "step": 1410
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00018690520103282923,
      "loss": 1.3931,
      "step": 1420
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00018681298413869422,
      "loss": 1.4689,
      "step": 1430
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001867207672445592,
      "loss": 1.4804,
      "step": 1440
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001866285503504242,
      "loss": 1.4743,
      "step": 1450
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001865363334562892,
      "loss": 1.386,
      "step": 1460
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00018644411656215418,
      "loss": 1.4222,
      "step": 1470
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001863518996680192,
      "loss": 1.3938,
      "step": 1480
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001862596827738842,
      "loss": 1.5226,
      "step": 1490
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001861674658797492,
      "loss": 1.4203,
      "step": 1500
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00018607524898561417,
      "loss": 1.5243,
      "step": 1510
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00018598303209147916,
      "loss": 1.4287,
      "step": 1520
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00018589081519734417,
      "loss": 1.2989,
      "step": 1530
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00018579859830320915,
      "loss": 1.4642,
      "step": 1540
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00018570638140907414,
      "loss": 1.5333,
      "step": 1550
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00018561416451493915,
      "loss": 1.4577,
      "step": 1560
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00018552194762080416,
      "loss": 1.409,
      "step": 1570
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00018542973072666915,
      "loss": 1.4131,
      "step": 1580
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00018533751383253413,
      "loss": 1.4178,
      "step": 1590
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018524529693839911,
      "loss": 1.4121,
      "step": 1600
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018515308004426413,
      "loss": 1.4046,
      "step": 1610
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001850608631501291,
      "loss": 1.4579,
      "step": 1620
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001849686462559941,
      "loss": 1.2591,
      "step": 1630
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001848764293618591,
      "loss": 1.3416,
      "step": 1640
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00018478421246772412,
      "loss": 1.3919,
      "step": 1650
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001846919955735891,
      "loss": 1.5162,
      "step": 1660
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018459977867945409,
      "loss": 1.32,
      "step": 1670
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018450756178531907,
      "loss": 1.4586,
      "step": 1680
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018441534489118408,
      "loss": 1.2827,
      "step": 1690
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00018432312799704907,
      "loss": 1.4172,
      "step": 1700
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00018423091110291405,
      "loss": 1.3311,
      "step": 1710
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00018413869420877906,
      "loss": 1.3658,
      "step": 1720
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00018404647731464407,
      "loss": 1.5157,
      "step": 1730
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00018395426042050906,
      "loss": 1.4216,
      "step": 1740
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00018386204352637404,
      "loss": 1.3597,
      "step": 1750
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00018376982663223903,
      "loss": 1.3921,
      "step": 1760
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00018367760973810404,
      "loss": 1.3468,
      "step": 1770
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00018358539284396902,
      "loss": 1.3247,
      "step": 1780
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.000183493175949834,
      "loss": 1.2005,
      "step": 1790
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00018340095905569902,
      "loss": 1.3875,
      "step": 1800
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00018330874216156403,
      "loss": 1.4271,
      "step": 1810
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00018321652526742901,
      "loss": 1.3405,
      "step": 1820
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.000183124308373294,
      "loss": 1.3398,
      "step": 1830
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00018303209147915898,
      "loss": 1.4113,
      "step": 1840
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.000182939874585024,
      "loss": 1.3913,
      "step": 1850
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00018284765769088898,
      "loss": 1.4525,
      "step": 1860
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00018275544079675396,
      "loss": 1.3711,
      "step": 1870
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00018266322390261898,
      "loss": 1.3856,
      "step": 1880
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.000182571007008484,
      "loss": 1.4341,
      "step": 1890
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00018247879011434897,
      "loss": 1.2842,
      "step": 1900
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00018238657322021396,
      "loss": 1.3662,
      "step": 1910
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00018229435632607894,
      "loss": 1.3707,
      "step": 1920
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00018220213943194392,
      "loss": 1.3719,
      "step": 1930
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00018210992253780894,
      "loss": 1.2909,
      "step": 1940
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00018201770564367392,
      "loss": 1.424,
      "step": 1950
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00018192548874953893,
      "loss": 1.291,
      "step": 1960
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00018183327185540392,
      "loss": 1.4188,
      "step": 1970
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00018174105496126893,
      "loss": 1.2462,
      "step": 1980
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001816488380671339,
      "loss": 1.4223,
      "step": 1990
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001815566211729989,
      "loss": 1.3103,
      "step": 2000
    },
    {
      "epoch": 0.37,
      "eval_accuracy": 0.5044732517801717,
      "eval_loss": 1.2796891927719116,
      "eval_runtime": 119.305,
      "eval_samples_per_second": 91.815,
      "eval_steps_per_second": 11.483,
      "step": 2000
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00018146440427886388,
      "loss": 1.4283,
      "step": 2010
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001813721873847289,
      "loss": 1.3649,
      "step": 2020
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00018127997049059388,
      "loss": 1.3352,
      "step": 2030
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0001811877535964589,
      "loss": 1.4464,
      "step": 2040
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00018109553670232387,
      "loss": 1.2952,
      "step": 2050
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00018100331980818888,
      "loss": 1.3101,
      "step": 2060
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00018091110291405387,
      "loss": 1.4286,
      "step": 2070
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00018081888601991885,
      "loss": 1.1986,
      "step": 2080
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00018072666912578384,
      "loss": 1.2121,
      "step": 2090
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00018063445223164885,
      "loss": 1.2574,
      "step": 2100
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00018054223533751383,
      "loss": 1.2758,
      "step": 2110
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00018045001844337885,
      "loss": 1.2478,
      "step": 2120
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00018035780154924383,
      "loss": 1.365,
      "step": 2130
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00018026558465510884,
      "loss": 1.4186,
      "step": 2140
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00018017336776097383,
      "loss": 1.326,
      "step": 2150
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001800811508668388,
      "loss": 1.2639,
      "step": 2160
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001799889339727038,
      "loss": 1.2415,
      "step": 2170
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001798967170785688,
      "loss": 1.2523,
      "step": 2180
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001798045001844338,
      "loss": 1.3794,
      "step": 2190
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001797122832902988,
      "loss": 1.2596,
      "step": 2200
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00017962006639616379,
      "loss": 1.2908,
      "step": 2210
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001795278495020288,
      "loss": 1.4256,
      "step": 2220
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00017943563260789378,
      "loss": 1.4267,
      "step": 2230
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00017934341571375877,
      "loss": 1.2581,
      "step": 2240
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00017925119881962375,
      "loss": 1.3161,
      "step": 2250
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00017915898192548876,
      "loss": 1.3534,
      "step": 2260
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00017906676503135375,
      "loss": 1.3234,
      "step": 2270
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00017897454813721876,
      "loss": 1.3659,
      "step": 2280
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00017888233124308374,
      "loss": 1.3122,
      "step": 2290
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00017879011434894875,
      "loss": 1.2602,
      "step": 2300
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00017869789745481374,
      "loss": 1.3454,
      "step": 2310
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00017860568056067872,
      "loss": 1.2308,
      "step": 2320
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001785134636665437,
      "loss": 1.3802,
      "step": 2330
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00017842124677240872,
      "loss": 1.328,
      "step": 2340
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001783290298782737,
      "loss": 1.2514,
      "step": 2350
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00017823681298413871,
      "loss": 1.3682,
      "step": 2360
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001781445960900037,
      "loss": 1.2913,
      "step": 2370
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001780523791958687,
      "loss": 1.2392,
      "step": 2380
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001779601623017337,
      "loss": 1.2229,
      "step": 2390
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00017786794540759868,
      "loss": 1.2137,
      "step": 2400
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00017777572851346366,
      "loss": 1.3575,
      "step": 2410
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00017768351161932868,
      "loss": 1.2537,
      "step": 2420
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00017759129472519366,
      "loss": 1.2323,
      "step": 2430
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00017749907783105864,
      "loss": 1.3776,
      "step": 2440
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00017740686093692366,
      "loss": 1.257,
      "step": 2450
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00017731464404278867,
      "loss": 1.2321,
      "step": 2460
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017722242714865365,
      "loss": 1.3849,
      "step": 2470
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017713021025451864,
      "loss": 1.2232,
      "step": 2480
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017703799336038362,
      "loss": 1.2499,
      "step": 2490
