{
  "best_metric": 0.44935256242752075,
  "best_model_checkpoint": "checkpoints/instrucode/with_input/1b_52k/checkpoint-4600",
  "epoch": 2.958199356913183,
  "eval_steps": 200,
  "global_step": 4600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.6999999999999996e-05,
      "loss": 1.27,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.6999999999999996e-05,
      "loss": 1.2423,
      "step": 20
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.699999999999999e-05,
      "loss": 1.0757,
      "step": 30
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.000117,
      "loss": 0.8238,
      "step": 40
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.000147,
      "loss": 0.7379,
      "step": 50
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00017699999999999997,
      "loss": 0.6487,
      "step": 60
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00020699999999999996,
      "loss": 0.5623,
      "step": 70
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.000237,
      "loss": 0.5591,
      "step": 80
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.000267,
      "loss": 0.5358,
      "step": 90
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029699999999999996,
      "loss": 0.5251,
      "step": 100
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029964820846905535,
      "loss": 0.5371,
      "step": 110
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.000299257328990228,
      "loss": 0.5192,
      "step": 120
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029886644951140066,
      "loss": 0.4993,
      "step": 130
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029847557003257326,
      "loss": 0.4993,
      "step": 140
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002980846905537459,
      "loss": 0.5304,
      "step": 150
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029769381107491857,
      "loss": 0.5094,
      "step": 160
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00029730293159609117,
      "loss": 0.492,
      "step": 170
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002969120521172638,
      "loss": 0.5123,
      "step": 180
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002965211726384364,
      "loss": 0.5018,
      "step": 190
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002961302931596091,
      "loss": 0.4899,
      "step": 200
    },
    {
      "epoch": 0.13,
      "eval_loss": 0.49921754002571106,
      "eval_runtime": 220.0251,
      "eval_samples_per_second": 9.09,
      "eval_steps_per_second": 1.136,
      "step": 200
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029573941368078174,
      "loss": 0.5064,
      "step": 210
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002953485342019544,
      "loss": 0.4656,
      "step": 220
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00029495765472312705,
      "loss": 0.4961,
      "step": 230
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00029456677524429965,
      "loss": 0.4909,
      "step": 240
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002941758957654723,
      "loss": 0.484,
      "step": 250
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002937850162866449,
      "loss": 0.5019,
      "step": 260
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00029339413680781756,
      "loss": 0.4887,
      "step": 270
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002930032573289902,
      "loss": 0.4761,
      "step": 280
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002926123778501628,
      "loss": 0.4874,
      "step": 290
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00029222149837133547,
      "loss": 0.479,
      "step": 300
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002918306188925081,
      "loss": 0.5078,
      "step": 310
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002914397394136808,
      "loss": 0.4875,
      "step": 320
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00029108794788273616,
      "loss": 0.4858,
      "step": 330
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002906970684039088,
      "loss": 0.4633,
      "step": 340
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002903061889250814,
      "loss": 0.4804,
      "step": 350
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00028991530944625407,
      "loss": 0.4981,
      "step": 360
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00028952442996742667,
      "loss": 0.4784,
      "step": 370
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002891335504885993,
      "loss": 0.4923,
      "step": 380
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.000288742671009772,
      "loss": 0.4892,
      "step": 390
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002883517915309446,
      "loss": 0.4902,
      "step": 400
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.48341748118400574,
      "eval_runtime": 220.2998,
      "eval_samples_per_second": 9.079,
      "eval_steps_per_second": 1.135,
      "step": 400
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00028796091205211723,
      "loss": 0.4725,
      "step": 410
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002875700325732899,
      "loss": 0.4873,
      "step": 420
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00028717915309446255,
      "loss": 0.483,
      "step": 430
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00028678827361563515,
      "loss": 0.4587,
      "step": 440
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002863973941368078,
      "loss": 0.4735,
      "step": 450
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002860065146579804,
      "loss": 0.5109,
      "step": 460
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00028561563517915306,
      "loss": 0.478,
      "step": 470
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002852247557003257,
      "loss": 0.4645,
      "step": 480
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00028483387622149837,
      "loss": 0.4932,
      "step": 490
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00028444299674267097,
      "loss": 0.4663,
      "step": 500
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002840521172638436,
      "loss": 0.478,
      "step": 510
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002836612377850163,
      "loss": 0.4567,
      "step": 520
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002832703583061889,
      "loss": 0.4811,
      "step": 530
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00028287947882736154,
      "loss": 0.4678,
      "step": 540
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00028248859934853414,
      "loss": 0.4714,
      "step": 550
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002820977198697068,
      "loss": 0.4714,
      "step": 560
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00028170684039087945,
      "loss": 0.4583,
      "step": 570
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002813159609120521,
      "loss": 0.4493,
      "step": 580
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00028092508143322476,
