{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.50003543837267,
  "eval_steps": 500,
  "global_step": 7055,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.0876745339854e-05,
      "eval_accuracy": 0.6605764392507908,
      "eval_loss": 1.4750796556472778,
      "eval_runtime": 7.2429,
      "eval_samples_per_second": 47.909,
      "eval_steps_per_second": 0.414,
      "step": 1
    },
    {
      "epoch": 0.00070876745339854,
      "grad_norm": 1.8515625,
      "learning_rate": 1e-05,
      "loss": 1.538,
      "step": 10
    },
    {
      "epoch": 0.00141753490679708,
      "grad_norm": 1.0,
      "learning_rate": 2e-05,
      "loss": 1.5311,
      "step": 20
    },
    {
      "epoch": 0.00212630236019562,
      "grad_norm": 0.57421875,
      "learning_rate": 3e-05,
      "loss": 1.5247,
      "step": 30
    },
    {
      "epoch": 0.00283506981359416,
      "grad_norm": 0.423828125,
      "learning_rate": 4e-05,
      "loss": 1.531,
      "step": 40
    },
    {
      "epoch": 0.0035438372669927,
      "grad_norm": 0.451171875,
      "learning_rate": 5e-05,
      "loss": 1.5146,
      "step": 50
    },
    {
      "epoch": 0.00425260472039124,
      "grad_norm": 0.453125,
      "learning_rate": 6e-05,
      "loss": 1.5055,
      "step": 60
    },
    {
      "epoch": 0.00496137217378978,
      "grad_norm": 0.515625,
      "learning_rate": 7e-05,
      "loss": 1.5345,
      "step": 70
    },
    {
      "epoch": 0.00567013962718832,
      "grad_norm": 0.53515625,
      "learning_rate": 8e-05,
      "loss": 1.5133,
      "step": 80
    },
    {
      "epoch": 0.00637890708058686,
      "grad_norm": 0.58984375,
      "learning_rate": 9e-05,
      "loss": 1.5043,
      "step": 90
    },
    {
      "epoch": 0.0070876745339854,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.513,
      "step": 100
    },
    {
      "epoch": 0.00779644198738394,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.5083,
      "step": 110
    },
    {
      "epoch": 0.00850520944078248,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0001,
      "loss": 1.5098,
      "step": 120
    },
    {
      "epoch": 0.00921397689418102,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0001,
      "loss": 1.4936,
      "step": 130
    },
    {
      "epoch": 0.00992274434757956,
      "grad_norm": 0.54296875,
      "learning_rate": 0.0001,
      "loss": 1.5091,
      "step": 140
    },
    {
      "epoch": 0.0106315118009781,
      "grad_norm": 0.439453125,
      "learning_rate": 0.0001,
      "loss": 1.5073,
      "step": 150
    },
    {
      "epoch": 0.01134027925437664,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.5161,
      "step": 160
    },
    {
      "epoch": 0.01204904670777518,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0001,
      "loss": 1.5067,
      "step": 170
    },
    {
      "epoch": 0.01275781416117372,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4984,
      "step": 180
    },
    {
      "epoch": 0.01346658161457226,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.5053,
      "step": 190
    },
    {
      "epoch": 0.0141753490679708,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0001,
      "loss": 1.4937,
      "step": 200
    },
    {
      "epoch": 0.01488411652136934,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.4946,
      "step": 210
    },
    {
      "epoch": 0.01559288397476788,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.5031,
      "step": 220
    },
    {
      "epoch": 0.01630165142816642,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0001,
      "loss": 1.4857,
      "step": 230
    },
    {
      "epoch": 0.01701041888156496,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.5134,
      "step": 240
    },
    {
      "epoch": 0.0177191863349635,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0001,
      "loss": 1.5069,
      "step": 250
    },
    {
      "epoch": 0.01842795378836204,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.5037,
      "step": 260
    },
    {
      "epoch": 0.01913672124176058,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0001,
      "loss": 1.4999,
      "step": 270
    },
    {
      "epoch": 0.01984548869515912,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0001,
      "loss": 1.5038,
      "step": 280
    },
    {
      "epoch": 0.02055425614855766,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4985,
      "step": 290
    },
    {
      "epoch": 0.0212630236019562,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.5112,
      "step": 300
    },
    {
      "epoch": 0.02197179105535474,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.5083,
      "step": 310
    },
    {
      "epoch": 0.02268055850875328,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.5104,
      "step": 320
    },
    {
      "epoch": 0.02338932596215182,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0001,
      "loss": 1.5132,
      "step": 330
    },
    {
      "epoch": 0.02409809341555036,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4956,
      "step": 340
    },
    {
      "epoch": 0.0248068608689489,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0001,
      "loss": 1.4949,
      "step": 350
    },
    {
      "epoch": 0.02551562832234744,
      "grad_norm": 0.46484375,
      "learning_rate": 0.0001,
      "loss": 1.5018,
      "step": 360
    },
    {
      "epoch": 0.02622439577574598,
      "grad_norm": 0.435546875,
      "learning_rate": 0.0001,
      "loss": 1.4821,
      "step": 370
    },
    {
      "epoch": 0.02693316322914452,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.5092,
      "step": 380
    },
    {
      "epoch": 0.02764193068254306,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0001,
      "loss": 1.4998,
      "step": 390
    },
    {
      "epoch": 0.0283506981359416,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4958,
      "step": 400
    },
    {
      "epoch": 0.02905946558934014,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.5031,
      "step": 410
    },
    {
      "epoch": 0.02976823304273868,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4929,
      "step": 420
    },
    {
      "epoch": 0.03047700049613722,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4987,
      "step": 430
    },
    {
      "epoch": 0.03118576794953576,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4853,
      "step": 440
    },
    {
      "epoch": 0.0318945354029343,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0001,
      "loss": 1.4968,
      "step": 450
    },
    {
      "epoch": 0.03260330285633284,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.4872,
      "step": 460
    },
    {
      "epoch": 0.03331207030973138,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0001,
      "loss": 1.5061,
      "step": 470
    },
    {
      "epoch": 0.03402083776312992,
      "grad_norm": 0.6484375,
      "learning_rate": 0.0001,
      "loss": 1.5078,
      "step": 480
    },
    {
      "epoch": 0.03472960521652846,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0001,
      "loss": 1.504,
      "step": 490
    },
    {
      "epoch": 0.035438372669927,
      "grad_norm": 0.470703125,
      "learning_rate": 0.0001,
      "loss": 1.5071,
      "step": 500
    },
    {
      "epoch": 0.035438372669927,
      "eval_accuracy": 0.6646616911746595,
      "eval_loss": 1.4113208055496216,
      "eval_runtime": 7.1977,
      "eval_samples_per_second": 48.21,
      "eval_steps_per_second": 0.417,
      "step": 500
    },
    {
      "epoch": 0.03614714012332554,
      "grad_norm": 0.5625,
      "learning_rate": 0.0001,
      "loss": 1.5028,
      "step": 510
    },
    {
      "epoch": 0.03685590757672408,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0001,
      "loss": 1.5042,
      "step": 520
    },
    {
      "epoch": 0.03756467503012262,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4969,
      "step": 530
    },
    {
      "epoch": 0.03827344248352116,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0001,
      "loss": 1.5016,
      "step": 540
    },
    {
      "epoch": 0.0389822099369197,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.5008,
      "step": 550
    },
    {
      "epoch": 0.03969097739031824,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.5126,
      "step": 560
    },
    {
      "epoch": 0.04039974484371678,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4949,
      "step": 570
    },
    {
      "epoch": 0.04110851229711532,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4912,
      "step": 580
    },
    {
      "epoch": 0.04181727975051386,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.4912,
      "step": 590
    },
    {
      "epoch": 0.0425260472039124,
      "grad_norm": 0.498046875,
      "learning_rate": 0.0001,
      "loss": 1.4945,
      "step": 600
    },
    {
      "epoch": 0.04323481465731094,
      "grad_norm": 0.455078125,
      "learning_rate": 0.0001,
      "loss": 1.4977,
      "step": 610
    },
    {
      "epoch": 0.04394358211070948,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.4908,
      "step": 620
    },
    {
      "epoch": 0.04465234956410802,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0001,
      "loss": 1.4937,
      "step": 630
    },
    {
      "epoch": 0.04536111701750656,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4966,
      "step": 640
    },
    {
      "epoch": 0.0460698844709051,
      "grad_norm": 0.58203125,
      "learning_rate": 0.0001,
      "loss": 1.4982,
      "step": 650
    },
    {
      "epoch": 0.04677865192430364,
      "grad_norm": 0.46484375,
      "learning_rate": 0.0001,
      "loss": 1.4889,
      "step": 660
    },
    {
      "epoch": 0.04748741937770218,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.4948,
      "step": 670
    },
    {
      "epoch": 0.04819618683110072,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0001,
      "loss": 1.5004,
      "step": 680
    },
    {
      "epoch": 0.04890495428449926,
      "grad_norm": 0.443359375,
      "learning_rate": 0.0001,
      "loss": 1.4888,
      "step": 690
    },
    {
      "epoch": 0.0496137217378978,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0001,
      "loss": 1.4981,
      "step": 700
    },
    {
      "epoch": 0.05032248919129634,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.504,
      "step": 710
    },
    {
      "epoch": 0.05103125664469488,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0001,
      "loss": 1.4909,
      "step": 720
    },
    {
      "epoch": 0.05174002409809342,
      "grad_norm": 0.498046875,
      "learning_rate": 0.0001,
      "loss": 1.4971,
      "step": 730
    },
    {
      "epoch": 0.05244879155149196,
      "grad_norm": 0.453125,
      "learning_rate": 0.0001,
      "loss": 1.4994,
      "step": 740
    },
    {
      "epoch": 0.0531575590048905,
      "grad_norm": 0.60546875,
      "learning_rate": 0.0001,
      "loss": 1.4926,
      "step": 750
    },
    {
      "epoch": 0.05386632645828904,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4945,
      "step": 760
    },
    {
      "epoch": 0.05457509391168758,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0001,
      "loss": 1.5003,
      "step": 770
    },
    {
      "epoch": 0.05528386136508612,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0001,
      "loss": 1.4998,
      "step": 780
    },
    {
      "epoch": 0.05599262881848466,
      "grad_norm": 0.59375,
      "learning_rate": 0.0001,
      "loss": 1.5005,
      "step": 790
    },
    {
      "epoch": 0.0567013962718832,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0001,
      "loss": 1.509,
      "step": 800
    },
    {
      "epoch": 0.05741016372528174,
      "grad_norm": 0.4375,
      "learning_rate": 0.0001,
      "loss": 1.4885,
      "step": 810
    },
    {
      "epoch": 0.05811893117868028,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4957,
      "step": 820
    },
    {
      "epoch": 0.05882769863207882,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.5006,
      "step": 830
    },
    {
      "epoch": 0.05953646608547736,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.493,
      "step": 840
    },
    {
      "epoch": 0.0602452335388759,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.5008,
      "step": 850
    },
    {
      "epoch": 0.06095400099227444,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0001,
      "loss": 1.4966,
      "step": 860
    },
    {
      "epoch": 0.06166276844567298,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.5013,
      "step": 870
    },
    {
      "epoch": 0.06237153589907152,
      "grad_norm": 0.4609375,
      "learning_rate": 0.0001,
      "loss": 1.4937,
      "step": 880
    },
    {
      "epoch": 0.06308030335247006,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.4815,
      "step": 890
    },
    {
      "epoch": 0.0637890708058686,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0001,
      "loss": 1.4878,
      "step": 900
    },
    {
