{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.982174688057041,
  "eval_steps": 500,
  "global_step": 2800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0035650623885918,
      "grad_norm": 544.0,
      "learning_rate": 7.142857142857143e-07,
      "loss": 53.7211,
      "step": 1
    },
    {
      "epoch": 0.017825311942959002,
      "grad_norm": 592.0,
      "learning_rate": 3.5714285714285714e-06,
      "loss": 57.6044,
      "step": 5
    },
    {
      "epoch": 0.035650623885918005,
      "grad_norm": 608.0,
      "learning_rate": 7.142857142857143e-06,
      "loss": 57.8426,
      "step": 10
    },
    {
      "epoch": 0.053475935828877004,
      "grad_norm": 440.0,
      "learning_rate": 1.0714285714285714e-05,
      "loss": 48.446,
      "step": 15
    },
    {
      "epoch": 0.07130124777183601,
      "grad_norm": 374.0,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 39.2254,
      "step": 20
    },
    {
      "epoch": 0.08912655971479501,
      "grad_norm": 66.5,
      "learning_rate": 1.785714285714286e-05,
      "loss": 27.6479,
      "step": 25
    },
    {
      "epoch": 0.10695187165775401,
      "grad_norm": 35.75,
      "learning_rate": 2.1428571428571428e-05,
      "loss": 24.2563,
      "step": 30
    },
    {
      "epoch": 0.12477718360071301,
      "grad_norm": 32.0,
      "learning_rate": 2.5e-05,
      "loss": 23.2193,
      "step": 35
    },
    {
      "epoch": 0.14260249554367202,
      "grad_norm": 19.75,
      "learning_rate": 2.857142857142857e-05,
      "loss": 21.5174,
      "step": 40
    },
    {
      "epoch": 0.16042780748663102,
      "grad_norm": 11.0625,
      "learning_rate": 3.2142857142857144e-05,
      "loss": 19.7135,
      "step": 45
    },
    {
      "epoch": 0.17825311942959002,
      "grad_norm": 7.1875,
      "learning_rate": 3.571428571428572e-05,
      "loss": 19.1551,
      "step": 50
    },
    {
      "epoch": 0.19607843137254902,
      "grad_norm": 9.0,
      "learning_rate": 3.928571428571429e-05,
      "loss": 18.2747,
      "step": 55
    },
    {
      "epoch": 0.21390374331550802,
      "grad_norm": 13.1875,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 17.2428,
      "step": 60
    },
    {
      "epoch": 0.23172905525846701,
      "grad_norm": 23.25,
      "learning_rate": 4.642857142857143e-05,
      "loss": 16.4017,
      "step": 65
    },
    {
      "epoch": 0.24955436720142601,
      "grad_norm": 57.0,
      "learning_rate": 5e-05,
      "loss": 14.1658,
      "step": 70
    },
    {
      "epoch": 0.26737967914438504,
      "grad_norm": 57.5,
      "learning_rate": 5.3571428571428575e-05,
      "loss": 9.052,
      "step": 75
    },
    {
      "epoch": 0.28520499108734404,
      "grad_norm": 10.875,
      "learning_rate": 5.714285714285714e-05,
      "loss": 3.1244,
      "step": 80
    },
    {
      "epoch": 0.30303030303030304,
      "grad_norm": 10.0625,
      "learning_rate": 6.0714285714285715e-05,
      "loss": 2.0444,
      "step": 85
    },
    {
      "epoch": 0.32085561497326204,
      "grad_norm": 2.640625,
      "learning_rate": 6.428571428571429e-05,
      "loss": 1.7451,
      "step": 90
    },
    {
      "epoch": 0.33868092691622104,
      "grad_norm": 2.96875,
      "learning_rate": 6.785714285714286e-05,
      "loss": 1.5311,
      "step": 95
    },
    {
      "epoch": 0.35650623885918004,
      "grad_norm": 5.90625,
      "learning_rate": 7.142857142857143e-05,
      "loss": 1.4329,
      "step": 100
    },
    {
      "epoch": 0.37433155080213903,
      "grad_norm": 2.78125,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.2936,
      "step": 105
    },
    {
      "epoch": 0.39215686274509803,
      "grad_norm": 1.8125,
      "learning_rate": 7.857142857142858e-05,
      "loss": 1.2525,
      "step": 110
    },
    {
      "epoch": 0.40998217468805703,
      "grad_norm": 2.578125,
      "learning_rate": 8.214285714285714e-05,
      "loss": 1.2163,
      "step": 115
    },
    {
      "epoch": 0.42780748663101603,
      "grad_norm": 10.3125,
      "learning_rate": 8.571428571428571e-05,
      "loss": 1.1466,
      "step": 120
    },
    {
      "epoch": 0.44563279857397503,
      "grad_norm": 8.5625,
      "learning_rate": 8.92857142857143e-05,
      "loss": 1.1372,
      "step": 125
    },
    {
      "epoch": 0.46345811051693403,
      "grad_norm": 1.375,
      "learning_rate": 9.285714285714286e-05,
      "loss": 1.0929,
      "step": 130
    },
    {
      "epoch": 0.48128342245989303,
      "grad_norm": 3.53125,
      "learning_rate": 9.642857142857143e-05,
      "loss": 1.0724,
      "step": 135
    },
    {
      "epoch": 0.49910873440285203,
      "grad_norm": 15.125,
      "learning_rate": 0.0001,
      "loss": 1.0491,
      "step": 140
    },
    {
      "epoch": 0.5169340463458111,
      "grad_norm": 1.84375,
      "learning_rate": 0.00010357142857142859,
      "loss": 1.03,
      "step": 145
    },
    {
      "epoch": 0.5347593582887701,
      "grad_norm": 2.734375,
      "learning_rate": 0.00010714285714285715,
      "loss": 0.9893,
      "step": 150
    },
    {
      "epoch": 0.5525846702317291,
      "grad_norm": 2.71875,
      "learning_rate": 0.00011071428571428572,
      "loss": 0.9907,
      "step": 155
    },
    {
      "epoch": 0.5704099821746881,
      "grad_norm": 7.21875,
      "learning_rate": 0.00011428571428571428,
      "loss": 0.9565,
      "step": 160
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 7.375,
      "learning_rate": 0.00011785714285714287,
      "loss": 0.937,
      "step": 165
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 2.59375,
      "learning_rate": 0.00012142857142857143,
      "loss": 0.9235,
      "step": 170
    },
    {
      "epoch": 0.6238859180035651,
      "grad_norm": 5.1875,
      "learning_rate": 0.000125,
      "loss": 0.9509,
      "step": 175
    },
    {
      "epoch": 0.6417112299465241,
      "grad_norm": 2.28125,
      "learning_rate": 0.00012857142857142858,
      "loss": 0.9249,
      "step": 180
    },
    {
      "epoch": 0.6595365418894831,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00013214285714285715,
      "loss": 0.9199,
      "step": 185
    },
    {
      "epoch": 0.6773618538324421,
      "grad_norm": 6.8125,
      "learning_rate": 0.00013571428571428572,
      "loss": 0.9006,
      "step": 190
    },
    {
      "epoch": 0.6951871657754011,
      "grad_norm": 2.140625,
      "learning_rate": 0.0001392857142857143,
      "loss": 0.9527,
      "step": 195
    },
    {
      "epoch": 0.7130124777183601,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00014285714285714287,
      "loss": 0.9255,
      "step": 200
    },
    {
      "epoch": 0.7308377896613191,
      "grad_norm": 4.875,
      "learning_rate": 0.00014642857142857141,
      "loss": 0.9124,
      "step": 205
    },
    {
      "epoch": 0.7486631016042781,
      "grad_norm": 3.328125,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.894,
      "step": 210
    },
    {
      "epoch": 0.7664884135472371,
      "grad_norm": 2.5625,
      "learning_rate": 0.0001535714285714286,
      "loss": 0.8926,
      "step": 215
    },
    {
      "epoch": 0.7843137254901961,
      "grad_norm": 5.5625,
      "learning_rate": 0.00015714285714285716,
      "loss": 0.8809,
      "step": 220
    },
    {
      "epoch": 0.8021390374331551,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00016071428571428573,
      "loss": 0.88,
      "step": 225
    },
    {
      "epoch": 0.8199643493761141,
      "grad_norm": 2.171875,
      "learning_rate": 0.00016428571428571428,
      "loss": 0.8727,
      "step": 230
    },
    {
      "epoch": 0.8377896613190731,
      "grad_norm": 1.875,
      "learning_rate": 0.00016785714285714288,
      "loss": 0.854,
      "step": 235
    },
    {
      "epoch": 0.8556149732620321,
      "grad_norm": 5.84375,
      "learning_rate": 0.00017142857142857143,
      "loss": 0.8838,
      "step": 240
    },
    {
      "epoch": 0.8734402852049911,
      "grad_norm": 1.7109375,
      "learning_rate": 0.000175,
      "loss": 0.8374,
      "step": 245
    },
    {
      "epoch": 0.8912655971479501,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001785714285714286,
      "loss": 0.8766,
      "step": 250
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 1.015625,
      "learning_rate": 0.00018214285714285714,
      "loss": 0.8761,
      "step": 255
    },
    {
      "epoch": 0.9269162210338681,
      "grad_norm": 2.5,
      "learning_rate": 0.00018571428571428572,
      "loss": 0.8501,
      "step": 260
    },
    {
      "epoch": 0.9447415329768271,
      "grad_norm": 3.78125,
      "learning_rate": 0.0001892857142857143,
      "loss": 0.8405,
      "step": 265
    },
    {
      "epoch": 0.9625668449197861,
      "grad_norm": 2.671875,
      "learning_rate": 0.00019285714285714286,
      "loss": 0.8425,
      "step": 270
    },
    {
      "epoch": 0.9803921568627451,
      "grad_norm": 4.1875,
      "learning_rate": 0.00019642857142857144,
      "loss": 0.8156,
      "step": 275
    },
    {
      "epoch": 0.9982174688057041,
      "grad_norm": 1.640625,
      "learning_rate": 0.0002,
      "loss": 0.8742,
      "step": 280
    },
    {
      "epoch": 0.9982174688057041,
      "eval_loss": 2.193777322769165,
      "eval_runtime": 1.0432,
      "eval_samples_per_second": 4.793,
      "eval_steps_per_second": 1.917,
      "step": 280
    },
    {
      "epoch": 1.0160427807486632,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00019999805729315381,
      "loss": 0.8053,
      "step": 285
    },
    {
      "epoch": 1.0338680926916222,
      "grad_norm": 15.875,
      "learning_rate": 0.0001999922292480975,
      "loss": 0.7993,
      "step": 290
    },
    {
      "epoch": 1.0516934046345812,
      "grad_norm": 0.703125,
      "learning_rate": 0.00019998251609127464,
      "loss": 0.7952,
      "step": 295
    },
    {
      "epoch": 1.0695187165775402,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00019996891820008164,
      "loss": 0.8014,
      "step": 300
    },
    {
      "epoch": 1.0873440285204992,
      "grad_norm": 2.9375,
      "learning_rate": 0.00019995143610285277,
      "loss": 0.8063,
      "step": 305
    },
    {
      "epoch": 1.1051693404634582,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019993007047883988,
      "loss": 0.7976,
      "step": 310
    },
    {
      "epoch": 1.1229946524064172,
      "grad_norm": 3.84375,
      "learning_rate": 0.0001999048221581858,
      "loss": 0.8032,
      "step": 315
    },
    {
      "epoch": 1.1408199643493762,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019987569212189224,
      "loss": 0.803,
      "step": 320
    },
    {
      "epoch": 1.1586452762923352,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00019984268150178167,
      "loss": 0.7784,
      "step": 325
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 2.03125,
      "learning_rate": 0.0001998057915804532,
      "loss": 0.8112,
      "step": 330
    },
    {
      "epoch": 1.1942959001782532,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0001997650237912329,
      "loss": 0.8125,
      "step": 335
    },
    {
      "epoch": 1.2121212121212122,
      "grad_norm": 1.703125,
      "learning_rate": 0.00019972037971811802,
      "loss": 0.8039,
      "step": 340
    },
    {
      "epoch": 1.2299465240641712,
      "grad_norm": 2.8125,
      "learning_rate": 0.00019967186109571552,
      "loss": 0.8025,
      "step": 345
    },
    {
      "epoch": 1.2477718360071302,
      "grad_norm": 1.046875,
      "learning_rate": 0.00019961946980917456,
      "loss": 0.7937,
      "step": 350
    },
    {