      "loss": 0.4551,
      "step": 590
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00028053420195439736,
      "loss": 0.5053,
      "step": 600
    },
    {
      "epoch": 0.39,
      "eval_loss": 0.4742714762687683,
      "eval_runtime": 220.0995,
      "eval_samples_per_second": 9.087,
      "eval_steps_per_second": 1.136,
      "step": 600
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00028014332247557,
      "loss": 0.4556,
      "step": 610
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00027975244299674267,
      "loss": 0.4819,
      "step": 620
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00027936156351791527,
      "loss": 0.4729,
      "step": 630
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002789706840390879,
      "loss": 0.4798,
      "step": 640
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002785798045602606,
      "loss": 0.4804,
      "step": 650
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002781889250814332,
      "loss": 0.4661,
      "step": 660
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00027779804560260584,
      "loss": 0.465,
      "step": 670
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002774071661237785,
      "loss": 0.4698,
      "step": 680
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00027701628664495115,
      "loss": 0.4824,
      "step": 690
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00027662540716612375,
      "loss": 0.4575,
      "step": 700
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002762345276872964,
      "loss": 0.4747,
      "step": 710
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.000275843648208469,
      "loss": 0.461,
      "step": 720
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00027545276872964166,
      "loss": 0.4588,
      "step": 730
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002750618892508143,
      "loss": 0.464,
      "step": 740
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00027467100977198697,
      "loss": 0.4829,
      "step": 750
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00027428013029315957,
      "loss": 0.4703,
      "step": 760
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00027388925081433223,
      "loss": 0.4697,
      "step": 770
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002734983713355049,
      "loss": 0.4774,
      "step": 780
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002731074918566775,
      "loss": 0.467,
      "step": 790
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00027271661237785014,
      "loss": 0.4924,
      "step": 800
    },
    {
      "epoch": 0.51,
      "eval_loss": 0.4694923460483551,
      "eval_runtime": 220.2001,
      "eval_samples_per_second": 9.083,
      "eval_steps_per_second": 1.135,
      "step": 800
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00027232573289902274,
      "loss": 0.4788,
      "step": 810
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002719348534201954,
      "loss": 0.4635,
      "step": 820
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00027154397394136805,
      "loss": 0.4816,
      "step": 830
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0002711530944625407,
      "loss": 0.4582,
      "step": 840
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00027076221498371336,
      "loss": 0.4541,
      "step": 850
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00027037133550488596,
      "loss": 0.4748,
      "step": 860
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002699804560260586,
      "loss": 0.4762,
      "step": 870
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002695895765472312,
      "loss": 0.4678,
      "step": 880
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002691986970684039,
      "loss": 0.4581,
      "step": 890
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00026880781758957653,
      "loss": 0.4439,
      "step": 900
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00026841693811074913,
      "loss": 0.4881,
      "step": 910
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002680260586319218,
      "loss": 0.4636,
      "step": 920
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00026763517915309444,
      "loss": 0.4649,
      "step": 930
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002672442996742671,
      "loss": 0.4769,
      "step": 940
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00026685342019543975,
      "loss": 0.4568,
      "step": 950
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00026646254071661235,
      "loss": 0.4673,
      "step": 960
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.000266071661237785,
      "loss": 0.4485,
      "step": 970
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002656807817589576,
      "loss": 0.471,
      "step": 980
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00026528990228013026,
      "loss": 0.4631,
      "step": 990
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002648990228013029,
      "loss": 0.4724,
      "step": 1000
    },
    {
      "epoch": 0.64,
      "eval_loss": 0.46559515595436096,
      "eval_runtime": 220.2191,
      "eval_samples_per_second": 9.082,
      "eval_steps_per_second": 1.135,
      "step": 1000
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002645081433224756,
      "loss": 0.465,
      "step": 1010
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002641172638436482,
      "loss": 0.4504,
      "step": 1020
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00026372638436482083,
      "loss": 0.4583,
      "step": 1030
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002633355048859935,
      "loss": 0.471,
      "step": 1040
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002629446254071661,
      "loss": 0.4779,
      "step": 1050
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00026255374592833874,
      "loss": 0.4543,
      "step": 1060
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00026216286644951134,
      "loss": 0.4629,
      "step": 1070
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.000261771986970684,
      "loss": 0.45,
      "step": 1080
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00026138110749185665,
      "loss": 0.4493,
      "step": 1090
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002609902280130293,
      "loss": 0.4695,
      "step": 1100
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00026059934853420196,
      "loss": 0.4538,
      "step": 1110
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00026020846905537456,
      "loss": 0.4672,
      "step": 1120
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002598175895765472,
      "loss": 0.4802,
      "step": 1130
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002594267100977198,
      "loss": 0.4656,
      "step": 1140
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0002590358306188925,
      "loss": 0.4466,
      "step": 1150
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00025864495114006513,
      "loss": 0.475,
      "step": 1160
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00025825407166123773,