      "epoch": 0.06449783825926714,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0001,
      "loss": 1.4948,
      "step": 910
    },
    {
      "epoch": 0.06520660571266568,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.5051,
      "step": 920
    },
    {
      "epoch": 0.06591537316606422,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0001,
      "loss": 1.4993,
      "step": 930
    },
    {
      "epoch": 0.06662414061946276,
      "grad_norm": 0.470703125,
      "learning_rate": 0.0001,
      "loss": 1.4984,
      "step": 940
    },
    {
      "epoch": 0.0673329080728613,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001,
      "loss": 1.5001,
      "step": 950
    },
    {
      "epoch": 0.06804167552625984,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0001,
      "loss": 1.4899,
      "step": 960
    },
    {
      "epoch": 0.06875044297965838,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4881,
      "step": 970
    },
    {
      "epoch": 0.06945921043305692,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001,
      "loss": 1.5096,
      "step": 980
    },
    {
      "epoch": 0.07016797788645546,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0001,
      "loss": 1.4902,
      "step": 990
    },
    {
      "epoch": 0.070876745339854,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0001,
      "loss": 1.5003,
      "step": 1000
    },
    {
      "epoch": 0.070876745339854,
      "eval_accuracy": 0.6648911127297294,
      "eval_loss": 1.4080075025558472,
      "eval_runtime": 7.1527,
      "eval_samples_per_second": 48.513,
      "eval_steps_per_second": 0.419,
      "step": 1000
    },
    {
      "epoch": 0.07158551279325254,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0001,
      "loss": 1.4882,
      "step": 1010
    },
    {
      "epoch": 0.07229428024665108,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0001,
      "loss": 1.5019,
      "step": 1020
    },
    {
      "epoch": 0.07300304770004962,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.488,
      "step": 1030
    },
    {
      "epoch": 0.07371181515344816,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.5065,
      "step": 1040
    },
    {
      "epoch": 0.0744205826068467,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0001,
      "loss": 1.4906,
      "step": 1050
    },
    {
      "epoch": 0.07512935006024524,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0001,
      "loss": 1.5036,
      "step": 1060
    },
    {
      "epoch": 0.07583811751364378,
      "grad_norm": 0.609375,
      "learning_rate": 0.0001,
      "loss": 1.4968,
      "step": 1070
    },
    {
      "epoch": 0.07654688496704232,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0001,
      "loss": 1.5076,
      "step": 1080
    },
    {
      "epoch": 0.07725565242044086,
      "grad_norm": 0.466796875,
      "learning_rate": 0.0001,
      "loss": 1.5008,
      "step": 1090
    },
    {
      "epoch": 0.0779644198738394,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0001,
      "loss": 1.4951,
      "step": 1100
    },
    {
      "epoch": 0.07867318732723794,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.4844,
      "step": 1110
    },
    {
      "epoch": 0.07938195478063648,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.499,
      "step": 1120
    },
    {
      "epoch": 0.08009072223403502,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0001,
      "loss": 1.4794,
      "step": 1130
    },
    {
      "epoch": 0.08079948968743356,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0001,
      "loss": 1.508,
      "step": 1140
    },
    {
      "epoch": 0.0815082571408321,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0001,
      "loss": 1.4997,
      "step": 1150
    },
    {
      "epoch": 0.08221702459423064,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0001,
      "loss": 1.4987,
      "step": 1160
    },
    {
      "epoch": 0.08292579204762918,
      "grad_norm": 0.578125,
      "learning_rate": 0.0001,
      "loss": 1.5028,
      "step": 1170
    },
    {
      "epoch": 0.08363455950102772,
      "grad_norm": 0.453125,
      "learning_rate": 0.0001,
      "loss": 1.4991,
      "step": 1180
    },
    {
      "epoch": 0.08434332695442626,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.5075,
      "step": 1190
    },
    {
      "epoch": 0.0850520944078248,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0001,
      "loss": 1.4845,
      "step": 1200
    },
    {
      "epoch": 0.08576086186122334,
      "grad_norm": 0.5703125,
      "learning_rate": 0.0001,
      "loss": 1.4993,
      "step": 1210
    },
    {
      "epoch": 0.08646962931462188,
      "grad_norm": 0.58984375,
      "learning_rate": 0.0001,
      "loss": 1.502,
      "step": 1220
    },
    {
      "epoch": 0.08717839676802042,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0001,
      "loss": 1.4859,
      "step": 1230
    },
    {
      "epoch": 0.08788716422141896,
      "grad_norm": 0.498046875,
      "learning_rate": 0.0001,
      "loss": 1.5029,
      "step": 1240
    },
    {
      "epoch": 0.0885959316748175,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0001,
      "loss": 1.5035,
      "step": 1250
    },
    {
      "epoch": 0.08930469912821604,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0001,
      "loss": 1.5057,
      "step": 1260
    },
    {
      "epoch": 0.09001346658161458,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4862,
      "step": 1270
    },
    {
      "epoch": 0.09072223403501312,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.4941,
      "step": 1280
    },
    {
      "epoch": 0.09143100148841166,
      "grad_norm": 0.427734375,
      "learning_rate": 0.0001,
      "loss": 1.4995,
      "step": 1290
    },
    {
      "epoch": 0.0921397689418102,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.4891,
      "step": 1300
    },
    {
      "epoch": 0.09284853639520874,
      "grad_norm": 0.4296875,
      "learning_rate": 0.0001,
      "loss": 1.5021,
      "step": 1310
    },
    {
      "epoch": 0.09355730384860728,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0001,
      "loss": 1.4896,
      "step": 1320
    },
    {
      "epoch": 0.09426607130200582,
      "grad_norm": 0.58203125,
      "learning_rate": 0.0001,
      "loss": 1.4889,
      "step": 1330
    },
    {
      "epoch": 0.09497483875540436,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4978,
      "step": 1340
    },
    {
      "epoch": 0.0956836062088029,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.505,
      "step": 1350
    },
    {
      "epoch": 0.09639237366220144,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0001,
      "loss": 1.4934,
      "step": 1360
    },
    {
      "epoch": 0.09710114111559998,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0001,
      "loss": 1.4956,
      "step": 1370
    },
    {
      "epoch": 0.09780990856899852,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001,
      "loss": 1.4973,
      "step": 1380
    },
    {
      "epoch": 0.09851867602239706,
      "grad_norm": 0.498046875,
      "learning_rate": 0.0001,
      "loss": 1.4906,
      "step": 1390
    },
    {
      "epoch": 0.0992274434757956,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4942,
      "step": 1400
    },
    {
      "epoch": 0.09993621092919414,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0001,
      "loss": 1.4881,
      "step": 1410
    },
    {
      "epoch": 0.10064497838259268,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0001,
      "loss": 1.4837,
      "step": 1420
    },
    {
      "epoch": 0.10135374583599122,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4954,
      "step": 1430
    },
    {
      "epoch": 0.10206251328938976,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4838,
      "step": 1440
    },
    {
      "epoch": 0.1027712807427883,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0001,
      "loss": 1.4998,
      "step": 1450
    },
    {
      "epoch": 0.10348004819618684,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4998,
      "step": 1460
    },
    {
      "epoch": 0.10418881564958538,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0001,
      "loss": 1.4876,
      "step": 1470
    },
    {
      "epoch": 0.10489758310298392,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0001,
      "loss": 1.4965,
      "step": 1480
    },
    {
      "epoch": 0.10560635055638246,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4875,
      "step": 1490
    },
    {
      "epoch": 0.106315118009781,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4959,
      "step": 1500
    },
    {
      "epoch": 0.106315118009781,
      "eval_accuracy": 0.6653851431949415,
      "eval_loss": 1.4062947034835815,
      "eval_runtime": 7.3356,
      "eval_samples_per_second": 47.304,
      "eval_steps_per_second": 0.409,
      "step": 1500
    },
    {
      "epoch": 0.10702388546317954,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.4954,
      "step": 1510
    },
    {
      "epoch": 0.10773265291657808,
      "grad_norm": 0.59375,
      "learning_rate": 0.0001,
      "loss": 1.4988,
      "step": 1520
    },
    {
      "epoch": 0.10844142036997662,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0001,
      "loss": 1.4895,
      "step": 1530
    },
    {
      "epoch": 0.10915018782337516,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.5043,
      "step": 1540
    },
    {
      "epoch": 0.1098589552767737,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.4746,
      "step": 1550
    },
    {
      "epoch": 0.11056772273017224,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4898,
      "step": 1560
    },
    {
      "epoch": 0.11127649018357078,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0001,
      "loss": 1.4876,
      "step": 1570
    },
    {
      "epoch": 0.11198525763696932,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0001,
      "loss": 1.4966,
      "step": 1580
    },
    {
      "epoch": 0.11269402509036786,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0001,
      "loss": 1.5048,
      "step": 1590
    },
    {
      "epoch": 0.1134027925437664,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.482,
      "step": 1600
    },
    {
      "epoch": 0.11411155999716494,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0001,
      "loss": 1.4951,
      "step": 1610
    },
    {
      "epoch": 0.11482032745056348,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.485,
      "step": 1620
    },
    {
      "epoch": 0.11552909490396202,
      "grad_norm": 0.455078125,
      "learning_rate": 0.0001,
      "loss": 1.491,
      "step": 1630
    },
    {
      "epoch": 0.11623786235736056,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0001,
      "loss": 1.4785,
      "step": 1640
    },
    {
      "epoch": 0.1169466298107591,
      "grad_norm": 0.5625,
      "learning_rate": 0.0001,
      "loss": 1.4971,
      "step": 1650
    },
    {
      "epoch": 0.11765539726415764,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0001,
      "loss": 1.5013,
      "step": 1660
    },
    {
      "epoch": 0.11836416471755618,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0001,
      "loss": 1.5077,
      "step": 1670
    },
    {
      "epoch": 0.11907293217095472,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0001,
      "loss": 1.4939,
      "step": 1680
    },
    {
      "epoch": 0.11978169962435326,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0001,
      "loss": 1.4829,
      "step": 1690
    },
    {
      "epoch": 0.1204904670777518,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4964,
      "step": 1700
    },
    {
      "epoch": 0.12119923453115033,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4969,
      "step": 1710
    },
    {
      "epoch": 0.12190800198454887,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4955,
      "step": 1720
    },
    {
      "epoch": 0.12261676943794741,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.4913,
      "step": 1730
    },
    {
      "epoch": 0.12332553689134595,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.4905,
      "step": 1740
    },
    {
      "epoch": 0.1240343043447445,
      "grad_norm": 0.4609375,
      "learning_rate": 0.0001,
      "loss": 1.5034,
      "step": 1750
    },
    {
      "epoch": 0.12474307179814303,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.4991,
      "step": 1760
    },
    {
      "epoch": 0.12545183925154157,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.5019,
      "step": 1770
    },
    {
      "epoch": 0.12616060670494011,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0001,
      "loss": 1.4943,
      "step": 1780
    },
    {
      "epoch": 0.12686937415833865,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4971,
      "step": 1790
    },
    {
      "epoch": 0.1275781416117372,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0001,
      "loss": 1.4961,
      "step": 1800
    },
    {
      "epoch": 0.12828690906513573,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0001,