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017694577646624863,
      "loss": 1.2204,
      "step": 2500
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00017685355957211362,
      "loss": 1.3772,
      "step": 2510
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0001767613426779786,
      "loss": 1.2534,
      "step": 2520
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001766691257838436,
      "loss": 1.2529,
      "step": 2530
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001765769088897086,
      "loss": 1.3371,
      "step": 2540
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001764846919955736,
      "loss": 1.2988,
      "step": 2550
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001763924751014386,
      "loss": 1.2163,
      "step": 2560
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017630025820730358,
      "loss": 1.3794,
      "step": 2570
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00017620804131316856,
      "loss": 1.1544,
      "step": 2580
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00017611582441903357,
      "loss": 1.1822,
      "step": 2590
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00017602360752489856,
      "loss": 1.0888,
      "step": 2600
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00017593139063076357,
      "loss": 1.2588,
      "step": 2610
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00017583917373662855,
      "loss": 1.3141,
      "step": 2620
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017574695684249356,
      "loss": 1.1886,
      "step": 2630
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017565473994835855,
      "loss": 1.3152,
      "step": 2640
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017556252305422353,
      "loss": 1.2039,
      "step": 2650
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017547030616008852,
      "loss": 1.2983,
      "step": 2660
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017537808926595353,
      "loss": 1.2921,
      "step": 2670
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00017528587237181851,
      "loss": 1.2147,
      "step": 2680
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00017519365547768353,
      "loss": 1.053,
      "step": 2690
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001751014385835485,
      "loss": 1.2289,
      "step": 2700
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00017500922168941352,
      "loss": 1.2486,
      "step": 2710
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001749170047952785,
      "loss": 1.164,
      "step": 2720
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001748247879011435,
      "loss": 1.3237,
      "step": 2730
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017473257100700847,
      "loss": 1.2586,
      "step": 2740
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017464035411287349,
      "loss": 1.2166,
      "step": 2750
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017454813721873847,
      "loss": 1.25,
      "step": 2760
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017445592032460348,
      "loss": 1.1968,
      "step": 2770
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017436370343046847,
      "loss": 1.2486,
      "step": 2780
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00017427148653633348,
      "loss": 1.2165,
      "step": 2790
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00017417926964219846,
      "loss": 1.2314,
      "step": 2800
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00017408705274806345,
      "loss": 1.1307,
      "step": 2810
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00017399483585392843,
      "loss": 1.2748,
      "step": 2820
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00017390261895979344,
      "loss": 1.2688,
      "step": 2830
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00017381040206565843,
      "loss": 1.3272,
      "step": 2840
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017371818517152344,
      "loss": 1.3796,
      "step": 2850
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017362596827738842,
      "loss": 1.2804,
      "step": 2860
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017353375138325343,
      "loss": 1.2676,
      "step": 2870
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00017344153448911842,
      "loss": 1.2247,
      "step": 2880
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001733493175949834,
      "loss": 1.2221,
      "step": 2890
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001732571007008484,
      "loss": 1.1982,
      "step": 2900
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001731648838067134,
      "loss": 1.1594,
      "step": 2910
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00017307266691257838,
      "loss": 1.0483,
      "step": 2920
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001729804500184434,
      "loss": 1.2851,
      "step": 2930
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00017288823312430838,
      "loss": 1.234,
      "step": 2940
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001727960162301734,
      "loss": 1.237,
      "step": 2950
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00017270379933603838,
      "loss": 1.2682,
      "step": 2960
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00017261158244190336,
      "loss": 1.2149,
      "step": 2970
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00017251936554776834,
      "loss": 1.2921,
      "step": 2980
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00017242714865363336,
      "loss": 1.2349,
      "step": 2990
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00017233493175949834,
      "loss": 1.2417,
      "step": 3000
    },
    {
      "epoch": 0.55,
      "eval_accuracy": 0.5450976812123425,
      "eval_loss": 1.1583396196365356,
      "eval_runtime": 118.0159,
      "eval_samples_per_second": 92.818,
      "eval_steps_per_second": 11.609,
      "step": 3000
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00017224271486536335,
      "loss": 1.2374,
      "step": 3010
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00017215049797122834,
      "loss": 1.2474,
      "step": 3020
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00017205828107709335,
      "loss": 1.2253,
      "step": 3030
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00017196606418295833,
      "loss": 1.2369,
      "step": 3040
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00017187384728882332,
      "loss": 1.1092,
      "step": 3050
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001717816303946883,
      "loss": 1.2028,
      "step": 3060
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001716894135005533,
      "loss": 1.2553,
      "step": 3070
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001715971966064183,
      "loss": 1.067,
      "step": 3080
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001715049797122833,
      "loss": 1.0817,
      "step": 3090
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001714127628181483,
      "loss": 1.179,
      "step": 3100
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001713205459240133,
      "loss": 1.251,
      "step": 3110
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001712283290298783,
      "loss": 1.2271,
      "step": 3120
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00017113611213574327,
      "loss": 1.1765,
      "step": 3130
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00017104389524160826,
      "loss": 1.1913,
      "step": 3140
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00017095167834747327,
      "loss": 1.361,
      "step": 3150
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00017085946145333825,
      "loss": 1.1744,
      "step": 3160
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00017076724455920326,
      "loss": 1.249,
      "step": 3170
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00017067502766506825,
      "loss": 1.1168,
      "step": 3180
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00017058281077093323,
      "loss": 1.0437,
      "step": 3190
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00017049059387679825,
      "loss": 1.1651,
      "step": 3200
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00017039837698266323,
      "loss": 1.1997,
      "step": 3210
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00017030616008852821,
      "loss": 1.0417,
      "step": 3220
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001702139431943932,
      "loss": 1.0223,
      "step": 3230
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001701217263002582,
      "loss": 1.1061,
      "step": 3240
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00017002950940612322,
      "loss": 1.3041,
      "step": 3250
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001699372925119882,
      "loss": 1.2808,
      "step": 3260
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001698450756178532,
      "loss": 1.3155,
      "step": 3270
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001697528587237182,
      "loss": 1.1121,
      "step": 3280
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00016966064182958319,
      "loss": 1.1627,
      "step": 3290
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00016956842493544817,
      "loss": 1.2635,
      "step": 3300
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00016947620804131316,
      "loss": 1.2897,
      "step": 3310
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00016938399114717817,
      "loss": 1.1851,
      "step": 3320
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00016929177425304318,
      "loss": 1.1192,
      "step": 3330
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00016919955735890816,
      "loss": 1.3427,
      "step": 3340
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00016910734046477315,
      "loss": 1.278,
      "step": 3350
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00016901512357063816,
      "loss": 1.0677,
      "step": 3360
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00016892290667650314,
      "loss": 1.1406,
      "step": 3370
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00016883068978236813,
      "loss": 1.0977,
      "step": 3380
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001687384728882331,
      "loss": 1.1641,
      "step": 3390
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00016864625599409812,
      "loss": 1.1614,
      "step": 3400
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00016855403909996313,
      "loss": 1.1798,
      "step": 3410
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00016846182220582812,
      "loss": 1.0466,
      "step": 3420
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001683696053116931,
      "loss": 1.1568,
      "step": 3430
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00016827738841755812,
      "loss": 1.1289,
      "step": 3440
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001681851715234231,
      "loss": 1.1635,
      "step": 3450
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00016809295462928808,
      "loss": 1.2475,
      "step": 3460
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00016800073773515307,
      "loss": 1.1312,
      "step": 3470
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00016790852084101808,
      "loss": 1.0037,
      "step": 3480
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001678163039468831,
      "loss": 1.2798,
      "step": 3490
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00016772408705274808,
      "loss": 1.1405,
      "step": 3500
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00016763187015861306,
      "loss": 1.239,
      "step": 3510
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00016753965326447807,
      "loss": 1.2206,
      "step": 3520
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00016744743637034306,
      "loss": 1.1298,
      "step": 3530
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00016735521947620804,
      "loss": 1.2235,
      "step": 3540
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00016726300258207302,
      "loss": 1.2435,
      "step": 3550
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00016717078568793804,
      "loss": 0.9712,
      "step": 3560
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00016707856879380305,
      "loss": 1.0133,
      "step": 3570
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00016698635189966803,