      "epoch": 1.2655971479500892,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00019956320789411337,
      "loss": 0.7789,
      "step": 355
    },
    {
      "epoch": 1.2834224598930482,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00019950307753654017,
      "loss": 0.7762,
      "step": 360
    },
    {
      "epoch": 1.3012477718360071,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00019943908107276798,
      "loss": 0.7617,
      "step": 365
    },
    {
      "epoch": 1.3190730837789661,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019937122098932428,
      "loss": 0.7645,
      "step": 370
    },
    {
      "epoch": 1.3368983957219251,
      "grad_norm": 0.65625,
      "learning_rate": 0.00019929949992285396,
      "loss": 0.764,
      "step": 375
    },
    {
      "epoch": 1.3547237076648841,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00019922392066001722,
      "loss": 0.781,
      "step": 380
    },
    {
      "epoch": 1.3725490196078431,
      "grad_norm": 0.71484375,
      "learning_rate": 0.00019914448613738106,
      "loss": 0.7643,
      "step": 385
    },
    {
      "epoch": 1.3903743315508021,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0001990611994413053,
      "loss": 0.7628,
      "step": 390
    },
    {
      "epoch": 1.4081996434937611,
      "grad_norm": 1.8125,
      "learning_rate": 0.00019897406380782261,
      "loss": 0.7739,
      "step": 395
    },
    {
      "epoch": 1.4260249554367201,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00019888308262251285,
      "loss": 0.774,
      "step": 400
    },
    {
      "epoch": 1.4438502673796791,
      "grad_norm": 2.28125,
      "learning_rate": 0.00019878825942037148,
      "loss": 0.7485,
      "step": 405
    },
    {
      "epoch": 1.4616755793226381,
      "grad_norm": 0.5859375,
      "learning_rate": 0.00019868959788567212,
      "loss": 0.7735,
      "step": 410
    },
    {
      "epoch": 1.4795008912655971,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001985871018518236,
      "loss": 0.7649,
      "step": 415
    },
    {
      "epoch": 1.4973262032085561,
      "grad_norm": 2.953125,
      "learning_rate": 0.00019848077530122083,
      "loss": 0.7589,
      "step": 420
    },
    {
      "epoch": 1.5151515151515151,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019837062236509014,
      "loss": 0.7574,
      "step": 425
    },
    {
      "epoch": 1.5329768270944741,
      "grad_norm": 2.609375,
      "learning_rate": 0.00019825664732332884,
      "loss": 0.7551,
      "step": 430
    },
    {
      "epoch": 1.5508021390374331,
      "grad_norm": 0.74609375,
      "learning_rate": 0.00019813885460433879,
      "loss": 0.7595,
      "step": 435
    },
    {
      "epoch": 1.5686274509803921,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00019801724878485438,
      "loss": 0.7716,
      "step": 440
    },
    {
      "epoch": 1.5864527629233511,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019789183458976484,
      "loss": 0.7629,
      "step": 445
    },
    {
      "epoch": 1.6042780748663101,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00019776261689193048,
      "loss": 0.7372,
      "step": 450
    },
    {
      "epoch": 1.6221033868092691,
      "grad_norm": 1.125,
      "learning_rate": 0.00019762960071199333,
      "loss": 0.7505,
      "step": 455
    },
    {
      "epoch": 1.6399286987522281,
      "grad_norm": 2.0625,
      "learning_rate": 0.00019749279121818235,
      "loss": 0.747,
      "step": 460
    },
    {
      "epoch": 1.6577540106951871,
      "grad_norm": 3.046875,
      "learning_rate": 0.00019735219372611233,
      "loss": 0.7641,
      "step": 465
    },
    {
      "epoch": 1.6755793226381461,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019720781369857746,
      "loss": 0.7568,
      "step": 470
    },
    {
      "epoch": 1.6934046345811051,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0001970596567453391,
      "loss": 0.7767,
      "step": 475
    },
    {
      "epoch": 1.7112299465240641,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001969077286229078,
      "loss": 0.7286,
      "step": 480
    },
    {
      "epoch": 1.7290552584670231,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019675203523431966,
      "loss": 0.742,
      "step": 485
    },
    {
      "epoch": 1.7468805704099821,
      "grad_norm": 3.15625,
      "learning_rate": 0.00019659258262890683,
      "loss": 0.7531,
      "step": 490
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 0.734375,
      "learning_rate": 0.0001964293770020628,
      "loss": 0.7452,
      "step": 495
    },
    {
      "epoch": 1.7825311942959001,
      "grad_norm": 4.25,
      "learning_rate": 0.0001962624246950012,
      "loss": 0.7418,
      "step": 500
    },
    {
      "epoch": 1.8003565062388591,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019609173219450998,
      "loss": 0.7584,
      "step": 505
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 5.0625,
      "learning_rate": 0.0001959173061326988,
      "loss": 0.7544,
      "step": 510
    },
    {
      "epoch": 1.8360071301247771,
      "grad_norm": 2.109375,
      "learning_rate": 0.0001957391532867418,
      "loss": 0.7601,
      "step": 515
    },
    {
      "epoch": 1.8538324420677363,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0001955572805786141,
      "loss": 0.7313,
      "step": 520
    },
    {
      "epoch": 1.8716577540106951,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0001953716950748227,
      "loss": 0.7241,
      "step": 525
    },
    {
      "epoch": 1.8894830659536543,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019518240398613227,
      "loss": 0.7512,
      "step": 530
    },
    {
      "epoch": 1.9073083778966131,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0001949894146672846,
      "loss": 0.7244,
      "step": 535
    },
    {
      "epoch": 1.9251336898395723,
      "grad_norm": 2.0625,
      "learning_rate": 0.0001947927346167132,
      "loss": 0.7434,
      "step": 540
    },
    {
      "epoch": 1.9429590017825311,
      "grad_norm": 1.28125,
      "learning_rate": 0.0001945923714762516,
      "loss": 0.7387,
      "step": 545
    },
    {
      "epoch": 1.9607843137254903,
      "grad_norm": 4.125,
      "learning_rate": 0.00019438833303083678,
      "loss": 0.7388,
      "step": 550
    },
    {
      "epoch": 1.9786096256684491,
      "grad_norm": 2.1875,
      "learning_rate": 0.00019418062720820637,
      "loss": 0.7361,
      "step": 555
    },
    {
      "epoch": 1.9964349376114083,
      "grad_norm": 1.84375,
      "learning_rate": 0.00019396926207859084,
      "loss": 0.7213,
      "step": 560
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.146196126937866,
      "eval_runtime": 1.0228,
      "eval_samples_per_second": 4.888,
      "eval_steps_per_second": 1.955,
      "step": 561
    },
    {
      "epoch": 2.014260249554367,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019375424585439994,
      "loss": 0.6961,
      "step": 565
    },
    {
      "epoch": 2.0320855614973263,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001935355868899034,
      "loss": 0.6878,
      "step": 570
    },
    {
      "epoch": 2.049910873440285,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019331329368090666,
      "loss": 0.6976,
      "step": 575
    },
    {
      "epoch": 2.0677361853832443,
      "grad_norm": 1.15625,
      "learning_rate": 0.00019308737486442045,
      "loss": 0.6947,
      "step": 580
    },
    {
      "epoch": 2.085561497326203,
      "grad_norm": 1.0625,
      "learning_rate": 0.00019285783921832537,
      "loss": 0.6682,
      "step": 585
    },
    {
      "epoch": 2.1033868092691623,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0001926246956610309,
      "loss": 0.6653,
      "step": 590
    },
    {
      "epoch": 2.121212121212121,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001923879532511287,
      "loss": 0.6826,
      "step": 595
    },
    {
      "epoch": 2.1390374331550803,
      "grad_norm": 1.015625,
      "learning_rate": 0.00019214762118704076,
      "loss": 0.6856,
      "step": 600
    },
    {
      "epoch": 2.156862745098039,
      "grad_norm": 0.828125,
      "learning_rate": 0.00019190370880666207,
      "loss": 0.7042,
      "step": 605
    },
    {
      "epoch": 2.1746880570409983,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00019165622558699763,
      "loss": 0.6811,
      "step": 610
    },
    {
      "epoch": 2.192513368983957,
      "grad_norm": 2.015625,
      "learning_rate": 0.00019140518114379434,
      "loss": 0.6949,
      "step": 615
    },
    {
      "epoch": 2.2103386809269163,
      "grad_norm": 0.55859375,
      "learning_rate": 0.00019115058523116733,
      "loss": 0.6886,
      "step": 620
    },
    {
      "epoch": 2.228163992869875,
      "grad_norm": 0.73046875,
      "learning_rate": 0.0001908924477412211,
      "loss": 0.6918,
      "step": 625
    },
    {
      "epoch": 2.2459893048128343,
      "grad_norm": 0.64453125,
      "learning_rate": 0.000190630778703665,
      "loss": 0.6844,
      "step": 630
    },
    {
      "epoch": 2.263814616755793,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00019036558828542367,
      "loss": 0.6939,
      "step": 635
    },
    {
      "epoch": 2.2816399286987523,
      "grad_norm": 0.6796875,
      "learning_rate": 0.0001900968867902419,
      "loss": 0.6966,
      "step": 640
    },
    {
      "epoch": 2.299465240641711,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00018982468465828442,
      "loss": 0.6894,
      "step": 645
    },
    {
      "epoch": 2.3172905525846703,
      "grad_norm": 0.65234375,
      "learning_rate": 0.0001895489924657301,
      "loss": 0.6878,
      "step": 650
    },
    {
      "epoch": 2.335115864527629,
      "grad_norm": 2.921875,
      "learning_rate": 0.00018926982092436117,
      "loss": 0.6878,
      "step": 655
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001889871808811469,
      "loss": 0.6829,
      "step": 660
    },
    {
      "epoch": 2.370766488413547,
      "grad_norm": 2.078125,
      "learning_rate": 0.00018870108331782217,
      "loss": 0.7115,
      "step": 665
    },
    {
      "epoch": 2.3885918003565063,
      "grad_norm": 1.53125,
      "learning_rate": 0.00018841153935046098,
      "loss": 0.6921,
      "step": 670
    },
    {
      "epoch": 2.406417112299465,
      "grad_norm": 0.6875,
      "learning_rate": 0.00018811856022904423,
      "loss": 0.6963,
      "step": 675
    },
    {
      "epoch": 2.4242424242424243,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00018782215733702286,
      "loss": 0.6968,
      "step": 680
    },
    {
      "epoch": 2.442067736185383,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00018752234219087538,
      "loss": 0.6953,
      "step": 685
    },
    {
      "epoch": 2.4598930481283423,
      "grad_norm": 1.09375,
      "learning_rate": 0.00018721912643966055,
      "loss": 0.6898,
      "step": 690
    },
    {
      "epoch": 2.477718360071301,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00018691252186456465,
      "loss": 0.6777,
      "step": 695
    },
    {
      "epoch": 2.4955436720142603,
      "grad_norm": 1.015625,
      "learning_rate": 0.00018660254037844388,
      "loss": 0.7143,
      "step": 700
    },
    {
      "epoch": 2.5133689839572195,
      "grad_norm": 0.6328125,
      "learning_rate": 0.0001862891940253613,
      "loss": 0.708,
      "step": 705
    },
    {
      "epoch": 2.5311942959001783,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00018597249498011903,
      "loss": 0.6775,
      "step": 710