      "loss": 1.4888,
      "step": 1810
    },
    {
      "epoch": 0.12899567651853427,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4974,
      "step": 1820
    },
    {
      "epoch": 0.12970444397193281,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4866,
      "step": 1830
    },
    {
      "epoch": 0.13041321142533135,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4878,
      "step": 1840
    },
    {
      "epoch": 0.1311219788787299,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4855,
      "step": 1850
    },
    {
      "epoch": 0.13183074633212843,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.503,
      "step": 1860
    },
    {
      "epoch": 0.13253951378552697,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4955,
      "step": 1870
    },
    {
      "epoch": 0.13324828123892551,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4956,
      "step": 1880
    },
    {
      "epoch": 0.13395704869232405,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0001,
      "loss": 1.4922,
      "step": 1890
    },
    {
      "epoch": 0.1346658161457226,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.5161,
      "step": 1900
    },
    {
      "epoch": 0.13537458359912113,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4909,
      "step": 1910
    },
    {
      "epoch": 0.13608335105251967,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.4936,
      "step": 1920
    },
    {
      "epoch": 0.13679211850591821,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0001,
      "loss": 1.4882,
      "step": 1930
    },
    {
      "epoch": 0.13750088595931675,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0001,
      "loss": 1.4745,
      "step": 1940
    },
    {
      "epoch": 0.1382096534127153,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.4964,
      "step": 1950
    },
    {
      "epoch": 0.13891842086611383,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.5041,
      "step": 1960
    },
    {
      "epoch": 0.13962718831951237,
      "grad_norm": 0.62890625,
      "learning_rate": 0.0001,
      "loss": 1.4916,
      "step": 1970
    },
    {
      "epoch": 0.14033595577291091,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.4992,
      "step": 1980
    },
    {
      "epoch": 0.14104472322630945,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.5001,
      "step": 1990
    },
    {
      "epoch": 0.141753490679708,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.5019,
      "step": 2000
    },
    {
      "epoch": 0.141753490679708,
      "eval_accuracy": 0.6655364488217514,
      "eval_loss": 1.4053725004196167,
      "eval_runtime": 7.1455,
      "eval_samples_per_second": 48.562,
      "eval_steps_per_second": 0.42,
      "step": 2000
    },
    {
      "epoch": 0.14246225813310653,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0001,
      "loss": 1.5015,
      "step": 2010
    },
    {
      "epoch": 0.14317102558650507,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4842,
      "step": 2020
    },
    {
      "epoch": 0.14387979303990361,
      "grad_norm": 0.466796875,
      "learning_rate": 0.0001,
      "loss": 1.4868,
      "step": 2030
    },
    {
      "epoch": 0.14458856049330215,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0001,
      "loss": 1.4947,
      "step": 2040
    },
    {
      "epoch": 0.1452973279467007,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0001,
      "loss": 1.4964,
      "step": 2050
    },
    {
      "epoch": 0.14600609540009923,
      "grad_norm": 0.578125,
      "learning_rate": 0.0001,
      "loss": 1.5029,
      "step": 2060
    },
    {
      "epoch": 0.14671486285349777,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0001,
      "loss": 1.4974,
      "step": 2070
    },
    {
      "epoch": 0.1474236303068963,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.488,
      "step": 2080
    },
    {
      "epoch": 0.14813239776029485,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.5008,
      "step": 2090
    },
    {
      "epoch": 0.1488411652136934,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.5018,
      "step": 2100
    },
    {
      "epoch": 0.14954993266709193,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.5048,
      "step": 2110
    },
    {
      "epoch": 0.15025870012049047,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.5077,
      "step": 2120
    },
    {
      "epoch": 0.150967467573889,
      "grad_norm": 0.466796875,
      "learning_rate": 0.0001,
      "loss": 1.4962,
      "step": 2130
    },
    {
      "epoch": 0.15167623502728755,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4966,
      "step": 2140
    },
    {
      "epoch": 0.1523850024806861,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0001,
      "loss": 1.4926,
      "step": 2150
    },
    {
      "epoch": 0.15309376993408463,
      "grad_norm": 0.640625,
      "learning_rate": 0.0001,
      "loss": 1.4895,
      "step": 2160
    },
    {
      "epoch": 0.15380253738748317,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.4877,
      "step": 2170
    },
    {
      "epoch": 0.1545113048408817,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0001,
      "loss": 1.4981,
      "step": 2180
    },
    {
      "epoch": 0.15522007229428025,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4947,
      "step": 2190
    },
    {
      "epoch": 0.1559288397476788,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0001,
      "loss": 1.4891,
      "step": 2200
    },
    {
      "epoch": 0.15663760720107733,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.5004,
      "step": 2210
    },
    {
      "epoch": 0.15734637465447587,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.4877,
      "step": 2220
    },
    {
      "epoch": 0.1580551421078744,
      "grad_norm": 0.455078125,
      "learning_rate": 0.0001,
      "loss": 1.4966,
      "step": 2230
    },
    {
      "epoch": 0.15876390956127295,
      "grad_norm": 0.609375,
      "learning_rate": 0.0001,
      "loss": 1.4934,
      "step": 2240
    },
    {
      "epoch": 0.1594726770146715,
      "grad_norm": 0.5625,
      "learning_rate": 0.0001,
      "loss": 1.4981,
      "step": 2250
    },
    {
      "epoch": 0.16018144446807003,
      "grad_norm": 0.6328125,
      "learning_rate": 0.0001,
      "loss": 1.5087,
      "step": 2260
    },
    {
      "epoch": 0.16089021192146857,
      "grad_norm": 0.5625,
      "learning_rate": 0.0001,
      "loss": 1.4989,
      "step": 2270
    },
    {
      "epoch": 0.1615989793748671,
      "grad_norm": 0.609375,
      "learning_rate": 0.0001,
      "loss": 1.4785,
      "step": 2280
    },
    {
      "epoch": 0.16230774682826565,
      "grad_norm": 0.498046875,
      "learning_rate": 0.0001,
      "loss": 1.4978,
      "step": 2290
    },
    {
      "epoch": 0.1630165142816642,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0001,
      "loss": 1.4931,
      "step": 2300
    },
    {
      "epoch": 0.16372528173506273,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0001,
      "loss": 1.5119,
      "step": 2310
    },
    {
      "epoch": 0.16443404918846127,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.5097,
      "step": 2320
    },
    {
      "epoch": 0.1651428166418598,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0001,
      "loss": 1.481,
      "step": 2330
    },
    {
      "epoch": 0.16585158409525835,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0001,
      "loss": 1.4884,
      "step": 2340
    },
    {
      "epoch": 0.1665603515486569,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4934,
      "step": 2350
    },
    {
      "epoch": 0.16726911900205543,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.4789,
      "step": 2360
    },
    {
      "epoch": 0.16797788645545397,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0001,
      "loss": 1.5063,
      "step": 2370
    },
    {
      "epoch": 0.1686866539088525,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.5095,
      "step": 2380
    },
    {
      "epoch": 0.16939542136225105,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0001,
      "loss": 1.4949,
      "step": 2390
    },
    {
      "epoch": 0.1701041888156496,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0001,
      "loss": 1.4918,
      "step": 2400
    },
    {
      "epoch": 0.17081295626904813,
      "grad_norm": 0.455078125,
      "learning_rate": 0.0001,
      "loss": 1.5029,
      "step": 2410
    },
    {
      "epoch": 0.17152172372244667,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0001,
      "loss": 1.4945,
      "step": 2420
    },
    {
      "epoch": 0.1722304911758452,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0001,
      "loss": 1.4933,
      "step": 2430
    },
    {
      "epoch": 0.17293925862924375,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0001,
      "loss": 1.4949,
      "step": 2440
    },
    {
      "epoch": 0.1736480260826423,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0001,
      "loss": 1.5043,
      "step": 2450
    },
    {
      "epoch": 0.17435679353604083,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0001,
      "loss": 1.4971,
      "step": 2460
    },
    {
      "epoch": 0.17506556098943937,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.49,
      "step": 2470
    },
    {
      "epoch": 0.1757743284428379,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0001,
      "loss": 1.4892,
      "step": 2480
    },
    {
      "epoch": 0.17648309589623645,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.5006,
      "step": 2490
    },
    {
      "epoch": 0.177191863349635,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0001,
      "loss": 1.4891,
      "step": 2500
    },
    {
      "epoch": 0.177191863349635,
      "eval_accuracy": 0.6656455296224749,
      "eval_loss": 1.4047455787658691,
      "eval_runtime": 7.1711,
      "eval_samples_per_second": 48.388,
      "eval_steps_per_second": 0.418,
      "step": 2500
    },
    {
      "epoch": 0.17790063080303353,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0001,
      "loss": 1.4883,
      "step": 2510
    },
    {
      "epoch": 0.17860939825643207,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4906,
      "step": 2520
    },
    {
      "epoch": 0.1793181657098306,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.4946,
      "step": 2530
    },
    {
      "epoch": 0.18002693316322915,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0001,
      "loss": 1.4933,
      "step": 2540
    },
    {
      "epoch": 0.1807357006166277,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0001,
      "loss": 1.4794,
      "step": 2550
    },
    {
      "epoch": 0.18144446807002623,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.5019,
      "step": 2560
    },
    {
      "epoch": 0.18215323552342477,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.4956,
      "step": 2570
    },
    {
      "epoch": 0.1828620029768233,
      "grad_norm": 0.4609375,
      "learning_rate": 0.0001,
      "loss": 1.5043,
      "step": 2580
    },
    {
      "epoch": 0.18357077043022185,
      "grad_norm": 0.455078125,
      "learning_rate": 0.0001,
      "loss": 1.4928,
      "step": 2590
    },
    {
      "epoch": 0.1842795378836204,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.5021,
      "step": 2600
    },
    {
      "epoch": 0.18498830533701893,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.5033,
      "step": 2610
    },
    {
      "epoch": 0.18569707279041747,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.4875,
      "step": 2620
    },
    {
      "epoch": 0.186405840243816,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.4913,
      "step": 2630
    },
    {
      "epoch": 0.18711460769721455,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.4929,
      "step": 2640
    },
    {
      "epoch": 0.1878233751506131,
      "grad_norm": 0.4609375,
      "learning_rate": 0.0001,
      "loss": 1.4948,
      "step": 2650
    },
    {
      "epoch": 0.18853214260401163,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.5015,
      "step": 2660
    },
    {
      "epoch": 0.18924091005741017,
      "grad_norm": 0.470703125,
      "learning_rate": 0.0001,
      "loss": 1.4996,
      "step": 2670
    },
    {
      "epoch": 0.1899496775108087,
      "grad_norm": 0.431640625,
      "learning_rate": 0.0001,
      "loss": 1.487,
      "step": 2680
    },
    {
      "epoch": 0.19065844496420725,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4944,
      "step": 2690
    },
    {
      "epoch": 0.1913672124176058,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4889,
      "step": 2700
    },
    {
      "epoch": 0.19207597987100433,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4996,
      "step": 2710
    },
    {
      "epoch": 0.19278474732440287,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.4908,
      "step": 2720
    },
    {
      "epoch": 0.1934935147778014,
      "grad_norm": 0.4453125,