      "loss": 0.4857,
      "step": 1170
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002578631921824104,
      "loss": 0.4479,
      "step": 1180
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00025747231270358304,
      "loss": 0.4488,
      "step": 1190
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002570814332247557,
      "loss": 0.4646,
      "step": 1200
    },
    {
      "epoch": 0.77,
      "eval_loss": 0.46300560235977173,
      "eval_runtime": 220.2387,
      "eval_samples_per_second": 9.081,
      "eval_steps_per_second": 1.135,
      "step": 1200
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00025669055374592835,
      "loss": 0.4722,
      "step": 1210
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00025629967426710095,
      "loss": 0.4661,
      "step": 1220
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002559087947882736,
      "loss": 0.4489,
      "step": 1230
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002555179153094462,
      "loss": 0.4769,
      "step": 1240
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00025512703583061887,
      "loss": 0.4707,
      "step": 1250
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0002547361563517915,
      "loss": 0.4448,
      "step": 1260
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002543452768729641,
      "loss": 0.4647,
      "step": 1270
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002539543973941368,
      "loss": 0.4694,
      "step": 1280
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00025356351791530943,
      "loss": 0.475,
      "step": 1290
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002531726384364821,
      "loss": 0.4589,
      "step": 1300
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002527817589576547,
      "loss": 0.483,
      "step": 1310
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00025239087947882734,
      "loss": 0.449,
      "step": 1320
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00025199999999999995,
      "loss": 0.4533,
      "step": 1330
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002516091205211726,
      "loss": 0.4429,
      "step": 1340
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00025121824104234526,
      "loss": 0.4623,
      "step": 1350
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002508273615635179,
      "loss": 0.4767,
      "step": 1360
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00025043648208469057,
      "loss": 0.4803,
      "step": 1370
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00025004560260586317,
      "loss": 0.4499,
      "step": 1380
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002496547231270358,
      "loss": 0.444,
      "step": 1390
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002492638436482084,
      "loss": 0.477,
      "step": 1400
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.46069058775901794,
      "eval_runtime": 220.249,
      "eval_samples_per_second": 9.081,
      "eval_steps_per_second": 1.135,
      "step": 1400
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002488729641693811,
      "loss": 0.4628,
      "step": 1410
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00024848208469055373,
      "loss": 0.4528,
      "step": 1420
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00024809120521172634,
      "loss": 0.4669,
      "step": 1430
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.000247700325732899,
      "loss": 0.4607,
      "step": 1440
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00024730944625407165,
      "loss": 0.4703,
      "step": 1450
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002469185667752443,
      "loss": 0.4613,
      "step": 1460
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00024652768729641696,
      "loss": 0.4593,
      "step": 1470
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00024613680781758956,
      "loss": 0.4515,
      "step": 1480
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0002457459283387622,
      "loss": 0.4493,
      "step": 1490
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0002453550488599348,
      "loss": 0.4572,
      "step": 1500
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00024496416938110747,
      "loss": 0.4396,
      "step": 1510
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0002445732899022801,
      "loss": 0.4572,
      "step": 1520
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0002441824104234527,
      "loss": 0.4652,
      "step": 1530
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00024379153094462538,
      "loss": 0.4449,
      "step": 1540
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00024340065146579804,
      "loss": 0.4574,
      "step": 1550
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00024300977198697066,
      "loss": 0.4336,
      "step": 1560
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00024261889250814332,
      "loss": 0.4368,
      "step": 1570
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00024222801302931592,
      "loss": 0.45,
      "step": 1580
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00024183713355048858,
      "loss": 0.4272,
      "step": 1590
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0002414462540716612,
      "loss": 0.4475,
      "step": 1600
    },
    {
      "epoch": 1.03,
      "eval_loss": 0.4588911533355713,
      "eval_runtime": 220.3526,
      "eval_samples_per_second": 9.076,
      "eval_steps_per_second": 1.135,
      "step": 1600
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00024105537459283386,
      "loss": 0.4423,
      "step": 1610
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00024066449511400651,
      "loss": 0.442,
      "step": 1620
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00024027361563517911,
      "loss": 0.4412,
      "step": 1630
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00023988273615635177,
      "loss": 0.4403,
      "step": 1640
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0002394918566775244,
      "loss": 0.4397,
      "step": 1650
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00023910097719869705,
      "loss": 0.44,
      "step": 1660
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002387100977198697,
      "loss": 0.4295,
      "step": 1670
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0002383192182410423,
      "loss": 0.4305,
      "step": 1680
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00023792833876221497,
      "loss": 0.433,
      "step": 1690
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002375374592833876,
      "loss": 0.4514,
      "step": 1700
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00023714657980456025,
      "loss": 0.4696,
      "step": 1710
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00023675570032573288,
      "loss": 0.4497,
      "step": 1720
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0002363648208469055,
      "loss": 0.433,
      "step": 1730
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00023597394136807813,