      "loss": 1.2278,
      "step": 3580
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00016689413500553302,
      "loss": 1.1089,
      "step": 3590
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00016680191811139803,
      "loss": 1.1549,
      "step": 3600
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.000166709701217263,
      "loss": 1.0782,
      "step": 3610
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.000166617484323128,
      "loss": 1.0607,
      "step": 3620
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00016652526742899298,
      "loss": 1.0517,
      "step": 3630
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.000166433050534858,
      "loss": 1.1832,
      "step": 3640
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.000166340833640723,
      "loss": 1.1528,
      "step": 3650
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.000166248616746588,
      "loss": 1.0889,
      "step": 3660
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00016615639985245297,
      "loss": 1.2566,
      "step": 3670
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00016606418295831798,
      "loss": 1.0122,
      "step": 3680
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00016597196606418297,
      "loss": 1.1981,
      "step": 3690
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00016587974917004795,
      "loss": 1.3029,
      "step": 3700
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00016578753227591294,
      "loss": 1.0953,
      "step": 3710
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00016569531538177795,
      "loss": 1.243,
      "step": 3720
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00016560309848764296,
      "loss": 1.2896,
      "step": 3730
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00016551088159350795,
      "loss": 1.281,
      "step": 3740
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00016541866469937293,
      "loss": 1.2114,
      "step": 3750
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00016532644780523794,
      "loss": 1.1415,
      "step": 3760
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00016523423091110293,
      "loss": 1.0854,
      "step": 3770
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001651420140169679,
      "loss": 1.2067,
      "step": 3780
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001650497971228329,
      "loss": 1.1454,
      "step": 3790
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001649575802286979,
      "loss": 1.227,
      "step": 3800
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00016486536333456292,
      "loss": 1.0913,
      "step": 3810
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001647731464404279,
      "loss": 1.2574,
      "step": 3820
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00016468092954629289,
      "loss": 1.2291,
      "step": 3830
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00016458871265215787,
      "loss": 1.2255,
      "step": 3840
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00016449649575802288,
      "loss": 1.2167,
      "step": 3850
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00016440427886388787,
      "loss": 1.0863,
      "step": 3860
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00016431206196975285,
      "loss": 1.1154,
      "step": 3870
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016421984507561786,
      "loss": 1.2457,
      "step": 3880
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016412762818148287,
      "loss": 1.1383,
      "step": 3890
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016403541128734786,
      "loss": 1.0618,
      "step": 3900
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016394319439321284,
      "loss": 1.1089,
      "step": 3910
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016385097749907783,
      "loss": 1.1526,
      "step": 3920
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016375876060494284,
      "loss": 1.155,
      "step": 3930
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00016366654371080782,
      "loss": 1.1888,
      "step": 3940
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001635743268166728,
      "loss": 1.0237,
      "step": 3950
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00016348210992253782,
      "loss": 1.006,
      "step": 3960
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00016338989302840283,
      "loss": 1.0144,
      "step": 3970
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00016329767613426782,
      "loss": 1.2536,
      "step": 3980
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001632054592401328,
      "loss": 1.1196,
      "step": 3990
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00016311324234599778,
      "loss": 1.2142,
      "step": 4000
    },
    {
      "epoch": 0.74,
      "eval_accuracy": 0.5779623881687055,
      "eval_loss": 1.0952427387237549,
      "eval_runtime": 117.9515,
      "eval_samples_per_second": 92.869,
      "eval_steps_per_second": 11.615,
      "step": 4000
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001630210254518628,
      "loss": 1.2744,
      "step": 4010
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00016292880855772778,
      "loss": 1.1747,
      "step": 4020
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00016283659166359276,
      "loss": 1.211,
      "step": 4030
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016274437476945778,
      "loss": 1.1793,
      "step": 4040
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001626521578753228,
      "loss": 1.1115,
      "step": 4050
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016255994098118777,
      "loss": 1.0671,
      "step": 4060
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016246772408705276,
      "loss": 1.1405,
      "step": 4070
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016237550719291774,
      "loss": 1.0318,
      "step": 4080
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016228329029878275,
      "loss": 1.091,
      "step": 4090
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00016219107340464774,
      "loss": 1.1943,
      "step": 4100
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00016209885651051272,
      "loss": 1.0205,
      "step": 4110
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00016200663961637773,
      "loss": 1.1909,
      "step": 4120
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00016191442272224274,
      "loss": 1.3308,
      "step": 4130
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00016182220582810773,
      "loss": 1.2898,
      "step": 4140
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001617299889339727,
      "loss": 1.2433,
      "step": 4150
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001616377720398377,
      "loss": 1.1034,
      "step": 4160
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001615455551457027,
      "loss": 1.1534,
      "step": 4170
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001614533382515677,
      "loss": 1.1877,
      "step": 4180
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00016136112135743268,
      "loss": 1.2006,
      "step": 4190
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001612689044632977,
      "loss": 1.1875,
      "step": 4200
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001611766875691627,
      "loss": 1.1954,
      "step": 4210
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00016108447067502768,
      "loss": 1.0619,
      "step": 4220
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00016099225378089267,
      "loss": 1.1578,
      "step": 4230
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00016090003688675765,
      "loss": 1.0976,
      "step": 4240
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00016080781999262267,
      "loss": 1.0075,
      "step": 4250
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00016071560309848765,
      "loss": 1.1182,
      "step": 4260
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00016062338620435263,
      "loss": 1.0065,
      "step": 4270
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00016053116931021765,
      "loss": 1.0983,
      "step": 4280
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00016043895241608266,
      "loss": 0.988,
      "step": 4290
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00016034673552194764,
      "loss": 1.1844,
      "step": 4300
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00016025451862781263,
      "loss": 1.1004,
      "step": 4310
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001601623017336776,
      "loss": 1.1675,
      "step": 4320
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00016007008483954262,
      "loss": 1.0223,
      "step": 4330
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001599778679454076,
      "loss": 1.1491,
      "step": 4340
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001598856510512726,
      "loss": 1.0693,
      "step": 4350
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001597934341571376,
      "loss": 1.2653,
      "step": 4360
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001597012172630026,
      "loss": 1.1791,
      "step": 4370
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001596090003688676,
      "loss": 1.0327,
      "step": 4380
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00015951678347473258,
      "loss": 1.1773,
      "step": 4390
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00015942456658059757,
      "loss": 1.2394,
      "step": 4400
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00015933234968646258,
      "loss": 1.1458,
      "step": 4410
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00015924013279232756,
      "loss": 1.0744,
      "step": 4420
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00015914791589819255,
      "loss": 1.2287,
      "step": 4430
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00015905569900405756,
      "loss": 1.0857,
      "step": 4440
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00015896348210992254,
      "loss": 1.2088,
      "step": 4450
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00015887126521578755,
      "loss": 1.0082,
      "step": 4460
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00015877904832165254,
      "loss": 1.054,
      "step": 4470
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00015868683142751752,
      "loss": 1.2285,
      "step": 4480
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001585946145333825,
      "loss": 1.0143,
      "step": 4490
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00015850239763924752,
      "loss": 1.0714,
      "step": 4500
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0001584101807451125,
      "loss": 1.0882,
      "step": 4510
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00015831796385097752,
      "loss": 1.1059,
      "step": 4520
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001582257469568425,
      "loss": 1.0658,
      "step": 4530
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001581335300627075,
      "loss": 1.1272,
      "step": 4540
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001580413131685725,
      "loss": 1.0791,
      "step": 4550
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00015794909627443748,
      "loss": 1.1533,
      "step": 4560
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00015785687938030246,
      "loss": 1.1378,
      "step": 4570
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00015776466248616748,
      "loss": 1.1286,
      "step": 4580
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00015767244559203246,
      "loss": 1.1265,
      "step": 4590
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00015758022869789747,
      "loss": 0.9253,
      "step": 4600
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00015748801180376246,
      "loss": 1.0376,
      "step": 4610
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00015739579490962747,
      "loss": 1.1497,
      "step": 4620
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00015730357801549245,
      "loss": 1.1336,
      "step": 4630
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00015721136112135744,
      "loss": 1.2066,
      "step": 4640
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00015711914422722242,
      "loss": 1.1973,
      "step": 4650
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00015702692733308743,