      "learning_rate": 0.0001,
      "loss": 1.4877,
      "step": 2730
    },
    {
      "epoch": 0.19420228223119995,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0001,
      "loss": 1.4824,
      "step": 2740
    },
    {
      "epoch": 0.1949110496845985,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.4925,
      "step": 2750
    },
    {
      "epoch": 0.19561981713799703,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4895,
      "step": 2760
    },
    {
      "epoch": 0.19632858459139557,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0001,
      "loss": 1.489,
      "step": 2770
    },
    {
      "epoch": 0.1970373520447941,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0001,
      "loss": 1.4968,
      "step": 2780
    },
    {
      "epoch": 0.19774611949819265,
      "grad_norm": 0.455078125,
      "learning_rate": 0.0001,
      "loss": 1.4985,
      "step": 2790
    },
    {
      "epoch": 0.1984548869515912,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.482,
      "step": 2800
    },
    {
      "epoch": 0.19916365440498973,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.5019,
      "step": 2810
    },
    {
      "epoch": 0.19987242185838827,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4914,
      "step": 2820
    },
    {
      "epoch": 0.2005811893117868,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4989,
      "step": 2830
    },
    {
      "epoch": 0.20128995676518535,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0001,
      "loss": 1.4927,
      "step": 2840
    },
    {
      "epoch": 0.2019987242185839,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001,
      "loss": 1.496,
      "step": 2850
    },
    {
      "epoch": 0.20270749167198243,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0001,
      "loss": 1.4828,
      "step": 2860
    },
    {
      "epoch": 0.20341625912538097,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.4945,
      "step": 2870
    },
    {
      "epoch": 0.2041250265787795,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4906,
      "step": 2880
    },
    {
      "epoch": 0.20483379403217805,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0001,
      "loss": 1.5005,
      "step": 2890
    },
    {
      "epoch": 0.2055425614855766,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0001,
      "loss": 1.5009,
      "step": 2900
    },
    {
      "epoch": 0.20625132893897513,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4862,
      "step": 2910
    },
    {
      "epoch": 0.20696009639237367,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0001,
      "loss": 1.4833,
      "step": 2920
    },
    {
      "epoch": 0.2076688638457722,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.4975,
      "step": 2930
    },
    {
      "epoch": 0.20837763129917075,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4992,
      "step": 2940
    },
    {
      "epoch": 0.2090863987525693,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0001,
      "loss": 1.4878,
      "step": 2950
    },
    {
      "epoch": 0.20979516620596783,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0001,
      "loss": 1.4978,
      "step": 2960
    },
    {
      "epoch": 0.21050393365936637,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.484,
      "step": 2970
    },
    {
      "epoch": 0.2112127011127649,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0001,
      "loss": 1.5065,
      "step": 2980
    },
    {
      "epoch": 0.21192146856616345,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.4829,
      "step": 2990
    },
    {
      "epoch": 0.212630236019562,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4916,
      "step": 3000
    },
    {
      "epoch": 0.212630236019562,
      "eval_accuracy": 0.6657292755275465,
      "eval_loss": 1.4040066003799438,
      "eval_runtime": 7.1205,
      "eval_samples_per_second": 48.733,
      "eval_steps_per_second": 0.421,
      "step": 3000
    },
    {
      "epoch": 0.21333900347296053,
      "grad_norm": 0.44140625,
      "learning_rate": 0.0001,
      "loss": 1.4859,
      "step": 3010
    },
    {
      "epoch": 0.21404777092635907,
      "grad_norm": 0.54296875,
      "learning_rate": 0.0001,
      "loss": 1.4907,
      "step": 3020
    },
    {
      "epoch": 0.2147565383797576,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0001,
      "loss": 1.4771,
      "step": 3030
    },
    {
      "epoch": 0.21546530583315615,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.4868,
      "step": 3040
    },
    {
      "epoch": 0.2161740732865547,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0001,
      "loss": 1.4941,
      "step": 3050
    },
    {
      "epoch": 0.21688284073995323,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0001,
      "loss": 1.4951,
      "step": 3060
    },
    {
      "epoch": 0.21759160819335177,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0001,
      "loss": 1.4915,
      "step": 3070
    },
    {
      "epoch": 0.2183003756467503,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.4869,
      "step": 3080
    },
    {
      "epoch": 0.21900914310014885,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.5057,
      "step": 3090
    },
    {
      "epoch": 0.2197179105535474,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4941,
      "step": 3100
    },
    {
      "epoch": 0.22042667800694593,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0001,
      "loss": 1.5017,
      "step": 3110
    },
    {
      "epoch": 0.22113544546034447,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0001,
      "loss": 1.511,
      "step": 3120
    },
    {
      "epoch": 0.221844212913743,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.4903,
      "step": 3130
    },
    {
      "epoch": 0.22255298036714155,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0001,
      "loss": 1.5098,
      "step": 3140
    },
    {
      "epoch": 0.2232617478205401,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4989,
      "step": 3150
    },
    {
      "epoch": 0.22397051527393863,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0001,
      "loss": 1.4935,
      "step": 3160
    },
    {
      "epoch": 0.22467928272733717,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4907,
      "step": 3170
    },
    {
      "epoch": 0.2253880501807357,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0001,
      "loss": 1.5033,
      "step": 3180
    },
    {
      "epoch": 0.22609681763413425,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4982,
      "step": 3190
    },
    {
      "epoch": 0.2268055850875328,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.4853,
      "step": 3200
    },
    {
      "epoch": 0.22751435254093133,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.498,
      "step": 3210
    },
    {
      "epoch": 0.22822311999432987,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0001,
      "loss": 1.489,
      "step": 3220
    },
    {
      "epoch": 0.2289318874477284,
      "grad_norm": 0.435546875,
      "learning_rate": 0.0001,
      "loss": 1.4758,
      "step": 3230
    },
    {
      "epoch": 0.22964065490112695,
      "grad_norm": 0.69921875,
      "learning_rate": 0.0001,
      "loss": 1.4896,
      "step": 3240
    },
    {
      "epoch": 0.2303494223545255,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.479,
      "step": 3250
    },
    {
      "epoch": 0.23105818980792403,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0001,
      "loss": 1.4968,
      "step": 3260
    },
    {
      "epoch": 0.23176695726132257,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.4913,
      "step": 3270
    },
    {
      "epoch": 0.2324757247147211,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0001,
      "loss": 1.4856,
      "step": 3280
    },
    {
      "epoch": 0.23318449216811965,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0001,
      "loss": 1.4969,
      "step": 3290
    },
    {
      "epoch": 0.2338932596215182,
      "grad_norm": 0.46484375,
      "learning_rate": 0.0001,
      "loss": 1.4968,
      "step": 3300
    },
    {
      "epoch": 0.23460202707491673,
      "grad_norm": 0.4453125,
      "learning_rate": 0.0001,
      "loss": 1.4904,
      "step": 3310
    },
    {
      "epoch": 0.23531079452831527,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0001,
      "loss": 1.484,
      "step": 3320
    },
    {
      "epoch": 0.2360195619817138,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4928,
      "step": 3330
    },
    {
      "epoch": 0.23672832943511235,
      "grad_norm": 0.54296875,
      "learning_rate": 0.0001,
      "loss": 1.4964,
      "step": 3340
    },
    {
      "epoch": 0.2374370968885109,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4974,
      "step": 3350
    },
    {
      "epoch": 0.23814586434190943,
      "grad_norm": 0.498046875,
      "learning_rate": 0.0001,
      "loss": 1.503,
      "step": 3360
    },
    {
      "epoch": 0.23885463179530797,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0001,
      "loss": 1.5029,
      "step": 3370
    },
    {
      "epoch": 0.2395633992487065,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001,
      "loss": 1.501,
      "step": 3380
    },
    {
      "epoch": 0.24027216670210505,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4885,
      "step": 3390
    },
    {
      "epoch": 0.2409809341555036,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.4904,
      "step": 3400
    },
    {
      "epoch": 0.24168970160890213,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0001,
      "loss": 1.4851,
      "step": 3410
    },
    {
      "epoch": 0.24239846906230067,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.4783,
      "step": 3420
    },
    {
      "epoch": 0.2431072365156992,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0001,
      "loss": 1.4948,
      "step": 3430
    },
    {
      "epoch": 0.24381600396909775,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.4754,
      "step": 3440
    },
    {
      "epoch": 0.2445247714224963,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4936,
      "step": 3450
    },
    {
      "epoch": 0.24523353887589483,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0001,
      "loss": 1.4814,
      "step": 3460
    },
    {
      "epoch": 0.24594230632929337,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0001,
      "loss": 1.505,
      "step": 3470
    },
    {
      "epoch": 0.2466510737826919,
      "grad_norm": 0.609375,
      "learning_rate": 0.0001,
      "loss": 1.4888,
      "step": 3480
    },
    {
      "epoch": 0.24735984123609045,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0001,
      "loss": 1.4974,
      "step": 3490
    },
    {
      "epoch": 0.248068608689489,
      "grad_norm": 0.4453125,
      "learning_rate": 0.0001,
      "loss": 1.496,
      "step": 3500
    },
    {
      "epoch": 0.248068608689489,
      "eval_accuracy": 0.665728571780445,
      "eval_loss": 1.403409481048584,
      "eval_runtime": 7.1308,
      "eval_samples_per_second": 48.662,
      "eval_steps_per_second": 0.421,
      "step": 3500
    },
    {
      "epoch": 0.24877737614288753,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0001,
      "loss": 1.4988,
      "step": 3510
    },
    {
      "epoch": 0.24948614359628607,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.5036,
      "step": 3520
    },
    {
      "epoch": 0.2501949110496846,
      "grad_norm": 0.609375,
      "learning_rate": 0.0001,
      "loss": 1.5038,
      "step": 3530
    },
    {
      "epoch": 0.25090367850308315,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0001,
      "loss": 1.4947,
      "step": 3540
    },
    {
      "epoch": 0.2516124459564817,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.4914,
      "step": 3550
    },
    {
      "epoch": 0.25232121340988023,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4855,
      "step": 3560
    },
    {
      "epoch": 0.25302998086327877,
      "grad_norm": 0.498046875,
      "learning_rate": 0.0001,
      "loss": 1.4845,
      "step": 3570
    },
    {
      "epoch": 0.2537387483166773,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.5041,
      "step": 3580
    },
    {
      "epoch": 0.25444751577007585,
      "grad_norm": 0.5703125,
      "learning_rate": 0.0001,
      "loss": 1.4803,
      "step": 3590
    },
    {
      "epoch": 0.2551562832234744,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4938,
      "step": 3600
    },
    {
      "epoch": 0.25586505067687293,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001,
      "loss": 1.4938,
      "step": 3610
    },
    {
      "epoch": 0.25657381813027147,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0001,
      "loss": 1.4972,
      "step": 3620
    },
    {
      "epoch": 0.25728258558367,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0001,
      "loss": 1.4941,
      "step": 3630
    },
    {
      "epoch": 0.25799135303706855,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0001,