      "loss": 0.4441,
      "step": 1740
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0002355830618892508,
      "loss": 0.4283,
      "step": 1750
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00023519218241042344,
      "loss": 0.46,
      "step": 1760
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00023480130293159607,
      "loss": 0.4386,
      "step": 1770
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00023441042345276873,
      "loss": 0.4548,
      "step": 1780
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00023401954397394133,
      "loss": 0.4546,
      "step": 1790
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00023362866449511398,
      "loss": 0.4482,
      "step": 1800
    },
    {
      "epoch": 1.16,
      "eval_loss": 0.4585164189338684,
      "eval_runtime": 220.4297,
      "eval_samples_per_second": 9.073,
      "eval_steps_per_second": 1.134,
      "step": 1800
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00023323778501628664,
      "loss": 0.4383,
      "step": 1810
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00023284690553745927,
      "loss": 0.4464,
      "step": 1820
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00023245602605863192,
      "loss": 0.4354,
      "step": 1830
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00023206514657980452,
      "loss": 0.4403,
      "step": 1840
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00023167426710097718,
      "loss": 0.4367,
      "step": 1850
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0002312833876221498,
      "loss": 0.4485,
      "step": 1860
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00023089250814332246,
      "loss": 0.4553,
      "step": 1870
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00023050162866449512,
      "loss": 0.4262,
      "step": 1880
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00023011074918566772,
      "loss": 0.4394,
      "step": 1890
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00022971986970684037,
      "loss": 0.4441,
      "step": 1900
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.000229328990228013,
      "loss": 0.4405,
      "step": 1910
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00022893811074918566,
      "loss": 0.434,
      "step": 1920
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002285472312703583,
      "loss": 0.4326,
      "step": 1930
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0002281563517915309,
      "loss": 0.4401,
      "step": 1940
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00022776547231270357,
      "loss": 0.4394,
      "step": 1950
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0002273745928338762,
      "loss": 0.4404,
      "step": 1960
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00022698371335504885,
      "loss": 0.4437,
      "step": 1970
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00022659283387622148,
      "loss": 0.4598,
      "step": 1980
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0002262019543973941,
      "loss": 0.4308,
      "step": 1990
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00022581107491856674,
      "loss": 0.4498,
      "step": 2000
    },
    {
      "epoch": 1.29,
      "eval_loss": 0.456809401512146,
      "eval_runtime": 220.2549,
      "eval_samples_per_second": 9.08,
      "eval_steps_per_second": 1.135,
      "step": 2000
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0002254201954397394,
      "loss": 0.4555,
      "step": 2010
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00022502931596091205,
      "loss": 0.4397,
      "step": 2020
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00022463843648208467,
      "loss": 0.4459,
      "step": 2030
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002242475570032573,
      "loss": 0.4373,
      "step": 2040
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00022385667752442993,
      "loss": 0.4346,
      "step": 2050
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00022346579804560259,
      "loss": 0.4224,
      "step": 2060
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00022307491856677524,
      "loss": 0.4539,
      "step": 2070
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00022268403908794787,
      "loss": 0.4431,
      "step": 2080
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0002222931596091205,
      "loss": 0.4436,
      "step": 2090
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00022190228013029313,
      "loss": 0.4453,
      "step": 2100
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00022151140065146578,
      "loss": 0.4268,
      "step": 2110
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0002211205211726384,
      "loss": 0.4121,
      "step": 2120
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00022072964169381106,
      "loss": 0.4614,
      "step": 2130
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00022033876221498372,
      "loss": 0.452,
      "step": 2140
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00021994788273615632,
      "loss": 0.4472,
      "step": 2150
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00021955700325732898,
      "loss": 0.4318,
      "step": 2160
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0002191661237785016,
      "loss": 0.4476,
      "step": 2170
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00021877524429967426,
      "loss": 0.4378,
      "step": 2180
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00021838436482084691,
      "loss": 0.425,
      "step": 2190
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00021799348534201952,
      "loss": 0.4392,
      "step": 2200
    },
    {
      "epoch": 1.41,
      "eval_loss": 0.45531219244003296,
      "eval_runtime": 220.3519,
      "eval_samples_per_second": 9.076,
      "eval_steps_per_second": 1.135,
      "step": 2200
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00021760260586319217,
      "loss": 0.4422,
      "step": 2210
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002172117263843648,
      "loss": 0.4403,
      "step": 2220
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00021682084690553745,
      "loss": 0.45,
      "step": 2230
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00021642996742671008,
      "loss": 0.4282,
      "step": 2240
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0002160390879478827,
      "loss": 0.4319,
      "step": 2250
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00021564820846905534,
      "loss": 0.4438,
      "step": 2260
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.000215257328990228,
      "loss": 0.4509,
      "step": 2270
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00021486644951140065,
      "loss": 0.4454,
      "step": 2280
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00021447557003257328,
      "loss": 0.4176,
      "step": 2290
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0002140846905537459,
      "loss": 0.4554,
      "step": 2300
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00021369381107491853,