    },
    {
      "epoch": 2.549019607843137,
      "grad_norm": 2.65625,
      "learning_rate": 0.00018565245554778517,
      "loss": 0.6673,
      "step": 715
    },
    {
      "epoch": 2.5668449197860963,
      "grad_norm": 0.53515625,
      "learning_rate": 0.00018532908816321558,
      "loss": 0.6801,
      "step": 720
    },
    {
      "epoch": 2.5846702317290555,
      "grad_norm": 0.96875,
      "learning_rate": 0.0001850024053905709,
      "loss": 0.6834,
      "step": 725
    },
    {
      "epoch": 2.6024955436720143,
      "grad_norm": 0.84375,
      "learning_rate": 0.00018467241992282843,
      "loss": 0.6809,
      "step": 730
    },
    {
      "epoch": 2.620320855614973,
      "grad_norm": 0.53125,
      "learning_rate": 0.0001843391445812886,
      "loss": 0.6793,
      "step": 735
    },
    {
      "epoch": 2.6381461675579323,
      "grad_norm": 0.56640625,
      "learning_rate": 0.00018400259231507717,
      "loss": 0.686,
      "step": 740
    },
    {
      "epoch": 2.6559714795008915,
      "grad_norm": 2.328125,
      "learning_rate": 0.00018366277620064197,
      "loss": 0.6829,
      "step": 745
    },
    {
      "epoch": 2.6737967914438503,
      "grad_norm": 1.5,
      "learning_rate": 0.0001833197094412449,
      "loss": 0.6862,
      "step": 750
    },
    {
      "epoch": 2.691622103386809,
      "grad_norm": 0.7265625,
      "learning_rate": 0.00018297340536644875,
      "loss": 0.6791,
      "step": 755
    },
    {
      "epoch": 2.7094474153297683,
      "grad_norm": 0.69921875,
      "learning_rate": 0.0001826238774315995,
      "loss": 0.6924,
      "step": 760
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 12.25,
      "learning_rate": 0.00018227113921730334,
      "loss": 0.6946,
      "step": 765
    },
    {
      "epoch": 2.7450980392156863,
      "grad_norm": 0.9375,
      "learning_rate": 0.0001819152044288992,
      "loss": 0.6999,
      "step": 770
    },
    {
      "epoch": 2.762923351158645,
      "grad_norm": 0.5234375,
      "learning_rate": 0.00018155608689592604,
      "loss": 0.694,
      "step": 775
    },
    {
      "epoch": 2.7807486631016043,
      "grad_norm": 0.58984375,
      "learning_rate": 0.00018119380057158568,
      "loss": 0.6958,
      "step": 780
    },
    {
      "epoch": 2.7985739750445635,
      "grad_norm": 0.5390625,
      "learning_rate": 0.00018082835953220054,
      "loss": 0.6951,
      "step": 785
    },
    {
      "epoch": 2.8163992869875223,
      "grad_norm": 0.72265625,
      "learning_rate": 0.00018045977797666684,
      "loss": 0.6826,
      "step": 790
    },
    {
      "epoch": 2.834224598930481,
      "grad_norm": 0.66015625,
      "learning_rate": 0.00018008807022590283,
      "loss": 0.6889,
      "step": 795
    },
    {
      "epoch": 2.8520499108734403,
      "grad_norm": 0.51171875,
      "learning_rate": 0.00017971325072229226,
      "loss": 0.6953,
      "step": 800
    },
    {
      "epoch": 2.8698752228163995,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00017933533402912354,
      "loss": 0.6755,
      "step": 805
    },
    {
      "epoch": 2.8877005347593583,
      "grad_norm": 0.51953125,
      "learning_rate": 0.00017895433483002354,
      "loss": 0.6981,
      "step": 810
    },
    {
      "epoch": 2.905525846702317,
      "grad_norm": 0.625,
      "learning_rate": 0.00017857026792838737,
      "loss": 0.6827,
      "step": 815
    },
    {
      "epoch": 2.9233511586452763,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000178183148246803,
      "loss": 0.6895,
      "step": 820
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 1.171875,
      "learning_rate": 0.00017779299082647148,
      "loss": 0.678,
      "step": 825
    },
    {
      "epoch": 2.9590017825311943,
      "grad_norm": 0.953125,
      "learning_rate": 0.00017739981082662276,
      "loss": 0.6959,
      "step": 830
    },
    {
      "epoch": 2.976827094474153,
      "grad_norm": 0.515625,
      "learning_rate": 0.0001770036235239263,
      "loss": 0.7065,
      "step": 835
    },
    {
      "epoch": 2.9946524064171123,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0001766044443118978,
      "loss": 0.675,
      "step": 840
    },
    {
      "epoch": 2.998217468805704,
      "eval_loss": 2.1484053134918213,
      "eval_runtime": 1.0318,
      "eval_samples_per_second": 4.846,
      "eval_steps_per_second": 1.938,
      "step": 841
    },
    {
      "epoch": 3.0124777183600715,
      "grad_norm": 0.68359375,
      "learning_rate": 0.00017620228870030108,
      "loss": 0.6463,
      "step": 845
    },
    {
      "epoch": 3.0303030303030303,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0001757971723145453,
      "loss": 0.6127,
      "step": 850
    },
    {
      "epoch": 3.0481283422459895,
      "grad_norm": 0.60546875,
      "learning_rate": 0.000175389110895078,
      "loss": 0.5995,
      "step": 855
    },
    {
      "epoch": 3.0659536541889483,
      "grad_norm": 0.6015625,
      "learning_rate": 0.00017497812029677344,
      "loss": 0.6274,
      "step": 860
    },
    {
      "epoch": 3.0837789661319075,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00017456421648831655,
      "loss": 0.6126,
      "step": 865
    },
    {
      "epoch": 3.1016042780748663,
      "grad_norm": 0.80859375,
      "learning_rate": 0.00017414741555158266,
      "loss": 0.632,
      "step": 870
    },
    {
      "epoch": 3.1194295900178255,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0001737277336810124,
      "loss": 0.6282,
      "step": 875
    },
    {
      "epoch": 3.1372549019607843,
      "grad_norm": 0.546875,
      "learning_rate": 0.00017330518718298264,
      "loss": 0.6381,
      "step": 880
    },
    {
      "epoch": 3.1550802139037435,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00017287979247517286,
      "loss": 0.6231,
      "step": 885
    },
    {
      "epoch": 3.1729055258467023,
      "grad_norm": 0.5703125,
      "learning_rate": 0.00017245156608592727,
      "loss": 0.619,
      "step": 890
    },
    {
      "epoch": 3.1907308377896615,
      "grad_norm": 0.5078125,
      "learning_rate": 0.00017202052465361268,
      "loss": 0.6245,
      "step": 895
    },
    {
      "epoch": 3.2085561497326203,
      "grad_norm": 0.56640625,
      "learning_rate": 0.00017158668492597186,
      "loss": 0.626,
      "step": 900
    },
    {
      "epoch": 3.2263814616755795,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00017115006375947303,
      "loss": 0.6136,
      "step": 905
    },
    {
      "epoch": 3.2442067736185383,
      "grad_norm": 0.58984375,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.631,
      "step": 910
    },
    {
      "epoch": 3.2620320855614975,
      "grad_norm": 0.57421875,
      "learning_rate": 0.00017026854507546692,
      "loss": 0.6254,
      "step": 915
    },
    {
      "epoch": 3.2798573975044563,
      "grad_norm": 0.484375,
      "learning_rate": 0.00016982368180860728,
      "loss": 0.6268,
      "step": 920
    },
    {
      "epoch": 3.2976827094474155,
      "grad_norm": 0.54296875,
      "learning_rate": 0.00016937610560285418,
      "loss": 0.6283,
      "step": 925
    },
    {
      "epoch": 3.3155080213903743,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0001689258338483947,
      "loss": 0.6185,
      "step": 930
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.451171875,
      "learning_rate": 0.00016847288404014935,
      "loss": 0.6363,
      "step": 935
    },
    {
      "epoch": 3.3511586452762923,
      "grad_norm": 0.5234375,
      "learning_rate": 0.00016801727377709194,
      "loss": 0.6231,
      "step": 940
    },
    {
      "epoch": 3.3689839572192515,
      "grad_norm": 0.73046875,
      "learning_rate": 0.00016755902076156604,
      "loss": 0.6178,
      "step": 945
    },
    {
      "epoch": 3.3868092691622103,
      "grad_norm": 0.67578125,
      "learning_rate": 0.00016709814279859702,
      "loss": 0.6196,
      "step": 950
    },
    {
      "epoch": 3.4046345811051695,
      "grad_norm": 0.5390625,
      "learning_rate": 0.0001666346577952004,
      "loss": 0.6196,
      "step": 955
    },
    {
      "epoch": 3.4224598930481283,
      "grad_norm": 0.5625,
      "learning_rate": 0.00016616858375968595,
      "loss": 0.6329,
      "step": 960
    },
    {
      "epoch": 3.4402852049910875,
      "grad_norm": 1.03125,
      "learning_rate": 0.00016569993880095806,
      "loss": 0.631,
      "step": 965
    },
    {
      "epoch": 3.4581105169340463,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016522874112781213,
      "loss": 0.6481,
      "step": 970
    },
    {
      "epoch": 3.4759358288770055,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00016475500904822706,
      "loss": 0.6349,
      "step": 975
    },
    {
      "epoch": 3.4937611408199643,
      "grad_norm": 0.6953125,
      "learning_rate": 0.00016427876096865394,
      "loss": 0.6409,
      "step": 980
    },
    {
      "epoch": 3.5115864527629235,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001638000153933009,
      "loss": 0.6424,
      "step": 985
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.79296875,
      "learning_rate": 0.000163318790923414,
      "loss": 0.6337,
      "step": 990
    },
    {
      "epoch": 3.5472370766488415,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00016283510625655472,
      "loss": 0.6283,
      "step": 995
    },
    {
      "epoch": 3.5650623885918002,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00016234898018587337,
      "loss": 0.6275,
      "step": 1000
    },
    {
      "epoch": 3.5828877005347595,
      "grad_norm": 1.125,
      "learning_rate": 0.00016186043159937882,
      "loss": 0.6319,
      "step": 1005
    },
    {
      "epoch": 3.6007130124777182,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00016136947947920476,
      "loss": 0.6268,
      "step": 1010
    },
    {
      "epoch": 3.6185383244206775,
      "grad_norm": 0.5859375,
      "learning_rate": 0.00016087614290087208,
      "loss": 0.6329,
      "step": 1015
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.7734375,
      "learning_rate": 0.00016038044103254775,
      "loss": 0.6183,
      "step": 1020
    },
    {
      "epoch": 3.6541889483065955,
      "grad_norm": 0.84375,
      "learning_rate": 0.00015988239313430004,
      "loss": 0.6359,
      "step": 1025
    },
    {
      "epoch": 3.6720142602495542,
      "grad_norm": 0.578125,
      "learning_rate": 0.00015938201855735014,
      "loss": 0.6376,
      "step": 1030
    },
    {
      "epoch": 3.6898395721925135,
      "grad_norm": 0.734375,
      "learning_rate": 0.00015887933674332046,
      "loss": 0.6314,
      "step": 1035
    },
    {
      "epoch": 3.7076648841354722,
      "grad_norm": 0.75390625,
      "learning_rate": 0.000158374367223479,
      "loss": 0.631,
      "step": 1040
    },
    {
      "epoch": 3.7254901960784315,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0001578671296179806,
      "loss": 0.6348,
      "step": 1045
    },
    {
      "epoch": 3.7433155080213902,
      "grad_norm": 0.75,
      "learning_rate": 0.0001573576436351046,
      "loss": 0.6371,
      "step": 1050
    },
    {
      "epoch": 3.7611408199643495,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00015684592907048926,
      "loss": 0.6365,
      "step": 1055
    },
    {
      "epoch": 3.7789661319073082,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001563320058063622,
      "loss": 0.6477,
      "step": 1060
    },
    {
      "epoch": 3.7967914438502675,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001558158938107684,
      "loss": 0.6472,