      "loss": 1.1119,
      "step": 4660
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00015693471043895242,
      "loss": 1.044,
      "step": 4670
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00015684249354481743,
      "loss": 1.1936,
      "step": 4680
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001567502766506824,
      "loss": 0.9883,
      "step": 4690
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00015665805975654742,
      "loss": 1.0117,
      "step": 4700
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001565658428624124,
      "loss": 1.0789,
      "step": 4710
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001564736259682774,
      "loss": 1.1306,
      "step": 4720
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00015638140907414238,
      "loss": 1.1597,
      "step": 4730
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001562891921800074,
      "loss": 1.0788,
      "step": 4740
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00015619697528587237,
      "loss": 1.0904,
      "step": 4750
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00015610475839173736,
      "loss": 1.155,
      "step": 4760
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00015601254149760237,
      "loss": 1.0715,
      "step": 4770
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00015592032460346738,
      "loss": 1.0461,
      "step": 4780
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00015582810770933237,
      "loss": 1.1844,
      "step": 4790
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00015573589081519735,
      "loss": 1.2037,
      "step": 4800
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00015564367392106233,
      "loss": 1.0438,
      "step": 4810
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00015555145702692735,
      "loss": 1.1288,
      "step": 4820
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00015545924013279233,
      "loss": 1.0254,
      "step": 4830
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00015536702323865731,
      "loss": 1.0471,
      "step": 4840
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00015527480634452233,
      "loss": 1.1056,
      "step": 4850
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00015518258945038734,
      "loss": 0.9244,
      "step": 4860
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00015509037255625232,
      "loss": 1.0792,
      "step": 4870
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001549981556621173,
      "loss": 1.1597,
      "step": 4880
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001549059387679823,
      "loss": 1.2357,
      "step": 4890
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001548137218738473,
      "loss": 1.0048,
      "step": 4900
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001547215049797123,
      "loss": 1.1648,
      "step": 4910
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00015462928808557727,
      "loss": 0.9634,
      "step": 4920
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00015453707119144228,
      "loss": 1.1526,
      "step": 4930
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001544448542973073,
      "loss": 1.0899,
      "step": 4940
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00015435263740317228,
      "loss": 1.1109,
      "step": 4950
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00015426042050903726,
      "loss": 1.0795,
      "step": 4960
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00015416820361490225,
      "loss": 0.9967,
      "step": 4970
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00015407598672076726,
      "loss": 1.0689,
      "step": 4980
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00015398376982663224,
      "loss": 1.0041,
      "step": 4990
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00015389155293249723,
      "loss": 1.0633,
      "step": 5000
    },
    {
      "epoch": 0.92,
      "eval_accuracy": 0.597316048931897,
      "eval_loss": 1.0508341789245605,
      "eval_runtime": 119.8159,
      "eval_samples_per_second": 91.424,
      "eval_steps_per_second": 11.434,
      "step": 5000
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00015379933603836224,
      "loss": 1.0456,
      "step": 5010
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00015370711914422725,
      "loss": 1.1123,
      "step": 5020
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00015361490225009223,
      "loss": 1.1132,
      "step": 5030
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00015352268535595722,
      "loss": 1.0581,
      "step": 5040
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001534304684618222,
      "loss": 1.0916,
      "step": 5050
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001533382515676872,
      "loss": 1.2104,
      "step": 5060
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001532460346735522,
      "loss": 1.0765,
      "step": 5070
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00015315381777941718,
      "loss": 1.0358,
      "step": 5080
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001530616008852822,
      "loss": 1.253,
      "step": 5090
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00015296938399114718,
      "loss": 1.0404,
      "step": 5100
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001528771670970122,
      "loss": 1.1361,
      "step": 5110
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00015278495020287718,
      "loss": 1.0879,
      "step": 5120
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00015269273330874216,
      "loss": 1.0397,
      "step": 5130
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00015260051641460714,
      "loss": 1.1521,
      "step": 5140
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00015250829952047216,
      "loss": 1.009,
      "step": 5150
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00015241608262633714,
      "loss": 1.0276,
      "step": 5160
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00015232386573220215,
      "loss": 1.2377,
      "step": 5170
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00015223164883806714,
      "loss": 1.24,
      "step": 5180
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00015213943194393215,
      "loss": 1.1269,
      "step": 5190
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00015204721504979713,
      "loss": 1.0245,
      "step": 5200
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00015195499815566212,
      "loss": 0.9721,
      "step": 5210
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001518627812615271,
      "loss": 1.1268,
      "step": 5220
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001517705643673921,
      "loss": 1.0673,
      "step": 5230
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001516783474732571,
      "loss": 0.9425,
      "step": 5240
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001515861305791221,
      "loss": 1.1363,
      "step": 5250
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001514939136849871,
      "loss": 1.1446,
      "step": 5260
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001514016967908521,
      "loss": 1.1441,
      "step": 5270
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001513094798967171,
      "loss": 1.1669,
      "step": 5280
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015121726300258207,
      "loss": 1.0665,
      "step": 5290
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015112504610844706,
      "loss": 0.9422,
      "step": 5300
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015103282921431207,
      "loss": 1.0679,
      "step": 5310
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015094061232017705,
      "loss": 1.1073,
      "step": 5320
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015084839542604207,
      "loss": 1.042,
      "step": 5330
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00015075617853190705,
      "loss": 1.1157,
      "step": 5340
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00015066396163777206,
      "loss": 1.1154,
      "step": 5350
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00015057174474363705,
      "loss": 1.2048,
      "step": 5360
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00015047952784950203,
      "loss": 1.0115,
      "step": 5370
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00015038731095536701,
      "loss": 1.1561,
      "step": 5380
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00015029509406123203,
      "loss": 1.0331,
      "step": 5390
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.000150202877167097,
      "loss": 1.0371,
      "step": 5400
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00015011066027296202,
      "loss": 0.9979,
      "step": 5410
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.000150018443378827,
      "loss": 1.1012,
      "step": 5420
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00014992622648469202,
      "loss": 1.0519,
      "step": 5430
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.000149834009590557,
      "loss": 0.9354,
      "step": 5440
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.000149741792696422,
      "loss": 0.9012,
      "step": 5450
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00014964957580228697,
      "loss": 1.0494,
      "step": 5460
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00014955735890815198,
      "loss": 0.9693,
      "step": 5470
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00014946514201401697,
      "loss": 1.1344,
      "step": 5480
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00014937292511988198,
      "loss": 1.0195,
      "step": 5490
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00014928070822574696,
      "loss": 0.8604,
      "step": 5500
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00014918849133161197,
      "loss": 1.045,
      "step": 5510
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00014909627443747696,
      "loss": 0.9119,
      "step": 5520
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00014900405754334194,
      "loss": 1.0778,
      "step": 5530
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00014891184064920693,
      "loss": 0.9471,
      "step": 5540
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00014881962375507194,
      "loss": 1.0131,
      "step": 5550
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00014872740686093692,
      "loss": 1.0563,
      "step": 5560
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00014863518996680193,
      "loss": 0.9712,
      "step": 5570
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00014854297307266692,
      "loss": 0.9751,
      "step": 5580
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00014845075617853193,
      "loss": 0.9858,
      "step": 5590
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00014835853928439692,
      "loss": 0.8813,
      "step": 5600
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0001482663223902619,
      "loss": 0.9845,
      "step": 5610
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00014817410549612688,
      "loss": 0.8825,
      "step": 5620
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001480818886019919,
      "loss": 1.1025,
      "step": 5630
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00014798967170785688,
      "loss": 1.0646,
      "step": 5640
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001478974548137219,
      "loss": 1.0291,
      "step": 5650
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00014780523791958688,
      "loss": 0.9306,
      "step": 5660
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0001477130210254519,
      "loss": 1.0153,
      "step": 5670
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00014762080413131687,
      "loss": 0.9032,
      "step": 5680
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00014752858723718186,
      "loss": 1.011,
      "step": 5690
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00014743637034304684,
      "loss": 0.9371,
      "step": 5700
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00014734415344891183,
      "loss": 1.0678,
      "step": 5710
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00014725193655477684,
      "loss": 1.0736,
      "step": 5720
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00014715971966064185,
      "loss": 1.0615,
      "step": 5730
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00014706750276650683,