      "loss": 1.4881,
      "step": 3640
    },
    {
      "epoch": 0.2587001204904671,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4906,
      "step": 3650
    },
    {
      "epoch": 0.25940888794386563,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0001,
      "loss": 1.4937,
      "step": 3660
    },
    {
      "epoch": 0.26011765539726417,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0001,
      "loss": 1.4904,
      "step": 3670
    },
    {
      "epoch": 0.2608264228506627,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.5034,
      "step": 3680
    },
    {
      "epoch": 0.26153519030406125,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0001,
      "loss": 1.4727,
      "step": 3690
    },
    {
      "epoch": 0.2622439577574598,
      "grad_norm": 0.421875,
      "learning_rate": 0.0001,
      "loss": 1.4919,
      "step": 3700
    },
    {
      "epoch": 0.26295272521085833,
      "grad_norm": 0.43359375,
      "learning_rate": 0.0001,
      "loss": 1.4842,
      "step": 3710
    },
    {
      "epoch": 0.26366149266425687,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0001,
      "loss": 1.4953,
      "step": 3720
    },
    {
      "epoch": 0.2643702601176554,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.4739,
      "step": 3730
    },
    {
      "epoch": 0.26507902757105395,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0001,
      "loss": 1.4913,
      "step": 3740
    },
    {
      "epoch": 0.2657877950244525,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001,
      "loss": 1.5003,
      "step": 3750
    },
    {
      "epoch": 0.26649656247785103,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0001,
      "loss": 1.4919,
      "step": 3760
    },
    {
      "epoch": 0.26720532993124957,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4818,
      "step": 3770
    },
    {
      "epoch": 0.2679140973846481,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.5003,
      "step": 3780
    },
    {
      "epoch": 0.26862286483804665,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4934,
      "step": 3790
    },
    {
      "epoch": 0.2693316322914452,
      "grad_norm": 0.59375,
      "learning_rate": 0.0001,
      "loss": 1.4869,
      "step": 3800
    },
    {
      "epoch": 0.27004039974484373,
      "grad_norm": 0.5625,
      "learning_rate": 0.0001,
      "loss": 1.4896,
      "step": 3810
    },
    {
      "epoch": 0.27074916719824227,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001,
      "loss": 1.5021,
      "step": 3820
    },
    {
      "epoch": 0.2714579346516408,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.5048,
      "step": 3830
    },
    {
      "epoch": 0.27216670210503935,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0001,
      "loss": 1.4963,
      "step": 3840
    },
    {
      "epoch": 0.2728754695584379,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4967,
      "step": 3850
    },
    {
      "epoch": 0.27358423701183643,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0001,
      "loss": 1.5036,
      "step": 3860
    },
    {
      "epoch": 0.27429300446523497,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.4944,
      "step": 3870
    },
    {
      "epoch": 0.2750017719186335,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0001,
      "loss": 1.4951,
      "step": 3880
    },
    {
      "epoch": 0.27571053937203205,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0001,
      "loss": 1.5006,
      "step": 3890
    },
    {
      "epoch": 0.2764193068254306,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0001,
      "loss": 1.4886,
      "step": 3900
    },
    {
      "epoch": 0.27712807427882913,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.4931,
      "step": 3910
    },
    {
      "epoch": 0.27783684173222767,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0001,
      "loss": 1.4926,
      "step": 3920
    },
    {
      "epoch": 0.2785456091856262,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.4962,
      "step": 3930
    },
    {
      "epoch": 0.27925437663902475,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0001,
      "loss": 1.4933,
      "step": 3940
    },
    {
      "epoch": 0.2799631440924233,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0001,
      "loss": 1.4959,
      "step": 3950
    },
    {
      "epoch": 0.28067191154582183,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0001,
      "loss": 1.49,
      "step": 3960
    },
    {
      "epoch": 0.28138067899922037,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0001,
      "loss": 1.5087,
      "step": 3970
    },
    {
      "epoch": 0.2820894464526189,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.4939,
      "step": 3980
    },
    {
      "epoch": 0.28279821390601745,
      "grad_norm": 0.4375,
      "learning_rate": 0.0001,
      "loss": 1.4913,
      "step": 3990
    },
    {
      "epoch": 0.283506981359416,
      "grad_norm": 0.443359375,
      "learning_rate": 0.0001,
      "loss": 1.495,
      "step": 4000
    },
    {
      "epoch": 0.283506981359416,
      "eval_accuracy": 0.6657313867688508,
      "eval_loss": 1.403212308883667,
      "eval_runtime": 7.176,
      "eval_samples_per_second": 48.355,
      "eval_steps_per_second": 0.418,
      "step": 4000
    },
    {
      "epoch": 0.28421574881281453,
      "grad_norm": 0.46484375,
      "learning_rate": 0.0001,
      "loss": 1.4793,
      "step": 4010
    },
    {
      "epoch": 0.28492451626621307,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0001,
      "loss": 1.484,
      "step": 4020
    },
    {
      "epoch": 0.2856332837196116,
      "grad_norm": 0.578125,
      "learning_rate": 0.0001,
      "loss": 1.4967,
      "step": 4030
    },
    {
      "epoch": 0.28634205117301015,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0001,
      "loss": 1.4952,
      "step": 4040
    },
    {
      "epoch": 0.2870508186264087,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0001,
      "loss": 1.494,
      "step": 4050
    },
    {
      "epoch": 0.28775958607980723,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0001,
      "loss": 1.4888,
      "step": 4060
    },
    {
      "epoch": 0.28846835353320577,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0001,
      "loss": 1.4784,
      "step": 4070
    },
    {
      "epoch": 0.2891771209866043,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0001,
      "loss": 1.4924,
      "step": 4080
    },
    {
      "epoch": 0.28988588844000285,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.5047,
      "step": 4090
    },
    {
      "epoch": 0.2905946558934014,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.4895,
      "step": 4100
    },
    {
      "epoch": 0.29130342334679993,
      "grad_norm": 0.466796875,
      "learning_rate": 0.0001,
      "loss": 1.4852,
      "step": 4110
    },
    {
      "epoch": 0.29201219080019847,
      "grad_norm": 0.5625,
      "learning_rate": 0.0001,
      "loss": 1.5075,
      "step": 4120
    },
    {
      "epoch": 0.292720958253597,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.5061,
      "step": 4130
    },
    {
      "epoch": 0.29342972570699555,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4953,
      "step": 4140
    },
    {
      "epoch": 0.2941384931603941,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4956,
      "step": 4150
    },
    {
      "epoch": 0.2948472606137926,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4978,
      "step": 4160
    },
    {
      "epoch": 0.29555602806719117,
      "grad_norm": 0.7421875,
      "learning_rate": 0.0001,
      "loss": 1.4759,
      "step": 4170
    },
    {
      "epoch": 0.2962647955205897,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4955,
      "step": 4180
    },
    {
      "epoch": 0.29697356297398825,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0001,
      "loss": 1.4955,
      "step": 4190
    },
    {
      "epoch": 0.2976823304273868,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0001,
      "loss": 1.482,
      "step": 4200
    },
    {
      "epoch": 0.2983910978807853,
      "grad_norm": 0.455078125,
      "learning_rate": 0.0001,
      "loss": 1.4774,
      "step": 4210
    },
    {
      "epoch": 0.29909986533418387,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.484,
      "step": 4220
    },
    {
      "epoch": 0.2998086327875824,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0001,
      "loss": 1.5073,
      "step": 4230
    },
    {
      "epoch": 0.30051740024098095,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.5047,
      "step": 4240
    },
    {
      "epoch": 0.3012261676943795,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.4898,
      "step": 4250
    },
    {
      "epoch": 0.301934935147778,
      "grad_norm": 0.4453125,
      "learning_rate": 0.0001,
      "loss": 1.4914,
      "step": 4260
    },
    {
      "epoch": 0.30264370260117657,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4838,
      "step": 4270
    },
    {
      "epoch": 0.3033524700545751,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.4908,
      "step": 4280
    },
    {
      "epoch": 0.30406123750797365,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0001,
      "loss": 1.4838,
      "step": 4290
    },
    {
      "epoch": 0.3047700049613722,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4871,
      "step": 4300
    },
    {
      "epoch": 0.3054787724147707,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0001,
      "loss": 1.5091,
      "step": 4310
    },
    {
      "epoch": 0.30618753986816927,
      "grad_norm": 0.4296875,
      "learning_rate": 0.0001,
      "loss": 1.4968,
      "step": 4320
    },
    {
      "epoch": 0.3068963073215678,
      "grad_norm": 0.5703125,
      "learning_rate": 0.0001,
      "loss": 1.4926,
      "step": 4330
    },
    {
      "epoch": 0.30760507477496635,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0001,
      "loss": 1.4982,
      "step": 4340
    },
    {
      "epoch": 0.3083138422283649,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0001,
      "loss": 1.5001,
      "step": 4350
    },
    {
      "epoch": 0.3090226096817634,
      "grad_norm": 0.43359375,
      "learning_rate": 0.0001,
      "loss": 1.4939,
      "step": 4360
    },
    {
      "epoch": 0.30973137713516197,
      "grad_norm": 0.49609375,
      "learning_rate": 0.0001,
      "loss": 1.4979,
      "step": 4370
    },
    {
      "epoch": 0.3104401445885605,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.4799,
      "step": 4380
    },
    {
      "epoch": 0.31114891204195905,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4945,
      "step": 4390
    },
    {
      "epoch": 0.3118576794953576,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0001,
      "loss": 1.4948,
      "step": 4400
    },
    {
      "epoch": 0.3125664469487561,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4942,
      "step": 4410
    },
    {
      "epoch": 0.31327521440215467,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.5039,
      "step": 4420
    },
    {
      "epoch": 0.3139839818555532,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.4882,
      "step": 4430
    },
    {
      "epoch": 0.31469274930895175,
      "grad_norm": 0.4296875,
      "learning_rate": 0.0001,
      "loss": 1.4925,
      "step": 4440
    },
    {
      "epoch": 0.3154015167623503,
      "grad_norm": 0.65625,
      "learning_rate": 0.0001,
      "loss": 1.5131,
      "step": 4450
    },
    {
      "epoch": 0.3161102842157488,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0001,
      "loss": 1.5026,
      "step": 4460
    },
    {
      "epoch": 0.31681905166914737,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4959,
      "step": 4470
    },
    {
      "epoch": 0.3175278191225459,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0001,
      "loss": 1.4972,
      "step": 4480
    },
    {
      "epoch": 0.31823658657594445,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4922,
      "step": 4490
    },
    {
      "epoch": 0.318945354029343,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0001,
      "loss": 1.4934,
      "step": 4500
    },
    {
      "epoch": 0.318945354029343,
      "eval_accuracy": 0.6658446900521828,
      "eval_loss": 1.4029836654663086,
      "eval_runtime": 7.1758,
      "eval_samples_per_second": 48.357,
      "eval_steps_per_second": 0.418,
      "step": 4500
    },
    {
      "epoch": 0.3196541214827415,
      "grad_norm": 0.57421875,
      "learning_rate": 0.0001,
      "loss": 1.4693,
      "step": 4510
    },
    {
      "epoch": 0.32036288893614007,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0001,
      "loss": 1.498,
      "step": 4520
    },
    {
      "epoch": 0.3210716563895386,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0001,
      "loss": 1.4872,
      "step": 4530
    },
    {
      "epoch": 0.32178042384293715,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0001,