      "step": 1065
    },
    {
      "epoch": 3.8146167557932262,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00015529761313679393,
      "loss": 0.6352,
      "step": 1070
    },
    {
      "epoch": 3.8324420677361855,
      "grad_norm": 0.73828125,
      "learning_rate": 0.00015477718392178716,
      "loss": 0.6299,
      "step": 1075
    },
    {
      "epoch": 3.8502673796791442,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00015425462638657595,
      "loss": 0.6255,
      "step": 1080
    },
    {
      "epoch": 3.8680926916221035,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001537299608346824,
      "loss": 0.6318,
      "step": 1085
    },
    {
      "epoch": 3.8859180035650622,
      "grad_norm": 0.62890625,
      "learning_rate": 0.00015320320765153367,
      "loss": 0.6308,
      "step": 1090
    },
    {
      "epoch": 3.9037433155080214,
      "grad_norm": 0.73046875,
      "learning_rate": 0.0001526743873036701,
      "loss": 0.6436,
      "step": 1095
    },
    {
      "epoch": 3.9215686274509802,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001521435203379498,
      "loss": 0.6412,
      "step": 1100
    },
    {
      "epoch": 3.9393939393939394,
      "grad_norm": 0.56640625,
      "learning_rate": 0.00015161062738075067,
      "loss": 0.6399,
      "step": 1105
    },
    {
      "epoch": 3.9572192513368982,
      "grad_norm": 0.859375,
      "learning_rate": 0.00015107572913716858,
      "loss": 0.631,
      "step": 1110
    },
    {
      "epoch": 3.9750445632798574,
      "grad_norm": 0.6171875,
      "learning_rate": 0.0001505388463902131,
      "loss": 0.6188,
      "step": 1115
    },
    {
      "epoch": 3.9928698752228167,
      "grad_norm": 0.546875,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.6439,
      "step": 1120
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.214867115020752,
      "eval_runtime": 1.0238,
      "eval_samples_per_second": 4.884,
      "eval_steps_per_second": 1.954,
      "step": 1122
    },
    {
      "epoch": 4.010695187165775,
      "grad_norm": 0.609375,
      "learning_rate": 0.00014945921090294076,
      "loss": 0.5755,
      "step": 1125
    },
    {
      "epoch": 4.028520499108734,
      "grad_norm": 0.80078125,
      "learning_rate": 0.00014891650011092896,
      "loss": 0.5648,
      "step": 1130
    },
    {
      "epoch": 4.046345811051693,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000148371888710524,
      "loss": 0.5641,
      "step": 1135
    },
    {
      "epoch": 4.064171122994653,
      "grad_norm": 0.64453125,
      "learning_rate": 0.00014782539786213183,
      "loss": 0.5618,
      "step": 1140
    },
    {
      "epoch": 4.081996434937611,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0001472770487991827,
      "loss": 0.5628,
      "step": 1145
    },
    {
      "epoch": 4.09982174688057,
      "grad_norm": 0.69140625,
      "learning_rate": 0.0001467268628273062,
      "loss": 0.558,
      "step": 1150
    },
    {
      "epoch": 4.117647058823529,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00014617486132350343,
      "loss": 0.5556,
      "step": 1155
    },
    {
      "epoch": 4.135472370766489,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001456210657353163,
      "loss": 0.565,
      "step": 1160
    },
    {
      "epoch": 4.153297682709447,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00014506549757999454,
      "loss": 0.5682,
      "step": 1165
    },
    {
      "epoch": 4.171122994652406,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00014450817844365921,
      "loss": 0.5699,
      "step": 1170
    },
    {
      "epoch": 4.188948306595365,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001439491299804645,
      "loss": 0.5665,
      "step": 1175
    },
    {
      "epoch": 4.206773618538325,
      "grad_norm": 1.15625,
      "learning_rate": 0.00014338837391175582,
      "loss": 0.5795,
      "step": 1180
    },
    {
      "epoch": 4.224598930481283,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00014282593202522627,
      "loss": 0.5707,
      "step": 1185
    },
    {
      "epoch": 4.242424242424242,
      "grad_norm": 0.765625,
      "learning_rate": 0.00014226182617406996,
      "loss": 0.5649,
      "step": 1190
    },
    {
      "epoch": 4.260249554367201,
      "grad_norm": 0.7265625,
      "learning_rate": 0.00014169607827613283,
      "loss": 0.5594,
      "step": 1195
    },
    {
      "epoch": 4.278074866310161,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00014112871031306119,
      "loss": 0.5871,
      "step": 1200
    },
    {
      "epoch": 4.295900178253119,
      "grad_norm": 0.96875,
      "learning_rate": 0.00014055974432944753,
      "loss": 0.5714,
      "step": 1205
    },
    {
      "epoch": 4.313725490196078,
      "grad_norm": 1.359375,
      "learning_rate": 0.00013998920243197407,
      "loss": 0.5731,
      "step": 1210
    },
    {
      "epoch": 4.331550802139038,
      "grad_norm": 1.109375,
      "learning_rate": 0.00013941710678855396,
      "loss": 0.5717,
      "step": 1215
    },
    {
      "epoch": 4.349376114081997,
      "grad_norm": 1.109375,
      "learning_rate": 0.00013884347962746948,
      "loss": 0.5773,
      "step": 1220
    },
    {
      "epoch": 4.367201426024955,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000138268343236509,
      "loss": 0.5818,
      "step": 1225
    },
    {
      "epoch": 4.385026737967914,
      "grad_norm": 1.234375,
      "learning_rate": 0.00013769171996210052,
      "loss": 0.5575,
      "step": 1230
    },
    {
      "epoch": 4.402852049910873,
      "grad_norm": 0.796875,
      "learning_rate": 0.00013711363220844379,
      "loss": 0.5638,
      "step": 1235
    },
    {
      "epoch": 4.420677361853833,
      "grad_norm": 0.625,
      "learning_rate": 0.00013653410243663952,
      "loss": 0.5702,
      "step": 1240
    },
    {
      "epoch": 4.438502673796791,
      "grad_norm": 0.55078125,
      "learning_rate": 0.00013595315316381677,
      "loss": 0.5773,
      "step": 1245
    },
    {
      "epoch": 4.45632798573975,
      "grad_norm": 0.73828125,
      "learning_rate": 0.00013537080696225814,
      "loss": 0.5686,
      "step": 1250
    },
    {
      "epoch": 4.47415329768271,
      "grad_norm": 0.80859375,
      "learning_rate": 0.00013478708645852272,
      "loss": 0.5714,
      "step": 1255
    },
    {
      "epoch": 4.491978609625669,
      "grad_norm": 0.76953125,
      "learning_rate": 0.00013420201433256689,
      "loss": 0.5718,
      "step": 1260
    },
    {
      "epoch": 4.509803921568627,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001336156133168631,
      "loss": 0.5846,
      "step": 1265
    },
    {
      "epoch": 4.527629233511586,
      "grad_norm": 0.7734375,
      "learning_rate": 0.00013302790619551674,
      "loss": 0.5761,
      "step": 1270
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 0.609375,
      "learning_rate": 0.00013243891580338072,
      "loss": 0.5737,
      "step": 1275
    },
    {
      "epoch": 4.563279857397505,
      "grad_norm": 0.56640625,
      "learning_rate": 0.00013184866502516845,
      "loss": 0.5854,
      "step": 1280
    },
    {
      "epoch": 4.581105169340463,
      "grad_norm": 0.57421875,
      "learning_rate": 0.00013125717679456447,
      "loss": 0.589,
      "step": 1285
    },
    {
      "epoch": 4.598930481283422,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00013066447409333345,
      "loss": 0.5706,
      "step": 1290
    },
    {
      "epoch": 4.616755793226382,
      "grad_norm": 0.5390625,
      "learning_rate": 0.00013007057995042732,
      "loss": 0.5733,
      "step": 1295
    },
    {
      "epoch": 4.634581105169341,
      "grad_norm": 0.54296875,
      "learning_rate": 0.00012947551744109043,
      "loss": 0.5871,
      "step": 1300
    },
    {
      "epoch": 4.652406417112299,
      "grad_norm": 0.51171875,
      "learning_rate": 0.000128879309685963,
      "loss": 0.5843,
      "step": 1305
    },
    {
      "epoch": 4.670231729055258,
      "grad_norm": 0.75390625,
      "learning_rate": 0.00012828197985018276,
      "loss": 0.5677,
      "step": 1310
    },
    {
      "epoch": 4.688057040998218,
      "grad_norm": 0.6015625,
      "learning_rate": 0.00012768355114248494,
      "loss": 0.5654,
      "step": 1315
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.58984375,
      "learning_rate": 0.00012708404681430053,
      "loss": 0.5741,
      "step": 1320
    },
    {
      "epoch": 4.723707664884135,
      "grad_norm": 0.58984375,
      "learning_rate": 0.00012648349015885273,
      "loss": 0.5678,
      "step": 1325
    },
    {
      "epoch": 4.741532976827094,
      "grad_norm": 0.70703125,
      "learning_rate": 0.00012588190451025207,
      "loss": 0.5756,
      "step": 1330
    },
    {
      "epoch": 4.759358288770054,
      "grad_norm": 0.625,
      "learning_rate": 0.00012527931324258975,
      "loss": 0.572,
      "step": 1335
    },
    {
      "epoch": 4.777183600713013,
      "grad_norm": 0.5703125,
      "learning_rate": 0.00012467573976902935,
      "loss": 0.5911,
      "step": 1340
    },
    {
      "epoch": 4.795008912655971,
      "grad_norm": 0.4765625,
      "learning_rate": 0.00012407120754089732,
      "loss": 0.5817,
      "step": 1345
    },
    {
      "epoch": 4.81283422459893,
      "grad_norm": 0.53515625,
      "learning_rate": 0.00012346574004677154,
      "loss": 0.5788,
      "step": 1350
    },
    {
      "epoch": 4.83065953654189,
      "grad_norm": 0.53125,
      "learning_rate": 0.00012285936081156897,
      "loss": 0.5816,
      "step": 1355
    },
    {
      "epoch": 4.848484848484849,
      "grad_norm": 0.53515625,
      "learning_rate": 0.00012225209339563145,
      "loss": 0.5847,
      "step": 1360
    },
    {
      "epoch": 4.866310160427807,
      "grad_norm": 0.5859375,
      "learning_rate": 0.00012164396139381029,
      "loss": 0.5869,
      "step": 1365
    },
    {
      "epoch": 4.884135472370766,
      "grad_norm": 0.82421875,
      "learning_rate": 0.00012103498843454959,
      "loss": 0.5827,
      "step": 1370
    },
    {
      "epoch": 4.901960784313726,
      "grad_norm": 0.5703125,
      "learning_rate": 0.00012042519817896804,
      "loss": 0.576,
      "step": 1375
    },
    {
      "epoch": 4.919786096256685,
      "grad_norm": 0.5546875,
      "learning_rate": 0.00011981461431993977,
      "loss": 0.5958,
      "step": 1380
    },
    {
      "epoch": 4.937611408199643,
      "grad_norm": 0.52734375,
      "learning_rate": 0.00011920326058117364,
      "loss": 0.585,
      "step": 1385
    },
    {
      "epoch": 4.955436720142602,
      "grad_norm": 0.96875,
      "learning_rate": 0.00011859116071629149,
      "loss": 0.5904,
      "step": 1390
    },
    {
      "epoch": 4.973262032085562,
      "grad_norm": 0.57421875,
      "learning_rate": 0.00011797833850790527,
      "loss": 0.5828,
      "step": 1395
    },
    {
      "epoch": 4.991087344028521,
      "grad_norm": 0.57421875,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.569,
      "step": 1400
    },
    {
      "epoch": 4.998217468805704,
      "eval_loss": 2.3224310874938965,
      "eval_runtime": 1.0378,
      "eval_samples_per_second": 4.818,
      "eval_steps_per_second": 1.927,
      "step": 1402
    },
    {
      "epoch": 5.008912655971479,
      "grad_norm": 0.54296875,
      "learning_rate": 0.00011675062233047364,
      "loss": 0.5598,
      "step": 1405
    },
    {
      "epoch": 5.026737967914438,
      "grad_norm": 0.78515625,
      "learning_rate": 0.00011613577606328068,
      "loss": 0.5118,
      "step": 1410
    },
    {
      "epoch": 5.044563279857398,