      "loss": 0.9433,
      "step": 5740
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00014697528587237182,
      "loss": 0.8013,
      "step": 5750
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00014688306897823683,
      "loss": 1.0051,
      "step": 5760
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0001467908520841018,
      "loss": 0.9543,
      "step": 5770
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001466986351899668,
      "loss": 0.918,
      "step": 5780
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00014660641829583178,
      "loss": 0.9612,
      "step": 5790
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001465142014016968,
      "loss": 0.9229,
      "step": 5800
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001464219845075618,
      "loss": 1.0071,
      "step": 5810
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001463297676134268,
      "loss": 1.0156,
      "step": 5820
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00014623755071929177,
      "loss": 0.9041,
      "step": 5830
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00014614533382515679,
      "loss": 0.8998,
      "step": 5840
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00014605311693102177,
      "loss": 1.0525,
      "step": 5850
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00014596090003688675,
      "loss": 1.0265,
      "step": 5860
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00014586868314275174,
      "loss": 0.8741,
      "step": 5870
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00014577646624861675,
      "loss": 1.0066,
      "step": 5880
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00014568424935448176,
      "loss": 1.1394,
      "step": 5890
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00014559203246034675,
      "loss": 0.9974,
      "step": 5900
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00014549981556621173,
      "loss": 0.9108,
      "step": 5910
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00014540759867207674,
      "loss": 0.9731,
      "step": 5920
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00014531538177794173,
      "loss": 0.914,
      "step": 5930
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001452231648838067,
      "loss": 1.1119,
      "step": 5940
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001451309479896717,
      "loss": 1.0655,
      "step": 5950
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001450387310955367,
      "loss": 0.982,
      "step": 5960
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00014494651420140172,
      "loss": 1.0214,
      "step": 5970
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001448542973072667,
      "loss": 0.9889,
      "step": 5980
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001447620804131317,
      "loss": 1.1089,
      "step": 5990
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001446698635189967,
      "loss": 0.9559,
      "step": 6000
    },
    {
      "epoch": 1.11,
      "eval_accuracy": 0.6106445134197553,
      "eval_loss": 1.017890453338623,
      "eval_runtime": 117.4367,
      "eval_samples_per_second": 93.276,
      "eval_steps_per_second": 11.666,
      "step": 6000
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00014457764662486168,
      "loss": 0.9921,
      "step": 6010
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00014448542973072667,
      "loss": 1.0913,
      "step": 6020
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00014439321283659165,
      "loss": 0.9651,
      "step": 6030
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00014430099594245666,
      "loss": 0.9044,
      "step": 6040
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00014420877904832167,
      "loss": 0.9205,
      "step": 6050
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00014411656215418666,
      "loss": 0.8574,
      "step": 6060
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00014402434526005164,
      "loss": 0.9738,
      "step": 6070
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00014393212836591665,
      "loss": 0.9524,
      "step": 6080
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00014383991147178164,
      "loss": 1.068,
      "step": 6090
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00014374769457764662,
      "loss": 0.9719,
      "step": 6100
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001436554776835116,
      "loss": 0.8633,
      "step": 6110
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00014356326078937662,
      "loss": 1.1786,
      "step": 6120
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00014347104389524163,
      "loss": 0.9662,
      "step": 6130
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00014337882700110662,
      "loss": 1.0083,
      "step": 6140
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001432866101069716,
      "loss": 0.9786,
      "step": 6150
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001431943932128366,
      "loss": 1.0162,
      "step": 6160
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001431021763187016,
      "loss": 1.1466,
      "step": 6170
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00014300995942456658,
      "loss": 0.9591,
      "step": 6180
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00014291774253043156,
      "loss": 0.9515,
      "step": 6190
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00014282552563629658,
      "loss": 0.9676,
      "step": 6200
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001427333087421616,
      "loss": 1.0068,
      "step": 6210
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00014264109184802657,
      "loss": 1.0432,
      "step": 6220
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00014254887495389156,
      "loss": 1.0162,
      "step": 6230
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00014245665805975657,
      "loss": 0.9487,
      "step": 6240
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00014236444116562155,
      "loss": 0.8694,
      "step": 6250
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00014227222427148654,
      "loss": 0.8753,
      "step": 6260
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00014218000737735152,
      "loss": 0.9522,
      "step": 6270
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00014208779048321653,
      "loss": 1.0287,
      "step": 6280
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00014199557358908154,
      "loss": 0.9371,
      "step": 6290
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00014190335669494653,
      "loss": 1.0457,
      "step": 6300
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001418111398008115,
      "loss": 0.9603,
      "step": 6310
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001417189229066765,
      "loss": 0.9226,
      "step": 6320
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001416267060125415,
      "loss": 0.9346,
      "step": 6330
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001415344891184065,
      "loss": 0.9726,
      "step": 6340
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00014144227222427148,
      "loss": 0.885,
      "step": 6350
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001413500553301365,
      "loss": 0.8587,
      "step": 6360
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001412578384360015,
      "loss": 1.0051,
      "step": 6370
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00014116562154186649,
      "loss": 1.1279,
      "step": 6380
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00014107340464773147,
      "loss": 0.9938,
      "step": 6390
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00014098118775359645,
      "loss": 0.884,
      "step": 6400
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00014088897085946147,
      "loss": 0.9564,
      "step": 6410
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00014079675396532645,
      "loss": 1.0019,
      "step": 6420
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00014070453707119143,
      "loss": 1.0467,
      "step": 6430
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00014061232017705645,
      "loss": 0.9473,
      "step": 6440
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00014052010328292146,
      "loss": 1.0038,
      "step": 6450
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00014042788638878644,
      "loss": 0.881,
      "step": 6460
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00014033566949465143,
      "loss": 0.9928,
      "step": 6470
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001402434526005164,
      "loss": 1.028,
      "step": 6480
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00014015123570638142,
      "loss": 0.8758,
      "step": 6490
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001400590188122464,
      "loss": 1.0071,
      "step": 6500
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001399668019181114,
      "loss": 0.9646,
      "step": 6510
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001398745850239764,
      "loss": 0.9039,
      "step": 6520
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00013978236812984141,
      "loss": 1.0133,
      "step": 6530
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001396901512357064,
      "loss": 0.9122,
      "step": 6540
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00013959793434157138,
      "loss": 0.9466,
      "step": 6550
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00013950571744743637,
      "loss": 1.0843,
      "step": 6560
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00013941350055330138,
      "loss": 0.9617,
      "step": 6570
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00013932128365916636,
      "loss": 0.9372,
      "step": 6580
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00013922906676503135,
      "loss": 0.8564,
      "step": 6590
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00013913684987089636,
      "loss": 1.0391,
      "step": 6600
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00013904463297676137,
      "loss": 1.0507,
      "step": 6610
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00013895241608262635,
      "loss": 1.116,
      "step": 6620
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00013886019918849134,
      "loss": 0.9702,
      "step": 6630
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00013876798229435632,
      "loss": 0.9306,
      "step": 6640
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00013867576540022134,
      "loss": 0.8996,
      "step": 6650
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00013858354850608632,
      "loss": 1.1706,
      "step": 6660
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001384913316119513,
      "loss": 1.0423,
      "step": 6670
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00013839911471781632,
      "loss": 1.0197,
      "step": 6680
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00013830689782368133,
      "loss": 0.885,
      "step": 6690
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001382146809295463,
      "loss": 1.0354,
      "step": 6700
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001381224640354113,
      "loss": 1.0241,
      "step": 6710
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00013803024714127628,
      "loss": 1.0995,
      "step": 6720
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001379380302471413,
      "loss": 0.9009,
      "step": 6730
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00013784581335300628,
      "loss": 1.0854,
      "step": 6740
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00013775359645887126,
      "loss": 1.0316,
      "step": 6750
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00013766137956473627,
      "loss": 0.9286,
      "step": 6760
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00013756916267060128,
      "loss": 0.7782,
      "step": 6770
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00013747694577646627,
      "loss": 1.0035,
      "step": 6780
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00013738472888233125,
      "loss": 1.0167,
      "step": 6790
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00013729251198819624,
      "loss": 0.8368,
      "step": 6800
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00013720029509406125,
      "loss": 0.9972,
      "step": 6810
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00013710807819992623,