      "loss": 1.4866,
      "step": 4540
    },
    {
      "epoch": 0.3224891912963357,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.4919,
      "step": 4550
    },
    {
      "epoch": 0.3231979587497342,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001,
      "loss": 1.4765,
      "step": 4560
    },
    {
      "epoch": 0.32390672620313277,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0001,
      "loss": 1.4919,
      "step": 4570
    },
    {
      "epoch": 0.3246154936565313,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0001,
      "loss": 1.5069,
      "step": 4580
    },
    {
      "epoch": 0.32532426110992985,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.5014,
      "step": 4590
    },
    {
      "epoch": 0.3260330285633284,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.5081,
      "step": 4600
    },
    {
      "epoch": 0.3267417960167269,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0001,
      "loss": 1.5007,
      "step": 4610
    },
    {
      "epoch": 0.32745056347012547,
      "grad_norm": 0.498046875,
      "learning_rate": 0.0001,
      "loss": 1.4755,
      "step": 4620
    },
    {
      "epoch": 0.328159330923524,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0001,
      "loss": 1.4875,
      "step": 4630
    },
    {
      "epoch": 0.32886809837692255,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.4871,
      "step": 4640
    },
    {
      "epoch": 0.3295768658303211,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.5034,
      "step": 4650
    },
    {
      "epoch": 0.3302856332837196,
      "grad_norm": 0.470703125,
      "learning_rate": 0.0001,
      "loss": 1.4828,
      "step": 4660
    },
    {
      "epoch": 0.33099440073711817,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0001,
      "loss": 1.489,
      "step": 4670
    },
    {
      "epoch": 0.3317031681905167,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0001,
      "loss": 1.5142,
      "step": 4680
    },
    {
      "epoch": 0.33241193564391525,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.4918,
      "step": 4690
    },
    {
      "epoch": 0.3331207030973138,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4862,
      "step": 4700
    },
    {
      "epoch": 0.3338294705507123,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0001,
      "loss": 1.4979,
      "step": 4710
    },
    {
      "epoch": 0.33453823800411087,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4968,
      "step": 4720
    },
    {
      "epoch": 0.3352470054575094,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0001,
      "loss": 1.4929,
      "step": 4730
    },
    {
      "epoch": 0.33595577291090795,
      "grad_norm": 0.54296875,
      "learning_rate": 0.0001,
      "loss": 1.4713,
      "step": 4740
    },
    {
      "epoch": 0.3366645403643065,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.4866,
      "step": 4750
    },
    {
      "epoch": 0.337373307817705,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4993,
      "step": 4760
    },
    {
      "epoch": 0.33808207527110357,
      "grad_norm": 0.58203125,
      "learning_rate": 0.0001,
      "loss": 1.4966,
      "step": 4770
    },
    {
      "epoch": 0.3387908427245021,
      "grad_norm": 0.46484375,
      "learning_rate": 0.0001,
      "loss": 1.5022,
      "step": 4780
    },
    {
      "epoch": 0.33949961017790065,
      "grad_norm": 0.453125,
      "learning_rate": 0.0001,
      "loss": 1.4908,
      "step": 4790
    },
    {
      "epoch": 0.3402083776312992,
      "grad_norm": 0.66015625,
      "learning_rate": 0.0001,
      "loss": 1.4884,
      "step": 4800
    },
    {
      "epoch": 0.3409171450846977,
      "grad_norm": 0.44140625,
      "learning_rate": 0.0001,
      "loss": 1.5059,
      "step": 4810
    },
    {
      "epoch": 0.34162591253809627,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0001,
      "loss": 1.4907,
      "step": 4820
    },
    {
      "epoch": 0.3423346799914948,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.4842,
      "step": 4830
    },
    {
      "epoch": 0.34304344744489335,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4984,
      "step": 4840
    },
    {
      "epoch": 0.3437522148982919,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4828,
      "step": 4850
    },
    {
      "epoch": 0.3444609823516904,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0001,
      "loss": 1.4865,
      "step": 4860
    },
    {
      "epoch": 0.34516974980508897,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.5001,
      "step": 4870
    },
    {
      "epoch": 0.3458785172584875,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.5026,
      "step": 4880
    },
    {
      "epoch": 0.34658728471188605,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.488,
      "step": 4890
    },
    {
      "epoch": 0.3472960521652846,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.4846,
      "step": 4900
    },
    {
      "epoch": 0.3480048196186831,
      "grad_norm": 0.5625,
      "learning_rate": 0.0001,
      "loss": 1.4972,
      "step": 4910
    },
    {
      "epoch": 0.34871358707208167,
      "grad_norm": 0.46484375,
      "learning_rate": 0.0001,
      "loss": 1.487,
      "step": 4920
    },
    {
      "epoch": 0.3494223545254802,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0001,
      "loss": 1.4945,
      "step": 4930
    },
    {
      "epoch": 0.35013112197887875,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.4942,
      "step": 4940
    },
    {
      "epoch": 0.3508398894322773,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4944,
      "step": 4950
    },
    {
      "epoch": 0.3515486568856758,
      "grad_norm": 0.59375,
      "learning_rate": 0.0001,
      "loss": 1.5099,
      "step": 4960
    },
    {
      "epoch": 0.35225742433907437,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0001,
      "loss": 1.4951,
      "step": 4970
    },
    {
      "epoch": 0.3529661917924729,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001,
      "loss": 1.5049,
      "step": 4980
    },
    {
      "epoch": 0.35367495924587145,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0001,
      "loss": 1.4764,
      "step": 4990
    },
    {
      "epoch": 0.35438372669927,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4849,
      "step": 5000
    },
    {
      "epoch": 0.35438372669927,
      "eval_accuracy": 0.6660431467347894,
      "eval_loss": 1.4028522968292236,
      "eval_runtime": 7.1426,
      "eval_samples_per_second": 48.581,
      "eval_steps_per_second": 0.42,
      "step": 5000
    },
    {
      "epoch": 0.3550924941526685,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.4884,
      "step": 5010
    },
    {
      "epoch": 0.35580126160606707,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.5002,
      "step": 5020
    },
    {
      "epoch": 0.3565100290594656,
      "grad_norm": 0.498046875,
      "learning_rate": 0.0001,
      "loss": 1.5019,
      "step": 5030
    },
    {
      "epoch": 0.35721879651286415,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0001,
      "loss": 1.4813,
      "step": 5040
    },
    {
      "epoch": 0.3579275639662627,
      "grad_norm": 0.439453125,
      "learning_rate": 0.0001,
      "loss": 1.4826,
      "step": 5050
    },
    {
      "epoch": 0.3586363314196612,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.5028,
      "step": 5060
    },
    {
      "epoch": 0.35934509887305977,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0001,
      "loss": 1.4953,
      "step": 5070
    },
    {
      "epoch": 0.3600538663264583,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001,
      "loss": 1.4826,
      "step": 5080
    },
    {
      "epoch": 0.36076263377985684,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0001,
      "loss": 1.4877,
      "step": 5090
    },
    {
      "epoch": 0.3614714012332554,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.4823,
      "step": 5100
    },
    {
      "epoch": 0.3621801686866539,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0001,
      "loss": 1.4957,
      "step": 5110
    },
    {
      "epoch": 0.36288893614005246,
      "grad_norm": 0.546875,
      "learning_rate": 0.0001,
      "loss": 1.4935,
      "step": 5120
    },
    {
      "epoch": 0.363597703593451,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.5076,
      "step": 5130
    },
    {
      "epoch": 0.36430647104684954,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.4848,
      "step": 5140
    },
    {
      "epoch": 0.3650152385002481,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0001,
      "loss": 1.4892,
      "step": 5150
    },
    {
      "epoch": 0.3657240059536466,
      "grad_norm": 0.490234375,
      "learning_rate": 0.0001,
      "loss": 1.4944,
      "step": 5160
    },
    {
      "epoch": 0.36643277340704516,
      "grad_norm": 0.609375,
      "learning_rate": 0.0001,
      "loss": 1.4946,
      "step": 5170
    },
    {
      "epoch": 0.3671415408604437,
      "grad_norm": 0.474609375,
      "learning_rate": 0.0001,
      "loss": 1.4826,
      "step": 5180
    },
    {
      "epoch": 0.36785030831384224,
      "grad_norm": 0.470703125,
      "learning_rate": 0.0001,
      "loss": 1.4874,
      "step": 5190
    },
    {
      "epoch": 0.3685590757672408,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.5055,
      "step": 5200
    },
    {
      "epoch": 0.3692678432206393,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0001,
      "loss": 1.5005,
      "step": 5210
    },
    {
      "epoch": 0.36997661067403786,
      "grad_norm": 0.46875,
      "learning_rate": 0.0001,
      "loss": 1.4842,
      "step": 5220
    },
    {
      "epoch": 0.3706853781274364,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0001,
      "loss": 1.501,
      "step": 5230
    },
    {
      "epoch": 0.37139414558083494,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0001,
      "loss": 1.5031,
      "step": 5240
    },
    {
      "epoch": 0.3721029130342335,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0001,
      "loss": 1.4887,
      "step": 5250
    },
    {
      "epoch": 0.372811680487632,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.4779,
      "step": 5260
    },
    {
      "epoch": 0.37352044794103056,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0001,
      "loss": 1.4779,
      "step": 5270
    },
    {
      "epoch": 0.3742292153944291,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.5006,
      "step": 5280
    },
    {
      "epoch": 0.37493798284782764,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.5109,
      "step": 5290
    },
    {
      "epoch": 0.3756467503012262,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0001,
      "loss": 1.4779,
      "step": 5300
    },
    {
      "epoch": 0.3763555177546247,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4861,
      "step": 5310
    },
    {
      "epoch": 0.37706428520802326,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.5001,
      "step": 5320
    },
    {
      "epoch": 0.3777730526614218,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4876,
      "step": 5330
    },
    {
      "epoch": 0.37848182011482034,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.486,
      "step": 5340
    },
    {
      "epoch": 0.3791905875682189,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0001,
      "loss": 1.4956,
      "step": 5350
    },
    {
      "epoch": 0.3798993550216174,
      "grad_norm": 0.61328125,
      "learning_rate": 0.0001,
      "loss": 1.4972,
      "step": 5360
    },
    {
      "epoch": 0.38060812247501596,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.4926,
      "step": 5370
    },
    {
      "epoch": 0.3813168899284145,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4982,
      "step": 5380
    },
    {
      "epoch": 0.38202565738181304,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4949,
      "step": 5390
    },
    {
      "epoch": 0.3827344248352116,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0001,
      "loss": 1.4928,
      "step": 5400
    },
    {
      "epoch": 0.3834431922886101,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.4841,
      "step": 5410
    },
    {
      "epoch": 0.38415195974200866,
      "grad_norm": 0.58984375,
      "learning_rate": 0.0001,
      "loss": 1.4975,
      "step": 5420
    },
    {
      "epoch": 0.3848607271954072,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0001,
      "loss": 1.4907,
      "step": 5430
    },
    {
      "epoch": 0.38556949464880574,
      "grad_norm": 0.453125,
      "learning_rate": 0.0001,
      "loss": 1.4933,
      "step": 5440
    },
    {
      "epoch": 0.3862782621022043,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.5083,
      "step": 5450
    },
    {
      "epoch": 0.3869870295556028,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.4766,
      "step": 5460
    },
    {
      "epoch": 0.38769579700900136,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.485,