      "grad_norm": 0.59765625,
      "learning_rate": 0.00011552030285443515,
      "loss": 0.5109,
      "step": 1415
    },
    {
      "epoch": 5.062388591800357,
      "grad_norm": 0.57421875,
      "learning_rate": 0.00011490422661761744,
      "loss": 0.5033,
      "step": 1420
    },
    {
      "epoch": 5.080213903743315,
      "grad_norm": 0.578125,
      "learning_rate": 0.00011428757128993802,
      "loss": 0.5113,
      "step": 1425
    },
    {
      "epoch": 5.098039215686274,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00011367036083100735,
      "loss": 0.4987,
      "step": 1430
    },
    {
      "epoch": 5.115864527629234,
      "grad_norm": 0.69140625,
      "learning_rate": 0.00011305261922200519,
      "loss": 0.5111,
      "step": 1435
    },
    {
      "epoch": 5.133689839572193,
      "grad_norm": 0.65234375,
      "learning_rate": 0.00011243437046474853,
      "loss": 0.5186,
      "step": 1440
    },
    {
      "epoch": 5.151515151515151,
      "grad_norm": 0.7109375,
      "learning_rate": 0.0001118156385807593,
      "loss": 0.5054,
      "step": 1445
    },
    {
      "epoch": 5.16934046345811,
      "grad_norm": 0.55859375,
      "learning_rate": 0.00011119644761033078,
      "loss": 0.5073,
      "step": 1450
    },
    {
      "epoch": 5.18716577540107,
      "grad_norm": 0.66015625,
      "learning_rate": 0.00011057682161159379,
      "loss": 0.5298,
      "step": 1455
    },
    {
      "epoch": 5.204991087344029,
      "grad_norm": 0.59765625,
      "learning_rate": 0.00010995678465958168,
      "loss": 0.504,
      "step": 1460
    },
    {
      "epoch": 5.222816399286987,
      "grad_norm": 0.5625,
      "learning_rate": 0.00010933636084529506,
      "loss": 0.5361,
      "step": 1465
    },
    {
      "epoch": 5.240641711229946,
      "grad_norm": 0.6171875,
      "learning_rate": 0.00010871557427476583,
      "loss": 0.521,
      "step": 1470
    },
    {
      "epoch": 5.258467023172906,
      "grad_norm": 0.66015625,
      "learning_rate": 0.00010809444906812033,
      "loss": 0.522,
      "step": 1475
    },
    {
      "epoch": 5.276292335115865,
      "grad_norm": 0.859375,
      "learning_rate": 0.00010747300935864243,
      "loss": 0.5188,
      "step": 1480
    },
    {
      "epoch": 5.294117647058823,
      "grad_norm": 0.78515625,
      "learning_rate": 0.00010685127929183567,
      "loss": 0.5203,
      "step": 1485
    },
    {
      "epoch": 5.311942959001782,
      "grad_norm": 0.62890625,
      "learning_rate": 0.00010622928302448523,
      "loss": 0.5182,
      "step": 1490
    },
    {
      "epoch": 5.329768270944742,
      "grad_norm": 0.5859375,
      "learning_rate": 0.00010560704472371919,
      "loss": 0.5317,
      "step": 1495
    },
    {
      "epoch": 5.347593582887701,
      "grad_norm": 0.75390625,
      "learning_rate": 0.00010498458856606972,
      "loss": 0.5186,
      "step": 1500
    },
    {
      "epoch": 5.365418894830659,
      "grad_norm": 0.63671875,
      "learning_rate": 0.00010436193873653361,
      "loss": 0.5155,
      "step": 1505
    },
    {
      "epoch": 5.383244206773618,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0001037391194276326,
      "loss": 0.5198,
      "step": 1510
    },
    {
      "epoch": 5.401069518716578,
      "grad_norm": 0.62109375,
      "learning_rate": 0.00010311615483847332,
      "loss": 0.5184,
      "step": 1515
    },
    {
      "epoch": 5.418894830659537,
      "grad_norm": 0.59765625,
      "learning_rate": 0.0001024930691738073,
      "loss": 0.5256,
      "step": 1520
    },
    {
      "epoch": 5.436720142602495,
      "grad_norm": 0.66015625,
      "learning_rate": 0.00010186988664309023,
      "loss": 0.5203,
      "step": 1525
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 0.7109375,
      "learning_rate": 0.00010124663145954152,
      "loss": 0.5166,
      "step": 1530
    },
    {
      "epoch": 5.472370766488414,
      "grad_norm": 0.62890625,
      "learning_rate": 0.00010062332783920336,
      "loss": 0.5168,
      "step": 1535
    },
    {
      "epoch": 5.490196078431373,
      "grad_norm": 0.70703125,
      "learning_rate": 0.0001,
      "loss": 0.5153,
      "step": 1540
    },
    {
      "epoch": 5.508021390374331,
      "grad_norm": 0.7109375,
      "learning_rate": 9.937667216079665e-05,
      "loss": 0.5308,
      "step": 1545
    },
    {
      "epoch": 5.52584670231729,
      "grad_norm": 0.609375,
      "learning_rate": 9.875336854045851e-05,
      "loss": 0.5148,
      "step": 1550
    },
    {
      "epoch": 5.54367201426025,
      "grad_norm": 0.6015625,
      "learning_rate": 9.813011335690981e-05,
      "loss": 0.5247,
      "step": 1555
    },
    {
      "epoch": 5.561497326203209,
      "grad_norm": 0.87890625,
      "learning_rate": 9.750693082619273e-05,
      "loss": 0.5219,
      "step": 1560
    },
    {
      "epoch": 5.579322638146167,
      "grad_norm": 0.71875,
      "learning_rate": 9.688384516152672e-05,
      "loss": 0.5213,
      "step": 1565
    },
    {
      "epoch": 5.597147950089127,
      "grad_norm": 0.734375,
      "learning_rate": 9.626088057236745e-05,
      "loss": 0.5279,
      "step": 1570
    },
    {
      "epoch": 5.614973262032086,
      "grad_norm": 0.69140625,
      "learning_rate": 9.563806126346642e-05,
      "loss": 0.5256,
      "step": 1575
    },
    {
      "epoch": 5.632798573975045,
      "grad_norm": 0.76171875,
      "learning_rate": 9.501541143393028e-05,
      "loss": 0.529,
      "step": 1580
    },
    {
      "epoch": 5.650623885918003,
      "grad_norm": 0.65234375,
      "learning_rate": 9.439295527628081e-05,
      "loss": 0.5163,
      "step": 1585
    },
    {
      "epoch": 5.668449197860962,
      "grad_norm": 0.5703125,
      "learning_rate": 9.37707169755148e-05,
      "loss": 0.5255,
      "step": 1590
    },
    {
      "epoch": 5.686274509803922,
      "grad_norm": 0.578125,
      "learning_rate": 9.314872070816434e-05,
      "loss": 0.5236,
      "step": 1595
    },
    {
      "epoch": 5.704099821746881,
      "grad_norm": 0.5703125,
      "learning_rate": 9.252699064135758e-05,
      "loss": 0.5272,
      "step": 1600
    },
    {
      "epoch": 5.721925133689839,
      "grad_norm": 0.546875,
      "learning_rate": 9.190555093187967e-05,
      "loss": 0.5276,
      "step": 1605
    },
    {
      "epoch": 5.739750445632799,
      "grad_norm": 0.58984375,
      "learning_rate": 9.128442572523417e-05,
      "loss": 0.519,
      "step": 1610
    },
    {
      "epoch": 5.757575757575758,
      "grad_norm": 0.58984375,
      "learning_rate": 9.066363915470495e-05,
      "loss": 0.5245,
      "step": 1615
    },
    {
      "epoch": 5.775401069518717,
      "grad_norm": 0.6640625,
      "learning_rate": 9.004321534041835e-05,
      "loss": 0.5296,
      "step": 1620
    },
    {
      "epoch": 5.793226381461675,
      "grad_norm": 0.578125,
      "learning_rate": 8.942317838840623e-05,
      "loss": 0.5191,
      "step": 1625
    },
    {
      "epoch": 5.811051693404634,
      "grad_norm": 0.6640625,
      "learning_rate": 8.880355238966923e-05,
      "loss": 0.5236,
      "step": 1630
    },
    {
      "epoch": 5.828877005347594,
      "grad_norm": 0.6875,
      "learning_rate": 8.818436141924072e-05,
      "loss": 0.5321,
      "step": 1635
    },
    {
      "epoch": 5.846702317290553,
      "grad_norm": 0.58203125,
      "learning_rate": 8.756562953525152e-05,
      "loss": 0.5367,
      "step": 1640
    },
    {
      "epoch": 5.864527629233511,
      "grad_norm": 0.55859375,
      "learning_rate": 8.694738077799488e-05,
      "loss": 0.5208,
      "step": 1645
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.5703125,
      "learning_rate": 8.632963916899268e-05,
      "loss": 0.5223,
      "step": 1650
    },
    {
      "epoch": 5.90017825311943,
      "grad_norm": 0.5703125,
      "learning_rate": 8.571242871006202e-05,
      "loss": 0.5221,
      "step": 1655
    },
    {
      "epoch": 5.918003565062389,
      "grad_norm": 0.59375,
      "learning_rate": 8.509577338238255e-05,
      "loss": 0.5299,
      "step": 1660
    },
    {
      "epoch": 5.935828877005347,
      "grad_norm": 0.6328125,
      "learning_rate": 8.447969714556484e-05,
      "loss": 0.5341,
      "step": 1665
    },
    {
      "epoch": 5.953654188948306,
      "grad_norm": 0.71484375,
      "learning_rate": 8.386422393671933e-05,
      "loss": 0.524,
      "step": 1670
    },
    {
      "epoch": 5.971479500891266,
      "grad_norm": 0.56640625,
      "learning_rate": 8.324937766952638e-05,
      "loss": 0.5377,
      "step": 1675
    },
    {
      "epoch": 5.989304812834225,
      "grad_norm": 0.765625,
      "learning_rate": 8.263518223330697e-05,
      "loss": 0.5317,
      "step": 1680
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.4839205741882324,
      "eval_runtime": 1.0218,
      "eval_samples_per_second": 4.893,
      "eval_steps_per_second": 1.957,
      "step": 1683
    },
    {
      "epoch": 6.007130124777183,
      "grad_norm": 0.6171875,
      "learning_rate": 8.202166149209474e-05,
      "loss": 0.4958,
      "step": 1685
    },
    {
      "epoch": 6.024955436720143,
      "grad_norm": 1.0703125,
      "learning_rate": 8.140883928370855e-05,
      "loss": 0.4602,
      "step": 1690
    },
    {
      "epoch": 6.042780748663102,
      "grad_norm": 0.63671875,
      "learning_rate": 8.07967394188264e-05,
      "loss": 0.4597,
      "step": 1695
    },
    {
      "epoch": 6.0606060606060606,
      "grad_norm": 0.91015625,
      "learning_rate": 8.018538568006027e-05,
      "loss": 0.4568,
      "step": 1700
    },
    {
      "epoch": 6.078431372549019,
      "grad_norm": 0.59375,
      "learning_rate": 7.957480182103198e-05,
      "loss": 0.4586,
      "step": 1705
    },
    {
      "epoch": 6.096256684491979,
      "grad_norm": 0.6640625,
      "learning_rate": 7.896501156545045e-05,
      "loss": 0.4695,
      "step": 1710
    },
    {
      "epoch": 6.114081996434938,
      "grad_norm": 0.64453125,
      "learning_rate": 7.835603860618972e-05,
      "loss": 0.4623,
      "step": 1715
    },
    {
      "epoch": 6.1319073083778965,
      "grad_norm": 0.6796875,
      "learning_rate": 7.774790660436858e-05,
      "loss": 0.4671,
      "step": 1720
    },
    {
      "epoch": 6.149732620320855,
      "grad_norm": 0.640625,
      "learning_rate": 7.714063918843106e-05,
      "loss": 0.4631,
      "step": 1725
    },
    {
      "epoch": 6.167557932263815,
      "grad_norm": 0.640625,
      "learning_rate": 7.653425995322851e-05,
      "loss": 0.4692,
      "step": 1730
    },
    {
      "epoch": 6.185383244206774,
      "grad_norm": 0.578125,
      "learning_rate": 7.592879245910273e-05,
      "loss": 0.468,
      "step": 1735
    },
    {
      "epoch": 6.2032085561497325,
      "grad_norm": 1.1328125,
      "learning_rate": 7.532426023097063e-05,
      "loss": 0.4651,
      "step": 1740
    },
    {
      "epoch": 6.221033868092691,
      "grad_norm": 0.58984375,
      "learning_rate": 7.472068675741025e-05,
      "loss": 0.4745,
      "step": 1745
    },
    {
      "epoch": 6.238859180035651,
      "grad_norm": 0.609375,
      "learning_rate": 7.411809548974792e-05,
      "loss": 0.4688,
      "step": 1750
    },
    {
      "epoch": 6.25668449197861,
      "grad_norm": 0.5859375,
      "learning_rate": 7.351650984114728e-05,
      "loss": 0.4645,
      "step": 1755
    },
    {
      "epoch": 6.2745098039215685,
      "grad_norm": 0.609375,
      "learning_rate": 7.291595318569951e-05,
      "loss": 0.4657,
      "step": 1760
    },
    {
      "epoch": 6.292335115864527,
      "grad_norm": 0.6484375,
      "learning_rate": 7.231644885751507e-05,