      "loss": 0.4388,
      "step": 2310
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0002133029315960912,
      "loss": 0.456,
      "step": 2320
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00021291205211726384,
      "loss": 0.4257,
      "step": 2330
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00021252117263843647,
      "loss": 0.4428,
      "step": 2340
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002121302931596091,
      "loss": 0.4266,
      "step": 2350
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00021173941368078173,
      "loss": 0.4591,
      "step": 2360
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00021134853420195438,
      "loss": 0.453,
      "step": 2370
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.000210957654723127,
      "loss": 0.4414,
      "step": 2380
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00021056677524429967,
      "loss": 0.4619,
      "step": 2390
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00021017589576547227,
      "loss": 0.4285,
      "step": 2400
    },
    {
      "epoch": 1.54,
      "eval_loss": 0.45426830649375916,
      "eval_runtime": 220.2316,
      "eval_samples_per_second": 9.081,
      "eval_steps_per_second": 1.135,
      "step": 2400
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00020978501628664492,
      "loss": 0.4476,
      "step": 2410
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00020939413680781758,
      "loss": 0.4361,
      "step": 2420
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0002090032573289902,
      "loss": 0.452,
      "step": 2430
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00020861237785016286,
      "loss": 0.4358,
      "step": 2440
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00020822149837133546,
      "loss": 0.447,
      "step": 2450
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00020783061889250812,
      "loss": 0.4592,
      "step": 2460
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00020743973941368077,
      "loss": 0.4529,
      "step": 2470
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0002070488599348534,
      "loss": 0.4545,
      "step": 2480
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00020665798045602606,
      "loss": 0.4303,
      "step": 2490
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00020626710097719866,
      "loss": 0.4489,
      "step": 2500
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0002058762214983713,
      "loss": 0.4306,
      "step": 2510
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00020548534201954394,
      "loss": 0.4441,
      "step": 2520
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0002050944625407166,
      "loss": 0.4132,
      "step": 2530
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00020470358306188925,
      "loss": 0.428,
      "step": 2540
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00020431270358306188,
      "loss": 0.4242,
      "step": 2550
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002039218241042345,
      "loss": 0.4372,
      "step": 2560
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00020353094462540714,
      "loss": 0.4339,
      "step": 2570
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0002031400651465798,
      "loss": 0.434,
      "step": 2580
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00020274918566775245,
      "loss": 0.4282,
      "step": 2590
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00020235830618892507,
      "loss": 0.4352,
      "step": 2600
    },
    {
      "epoch": 1.67,
      "eval_loss": 0.4528008699417114,
      "eval_runtime": 220.0694,
      "eval_samples_per_second": 9.088,
      "eval_steps_per_second": 1.136,
      "step": 2600
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002019674267100977,
      "loss": 0.4478,
      "step": 2610
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00020157654723127033,
      "loss": 0.4317,
      "step": 2620
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00020118566775244299,
      "loss": 0.4447,
      "step": 2630
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00020079478827361561,
      "loss": 0.4454,
      "step": 2640
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00020040390879478827,
      "loss": 0.4422,
      "step": 2650
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00020001302931596087,
      "loss": 0.4504,
      "step": 2660
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00019962214983713353,
      "loss": 0.4322,
      "step": 2670
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00019923127035830618,
      "loss": 0.4387,
      "step": 2680
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0001988403908794788,
      "loss": 0.4398,
      "step": 2690
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00019844951140065146,
      "loss": 0.4384,
      "step": 2700
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00019805863192182407,
      "loss": 0.4412,
      "step": 2710
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00019766775244299672,
      "loss": 0.4535,
      "step": 2720
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00019727687296416938,
      "loss": 0.437,
      "step": 2730
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.000196885993485342,
      "loss": 0.4403,
      "step": 2740
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00019649511400651466,
      "loss": 0.4343,
      "step": 2750
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00019610423452768726,
      "loss": 0.4281,
      "step": 2760
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00019571335504885992,
      "loss": 0.4364,
      "step": 2770
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00019532247557003254,
      "loss": 0.4473,
      "step": 2780
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0001949315960912052,
      "loss": 0.4421,
      "step": 2790
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00019454071661237785,
      "loss": 0.4427,
      "step": 2800
    },
    {
      "epoch": 1.8,
      "eval_loss": 0.4516027867794037,
      "eval_runtime": 220.0573,
      "eval_samples_per_second": 9.089,
      "eval_steps_per_second": 1.136,
      "step": 2800
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00019414983713355046,
      "loss": 0.449,
      "step": 2810
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0001937589576547231,
      "loss": 0.433,
      "step": 2820
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00019336807817589574,
      "loss": 0.4368,
      "step": 2830
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0001929771986970684,
      "loss": 0.4432,
      "step": 2840
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00019258631921824105,
      "loss": 0.4394,
      "step": 2850
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00019219543973941365,
      "loss": 0.4421,
      "step": 2860
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0001918045602605863,
      "loss": 0.4526,
      "step": 2870
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00019141368078175893,
      "loss": 0.4322,