      "loss": 0.958,
      "step": 6820
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00013701586130579122,
      "loss": 1.0516,
      "step": 6830
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00013692364441165623,
      "loss": 0.8795,
      "step": 6840
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00013683142751752124,
      "loss": 1.0168,
      "step": 6850
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00013673921062338622,
      "loss": 0.9276,
      "step": 6860
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001366469937292512,
      "loss": 1.1021,
      "step": 6870
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001365547768351162,
      "loss": 0.975,
      "step": 6880
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001364625599409812,
      "loss": 0.8585,
      "step": 6890
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001363703430468462,
      "loss": 1.0498,
      "step": 6900
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00013627812615271117,
      "loss": 1.0257,
      "step": 6910
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00013618590925857619,
      "loss": 0.9268,
      "step": 6920
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001360936923644412,
      "loss": 1.0438,
      "step": 6930
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00013600147547030618,
      "loss": 0.8607,
      "step": 6940
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00013590925857617117,
      "loss": 0.9124,
      "step": 6950
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00013581704168203615,
      "loss": 0.9851,
      "step": 6960
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00013572482478790113,
      "loss": 0.9737,
      "step": 6970
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00013563260789376615,
      "loss": 0.8973,
      "step": 6980
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00013554039099963113,
      "loss": 1.0597,
      "step": 6990
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00013544817410549614,
      "loss": 0.9984,
      "step": 7000
    },
    {
      "epoch": 1.29,
      "eval_accuracy": 0.6178564907796239,
      "eval_loss": 0.9958341121673584,
      "eval_runtime": 116.6094,
      "eval_samples_per_second": 93.938,
      "eval_steps_per_second": 11.749,
      "step": 7000
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00013535595721136113,
      "loss": 0.9331,
      "step": 7010
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00013526374031722614,
      "loss": 0.9406,
      "step": 7020
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00013517152342309112,
      "loss": 1.072,
      "step": 7030
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001350793065289561,
      "loss": 0.9313,
      "step": 7040
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001349870896348211,
      "loss": 0.9284,
      "step": 7050
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001348948727406861,
      "loss": 1.0917,
      "step": 7060
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001348026558465511,
      "loss": 0.9193,
      "step": 7070
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001347104389524161,
      "loss": 0.9261,
      "step": 7080
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00013461822205828108,
      "loss": 0.9229,
      "step": 7090
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001345260051641461,
      "loss": 0.8888,
      "step": 7100
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00013443378827001108,
      "loss": 1.0265,
      "step": 7110
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00013434157137587606,
      "loss": 0.913,
      "step": 7120
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013424935448174105,
      "loss": 0.8745,
      "step": 7130
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013415713758760606,
      "loss": 1.0292,
      "step": 7140
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013406492069347104,
      "loss": 0.9958,
      "step": 7150
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013397270379933603,
      "loss": 0.8609,
      "step": 7160
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013388048690520104,
      "loss": 1.0622,
      "step": 7170
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013378827001106605,
      "loss": 0.959,
      "step": 7180
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00013369605311693104,
      "loss": 1.009,
      "step": 7190
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00013360383622279602,
      "loss": 0.8563,
      "step": 7200
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.000133511619328661,
      "loss": 0.9683,
      "step": 7210
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00013341940243452602,
      "loss": 0.9321,
      "step": 7220
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.000133327185540391,
      "loss": 0.836,
      "step": 7230
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00013323496864625598,
      "loss": 1.0297,
      "step": 7240
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.000133142751752121,
      "loss": 0.8333,
      "step": 7250
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.000133050534857986,
      "loss": 0.8996,
      "step": 7260
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.000132958317963851,
      "loss": 0.8785,
      "step": 7270
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00013286610106971598,
      "loss": 0.9904,
      "step": 7280
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00013277388417558096,
      "loss": 1.0555,
      "step": 7290
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00013268166728144597,
      "loss": 1.0595,
      "step": 7300
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00013258945038731096,
      "loss": 0.9628,
      "step": 7310
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00013249723349317594,
      "loss": 0.9414,
      "step": 7320
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00013240501659904095,
      "loss": 1.0532,
      "step": 7330
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00013231279970490596,
      "loss": 0.796,
      "step": 7340
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00013222058281077095,
      "loss": 0.8783,
      "step": 7350
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00013212836591663593,
      "loss": 0.9517,
      "step": 7360
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00013203614902250092,
      "loss": 1.0121,
      "step": 7370
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00013194393212836593,
      "loss": 0.8807,
      "step": 7380
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001318517152342309,
      "loss": 0.9829,
      "step": 7390
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001317594983400959,
      "loss": 1.1037,
      "step": 7400
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001316672814459609,
      "loss": 0.9892,
      "step": 7410
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00013157506455182592,
      "loss": 0.8818,
      "step": 7420
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001314828476576909,
      "loss": 0.8415,
      "step": 7430
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001313906307635559,
      "loss": 0.9181,
      "step": 7440
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00013129841386942087,
      "loss": 1.0715,
      "step": 7450
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013120619697528589,
      "loss": 0.9243,
      "step": 7460
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013111398008115087,
      "loss": 0.911,
      "step": 7470
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013102176318701585,
      "loss": 0.8901,
      "step": 7480
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013092954629288087,
      "loss": 1.0062,
      "step": 7490
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013083732939874588,
      "loss": 0.9856,
      "step": 7500
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00013074511250461086,
      "loss": 1.0214,
      "step": 7510
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00013065289561047585,
      "loss": 0.8438,
      "step": 7520
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00013056067871634083,
      "loss": 0.8056,
      "step": 7530
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00013046846182220584,
      "loss": 1.0268,
      "step": 7540
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00013037624492807083,
      "loss": 0.9718,
      "step": 7550
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001302840280339358,
      "loss": 1.0429,
      "step": 7560
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00013019181113980082,
      "loss": 0.9091,
      "step": 7570
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001300995942456658,
      "loss": 0.9045,
      "step": 7580
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00013000737735153082,
      "loss": 1.0823,
      "step": 7590
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001299151604573958,
      "loss": 0.8333,
      "step": 7600
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001298229435632608,
      "loss": 0.8417,
      "step": 7610
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00012973072666912577,
      "loss": 0.9994,
      "step": 7620
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00012963850977499078,
      "loss": 1.0543,
      "step": 7630
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00012954629288085577,
      "loss": 0.9443,
      "step": 7640
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00012945407598672078,
      "loss": 0.9332,
      "step": 7650
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00012936185909258576,
      "loss": 1.0525,
      "step": 7660
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00012926964219845077,
      "loss": 1.0072,
      "step": 7670
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00012917742530431576,
      "loss": 0.8867,
      "step": 7680
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00012908520841018074,
      "loss": 1.0054,
      "step": 7690
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00012899299151604573,
      "loss": 1.0029,
      "step": 7700
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00012890077462191074,
      "loss": 1.1104,
      "step": 7710
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00012880855772777572,
      "loss": 1.0245,
      "step": 7720
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00012871634083364074,
      "loss": 0.8354,
      "step": 7730
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00012862412393950572,
      "loss": 0.8816,
      "step": 7740
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00012853190704537073,
      "loss": 0.8503,
      "step": 7750
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00012843969015123572,
      "loss": 0.9247,
      "step": 7760
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0001283474732571007,
      "loss": 0.9184,
      "step": 7770
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00012825525636296568,
      "loss": 0.9498,
      "step": 7780
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001281630394688307,
      "loss": 0.8573,
      "step": 7790
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00012807082257469568,
      "loss": 0.9359,
      "step": 7800
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001279786056805607,
      "loss": 0.8813,
      "step": 7810
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00012788638878642568,
      "loss": 0.9125,
      "step": 7820
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001277941718922907,
      "loss": 0.8091,
      "step": 7830
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00012770195499815567,
      "loss": 0.8828,
      "step": 7840
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00012760973810402066,
      "loss": 0.8689,
      "step": 7850
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00012751752120988564,
      "loss": 0.9063,
      "step": 7860
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00012742530431575065,
      "loss": 0.9397,
      "step": 7870
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00012733308742161564,
      "loss": 0.8055,
      "step": 7880
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00012724087052748065,
      "loss": 1.0647,
      "step": 7890
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00012714865363334563,
      "loss": 0.9543,