      "loss": 0.4659,
      "step": 1765
    },
    {
      "epoch": 6.310160427807487,
      "grad_norm": 0.87890625,
      "learning_rate": 7.171802014981726e-05,
      "loss": 0.4715,
      "step": 1770
    },
    {
      "epoch": 6.327985739750446,
      "grad_norm": 0.65234375,
      "learning_rate": 7.112069031403704e-05,
      "loss": 0.4681,
      "step": 1775
    },
    {
      "epoch": 6.3458110516934045,
      "grad_norm": 0.625,
      "learning_rate": 7.052448255890957e-05,
      "loss": 0.4625,
      "step": 1780
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 0.69921875,
      "learning_rate": 6.992942004957271e-05,
      "loss": 0.4675,
      "step": 1785
    },
    {
      "epoch": 6.381461675579323,
      "grad_norm": 0.66015625,
      "learning_rate": 6.933552590666659e-05,
      "loss": 0.4713,
      "step": 1790
    },
    {
      "epoch": 6.399286987522282,
      "grad_norm": 0.57421875,
      "learning_rate": 6.874282320543557e-05,
      "loss": 0.4707,
      "step": 1795
    },
    {
      "epoch": 6.4171122994652405,
      "grad_norm": 0.625,
      "learning_rate": 6.815133497483157e-05,
      "loss": 0.4678,
      "step": 1800
    },
    {
      "epoch": 6.434937611408199,
      "grad_norm": 0.609375,
      "learning_rate": 6.756108419661931e-05,
      "loss": 0.4701,
      "step": 1805
    },
    {
      "epoch": 6.452762923351159,
      "grad_norm": 0.6484375,
      "learning_rate": 6.697209380448333e-05,
      "loss": 0.4721,
      "step": 1810
    },
    {
      "epoch": 6.470588235294118,
      "grad_norm": 0.67578125,
      "learning_rate": 6.638438668313695e-05,
      "loss": 0.4767,
      "step": 1815
    },
    {
      "epoch": 6.4884135472370765,
      "grad_norm": 0.68359375,
      "learning_rate": 6.579798566743314e-05,
      "loss": 0.4772,
      "step": 1820
    },
    {
      "epoch": 6.506238859180035,
      "grad_norm": 0.71875,
      "learning_rate": 6.521291354147727e-05,
      "loss": 0.4749,
      "step": 1825
    },
    {
      "epoch": 6.524064171122995,
      "grad_norm": 0.61328125,
      "learning_rate": 6.462919303774186e-05,
      "loss": 0.4686,
      "step": 1830
    },
    {
      "epoch": 6.541889483065954,
      "grad_norm": 0.609375,
      "learning_rate": 6.404684683618325e-05,
      "loss": 0.4761,
      "step": 1835
    },
    {
      "epoch": 6.5597147950089125,
      "grad_norm": 0.59765625,
      "learning_rate": 6.34658975633605e-05,
      "loss": 0.4785,
      "step": 1840
    },
    {
      "epoch": 6.577540106951871,
      "grad_norm": 0.64453125,
      "learning_rate": 6.288636779155621e-05,
      "loss": 0.473,
      "step": 1845
    },
    {
      "epoch": 6.595365418894831,
      "grad_norm": 0.609375,
      "learning_rate": 6.230828003789949e-05,
      "loss": 0.4684,
      "step": 1850
    },
    {
      "epoch": 6.61319073083779,
      "grad_norm": 0.63671875,
      "learning_rate": 6.173165676349103e-05,
      "loss": 0.4638,
      "step": 1855
    },
    {
      "epoch": 6.6310160427807485,
      "grad_norm": 0.58203125,
      "learning_rate": 6.115652037253053e-05,
      "loss": 0.473,
      "step": 1860
    },
    {
      "epoch": 6.648841354723707,
      "grad_norm": 0.6328125,
      "learning_rate": 6.058289321144608e-05,
      "loss": 0.4724,
      "step": 1865
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.65625,
      "learning_rate": 6.001079756802592e-05,
      "loss": 0.4811,
      "step": 1870
    },
    {
      "epoch": 6.684491978609626,
      "grad_norm": 0.62890625,
      "learning_rate": 5.944025567055251e-05,
      "loss": 0.4698,
      "step": 1875
    },
    {
      "epoch": 6.7023172905525845,
      "grad_norm": 0.69140625,
      "learning_rate": 5.887128968693887e-05,
      "loss": 0.4789,
      "step": 1880
    },
    {
      "epoch": 6.720142602495543,
      "grad_norm": 0.6171875,
      "learning_rate": 5.830392172386723e-05,
      "loss": 0.459,
      "step": 1885
    },
    {
      "epoch": 6.737967914438503,
      "grad_norm": 0.625,
      "learning_rate": 5.773817382593008e-05,
      "loss": 0.4668,
      "step": 1890
    },
    {
      "epoch": 6.755793226381462,
      "grad_norm": 0.69140625,
      "learning_rate": 5.717406797477372e-05,
      "loss": 0.467,
      "step": 1895
    },
    {
      "epoch": 6.7736185383244205,
      "grad_norm": 0.65234375,
      "learning_rate": 5.6611626088244194e-05,
      "loss": 0.4729,
      "step": 1900
    },
    {
      "epoch": 6.791443850267379,
      "grad_norm": 0.6796875,
      "learning_rate": 5.6050870019535494e-05,
      "loss": 0.4677,
      "step": 1905
    },
    {
      "epoch": 6.809269162210339,
      "grad_norm": 0.66796875,
      "learning_rate": 5.549182155634076e-05,
      "loss": 0.4672,
      "step": 1910
    },
    {
      "epoch": 6.827094474153298,
      "grad_norm": 0.70703125,
      "learning_rate": 5.493450242000546e-05,
      "loss": 0.4764,
      "step": 1915
    },
    {
      "epoch": 6.8449197860962565,
      "grad_norm": 0.7109375,
      "learning_rate": 5.43789342646837e-05,
      "loss": 0.4708,
      "step": 1920
    },
    {
      "epoch": 6.862745098039216,
      "grad_norm": 0.66796875,
      "learning_rate": 5.382513867649663e-05,
      "loss": 0.4603,
      "step": 1925
    },
    {
      "epoch": 6.880570409982175,
      "grad_norm": 0.63671875,
      "learning_rate": 5.32731371726938e-05,
      "loss": 0.4689,
      "step": 1930
    },
    {
      "epoch": 6.898395721925134,
      "grad_norm": 0.67578125,
      "learning_rate": 5.272295120081732e-05,
      "loss": 0.4701,
      "step": 1935
    },
    {
      "epoch": 6.9162210338680925,
      "grad_norm": 0.62109375,
      "learning_rate": 5.217460213786821e-05,
      "loss": 0.472,
      "step": 1940
    },
    {
      "epoch": 6.934046345811051,
      "grad_norm": 0.60546875,
      "learning_rate": 5.162811128947602e-05,
      "loss": 0.4717,
      "step": 1945
    },
    {
      "epoch": 6.951871657754011,
      "grad_norm": 0.70703125,
      "learning_rate": 5.108349988907111e-05,
      "loss": 0.4766,
      "step": 1950
    },
    {
      "epoch": 6.96969696969697,
      "grad_norm": 0.6015625,
      "learning_rate": 5.054078909705926e-05,
      "loss": 0.465,
      "step": 1955
    },
    {
      "epoch": 6.9875222816399285,
      "grad_norm": 0.6640625,
      "learning_rate": 5.000000000000002e-05,
      "loss": 0.472,
      "step": 1960
    },
    {
      "epoch": 6.998217468805704,
      "eval_loss": 2.6540169715881348,
      "eval_runtime": 1.0444,
      "eval_samples_per_second": 4.787,
      "eval_steps_per_second": 1.915,
      "step": 1963
    },
    {
      "epoch": 7.005347593582887,
      "grad_norm": 0.5625,
      "learning_rate": 4.946115360978696e-05,
      "loss": 0.4574,
      "step": 1965
    },
    {
      "epoch": 7.023172905525847,
      "grad_norm": 0.6015625,
      "learning_rate": 4.892427086283147e-05,
      "loss": 0.4256,
      "step": 1970
    },
    {
      "epoch": 7.040998217468806,
      "grad_norm": 0.6875,
      "learning_rate": 4.8389372619249326e-05,
      "loss": 0.4282,
      "step": 1975
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.66796875,
      "learning_rate": 4.78564796620502e-05,
      "loss": 0.4247,
      "step": 1980
    },
    {
      "epoch": 7.076648841354723,
      "grad_norm": 0.6328125,
      "learning_rate": 4.732561269632992e-05,
      "loss": 0.4263,
      "step": 1985
    },
    {
      "epoch": 7.094474153297683,
      "grad_norm": 0.703125,
      "learning_rate": 4.6796792348466356e-05,
      "loss": 0.4238,
      "step": 1990
    },
    {
      "epoch": 7.112299465240642,
      "grad_norm": 0.67578125,
      "learning_rate": 4.6270039165317605e-05,
      "loss": 0.4357,
      "step": 1995
    },
    {
      "epoch": 7.1301247771836005,
      "grad_norm": 0.74609375,
      "learning_rate": 4.574537361342407e-05,
      "loss": 0.4243,
      "step": 2000
    },
    {
      "epoch": 7.14795008912656,
      "grad_norm": 0.6328125,
      "learning_rate": 4.522281607821288e-05,
      "loss": 0.4226,
      "step": 2005
    },
    {
      "epoch": 7.165775401069519,
      "grad_norm": 0.7109375,
      "learning_rate": 4.470238686320606e-05,
      "loss": 0.4256,
      "step": 2010
    },
    {
      "epoch": 7.183600713012478,
      "grad_norm": 0.58984375,
      "learning_rate": 4.4184106189231625e-05,
      "loss": 0.4238,
      "step": 2015
    },
    {
      "epoch": 7.2014260249554365,
      "grad_norm": 0.609375,
      "learning_rate": 4.3667994193637796e-05,
      "loss": 0.429,
      "step": 2020
    },
    {
      "epoch": 7.219251336898395,
      "grad_norm": 0.640625,
      "learning_rate": 4.315407092951078e-05,
      "loss": 0.4252,
      "step": 2025
    },
    {
      "epoch": 7.237076648841355,
      "grad_norm": 0.66015625,
      "learning_rate": 4.264235636489542e-05,
      "loss": 0.425,
      "step": 2030
    },
    {
      "epoch": 7.254901960784314,
      "grad_norm": 0.63671875,
      "learning_rate": 4.213287038201943e-05,
      "loss": 0.4201,
      "step": 2035
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 0.62890625,
      "learning_rate": 4.1625632776521037e-05,
      "loss": 0.4321,
      "step": 2040
    },
    {
      "epoch": 7.290552584670232,
      "grad_norm": 0.62109375,
      "learning_rate": 4.112066325667954e-05,
      "loss": 0.4215,
      "step": 2045
    },
    {
      "epoch": 7.308377896613191,
      "grad_norm": 0.63671875,
      "learning_rate": 4.0617981442649855e-05,
      "loss": 0.4248,
      "step": 2050
    },
    {
      "epoch": 7.32620320855615,
      "grad_norm": 0.73828125,
      "learning_rate": 4.0117606865699975e-05,
      "loss": 0.4308,
      "step": 2055
    },
    {
      "epoch": 7.3440285204991085,
      "grad_norm": 0.6640625,
      "learning_rate": 3.961955896745224e-05,
      "loss": 0.4227,
      "step": 2060
    },
    {
      "epoch": 7.361853832442068,
      "grad_norm": 0.66015625,
      "learning_rate": 3.9123857099127936e-05,
      "loss": 0.4292,
      "step": 2065
    },
    {
      "epoch": 7.379679144385027,
      "grad_norm": 0.62890625,
      "learning_rate": 3.863052052079528e-05,
      "loss": 0.4275,
      "step": 2070
    },
    {
      "epoch": 7.397504456327986,
      "grad_norm": 0.62109375,
      "learning_rate": 3.813956840062118e-05,
      "loss": 0.429,
      "step": 2075
    },
    {
      "epoch": 7.4153297682709445,
      "grad_norm": 0.68359375,
      "learning_rate": 3.7651019814126654e-05,
      "loss": 0.4301,
      "step": 2080
    },
    {
      "epoch": 7.433155080213904,
      "grad_norm": 0.6328125,
      "learning_rate": 3.716489374344527e-05,
      "loss": 0.4264,
      "step": 2085
    },
    {
      "epoch": 7.450980392156863,
      "grad_norm": 0.64453125,
      "learning_rate": 3.668120907658603e-05,
      "loss": 0.4255,
      "step": 2090
    },
    {
      "epoch": 7.468805704099822,
      "grad_norm": 0.7421875,
      "learning_rate": 3.6199984606699155e-05,
      "loss": 0.4261,
      "step": 2095
    },
    {
      "epoch": 7.4866310160427805,
      "grad_norm": 0.640625,
      "learning_rate": 3.5721239031346066e-05,
      "loss": 0.4319,
      "step": 2100
    },
    {
      "epoch": 7.50445632798574,
      "grad_norm": 0.640625,
      "learning_rate": 3.5244990951772974e-05,
      "loss": 0.4233,
      "step": 2105
    },
    {
      "epoch": 7.522281639928699,
      "grad_norm": 0.65625,
      "learning_rate": 3.477125887218792e-05,
      "loss": 0.4264,
      "step": 2110
    },
    {
      "epoch": 7.540106951871658,
      "grad_norm": 0.63671875,
      "learning_rate": 3.430006119904197e-05,
      "loss": 0.4343,
      "step": 2115
    },