      "step": 2880
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0001910228013029316,
      "loss": 0.4376,
      "step": 2890
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00019063192182410422,
      "loss": 0.4496,
      "step": 2900
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00019024104234527687,
      "loss": 0.4429,
      "step": 2910
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00018985016286644947,
      "loss": 0.4269,
      "step": 2920
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00018945928338762213,
      "loss": 0.4409,
      "step": 2930
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00018906840390879478,
      "loss": 0.4328,
      "step": 2940
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0001886775244299674,
      "loss": 0.4438,
      "step": 2950
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00018828664495114007,
      "loss": 0.4238,
      "step": 2960
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00018789576547231267,
      "loss": 0.4288,
      "step": 2970
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00018750488599348532,
      "loss": 0.4356,
      "step": 2980
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00018711400651465798,
      "loss": 0.453,
      "step": 2990
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0001867231270358306,
      "loss": 0.4574,
      "step": 3000
    },
    {
      "epoch": 1.93,
      "eval_loss": 0.4507245421409607,
      "eval_runtime": 219.8338,
      "eval_samples_per_second": 9.098,
      "eval_steps_per_second": 1.137,
      "step": 3000
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00018633224755700326,
      "loss": 0.4401,
      "step": 3010
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00018594136807817586,
      "loss": 0.4562,
      "step": 3020
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00018555048859934852,
      "loss": 0.4458,
      "step": 3030
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00018515960912052115,
      "loss": 0.4391,
      "step": 3040
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0001847687296416938,
      "loss": 0.4328,
      "step": 3050
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00018437785016286646,
      "loss": 0.4397,
      "step": 3060
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00018398697068403906,
      "loss": 0.4199,
      "step": 3070
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0001835960912052117,
      "loss": 0.439,
      "step": 3080
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00018320521172638434,
      "loss": 0.4354,
      "step": 3090
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.000182814332247557,
      "loss": 0.4265,
      "step": 3100
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.00018242345276872965,
      "loss": 0.4338,
      "step": 3110
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.00018203257328990225,
      "loss": 0.4107,
      "step": 3120
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0001816416938110749,
      "loss": 0.4254,
      "step": 3130
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.00018125081433224754,
      "loss": 0.4312,
      "step": 3140
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0001808599348534202,
      "loss": 0.4049,
      "step": 3150
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.00018046905537459282,
      "loss": 0.4226,
      "step": 3160
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.00018007817589576545,
      "loss": 0.4172,
      "step": 3170
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.00017968729641693808,
      "loss": 0.4047,
      "step": 3180
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.00017929641693811073,
      "loss": 0.4269,
      "step": 3190
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.00017890553745928339,
      "loss": 0.4027,
      "step": 3200
    },
    {
      "epoch": 2.06,
      "eval_loss": 0.45280253887176514,
      "eval_runtime": 219.8269,
      "eval_samples_per_second": 9.098,
      "eval_steps_per_second": 1.137,
      "step": 3200
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.00017851465798045601,
      "loss": 0.4149,
      "step": 3210
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.00017812377850162864,
      "loss": 0.4317,
      "step": 3220
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.00017773289902280127,
      "loss": 0.4243,
      "step": 3230
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.00017734201954397393,
      "loss": 0.4036,
      "step": 3240
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.00017695114006514658,
      "loss": 0.4113,
      "step": 3250
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0001765602605863192,
      "loss": 0.4206,
      "step": 3260
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.00017616938110749184,
      "loss": 0.4194,
      "step": 3270
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.00017577850162866447,
      "loss": 0.4203,
      "step": 3280
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.00017538762214983712,
      "loss": 0.4149,
      "step": 3290
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.00017499674267100975,
      "loss": 0.3972,
      "step": 3300
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0001746058631921824,
      "loss": 0.3971,
      "step": 3310
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.00017421498371335506,
      "loss": 0.4309,
      "step": 3320
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.00017382410423452766,
      "loss": 0.4216,
      "step": 3330
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.00017343322475570032,
      "loss": 0.4291,
      "step": 3340
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.00017304234527687294,
      "loss": 0.4186,
      "step": 3350
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0001726514657980456,
      "loss": 0.4258,
      "step": 3360
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.00017226058631921825,
      "loss": 0.4114,
      "step": 3370
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.00017186970684039086,
      "loss": 0.4166,
      "step": 3380
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0001714788273615635,
      "loss": 0.4157,
      "step": 3390
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.00017108794788273614,
      "loss": 0.4302,
      "step": 3400
    },
    {
      "epoch": 2.19,
      "eval_loss": 0.4515763223171234,
      "eval_runtime": 219.868,
      "eval_samples_per_second": 9.096,
      "eval_steps_per_second": 1.137,
      "step": 3400
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0001706970684039088,
      "loss": 0.4098,
      "step": 3410
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.00017030618892508142,
      "loss": 0.4326,
      "step": 3420
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.00016991530944625405,
      "loss": 0.4303,
      "step": 3430
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.00016952442996742668,
      "loss": 0.4325,
      "step": 3440
    },
    {
      "epoch": 2.22,