      "step": 7900
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00012705643673921064,
      "loss": 0.9021,
      "step": 7910
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00012696421984507563,
      "loss": 0.8933,
      "step": 7920
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001268720029509406,
      "loss": 0.9597,
      "step": 7930
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001267797860568056,
      "loss": 0.8972,
      "step": 7940
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001266875691626706,
      "loss": 0.8196,
      "step": 7950
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001265953522685356,
      "loss": 0.9674,
      "step": 7960
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001265031353744006,
      "loss": 0.9939,
      "step": 7970
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001264109184802656,
      "loss": 0.9279,
      "step": 7980
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0001263187015861306,
      "loss": 0.9121,
      "step": 7990
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00012622648469199559,
      "loss": 0.9843,
      "step": 8000
    },
    {
      "epoch": 1.48,
      "eval_accuracy": 0.6314588278254519,
      "eval_loss": 0.978008508682251,
      "eval_runtime": 119.0456,
      "eval_samples_per_second": 92.015,
      "eval_steps_per_second": 11.508,
      "step": 8000
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00012613426779786057,
      "loss": 1.0928,
      "step": 8010
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00012604205090372555,
      "loss": 1.008,
      "step": 8020
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00012594983400959057,
      "loss": 0.8691,
      "step": 8030
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00012585761711545555,
      "loss": 0.9296,
      "step": 8040
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00012576540022132056,
      "loss": 0.9163,
      "step": 8050
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00012567318332718555,
      "loss": 0.8996,
      "step": 8060
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00012558096643305056,
      "loss": 0.8988,
      "step": 8070
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00012548874953891554,
      "loss": 1.0093,
      "step": 8080
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00012539653264478053,
      "loss": 0.9081,
      "step": 8090
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0001253043157506455,
      "loss": 0.9128,
      "step": 8100
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00012521209885651052,
      "loss": 0.8395,
      "step": 8110
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0001251198819623755,
      "loss": 0.9784,
      "step": 8120
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00012502766506824052,
      "loss": 0.9919,
      "step": 8130
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0001249354481741055,
      "loss": 1.0478,
      "step": 8140
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00012484323127997051,
      "loss": 1.0693,
      "step": 8150
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.0001247510143858355,
      "loss": 0.9146,
      "step": 8160
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012465879749170048,
      "loss": 1.0575,
      "step": 8170
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012456658059756547,
      "loss": 0.9678,
      "step": 8180
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012447436370343048,
      "loss": 0.9004,
      "step": 8190
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012438214680929546,
      "loss": 0.9842,
      "step": 8200
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00012428992991516047,
      "loss": 0.9282,
      "step": 8210
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00012419771302102546,
      "loss": 0.9722,
      "step": 8220
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00012410549612689044,
      "loss": 1.1208,
      "step": 8230
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00012401327923275546,
      "loss": 0.8511,
      "step": 8240
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00012392106233862044,
      "loss": 0.9338,
      "step": 8250
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00012382884544448542,
      "loss": 0.8577,
      "step": 8260
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001237366285503504,
      "loss": 0.9919,
      "step": 8270
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00012364441165621542,
      "loss": 0.9534,
      "step": 8280
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00012355219476208043,
      "loss": 0.9393,
      "step": 8290
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00012345997786794542,
      "loss": 0.8468,
      "step": 8300
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001233677609738104,
      "loss": 0.8686,
      "step": 8310
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001232755440796754,
      "loss": 1.0077,
      "step": 8320
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0001231833271855404,
      "loss": 0.8607,
      "step": 8330
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00012309111029140538,
      "loss": 1.1106,
      "step": 8340
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00012299889339727036,
      "loss": 1.0332,
      "step": 8350
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00012290667650313538,
      "loss": 0.8054,
      "step": 8360
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0001228144596090004,
      "loss": 0.8144,
      "step": 8370
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00012272224271486537,
      "loss": 0.8902,
      "step": 8380
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00012263002582073036,
      "loss": 0.9698,
      "step": 8390
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00012253780892659537,
      "loss": 0.919,
      "step": 8400
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00012244559203246035,
      "loss": 0.8431,
      "step": 8410
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00012235337513832534,
      "loss": 1.0242,
      "step": 8420
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00012226115824419032,
      "loss": 0.9033,
      "step": 8430
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00012216894135005533,
      "loss": 0.9966,
      "step": 8440
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00012207672445592034,
      "loss": 0.8997,
      "step": 8450
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00012198450756178532,
      "loss": 0.9247,
      "step": 8460
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00012189229066765031,
      "loss": 0.7221,
      "step": 8470
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00012180007377351532,
      "loss": 0.8302,
      "step": 8480
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00012170785687938031,
      "loss": 1.0828,
      "step": 8490
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0001216156399852453,
      "loss": 0.9364,
      "step": 8500
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00012152342309111029,
      "loss": 0.927,
      "step": 8510
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0001214312061969753,
      "loss": 1.0354,
      "step": 8520
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00012133898930284029,
      "loss": 0.9749,
      "step": 8530
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012124677240870527,
      "loss": 0.8811,
      "step": 8540
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012115455551457027,
      "loss": 1.0251,
      "step": 8550
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012106233862043528,
      "loss": 0.9448,
      "step": 8560
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012097012172630027,
      "loss": 0.9655,
      "step": 8570
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012087790483216525,
      "loss": 0.9447,
      "step": 8580
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012078568793803025,
      "loss": 0.9376,
      "step": 8590
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00012069347104389526,
      "loss": 0.9657,
      "step": 8600
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00012060125414976024,
      "loss": 0.8257,
      "step": 8610
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00012050903725562523,
      "loss": 0.9139,
      "step": 8620
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00012041682036149023,
      "loss": 0.8469,
      "step": 8630
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00012032460346735524,
      "loss": 0.8781,
      "step": 8640
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00012023238657322022,
      "loss": 0.9082,
      "step": 8650
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0001201401696790852,
      "loss": 0.9517,
      "step": 8660
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0001200479527849502,
      "loss": 0.8731,
      "step": 8670
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00011995573589081522,
      "loss": 1.0299,
      "step": 8680
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0001198635189966802,
      "loss": 0.8816,
      "step": 8690
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00011977130210254518,
      "loss": 0.8483,
      "step": 8700
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00011967908520841018,
      "loss": 0.865,
      "step": 8710
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0001195868683142752,
      "loss": 0.8243,
      "step": 8720
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00011949465142014018,
      "loss": 0.8385,
      "step": 8730
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00011940243452600516,
      "loss": 1.0068,
      "step": 8740
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00011931021763187016,
      "loss": 0.8506,
      "step": 8750
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00011921800073773517,
      "loss": 0.9249,
      "step": 8760
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00011912578384360016,
      "loss": 1.1078,
      "step": 8770
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00011903356694946514,
      "loss": 1.0514,
      "step": 8780
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00011894135005533014,
      "loss": 0.9401,
      "step": 8790
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00011884913316119515,
      "loss": 0.8352,
      "step": 8800
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00011875691626706014,
      "loss": 0.9655,
      "step": 8810
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00011866469937292512,
      "loss": 1.1827,
      "step": 8820
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00011857248247879012,
      "loss": 0.9401,
      "step": 8830
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0001184802655846551,
      "loss": 0.9116,
      "step": 8840
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00011838804869052011,
      "loss": 0.8191,
      "step": 8850
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0001182958317963851,
      "loss": 0.8648,
      "step": 8860
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0001182036149022501,
      "loss": 0.9634,
      "step": 8870
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00011811139800811508,
      "loss": 0.981,
      "step": 8880
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00011801918111398009,
      "loss": 0.8884,
      "step": 8890
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00011792696421984508,
      "loss": 0.8395,
      "step": 8900
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00011783474732571007,
      "loss": 0.9688,
      "step": 8910
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00011774253043157506,
      "loss": 0.9953,
      "step": 8920
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00011765031353744007,
      "loss": 0.9366,
      "step": 8930
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00011755809664330505,
      "loss": 1.0527,
      "step": 8940
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00011746587974917005,
      "loss": 0.9375,
      "step": 8950
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00011737366285503504,
      "loss": 0.8931,
      "step": 8960
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00011728144596090005,
      "loss": 0.9988,
      "step": 8970
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00011718922906676503,
      "loss": 0.7768,