      "step": 5470
    },
    {
      "epoch": 0.3884045644623999,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0001,
      "loss": 1.4891,
      "step": 5480
    },
    {
      "epoch": 0.38911333191579844,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.495,
      "step": 5490
    },
    {
      "epoch": 0.389822099369197,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.4833,
      "step": 5500
    },
    {
      "epoch": 0.389822099369197,
      "eval_accuracy": 0.6661451900644985,
      "eval_loss": 1.4023902416229248,
      "eval_runtime": 7.2106,
      "eval_samples_per_second": 48.124,
      "eval_steps_per_second": 0.416,
      "step": 5500
    },
    {
      "epoch": 0.3905308668225955,
      "grad_norm": 0.466796875,
      "learning_rate": 0.0001,
      "loss": 1.5,
      "step": 5510
    },
    {
      "epoch": 0.39123963427599406,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0001,
      "loss": 1.4789,
      "step": 5520
    },
    {
      "epoch": 0.3919484017293926,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.498,
      "step": 5530
    },
    {
      "epoch": 0.39265716918279114,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.5051,
      "step": 5540
    },
    {
      "epoch": 0.3933659366361897,
      "grad_norm": 0.66015625,
      "learning_rate": 0.0001,
      "loss": 1.5041,
      "step": 5550
    },
    {
      "epoch": 0.3940747040895882,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0001,
      "loss": 1.4736,
      "step": 5560
    },
    {
      "epoch": 0.39478347154298676,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.4863,
      "step": 5570
    },
    {
      "epoch": 0.3954922389963853,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.4962,
      "step": 5580
    },
    {
      "epoch": 0.39620100644978384,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0001,
      "loss": 1.4904,
      "step": 5590
    },
    {
      "epoch": 0.3969097739031824,
      "grad_norm": 0.54296875,
      "learning_rate": 0.0001,
      "loss": 1.4955,
      "step": 5600
    },
    {
      "epoch": 0.3976185413565809,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4925,
      "step": 5610
    },
    {
      "epoch": 0.39832730880997946,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0001,
      "loss": 1.4945,
      "step": 5620
    },
    {
      "epoch": 0.399036076263378,
      "grad_norm": 0.5625,
      "learning_rate": 0.0001,
      "loss": 1.5,
      "step": 5630
    },
    {
      "epoch": 0.39974484371677654,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0001,
      "loss": 1.4946,
      "step": 5640
    },
    {
      "epoch": 0.4004536111701751,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.4868,
      "step": 5650
    },
    {
      "epoch": 0.4011623786235736,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0001,
      "loss": 1.4841,
      "step": 5660
    },
    {
      "epoch": 0.40187114607697216,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0001,
      "loss": 1.494,
      "step": 5670
    },
    {
      "epoch": 0.4025799135303707,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001,
      "loss": 1.4844,
      "step": 5680
    },
    {
      "epoch": 0.40328868098376924,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0001,
      "loss": 1.4932,
      "step": 5690
    },
    {
      "epoch": 0.4039974484371678,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.5066,
      "step": 5700
    },
    {
      "epoch": 0.4047062158905663,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001,
      "loss": 1.498,
      "step": 5710
    },
    {
      "epoch": 0.40541498334396486,
      "grad_norm": 0.482421875,
      "learning_rate": 0.0001,
      "loss": 1.5048,
      "step": 5720
    },
    {
      "epoch": 0.4061237507973634,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4863,
      "step": 5730
    },
    {
      "epoch": 0.40683251825076194,
      "grad_norm": 0.66796875,
      "learning_rate": 0.0001,
      "loss": 1.4899,
      "step": 5740
    },
    {
      "epoch": 0.4075412857041605,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0001,
      "loss": 1.4792,
      "step": 5750
    },
    {
      "epoch": 0.408250053157559,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.4915,
      "step": 5760
    },
    {
      "epoch": 0.40895882061095756,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001,
      "loss": 1.4923,
      "step": 5770
    },
    {
      "epoch": 0.4096675880643561,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0001,
      "loss": 1.4757,
      "step": 5780
    },
    {
      "epoch": 0.41037635551775464,
      "grad_norm": 0.484375,
      "learning_rate": 0.0001,
      "loss": 1.4972,
      "step": 5790
    },
    {
      "epoch": 0.4110851229711532,
      "grad_norm": 0.5625,
      "learning_rate": 0.0001,
      "loss": 1.4846,
      "step": 5800
    },
    {
      "epoch": 0.4117938904245517,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0001,
      "loss": 1.4888,
      "step": 5810
    },
    {
      "epoch": 0.41250265787795026,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0001,
      "loss": 1.4996,
      "step": 5820
    },
    {
      "epoch": 0.4132114253313488,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0001,
      "loss": 1.4927,
      "step": 5830
    },
    {
      "epoch": 0.41392019278474734,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0001,
      "loss": 1.4861,
      "step": 5840
    },
    {
      "epoch": 0.4146289602381459,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001,
      "loss": 1.4917,
      "step": 5850
    },
    {
      "epoch": 0.4153377276915444,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0001,
      "loss": 1.5034,
      "step": 5860
    },
    {
      "epoch": 0.41604649514494296,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.5052,
      "step": 5870
    },
    {
      "epoch": 0.4167552625983415,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0001,
      "loss": 1.5002,
      "step": 5880
    },
    {
      "epoch": 0.41746403005174004,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0001,
      "loss": 1.5054,
      "step": 5890
    },
    {
      "epoch": 0.4181727975051386,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0001,
      "loss": 1.498,
      "step": 5900
    },
    {
      "epoch": 0.4188815649585371,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0001,
      "loss": 1.4894,
      "step": 5910
    },
    {
      "epoch": 0.41959033241193566,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0001,
      "loss": 1.5121,
      "step": 5920
    },
    {
      "epoch": 0.4202990998653342,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0001,
      "loss": 1.4975,
      "step": 5930
    },
    {
      "epoch": 0.42100786731873274,
      "grad_norm": 0.6484375,
      "learning_rate": 0.0001,
      "loss": 1.4825,
      "step": 5940
    },
    {
      "epoch": 0.4217166347721313,
      "grad_norm": 0.609375,
      "learning_rate": 0.0001,
      "loss": 1.4985,
      "step": 5950
    },
    {
      "epoch": 0.4224254022255298,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001,
      "loss": 1.4922,
      "step": 5960
    },
    {
      "epoch": 0.42313416967892836,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001,
      "loss": 1.4964,
      "step": 5970
    },
    {
      "epoch": 0.4238429371323269,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0001,
      "loss": 1.4982,
      "step": 5980
    },
    {
      "epoch": 0.42455170458572544,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4778,
      "step": 5990
    },
    {
      "epoch": 0.425260472039124,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0001,
      "loss": 1.4909,
      "step": 6000
    },
    {
      "epoch": 0.425260472039124,
      "eval_accuracy": 0.6661219664101509,
      "eval_loss": 1.4022555351257324,
      "eval_runtime": 7.2496,
      "eval_samples_per_second": 47.865,
      "eval_steps_per_second": 0.414,
      "step": 6000
    },
    {
      "epoch": 0.4259692394925225,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0001,
      "loss": 1.5043,
      "step": 6010
    },
    {
      "epoch": 0.42667800694592106,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.507,
      "step": 6020
    },
    {
      "epoch": 0.4273867743993196,
      "grad_norm": 0.4453125,
      "learning_rate": 0.0001,
      "loss": 1.4907,
      "step": 6030
    },
    {
      "epoch": 0.42809554185271814,
      "grad_norm": 0.5,
      "learning_rate": 0.0001,
      "loss": 1.491,
      "step": 6040
    },
    {
      "epoch": 0.4288043093061167,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0001,
      "loss": 1.5023,
      "step": 6050
    },
    {
      "epoch": 0.4295130767595152,
      "grad_norm": 0.486328125,
      "learning_rate": 9.997960964140947e-05,
      "loss": 1.4967,
      "step": 6060
    },
    {
      "epoch": 0.43022184421291376,
      "grad_norm": 0.4921875,
      "learning_rate": 9.991845519630678e-05,
      "loss": 1.4897,
      "step": 6070
    },
    {
      "epoch": 0.4309306116663123,
      "grad_norm": 0.5703125,
      "learning_rate": 9.981658654313457e-05,
      "loss": 1.4917,
      "step": 6080
    },
    {
      "epoch": 0.43163937911971084,
      "grad_norm": 0.5,
      "learning_rate": 9.967408676742751e-05,
      "loss": 1.5015,
      "step": 6090
    },
    {
      "epoch": 0.4323481465731094,
      "grad_norm": 0.486328125,
      "learning_rate": 9.949107209404665e-05,
      "loss": 1.4798,
      "step": 6100
    },
    {
      "epoch": 0.4330569140265079,
      "grad_norm": 0.4453125,
      "learning_rate": 9.926769179238466e-05,
      "loss": 1.4874,
      "step": 6110
    },
    {
      "epoch": 0.43376568147990646,
      "grad_norm": 0.62109375,
      "learning_rate": 9.900412805461967e-05,
      "loss": 1.4951,
      "step": 6120
    },
    {
      "epoch": 0.434474448933305,
      "grad_norm": 0.47265625,
      "learning_rate": 9.870059584711668e-05,
      "loss": 1.4917,
      "step": 6130
    },
    {
      "epoch": 0.43518321638670354,
      "grad_norm": 0.51953125,
      "learning_rate": 9.835734273509786e-05,
      "loss": 1.5102,
      "step": 6140
    },
    {
      "epoch": 0.4358919838401021,
      "grad_norm": 0.482421875,
      "learning_rate": 9.797464868072488e-05,
      "loss": 1.485,
      "step": 6150
    },
    {
      "epoch": 0.4366007512935006,
      "grad_norm": 0.5234375,
      "learning_rate": 9.755282581475769e-05,
      "loss": 1.4967,
      "step": 6160
    },
    {
      "epoch": 0.43730951874689916,
      "grad_norm": 0.4609375,
      "learning_rate": 9.709221818197624e-05,
      "loss": 1.4827,
      "step": 6170
    },
    {
      "epoch": 0.4380182862002977,
      "grad_norm": 0.56640625,
      "learning_rate": 9.659320146057262e-05,
      "loss": 1.4919,
      "step": 6180
    },
    {
      "epoch": 0.43872705365369624,
      "grad_norm": 0.4921875,
      "learning_rate": 9.60561826557425e-05,
      "loss": 1.5128,
      "step": 6190
    },
    {
      "epoch": 0.4394358211070948,
      "grad_norm": 0.44921875,
      "learning_rate": 9.548159976772592e-05,
      "loss": 1.496,
      "step": 6200
    },
    {
      "epoch": 0.4401445885604933,
      "grad_norm": 0.51953125,
      "learning_rate": 9.486992143456792e-05,
      "loss": 1.4835,
      "step": 6210
    },
    {
      "epoch": 0.44085335601389186,
      "grad_norm": 0.45703125,
      "learning_rate": 9.422164654989072e-05,
      "loss": 1.4907,
      "step": 6220
    },
    {
      "epoch": 0.4415621234672904,
      "grad_norm": 0.5703125,
      "learning_rate": 9.353730385598887e-05,
      "loss": 1.4982,
      "step": 6230
    },
    {
      "epoch": 0.44227089092068894,
      "grad_norm": 0.447265625,
      "learning_rate": 9.281745151257946e-05,
      "loss": 1.493,
      "step": 6240
    },
    {
      "epoch": 0.4429796583740875,
      "grad_norm": 0.48046875,
      "learning_rate": 9.206267664155907e-05,
      "loss": 1.4849,
      "step": 6250
    },
    {
      "epoch": 0.443688425827486,
      "grad_norm": 0.484375,
      "learning_rate": 9.12735948481387e-05,
      "loss": 1.495,
      "step": 6260
    },
    {
      "epoch": 0.44439719328088456,
      "grad_norm": 0.47265625,
      "learning_rate": 9.045084971874738e-05,
      "loss": 1.4908,
      "step": 6270
    },
    {
      "epoch": 0.4451059607342831,
      "grad_norm": 0.462890625,
      "learning_rate": 8.959511229611376e-05,
      "loss": 1.4864,
      "step": 6280
    },
    {
      "epoch": 0.44581472818768164,
      "grad_norm": 0.48046875,
      "learning_rate": 8.870708053195413e-05,
      "loss": 1.4839,
      "step": 6290
    },
    {
      "epoch": 0.4465234956410802,
      "grad_norm": 0.4453125,
      "learning_rate": 8.778747871771292e-05,
      "loss": 1.4995,
      "step": 6300
    },
    {
      "epoch": 0.4472322630944787,
      "grad_norm": 0.45703125,
      "learning_rate": 8.683705689382024e-05,
      "loss": 1.4856,
      "step": 6310
    },
    {
      "epoch": 0.44794103054787726,