      "learning_rate": 0.00016913355048859933,
      "loss": 0.4164,
      "step": 3450
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.000168742671009772,
      "loss": 0.4195,
      "step": 3460
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.00016835179153094462,
      "loss": 0.408,
      "step": 3470
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.00016796091205211725,
      "loss": 0.4185,
      "step": 3480
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.00016757003257328987,
      "loss": 0.4155,
      "step": 3490
    },
    {
      "epoch": 2.25,
      "learning_rate": 0.00016717915309446253,
      "loss": 0.4004,
      "step": 3500
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.00016678827361563518,
      "loss": 0.4078,
      "step": 3510
    },
    {
      "epoch": 2.26,
      "learning_rate": 0.0001663973941368078,
      "loss": 0.411,
      "step": 3520
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.00016600651465798044,
      "loss": 0.4242,
      "step": 3530
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.00016561563517915307,
      "loss": 0.4092,
      "step": 3540
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.00016522475570032572,
      "loss": 0.4193,
      "step": 3550
    },
    {
      "epoch": 2.29,
      "learning_rate": 0.00016483387622149835,
      "loss": 0.419,
      "step": 3560
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.000164442996742671,
      "loss": 0.4173,
      "step": 3570
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.0001640521172638436,
      "loss": 0.414,
      "step": 3580
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.00016366123778501626,
      "loss": 0.4115,
      "step": 3590
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.00016327035830618892,
      "loss": 0.4227,
      "step": 3600
    },
    {
      "epoch": 2.32,
      "eval_loss": 0.45233967900276184,
      "eval_runtime": 220.0403,
      "eval_samples_per_second": 9.089,
      "eval_steps_per_second": 1.136,
      "step": 3600
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.00016287947882736155,
      "loss": 0.4238,
      "step": 3610
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0001624885993485342,
      "loss": 0.4066,
      "step": 3620
    },
    {
      "epoch": 2.33,
      "learning_rate": 0.0001620977198697068,
      "loss": 0.4226,
      "step": 3630
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.00016170684039087946,
      "loss": 0.429,
      "step": 3640
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0001613159609120521,
      "loss": 0.4181,
      "step": 3650
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.00016092508143322474,
      "loss": 0.4276,
      "step": 3660
    },
    {
      "epoch": 2.36,
      "learning_rate": 0.0001605342019543974,
      "loss": 0.4169,
      "step": 3670
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.00016014332247557003,
      "loss": 0.4223,
      "step": 3680
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.00015975244299674265,
      "loss": 0.4136,
      "step": 3690
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.00015936156351791528,
      "loss": 0.4137,
      "step": 3700
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.00015897068403908794,
      "loss": 0.4247,
      "step": 3710
    },
    {
      "epoch": 2.39,
      "learning_rate": 0.0001585798045602606,
      "loss": 0.4052,
      "step": 3720
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.00015818892508143322,
      "loss": 0.4308,
      "step": 3730
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.00015779804560260585,
      "loss": 0.4115,
      "step": 3740
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.00015740716612377848,
      "loss": 0.423,
      "step": 3750
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.00015701628664495113,
      "loss": 0.4181,
      "step": 3760
    },
    {
      "epoch": 2.42,
      "learning_rate": 0.00015662540716612376,
      "loss": 0.4069,
      "step": 3770
    },
    {
      "epoch": 2.43,
      "learning_rate": 0.00015623452768729641,
      "loss": 0.4112,
      "step": 3780
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.00015584364820846904,
      "loss": 0.4221,
      "step": 3790
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.00015545276872964167,
      "loss": 0.4318,
      "step": 3800
    },
    {
      "epoch": 2.44,
      "eval_loss": 0.4517545998096466,
      "eval_runtime": 220.1454,
      "eval_samples_per_second": 9.085,
      "eval_steps_per_second": 1.136,
      "step": 3800
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.00015506188925081433,
      "loss": 0.419,
      "step": 3810
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.00015467100977198695,
      "loss": 0.4357,
      "step": 3820
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.0001542801302931596,
      "loss": 0.4197,
      "step": 3830
    },
    {
      "epoch": 2.47,
      "learning_rate": 0.0001538892508143322,
      "loss": 0.4064,
      "step": 3840
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.00015349837133550487,
      "loss": 0.421,
      "step": 3850
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.00015310749185667752,
      "loss": 0.416,
      "step": 3860
    },
    {
      "epoch": 2.49,
      "learning_rate": 0.00015271661237785015,
      "loss": 0.3995,
      "step": 3870
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0001523257328990228,
      "loss": 0.4222,
      "step": 3880
    },
    {
      "epoch": 2.5,
      "learning_rate": 0.0001519348534201954,
      "loss": 0.3989,
      "step": 3890
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.00015154397394136806,
      "loss": 0.3986,
      "step": 3900
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0001511530944625407,
      "loss": 0.4256,
      "step": 3910
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.00015076221498371334,
      "loss": 0.4289,
      "step": 3920
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.000150371335504886,
      "loss": 0.4138,
      "step": 3930
    },
    {
      "epoch": 2.53,
      "learning_rate": 0.00014998045602605863,
      "loss": 0.4125,
      "step": 3940
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.00014958957654723126,
      "loss": 0.4236,
      "step": 3950
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.00014919869706840388,
      "loss": 0.4189,
      "step": 3960
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.00014880781758957654,
      "loss": 0.4155,
      "step": 3970
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.00014841693811074917,
      "loss": 0.4144,
      "step": 3980
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.00014802605863192182,
      "loss": 0.4295,
      "step": 3990
    },
    {
      "epoch": 2.57,
      "learning_rate": 0.00014763517915309445,
      "loss": 0.3985,
      "step": 4000
    },
    {
      "epoch": 2.57,
      "eval_loss": 0.4515686333179474,
      "eval_runtime": 220.1634,
      "eval_samples_per_second": 9.084,
      "eval_steps_per_second": 1.136,
      "step": 4000
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.00014724429967426708,