      "step": 8980
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00011709701217263003,
      "loss": 0.8878,
      "step": 8990
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00011700479527849502,
      "loss": 0.8943,
      "step": 9000
    },
    {
      "epoch": 1.66,
      "eval_accuracy": 0.6385795143326639,
      "eval_loss": 0.9470136761665344,
      "eval_runtime": 115.3232,
      "eval_samples_per_second": 94.985,
      "eval_steps_per_second": 11.88,
      "step": 9000
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00011691257838436003,
      "loss": 0.8918,
      "step": 9010
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00011682036149022501,
      "loss": 0.7877,
      "step": 9020
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00011672814459609001,
      "loss": 0.7868,
      "step": 9030
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.000116635927701955,
      "loss": 0.9623,
      "step": 9040
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00011654371080782,
      "loss": 0.9734,
      "step": 9050
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00011645149391368499,
      "loss": 0.9686,
      "step": 9060
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00011635927701954999,
      "loss": 0.9404,
      "step": 9070
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00011626706012541497,
      "loss": 0.9221,
      "step": 9080
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00011617484323127998,
      "loss": 0.9959,
      "step": 9090
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00011608262633714497,
      "loss": 0.8569,
      "step": 9100
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00011599040944300997,
      "loss": 0.8794,
      "step": 9110
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00011589819254887495,
      "loss": 0.8572,
      "step": 9120
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00011580597565473996,
      "loss": 0.9671,
      "step": 9130
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00011571375876060495,
      "loss": 0.953,
      "step": 9140
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00011562154186646994,
      "loss": 0.9218,
      "step": 9150
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00011552932497233493,
      "loss": 1.1192,
      "step": 9160
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00011543710807819994,
      "loss": 0.9314,
      "step": 9170
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00011534489118406492,
      "loss": 0.9793,
      "step": 9180
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00011525267428992992,
      "loss": 0.9013,
      "step": 9190
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001151604573957949,
      "loss": 0.9303,
      "step": 9200
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00011506824050165992,
      "loss": 0.868,
      "step": 9210
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001149760236075249,
      "loss": 0.9062,
      "step": 9220
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001148838067133899,
      "loss": 1.021,
      "step": 9230
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00011479158981925489,
      "loss": 0.8758,
      "step": 9240
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0001146993729251199,
      "loss": 0.9029,
      "step": 9250
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00011460715603098488,
      "loss": 0.9457,
      "step": 9260
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00011451493913684988,
      "loss": 0.9978,
      "step": 9270
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00011442272224271486,
      "loss": 0.9076,
      "step": 9280
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00011433050534857987,
      "loss": 0.996,
      "step": 9290
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00011423828845444486,
      "loss": 0.8963,
      "step": 9300
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00011414607156030986,
      "loss": 0.913,
      "step": 9310
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00011405385466617484,
      "loss": 1.0293,
      "step": 9320
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00011396163777203985,
      "loss": 0.9292,
      "step": 9330
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00011386942087790484,
      "loss": 0.8293,
      "step": 9340
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00011377720398376984,
      "loss": 1.0578,
      "step": 9350
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00011368498708963482,
      "loss": 0.9555,
      "step": 9360
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00011359277019549983,
      "loss": 0.7886,
      "step": 9370
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00011350055330136482,
      "loss": 0.8831,
      "step": 9380
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00011340833640722981,
      "loss": 0.9281,
      "step": 9390
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0001133161195130948,
      "loss": 0.8645,
      "step": 9400
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00011322390261895981,
      "loss": 0.9587,
      "step": 9410
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0001131316857248248,
      "loss": 0.9112,
      "step": 9420
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00011303946883068979,
      "loss": 0.8635,
      "step": 9430
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00011294725193655478,
      "loss": 1.012,
      "step": 9440
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00011285503504241979,
      "loss": 0.9985,
      "step": 9450
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00011276281814828477,
      "loss": 0.9016,
      "step": 9460
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011267060125414977,
      "loss": 1.0043,
      "step": 9470
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011257838436001475,
      "loss": 0.8155,
      "step": 9480
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011248616746587974,
      "loss": 0.9456,
      "step": 9490
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011239395057174475,
      "loss": 0.8511,
      "step": 9500
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011230173367760975,
      "loss": 0.8481,
      "step": 9510
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011220951678347473,
      "loss": 0.9624,
      "step": 9520
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011211729988933972,
      "loss": 1.0411,
      "step": 9530
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011202508299520473,
      "loss": 0.8652,
      "step": 9540
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011193286610106973,
      "loss": 0.968,
      "step": 9550
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011184064920693471,
      "loss": 0.8184,
      "step": 9560
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0001117484323127997,
      "loss": 0.9174,
      "step": 9570
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011165621541866471,
      "loss": 1.0584,
      "step": 9580
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0001115639985245297,
      "loss": 0.9099,
      "step": 9590
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011147178163039469,
      "loss": 0.8901,
      "step": 9600
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011137956473625967,
      "loss": 0.8803,
      "step": 9610
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011128734784212469,
      "loss": 0.9376,
      "step": 9620
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00011119513094798968,
      "loss": 1.0229,
      "step": 9630
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00011110291405385467,
      "loss": 0.8243,
      "step": 9640
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00011101069715971965,
      "loss": 1.0061,
      "step": 9650
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00011091848026558466,
      "loss": 0.9275,
      "step": 9660
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00011082626337144966,
      "loss": 0.8948,
      "step": 9670
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011073404647731465,
      "loss": 1.0331,
      "step": 9680
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011064182958317963,
      "loss": 0.9624,
      "step": 9690
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011054961268904464,
      "loss": 0.9445,
      "step": 9700
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011045739579490964,
      "loss": 0.88,
      "step": 9710
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011036517890077462,
      "loss": 0.9094,
      "step": 9720
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011027296200663961,
      "loss": 1.03,
      "step": 9730
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011018074511250462,
      "loss": 1.0257,
      "step": 9740
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011008852821836962,
      "loss": 0.9672,
      "step": 9750
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0001099963113242346,
      "loss": 0.9712,
      "step": 9760
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00010990409443009959,
      "loss": 0.9725,
      "step": 9770
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0001098118775359646,
      "loss": 0.8735,
      "step": 9780
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0001097196606418296,
      "loss": 0.8384,
      "step": 9790
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00010962744374769458,
      "loss": 0.9615,
      "step": 9800
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00010953522685355957,
      "loss": 0.99,
      "step": 9810
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00010944300995942458,
      "loss": 0.867,
      "step": 9820
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00010935079306528957,
      "loss": 0.9244,
      "step": 9830
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00010925857617115456,
      "loss": 0.8202,
      "step": 9840
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00010916635927701954,
      "loss": 0.905,
      "step": 9850
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00010907414238288456,
      "loss": 0.8523,
      "step": 9860
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00010898192548874955,
      "loss": 1.079,
      "step": 9870
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00010888970859461454,
      "loss": 0.9162,
      "step": 9880
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00010879749170047952,
      "loss": 0.9556,
      "step": 9890
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00010870527480634453,
      "loss": 0.9201,
      "step": 9900
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00010861305791220953,
      "loss": 0.8199,
      "step": 9910
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00010852084101807452,
      "loss": 0.9409,
      "step": 9920
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0001084286241239395,
      "loss": 0.9418,
      "step": 9930
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00010833640722980451,
      "loss": 0.9588,
      "step": 9940
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00010824419033566951,
      "loss": 0.9613,
      "step": 9950
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0001081519734415345,
      "loss": 0.9023,
      "step": 9960
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00010805975654739948,
      "loss": 0.841,
      "step": 9970
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00010796753965326449,
      "loss": 0.9361,
      "step": 9980
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00010787532275912949,
      "loss": 0.7094,
      "step": 9990
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00010778310586499447,
      "loss": 0.9941,
      "step": 10000
    },
    {
      "epoch": 1.84,
      "eval_accuracy": 0.634380135110462,
      "eval_loss": 0.9495174288749695,
      "eval_runtime": 113.8999,
      "eval_samples_per_second": 96.172,
      "eval_steps_per_second": 12.028,
      "step": 10000
    },
    {
      "epoch": 1.84,
      "step": 10000,
      "total_flos": 3.692982298032341e+18,
      "train_loss": 1.1900101467609405,
      "train_runtime": 3440.7653,
      "train_samples_per_second": 100.843,
      "train_steps_per_second": 6.303
    }
  ],
  "logging_steps": 10,
  "max_steps": 21688,
  "num_train_epochs": 4,
  "save_steps": 1000,
  "total_flos": 3.692982298032341e+18,
  "trial_name": null,
  "trial_params": null
}