      "grad_norm": 0.51953125,
      "learning_rate": 8.585659023794818e-05,
      "loss": 1.4933,
      "step": 6320
    },
    {
      "epoch": 0.4486497980012758,
      "grad_norm": 0.490234375,
      "learning_rate": 8.484687843276469e-05,
      "loss": 1.4798,
      "step": 6330
    },
    {
      "epoch": 0.44935856545467434,
      "grad_norm": 0.439453125,
      "learning_rate": 8.380874501370097e-05,
      "loss": 1.4972,
      "step": 6340
    },
    {
      "epoch": 0.4500673329080729,
      "grad_norm": 0.45703125,
      "learning_rate": 8.274303669726426e-05,
      "loss": 1.482,
      "step": 6350
    },
    {
      "epoch": 0.4507761003614714,
      "grad_norm": 0.44140625,
      "learning_rate": 8.165062269044353e-05,
      "loss": 1.4946,
      "step": 6360
    },
    {
      "epoch": 0.45148486781486996,
      "grad_norm": 0.443359375,
      "learning_rate": 8.053239398177191e-05,
      "loss": 1.487,
      "step": 6370
    },
    {
      "epoch": 0.4521936352682685,
      "grad_norm": 0.51953125,
      "learning_rate": 7.938926261462366e-05,
      "loss": 1.5043,
      "step": 6380
    },
    {
      "epoch": 0.45290240272166704,
      "grad_norm": 0.4921875,
      "learning_rate": 7.822216094333847e-05,
      "loss": 1.4912,
      "step": 6390
    },
    {
      "epoch": 0.4536111701750656,
      "grad_norm": 0.455078125,
      "learning_rate": 7.703204087277988e-05,
      "loss": 1.4895,
      "step": 6400
    },
    {
      "epoch": 0.4543199376284641,
      "grad_norm": 0.470703125,
      "learning_rate": 7.58198730819481e-05,
      "loss": 1.5003,
      "step": 6410
    },
    {
      "epoch": 0.45502870508186266,
      "grad_norm": 0.5078125,
      "learning_rate": 7.45866462322802e-05,
      "loss": 1.4872,
      "step": 6420
    },
    {
      "epoch": 0.4557374725352612,
      "grad_norm": 0.435546875,
      "learning_rate": 7.333336616128369e-05,
      "loss": 1.4993,
      "step": 6430
    },
    {
      "epoch": 0.45644623998865974,
      "grad_norm": 0.396484375,
      "learning_rate": 7.206105506216106e-05,
      "loss": 1.5037,
      "step": 6440
    },
    {
      "epoch": 0.4571550074420583,
      "grad_norm": 0.43359375,
      "learning_rate": 7.077075065009433e-05,
      "loss": 1.5001,
      "step": 6450
    },
    {
      "epoch": 0.4578637748954568,
      "grad_norm": 0.478515625,
      "learning_rate": 6.946350531586959e-05,
      "loss": 1.5007,
      "step": 6460
    },
    {
      "epoch": 0.45857254234885536,
      "grad_norm": 0.40625,
      "learning_rate": 6.814038526753205e-05,
      "loss": 1.4763,
      "step": 6470
    },
    {
      "epoch": 0.4592813098022539,
      "grad_norm": 0.4453125,
      "learning_rate": 6.680246966077151e-05,
      "loss": 1.4987,
      "step": 6480
    },
    {
      "epoch": 0.45999007725565244,
      "grad_norm": 0.3984375,
      "learning_rate": 6.545084971874738e-05,
      "loss": 1.4925,
      "step": 6490
    },
    {
      "epoch": 0.460698844709051,
      "grad_norm": 0.40234375,
      "learning_rate": 6.408662784207149e-05,
      "loss": 1.4923,
      "step": 6500
    },
    {
      "epoch": 0.460698844709051,
      "eval_accuracy": 0.666518879775364,
      "eval_loss": 1.39996337890625,
      "eval_runtime": 7.2215,
      "eval_samples_per_second": 48.051,
      "eval_steps_per_second": 0.415,
      "step": 6500
    },
    {
      "epoch": 0.4614076121624495,
      "grad_norm": 0.42578125,
      "learning_rate": 6.271091670967436e-05,
      "loss": 1.4809,
      "step": 6510
    },
    {
      "epoch": 0.46211637961584806,
      "grad_norm": 0.41015625,
      "learning_rate": 6.132483837128823e-05,
      "loss": 1.4809,
      "step": 6520
    },
    {
      "epoch": 0.4628251470692466,
      "grad_norm": 0.41796875,
      "learning_rate": 5.992952333228728e-05,
      "loss": 1.5065,
      "step": 6530
    },
    {
      "epoch": 0.46353391452264514,
      "grad_norm": 0.453125,
      "learning_rate": 5.85261096316312e-05,
      "loss": 1.4931,
      "step": 6540
    },
    {
      "epoch": 0.4642426819760437,
      "grad_norm": 0.431640625,
      "learning_rate": 5.7115741913664264e-05,
      "loss": 1.4766,
      "step": 6550
    },
    {
      "epoch": 0.4649514494294422,
      "grad_norm": 0.412109375,
      "learning_rate": 5.569957049452703e-05,
      "loss": 1.49,
      "step": 6560
    },
    {
      "epoch": 0.46566021688284076,
      "grad_norm": 0.423828125,
      "learning_rate": 5.427875042394199e-05,
      "loss": 1.4955,
      "step": 6570
    },
    {
      "epoch": 0.4663689843362393,
      "grad_norm": 0.4375,
      "learning_rate": 5.2854440543138406e-05,
      "loss": 1.4972,
      "step": 6580
    },
    {
      "epoch": 0.46707775178963784,
      "grad_norm": 0.408203125,
      "learning_rate": 5.142780253968481e-05,
      "loss": 1.4879,
      "step": 6590
    },
    {
      "epoch": 0.4677865192430364,
      "grad_norm": 0.380859375,
      "learning_rate": 5e-05,
      "loss": 1.493,
      "step": 6600
    },
    {
      "epoch": 0.4684952866964349,
      "grad_norm": 0.3671875,
      "learning_rate": 4.85721974603152e-05,
      "loss": 1.4995,
      "step": 6610
    },
    {
      "epoch": 0.46920405414983346,
      "grad_norm": 0.408203125,
      "learning_rate": 4.71455594568616e-05,
      "loss": 1.4771,
      "step": 6620
    },
    {
      "epoch": 0.469912821603232,
      "grad_norm": 0.419921875,
      "learning_rate": 4.5721249576058027e-05,
      "loss": 1.4815,
      "step": 6630
    },
    {
      "epoch": 0.47062158905663054,
      "grad_norm": 0.396484375,
      "learning_rate": 4.4300429505472976e-05,
      "loss": 1.4733,
      "step": 6640
    },
    {
      "epoch": 0.4713303565100291,
      "grad_norm": 0.365234375,
      "learning_rate": 4.288425808633575e-05,
      "loss": 1.4976,
      "step": 6650
    },
    {
      "epoch": 0.4720391239634276,
      "grad_norm": 0.35546875,
      "learning_rate": 4.147389036836881e-05,
      "loss": 1.4884,
      "step": 6660
    },
    {
      "epoch": 0.47274789141682616,
      "grad_norm": 0.375,
      "learning_rate": 4.007047666771274e-05,
      "loss": 1.4969,
      "step": 6670
    },
    {
      "epoch": 0.4734566588702247,
      "grad_norm": 0.375,
      "learning_rate": 3.8675161628711776e-05,
      "loss": 1.5137,
      "step": 6680
    },
    {
      "epoch": 0.47416542632362324,
      "grad_norm": 0.365234375,
      "learning_rate": 3.728908329032567e-05,
      "loss": 1.4889,
      "step": 6690
    },
    {
      "epoch": 0.4748741937770218,
      "grad_norm": 0.408203125,
      "learning_rate": 3.591337215792852e-05,
      "loss": 1.4875,
      "step": 6700
    },
    {
      "epoch": 0.4755829612304203,
      "grad_norm": 0.349609375,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 1.5153,
      "step": 6710
    },
    {
      "epoch": 0.47629172868381886,
      "grad_norm": 0.353515625,
      "learning_rate": 3.3197530339228487e-05,
      "loss": 1.4904,
      "step": 6720
    },
    {
      "epoch": 0.4770004961372174,
      "grad_norm": 0.349609375,
      "learning_rate": 3.1859614732467954e-05,
      "loss": 1.4851,
      "step": 6730
    },
    {
      "epoch": 0.47770926359061594,
      "grad_norm": 0.34765625,
      "learning_rate": 3.053649468413043e-05,
      "loss": 1.4979,
      "step": 6740
    },
    {
      "epoch": 0.4784180310440145,
      "grad_norm": 0.33203125,
      "learning_rate": 2.9229249349905684e-05,
      "loss": 1.492,
      "step": 6750
    },
    {
      "epoch": 0.479126798497413,
      "grad_norm": 0.345703125,
      "learning_rate": 2.7938944937838923e-05,
      "loss": 1.4843,
      "step": 6760
    },
    {
      "epoch": 0.47983556595081156,
      "grad_norm": 0.34765625,
      "learning_rate": 2.6666633838716314e-05,
      "loss": 1.4781,
      "step": 6770
    },
    {
      "epoch": 0.4805443334042101,
      "grad_norm": 0.349609375,
      "learning_rate": 2.5413353767719805e-05,
      "loss": 1.4949,
      "step": 6780
    },
    {
      "epoch": 0.48125310085760864,
      "grad_norm": 0.35546875,
      "learning_rate": 2.418012691805191e-05,
      "loss": 1.4961,
      "step": 6790
    },
    {
      "epoch": 0.4819618683110072,
      "grad_norm": 0.38671875,
      "learning_rate": 2.296795912722014e-05,
      "loss": 1.4942,
      "step": 6800
    },
    {
      "epoch": 0.4826706357644057,
      "grad_norm": 0.333984375,
      "learning_rate": 2.1777839056661554e-05,
      "loss": 1.49,
      "step": 6810
    },
    {
      "epoch": 0.48337940321780426,
      "grad_norm": 0.361328125,
      "learning_rate": 2.061073738537635e-05,
      "loss": 1.5085,
      "step": 6820
    },
    {
      "epoch": 0.4840881706712028,
      "grad_norm": 0.34765625,
      "learning_rate": 1.946760601822809e-05,
      "loss": 1.4926,
      "step": 6830
    },
    {
      "epoch": 0.48479693812460134,
      "grad_norm": 0.3203125,
      "learning_rate": 1.8349377309556486e-05,
      "loss": 1.4815,
      "step": 6840
    },
    {
      "epoch": 0.4855057055779999,
      "grad_norm": 0.328125,
      "learning_rate": 1.725696330273575e-05,
      "loss": 1.5001,
      "step": 6850
    },
    {
      "epoch": 0.4862144730313984,
      "grad_norm": 0.3359375,
      "learning_rate": 1.619125498629904e-05,
      "loss": 1.481,
      "step": 6860
    },
    {
      "epoch": 0.48692324048479696,
      "grad_norm": 0.353515625,
      "learning_rate": 1.5153121567235335e-05,
      "loss": 1.4781,
      "step": 6870
    },
    {
      "epoch": 0.4876320079381955,
      "grad_norm": 0.31640625,
      "learning_rate": 1.414340976205183e-05,
      "loss": 1.4906,
      "step": 6880
    },
    {
      "epoch": 0.48834077539159404,
      "grad_norm": 0.328125,
      "learning_rate": 1.3162943106179749e-05,
      "loss": 1.5031,
      "step": 6890
    },
    {
      "epoch": 0.4890495428449926,
      "grad_norm": 0.328125,
      "learning_rate": 1.2212521282287092e-05,
      "loss": 1.4878,
      "step": 6900
    },
    {
      "epoch": 0.4897583102983911,
      "grad_norm": 0.322265625,
      "learning_rate": 1.1292919468045877e-05,
      "loss": 1.4929,
      "step": 6910
    },
    {
      "epoch": 0.49046707775178966,
      "grad_norm": 0.33984375,
      "learning_rate": 1.0404887703886251e-05,
      "loss": 1.4889,
      "step": 6920
    },
    {
      "epoch": 0.4911758452051882,
      "grad_norm": 0.359375,
      "learning_rate": 9.549150281252633e-06,
      "loss": 1.4783,
      "step": 6930
    },
    {
      "epoch": 0.49188461265858674,
      "grad_norm": 0.322265625,
      "learning_rate": 8.7264051518613e-06,
      "loss": 1.4928,
      "step": 6940
    },
    {
      "epoch": 0.4925933801119853,
      "grad_norm": 0.3359375,
      "learning_rate": 7.937323358440935e-06,
      "loss": 1.489,
      "step": 6950
    },
    {
      "epoch": 0.4933021475653838,
      "grad_norm": 0.328125,
      "learning_rate": 7.182548487420554e-06,
      "loss": 1.4899,
      "step": 6960
    },
    {
      "epoch": 0.49401091501878236,
      "grad_norm": 0.33984375,
      "learning_rate": 6.462696144011149e-06,
      "loss": 1.4859,
      "step": 6970
    },
    {
      "epoch": 0.4947196824721809,
      "grad_norm": 0.34375,
      "learning_rate": 5.778353450109286e-06,
      "loss": 1.4944,
      "step": 6980
    },
    {
      "epoch": 0.49542844992557944,
      "grad_norm": 0.33203125,
      "learning_rate": 5.13007856543209e-06,
      "loss": 1.481,
      "step": 6990
    },
    {
      "epoch": 0.496137217378978,
      "grad_norm": 0.333984375,
      "learning_rate": 4.5184002322740785e-06,
      "loss": 1.4965,
      "step": 7000
    },
    {
      "epoch": 0.496137217378978,
      "eval_accuracy": 0.6668890507507222,
      "eval_loss": 1.397897720336914,
      "eval_runtime": 7.477,
      "eval_samples_per_second": 46.409,
      "eval_steps_per_second": 0.401,
      "step": 7000
    },
    {
      "epoch": 0.4968459848323765,
      "grad_norm": 0.373046875,
      "learning_rate": 3.9438173442575e-06,
      "loss": 1.492,
      "step": 7010
    },
    {
      "epoch": 0.49755475228577506,
      "grad_norm": 0.310546875,
      "learning_rate": 3.406798539427386e-06,
      "loss": 1.4714,
      "step": 7020
    },
    {
      "epoch": 0.4982635197391736,
      "grad_norm": 0.31640625,
      "learning_rate": 2.9077818180237693e-06,
      "loss": 1.4839,
      "step": 7030
    },
    {
      "epoch": 0.49897228719257214,
      "grad_norm": 0.33203125,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 1.4976,
      "step": 7040
    },
    {
      "epoch": 0.4996810546459707,
      "grad_norm": 0.328125,
      "learning_rate": 2.0253513192751373e-06,
      "loss": 1.4857,
      "step": 7050
    },
    {
      "epoch": 0.50003543837267,
      "step": 7055,
      "total_flos": 1.568873006179696e+20,
      "train_loss": 1.4944760516549913,
      "train_runtime": 7219.1988,
      "train_samples_per_second": 125.076,
      "train_steps_per_second": 0.977
    }
  ],
  "logging_steps": 10,
  "max_steps": 7055,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.568873006179696e+20,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}