    {
      "epoch": 7.5579322638146165,
      "grad_norm": 0.66015625,
      "learning_rate": 3.383141624031408e-05,
      "loss": 0.434,
      "step": 2120
    },
    {
      "epoch": 7.575757575757576,
      "grad_norm": 0.60546875,
      "learning_rate": 3.336534220479961e-05,
      "loss": 0.4326,
      "step": 2125
    },
    {
      "epoch": 7.593582887700535,
      "grad_norm": 0.6796875,
      "learning_rate": 3.290185720140301e-05,
      "loss": 0.4299,
      "step": 2130
    },
    {
      "epoch": 7.611408199643494,
      "grad_norm": 0.625,
      "learning_rate": 3.244097923843398e-05,
      "loss": 0.4219,
      "step": 2135
    },
    {
      "epoch": 7.6292335115864525,
      "grad_norm": 0.62890625,
      "learning_rate": 3.198272622290804e-05,
      "loss": 0.4262,
      "step": 2140
    },
    {
      "epoch": 7.647058823529412,
      "grad_norm": 0.63671875,
      "learning_rate": 3.152711595985065e-05,
      "loss": 0.428,
      "step": 2145
    },
    {
      "epoch": 7.664884135472371,
      "grad_norm": 0.63671875,
      "learning_rate": 3.10741661516053e-05,
      "loss": 0.4282,
      "step": 2150
    },
    {
      "epoch": 7.68270944741533,
      "grad_norm": 0.63671875,
      "learning_rate": 3.062389439714584e-05,
      "loss": 0.4346,
      "step": 2155
    },
    {
      "epoch": 7.7005347593582885,
      "grad_norm": 0.578125,
      "learning_rate": 3.0176318191392726e-05,
      "loss": 0.4253,
      "step": 2160
    },
    {
      "epoch": 7.718360071301248,
      "grad_norm": 0.609375,
      "learning_rate": 2.9731454924533086e-05,
      "loss": 0.4218,
      "step": 2165
    },
    {
      "epoch": 7.736185383244207,
      "grad_norm": 0.640625,
      "learning_rate": 2.9289321881345254e-05,
      "loss": 0.4284,
      "step": 2170
    },
    {
      "epoch": 7.754010695187166,
      "grad_norm": 0.62890625,
      "learning_rate": 2.8849936240527008e-05,
      "loss": 0.4359,
      "step": 2175
    },
    {
      "epoch": 7.7718360071301245,
      "grad_norm": 0.69921875,
      "learning_rate": 2.8413315074028158e-05,
      "loss": 0.4255,
      "step": 2180
    },
    {
      "epoch": 7.789661319073084,
      "grad_norm": 0.625,
      "learning_rate": 2.797947534638736e-05,
      "loss": 0.4396,
      "step": 2185
    },
    {
      "epoch": 7.807486631016043,
      "grad_norm": 0.64453125,
      "learning_rate": 2.7548433914072734e-05,
      "loss": 0.4244,
      "step": 2190
    },
    {
      "epoch": 7.825311942959002,
      "grad_norm": 0.671875,
      "learning_rate": 2.7120207524827168e-05,
      "loss": 0.4363,
      "step": 2195
    },
    {
      "epoch": 7.8431372549019605,
      "grad_norm": 0.6875,
      "learning_rate": 2.669481281701739e-05,
      "loss": 0.4212,
      "step": 2200
    },
    {
      "epoch": 7.86096256684492,
      "grad_norm": 0.64453125,
      "learning_rate": 2.6272266318987603e-05,
      "loss": 0.4321,
      "step": 2205
    },
    {
      "epoch": 7.878787878787879,
      "grad_norm": 0.640625,
      "learning_rate": 2.5852584448417328e-05,
      "loss": 0.4362,
      "step": 2210
    },
    {
      "epoch": 7.896613190730838,
      "grad_norm": 0.671875,
      "learning_rate": 2.5435783511683443e-05,
      "loss": 0.4327,
      "step": 2215
    },
    {
      "epoch": 7.9144385026737964,
      "grad_norm": 0.62890625,
      "learning_rate": 2.502187970322657e-05,
      "loss": 0.4261,
      "step": 2220
    },
    {
      "epoch": 7.932263814616756,
      "grad_norm": 0.6875,
      "learning_rate": 2.461088910492202e-05,
      "loss": 0.4314,
      "step": 2225
    },
    {
      "epoch": 7.950089126559715,
      "grad_norm": 0.6796875,
      "learning_rate": 2.420282768545469e-05,
      "loss": 0.4356,
      "step": 2230
    },
    {
      "epoch": 7.967914438502674,
      "grad_norm": 0.64453125,
      "learning_rate": 2.3797711299698923e-05,
      "loss": 0.429,
      "step": 2235
    },
    {
      "epoch": 7.9857397504456324,
      "grad_norm": 0.69140625,
      "learning_rate": 2.339555568810221e-05,
      "loss": 0.4306,
      "step": 2240
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.8791282176971436,
      "eval_runtime": 1.0241,
      "eval_samples_per_second": 4.883,
      "eval_steps_per_second": 1.953,
      "step": 2244
    },
    {
      "epoch": 8.003565062388592,
      "grad_norm": 0.578125,
      "learning_rate": 2.2996376476073723e-05,
      "loss": 0.4213,
      "step": 2245
    },
    {
      "epoch": 8.02139037433155,
      "grad_norm": 0.58984375,
      "learning_rate": 2.260018917337726e-05,
      "loss": 0.4101,
      "step": 2250
    },
    {
      "epoch": 8.03921568627451,
      "grad_norm": 0.625,
      "learning_rate": 2.2207009173528527e-05,
      "loss": 0.4112,
      "step": 2255
    },
    {
      "epoch": 8.057040998217468,
      "grad_norm": 0.6015625,
      "learning_rate": 2.181685175319702e-05,
      "loss": 0.4117,
      "step": 2260
    },
    {
      "epoch": 8.074866310160427,
      "grad_norm": 0.6328125,
      "learning_rate": 2.142973207161265e-05,
      "loss": 0.4099,
      "step": 2265
    },
    {
      "epoch": 8.092691622103386,
      "grad_norm": 0.62109375,
      "learning_rate": 2.1045665169976468e-05,
      "loss": 0.4074,
      "step": 2270
    },
    {
      "epoch": 8.110516934046347,
      "grad_norm": 0.65625,
      "learning_rate": 2.0664665970876496e-05,
      "loss": 0.4103,
      "step": 2275
    },
    {
      "epoch": 8.128342245989305,
      "grad_norm": 0.57421875,
      "learning_rate": 2.0286749277707782e-05,
      "loss": 0.4073,
      "step": 2280
    },
    {
      "epoch": 8.146167557932264,
      "grad_norm": 0.66796875,
      "learning_rate": 1.9911929774097215e-05,
      "loss": 0.4024,
      "step": 2285
    },
    {
      "epoch": 8.163992869875223,
      "grad_norm": 0.62890625,
      "learning_rate": 1.9540222023333166e-05,
      "loss": 0.4016,
      "step": 2290
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 0.63671875,
      "learning_rate": 1.917164046779948e-05,
      "loss": 0.414,
      "step": 2295
    },
    {
      "epoch": 8.19964349376114,
      "grad_norm": 0.703125,
      "learning_rate": 1.880619942841435e-05,
      "loss": 0.4106,
      "step": 2300
    },
    {
      "epoch": 8.2174688057041,
      "grad_norm": 0.61328125,
      "learning_rate": 1.8443913104073983e-05,
      "loss": 0.3989,
      "step": 2305
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.62890625,
      "learning_rate": 1.808479557110081e-05,
      "loss": 0.4134,
      "step": 2310
    },
    {
      "epoch": 8.253119429590019,
      "grad_norm": 0.62109375,
      "learning_rate": 1.7728860782696664e-05,
      "loss": 0.4121,
      "step": 2315
    },
    {
      "epoch": 8.270944741532977,
      "grad_norm": 0.69921875,
      "learning_rate": 1.7376122568400532e-05,
      "loss": 0.4034,
      "step": 2320
    },
    {
      "epoch": 8.288770053475936,
      "grad_norm": 0.671875,
      "learning_rate": 1.702659463355125e-05,
      "loss": 0.4085,
      "step": 2325
    },
    {
      "epoch": 8.306595365418895,
      "grad_norm": 0.6484375,
      "learning_rate": 1.668029055875512e-05,
      "loss": 0.4109,
      "step": 2330
    },
    {
      "epoch": 8.324420677361854,
      "grad_norm": 0.625,
      "learning_rate": 1.6337223799358026e-05,
      "loss": 0.4103,
      "step": 2335
    },
    {
      "epoch": 8.342245989304812,
      "grad_norm": 0.61328125,
      "learning_rate": 1.5997407684922862e-05,
      "loss": 0.4104,
      "step": 2340
    },
    {
      "epoch": 8.360071301247771,
      "grad_norm": 0.66796875,
      "learning_rate": 1.566085541871145e-05,
      "loss": 0.408,
      "step": 2345
    },
    {
      "epoch": 8.37789661319073,
      "grad_norm": 0.6171875,
      "learning_rate": 1.5327580077171587e-05,
      "loss": 0.4095,
      "step": 2350
    },
    {
      "epoch": 8.39572192513369,
      "grad_norm": 0.6328125,
      "learning_rate": 1.4997594609429088e-05,
      "loss": 0.4044,
      "step": 2355
    },
    {
      "epoch": 8.41354723707665,
      "grad_norm": 0.64453125,
      "learning_rate": 1.467091183678444e-05,
      "loss": 0.4029,
      "step": 2360
    },
    {
      "epoch": 8.431372549019608,
      "grad_norm": 0.609375,
      "learning_rate": 1.4347544452214868e-05,
      "loss": 0.4042,
      "step": 2365
    },
    {
      "epoch": 8.449197860962567,
      "grad_norm": 0.609375,
      "learning_rate": 1.402750501988097e-05,
      "loss": 0.4043,
      "step": 2370
    },
    {
      "epoch": 8.467023172905526,
      "grad_norm": 0.609375,
      "learning_rate": 1.3710805974638696e-05,
      "loss": 0.4041,
      "step": 2375
    },
    {
      "epoch": 8.484848484848484,
      "grad_norm": 0.59375,
      "learning_rate": 1.339745962155613e-05,
      "loss": 0.4155,
      "step": 2380
    },
    {
      "epoch": 8.502673796791443,
      "grad_norm": 0.6875,
      "learning_rate": 1.308747813543536e-05,
      "loss": 0.4106,
      "step": 2385
    },
    {
      "epoch": 8.520499108734402,
      "grad_norm": 0.6640625,
      "learning_rate": 1.2780873560339468e-05,
      "loss": 0.4112,
      "step": 2390
    },
    {
      "epoch": 8.538324420677363,
      "grad_norm": 0.640625,
      "learning_rate": 1.2477657809124631e-05,
      "loss": 0.4062,
      "step": 2395
    },
    {
      "epoch": 8.556149732620321,
      "grad_norm": 0.65234375,
      "learning_rate": 1.2177842662977135e-05,
      "loss": 0.4053,
      "step": 2400
    },
    {
      "epoch": 8.57397504456328,
      "grad_norm": 0.63671875,
      "learning_rate": 1.188143977095576e-05,
      "loss": 0.4135,
      "step": 2405
    },
    {
      "epoch": 8.591800356506239,
      "grad_norm": 0.6015625,
      "learning_rate": 1.1588460649539035e-05,
      "loss": 0.4011,
      "step": 2410
    },
    {
      "epoch": 8.609625668449198,
      "grad_norm": 0.66796875,
      "learning_rate": 1.129891668217783e-05,
      "loss": 0.4061,
      "step": 2415
    },
    {
      "epoch": 8.627450980392156,
      "grad_norm": 0.62109375,
      "learning_rate": 1.1012819118853147e-05,
      "loss": 0.4119,
      "step": 2420
    },
    {
      "epoch": 8.645276292335115,
      "grad_norm": 0.66796875,
      "learning_rate": 1.0730179075638868e-05,
      "loss": 0.409,
      "step": 2425
    },
    {
      "epoch": 8.663101604278076,
      "grad_norm": 0.60546875,
      "learning_rate": 1.0451007534269907e-05,
      "loss": 0.4089,
      "step": 2430
    },
    {
      "epoch": 8.680926916221035,
      "grad_norm": 0.61328125,
      "learning_rate": 1.0175315341715597e-05,
      "loss": 0.4066,
      "step": 2435
    },
    {
      "epoch": 8.698752228163993,
      "grad_norm": 0.59375,
      "learning_rate": 9.903113209758096e-06,
      "loss": 0.4117,
      "step": 2440
    },
    {
      "epoch": 8.716577540106952,
      "grad_norm": 0.60546875,
      "learning_rate": 9.634411714576353e-06,
      "loss": 0.4017,
      "step": 2445
    },
    {
      "epoch": 8.73440285204991,
      "grad_norm": 0.625,
      "learning_rate": 9.369221296335006e-06,
      "loss": 0.4036,
      "step": 2450
    },
    {
      "epoch": 8.75222816399287,
      "grad_norm": 0.62109375,
      "learning_rate": 9.107552258778907e-06,
      "loss": 0.4019,
      "step": 2455
    },
    {
      "epoch": 8.770053475935828,
      "grad_norm": 0.65234375,
      "learning_rate": 8.849414768832687e-06,
      "loss": 0.4119,
      "step": 2460
    },
    {
      "epoch": 8.787878787878787,
      "grad_norm": 0.65625,
      "learning_rate": 8.5948188562057e-06,
      "loss": 0.4065,
      "step": 2465
    },
    {
      "epoch": 8.805704099821746,
      "grad_norm": 0.625,
      "learning_rate": 8.343774413002381e-06,
      "loss": 0.405,
      "step": 2470
    },
    {
      "epoch": 8.823529411764707,
      "grad_norm": 0.6328125,
      "learning_rate": 8.096291193337934e-06,
      "loss": 0.4049,
      "step": 2475
    },
    {
      "epoch": 8.841354723707665,
      "grad_norm": 0.6328125,
      "learning_rate": 7.852378812959227e-06,
      "loss": 0.4033,
      "step": 2480
    },
    {
      "epoch": 8.859180035650624,
      "grad_norm": 0.58203125,
      "learning_rate": 7.612046748871327e-06,
      "loss": 0.4044,
      "step": 2485
    },
    {
      "epoch": 8.877005347593583,
      "grad_norm": 0.71484375,
      "learning_rate": 7.375304338969136e-06,
      "loss": 0.4068,
      "step": 2490
    },
    {
      "epoch": 8.894830659536542,
      "grad_norm": 0.6171875,
      "learning_rate": 7.142160781674645e-06,
      "loss": 0.4035,
      "step": 2495
    },
    {
      "epoch": 8.9126559714795,
      "grad_norm": 0.6328125,
      "learning_rate": 6.9126251355795864e-06,
      "loss": 0.4083,
      "step": 2500
    },
    {
      "epoch": 8.93048128342246,
      "grad_norm": 0.62109375,
      "learning_rate": 6.68670631909335e-06,
      "loss": 0.4059,
      "step": 2505
    },
    {
      "epoch": 8.94830659536542,
      "grad_norm": 0.66796875,
      "learning_rate": 6.464413110096601e-06,
      "loss": 0.4114,
      "step": 2510
    },
    {
      "epoch": 8.966131907308379,
      "grad_norm": 0.625,
      "learning_rate": 6.245754145600091e-06,
      "loss": 0.4132,
      "step": 2515
    },
    {
      "epoch": 8.983957219251337,
      "grad_norm": 0.640625,
      "learning_rate": 6.030737921409169e-06,
      "loss": 0.4106,
      "step": 2520
    },
    {
      "epoch": 8.998217468805704,
      "eval_loss": 3.001107692718506,
      "eval_runtime": 1.0503,
      "eval_samples_per_second": 4.76,
      "eval_steps_per_second": 1.904,
      "step": 2524
    },
    {
      "epoch": 9.001782531194296,
      "grad_norm": 0.578125,
      "learning_rate": 5.8193727917936536e-06,
      "loss": 0.4088,
      "step": 2525
    },
    {
      "epoch": 9.019607843137255,
      "grad_norm": 0.65234375,
      "learning_rate": 5.611666969163243e-06,
      "loss": 0.4124,
      "step": 2530
    },
    {
      "epoch": 9.037433155080214,
      "grad_norm": 0.5859375,
      "learning_rate": 5.4076285237483984e-06,
      "loss": 0.405,
      "step": 2535
    },
    {
      "epoch": 9.055258467023172,
      "grad_norm": 0.62890625,
      "learning_rate": 5.20726538328683e-06,
      "loss": 0.3947,
      "step": 2540
    },
    {
      "epoch": 9.073083778966131,
      "grad_norm": 0.6015625,
      "learning_rate": 5.0105853327154e-06,
      "loss": 0.4077,
      "step": 2545
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 0.58984375,
      "learning_rate": 4.817596013867764e-06,
      "loss": 0.402,
      "step": 2550
    },
    {
      "epoch": 9.10873440285205,
      "grad_norm": 0.58984375,
      "learning_rate": 4.628304925177318e-06,
      "loss": 0.4053,
      "step": 2555
    },
    {
      "epoch": 9.12655971479501,
      "grad_norm": 0.6015625,
      "learning_rate": 4.442719421385922e-06,
      "loss": 0.4063,
      "step": 2560
    },
    {
      "epoch": 9.144385026737968,
      "grad_norm": 0.6171875,
      "learning_rate": 4.260846713258193e-06,
      "loss": 0.4066,
      "step": 2565
    },
    {
      "epoch": 9.162210338680927,
      "grad_norm": 0.6484375,
      "learning_rate": 4.082693867301224e-06,
      "loss": 0.4103,
      "step": 2570
    },
    {
      "epoch": 9.180035650623886,
      "grad_norm": 0.61328125,
      "learning_rate": 3.908267805490051e-06,
      "loss": 0.4052,
      "step": 2575
    },
    {
      "epoch": 9.197860962566844,
      "grad_norm": 0.59765625,
      "learning_rate": 3.7375753049987973e-06,
      "loss": 0.4004,
      "step": 2580
    },
    {
      "epoch": 9.215686274509803,
      "grad_norm": 0.62109375,
      "learning_rate": 3.570622997937234e-06,
      "loss": 0.4049,
      "step": 2585
    },
    {
      "epoch": 9.233511586452764,
      "grad_norm": 0.62109375,
      "learning_rate": 3.40741737109318e-06,
      "loss": 0.4048,
      "step": 2590
    },
    {
      "epoch": 9.251336898395722,
      "grad_norm": 0.6171875,
      "learning_rate": 3.247964765680389e-06,
      "loss": 0.4052,
      "step": 2595
    },
    {
      "epoch": 9.269162210338681,
      "grad_norm": 0.6640625,
      "learning_rate": 3.092271377092215e-06,
      "loss": 0.4098,
      "step": 2600
    },
    {
      "epoch": 9.28698752228164,
      "grad_norm": 0.6171875,
      "learning_rate": 2.9403432546609043e-06,
      "loss": 0.4011,
      "step": 2605
    },
    {
      "epoch": 9.304812834224599,
      "grad_norm": 0.6015625,
      "learning_rate": 2.7921863014225503e-06,
      "loss": 0.4075,
      "step": 2610
    },
    {
      "epoch": 9.322638146167558,
      "grad_norm": 0.59375,
      "learning_rate": 2.647806273887665e-06,
      "loss": 0.4045,
      "step": 2615
    },
    {
      "epoch": 9.340463458110516,
      "grad_norm": 0.609375,
      "learning_rate": 2.5072087818176382e-06,
      "loss": 0.399,
      "step": 2620
    },
    {
      "epoch": 9.358288770053475,
      "grad_norm": 0.625,
      "learning_rate": 2.3703992880066638e-06,
      "loss": 0.4012,
      "step": 2625
    },
    {
      "epoch": 9.376114081996436,
      "grad_norm": 0.62109375,
      "learning_rate": 2.237383108069546e-06,
      "loss": 0.4092,
      "step": 2630
    },
    {
      "epoch": 9.393939393939394,
      "grad_norm": 0.58984375,
      "learning_rate": 2.1081654102351635e-06,
      "loss": 0.3956,
      "step": 2635
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.60546875,
      "learning_rate": 1.9827512151456173e-06,
      "loss": 0.4007,
      "step": 2640
    },
    {
      "epoch": 9.429590017825312,
      "grad_norm": 0.6328125,
      "learning_rate": 1.8611453956612347e-06,
      "loss": 0.4005,
      "step": 2645
    },
    {
      "epoch": 9.44741532976827,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7433526766711728e-06,
      "loss": 0.4045,
      "step": 2650
    },
    {
      "epoch": 9.46524064171123,
      "grad_norm": 0.58984375,
      "learning_rate": 1.6293776349098677e-06,
      "loss": 0.4041,
      "step": 2655
    },
    {
      "epoch": 9.483065953654188,
      "grad_norm": 0.6015625,
      "learning_rate": 1.5192246987791981e-06,
      "loss": 0.403,
      "step": 2660
    },
    {
      "epoch": 9.500891265597147,
      "grad_norm": 0.64453125,
      "learning_rate": 1.4128981481764115e-06,
      "loss": 0.3989,
      "step": 2665
    },
    {
      "epoch": 9.518716577540108,
      "grad_norm": 0.65625,
      "learning_rate": 1.3104021143278911e-06,
      "loss": 0.4132,
      "step": 2670
    },
    {
      "epoch": 9.536541889483066,
      "grad_norm": 0.640625,
      "learning_rate": 1.2117405796285286e-06,
      "loss": 0.401,
      "step": 2675
    },
    {
      "epoch": 9.554367201426025,
      "grad_norm": 0.6640625,
      "learning_rate": 1.1169173774871478e-06,
      "loss": 0.3957,
      "step": 2680
    },
    {
      "epoch": 9.572192513368984,
      "grad_norm": 0.62890625,
      "learning_rate": 1.0259361921774013e-06,
      "loss": 0.4003,
      "step": 2685
    },
    {
      "epoch": 9.590017825311943,
      "grad_norm": 0.640625,
      "learning_rate": 9.388005586947191e-07,
      "loss": 0.4009,
      "step": 2690
    },
    {
      "epoch": 9.607843137254902,
      "grad_norm": 0.66796875,
      "learning_rate": 8.555138626189618e-07,
      "loss": 0.4052,
      "step": 2695
    },
    {
      "epoch": 9.62566844919786,
      "grad_norm": 0.5859375,
      "learning_rate": 7.760793399827937e-07,
      "loss": 0.4009,
      "step": 2700
    },
    {
      "epoch": 9.643493761140821,
      "grad_norm": 0.609375,
      "learning_rate": 7.00500077146038e-07,
      "loss": 0.4017,
      "step": 2705
    },
    {
      "epoch": 9.66131907308378,
      "grad_norm": 0.6171875,
      "learning_rate": 6.287790106757396e-07,
      "loss": 0.3964,
      "step": 2710
    },
    {
      "epoch": 9.679144385026738,
      "grad_norm": 0.62109375,
      "learning_rate": 5.609189272320237e-07,
      "loss": 0.4022,
      "step": 2715
    },
    {
      "epoch": 9.696969696969697,
      "grad_norm": 0.625,
      "learning_rate": 4.969224634598591e-07,
      "loss": 0.404,
      "step": 2720
    },
    {
      "epoch": 9.714795008912656,
      "grad_norm": 0.609375,
      "learning_rate": 4.3679210588661866e-07,
      "loss": 0.409,
      "step": 2725
    },
    {
      "epoch": 9.732620320855615,
      "grad_norm": 0.6328125,
      "learning_rate": 3.805301908254455e-07,
      "loss": 0.4033,
      "step": 2730
    },
    {
      "epoch": 9.750445632798574,
      "grad_norm": 0.62890625,
      "learning_rate": 3.281389042844918e-07,
      "loss": 0.4111,
      "step": 2735
    },
    {
      "epoch": 9.768270944741532,
      "grad_norm": 0.61328125,
      "learning_rate": 2.7962028188198706e-07,
      "loss": 0.3937,
      "step": 2740
    },
    {
      "epoch": 9.786096256684491,
      "grad_norm": 0.58984375,
      "learning_rate": 2.3497620876711257e-07,
      "loss": 0.4115,
      "step": 2745
    },
    {
      "epoch": 9.803921568627452,
      "grad_norm": 0.625,
      "learning_rate": 1.9420841954681525e-07,
      "loss": 0.4025,
      "step": 2750
    },
    {
      "epoch": 9.82174688057041,
      "grad_norm": 0.640625,
      "learning_rate": 1.5731849821833954e-07,
      "loss": 0.4012,
      "step": 2755
    },
    {
      "epoch": 9.83957219251337,
      "grad_norm": 0.63671875,
      "learning_rate": 1.2430787810776555e-07,
      "loss": 0.402,
      "step": 2760
    },
    {
      "epoch": 9.857397504456328,
      "grad_norm": 0.59375,
      "learning_rate": 9.517784181422019e-08,
      "loss": 0.4128,
      "step": 2765
    },
    {
      "epoch": 9.875222816399287,
      "grad_norm": 0.58984375,
      "learning_rate": 6.992952116013918e-08,
      "loss": 0.4023,
      "step": 2770
    },
    {
      "epoch": 9.893048128342246,
      "grad_norm": 0.60546875,
      "learning_rate": 4.856389714723575e-08,
      "loss": 0.4023,
      "step": 2775
    },
    {
      "epoch": 9.910873440285204,
      "grad_norm": 0.61328125,
      "learning_rate": 3.1081799918375454e-08,
      "loss": 0.4105,
      "step": 2780
    },
    {
      "epoch": 9.928698752228165,
      "grad_norm": 0.625,
      "learning_rate": 1.7483908725357545e-08,
      "loss": 0.4085,
      "step": 2785
    },
    {
      "epoch": 9.946524064171124,
      "grad_norm": 0.5859375,
      "learning_rate": 7.770751902513862e-09,
      "loss": 0.3979,
      "step": 2790
    },
    {
      "epoch": 9.964349376114082,
      "grad_norm": 0.65234375,
      "learning_rate": 1.9427068461808083e-09,
      "loss": 0.4084,
      "step": 2795
    },
    {
      "epoch": 9.982174688057041,
      "grad_norm": 0.609375,
      "learning_rate": 0.0,
      "loss": 0.4021,
      "step": 2800
    },
    {
      "epoch": 9.982174688057041,
      "eval_loss": 3.022949695587158,
      "eval_runtime": 1.033,
      "eval_samples_per_second": 4.84,
      "eval_steps_per_second": 1.936,
      "step": 2800
    },
    {
      "epoch": 9.982174688057041,
      "step": 2800,
      "total_flos": 4.2808448846481326e+18,
      "train_loss": 1.3059567311831883,
      "train_runtime": 23112.9867,
      "train_samples_per_second": 1.939,
      "train_steps_per_second": 0.121
    }
  ],
  "logging_steps": 5,
  "max_steps": 2800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.2808448846481326e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}