      "loss": 0.418,
      "step": 4010
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.00014685342019543973,
      "loss": 0.4267,
      "step": 4020
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.00014646254071661236,
      "loss": 0.4231,
      "step": 4030
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.000146071661237785,
      "loss": 0.4153,
      "step": 4040
    },
    {
      "epoch": 2.6,
      "learning_rate": 0.00014568078175895765,
      "loss": 0.4228,
      "step": 4050
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.00014528990228013027,
      "loss": 0.4041,
      "step": 4060
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.00014489902280130293,
      "loss": 0.4361,
      "step": 4070
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.00014450814332247556,
      "loss": 0.4158,
      "step": 4080
    },
    {
      "epoch": 2.63,
      "learning_rate": 0.00014411726384364819,
      "loss": 0.4235,
      "step": 4090
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.00014372638436482084,
      "loss": 0.4157,
      "step": 4100
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.00014333550488599347,
      "loss": 0.4233,
      "step": 4110
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.00014294462540716612,
      "loss": 0.4138,
      "step": 4120
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.00014255374592833875,
      "loss": 0.422,
      "step": 4130
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.00014216286644951138,
      "loss": 0.4234,
      "step": 4140
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.00014177198697068404,
      "loss": 0.4224,
      "step": 4150
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.00014138110749185666,
      "loss": 0.4256,
      "step": 4160
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0001409902280130293,
      "loss": 0.4216,
      "step": 4170
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.00014059934853420195,
      "loss": 0.4139,
      "step": 4180
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.00014020846905537458,
      "loss": 0.4112,
      "step": 4190
    },
    {
      "epoch": 2.7,
      "learning_rate": 0.00013981758957654723,
      "loss": 0.4203,
      "step": 4200
    },
    {
      "epoch": 2.7,
      "eval_loss": 0.45124533772468567,
      "eval_runtime": 220.1249,
      "eval_samples_per_second": 9.086,
      "eval_steps_per_second": 1.136,
      "step": 4200
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.00013942671009771986,
      "loss": 0.4072,
      "step": 4210
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0001390358306188925,
      "loss": 0.4134,
      "step": 4220
    },
    {
      "epoch": 2.72,
      "learning_rate": 0.00013864495114006514,
      "loss": 0.4175,
      "step": 4230
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.00013825407166123777,
      "loss": 0.4215,
      "step": 4240
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.00013786319218241043,
      "loss": 0.4226,
      "step": 4250
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.00013747231270358305,
      "loss": 0.4003,
      "step": 4260
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.00013708143322475568,
      "loss": 0.4198,
      "step": 4270
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.00013669055374592834,
      "loss": 0.4225,
      "step": 4280
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.00013629967426710097,
      "loss": 0.418,
      "step": 4290
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.0001359087947882736,
      "loss": 0.4234,
      "step": 4300
    },
    {
      "epoch": 2.77,
      "learning_rate": 0.00013551791530944622,
      "loss": 0.421,
      "step": 4310
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.00013512703583061888,
      "loss": 0.4162,
      "step": 4320
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.00013473615635179153,
      "loss": 0.4103,
      "step": 4330
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.00013434527687296416,
      "loss": 0.4157,
      "step": 4340
    },
    {
      "epoch": 2.8,
      "learning_rate": 0.0001339543973941368,
      "loss": 0.4144,
      "step": 4350
    },
    {
      "epoch": 2.8,
      "learning_rate": 0.00013356351791530944,
      "loss": 0.4096,
      "step": 4360
    },
    {
      "epoch": 2.81,
      "learning_rate": 0.00013317263843648207,
      "loss": 0.4134,
      "step": 4370
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.00013278175895765473,
      "loss": 0.42,
      "step": 4380
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.00013239087947882735,
      "loss": 0.4187,
      "step": 4390
    },
    {
      "epoch": 2.83,
      "learning_rate": 0.00013199999999999998,
      "loss": 0.4194,
      "step": 4400
    },
    {
      "epoch": 2.83,
      "eval_loss": 0.4500805735588074,
      "eval_runtime": 220.1289,
      "eval_samples_per_second": 9.086,
      "eval_steps_per_second": 1.136,
      "step": 4400
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.00013160912052117264,
      "loss": 0.4193,
      "step": 4410
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.00013121824104234527,
      "loss": 0.4249,
      "step": 4420
    },
    {
      "epoch": 2.85,
      "learning_rate": 0.0001308273615635179,
      "loss": 0.4187,
      "step": 4430
    },
    {
      "epoch": 2.86,
      "learning_rate": 0.00013043648208469052,
      "loss": 0.4215,
      "step": 4440
    },
    {
      "epoch": 2.86,
      "learning_rate": 0.00013004560260586318,
      "loss": 0.4075,
      "step": 4450
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.00012965472312703583,
      "loss": 0.4165,
      "step": 4460
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.00012926384364820846,
      "loss": 0.4197,
      "step": 4470
    },
    {
      "epoch": 2.88,
      "learning_rate": 0.0001288729641693811,
      "loss": 0.4392,
      "step": 4480
    },
    {
      "epoch": 2.89,
      "learning_rate": 0.00012848208469055372,
      "loss": 0.4064,
      "step": 4490
    },
    {
      "epoch": 2.89,
      "learning_rate": 0.00012809120521172637,
      "loss": 0.4176,
      "step": 4500
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.00012770032573289903,
      "loss": 0.4116,
      "step": 4510
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.00012730944625407166,
      "loss": 0.4177,
      "step": 4520
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.00012691856677524428,
      "loss": 0.4109,
      "step": 4530
    },
    {
      "epoch": 2.92,
      "learning_rate": 0.00012652768729641694,
      "loss": 0.4072,
      "step": 4540
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.00012613680781758957,
      "loss": 0.435,
      "step": 4550
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.0001257459283387622,
      "loss": 0.401,
      "step": 4560
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.00012535504885993482,
      "loss": 0.4175,
      "step": 4570
    },
    {
      "epoch": 2.95,
      "learning_rate": 0.00012496416938110748,
      "loss": 0.4245,
      "step": 4580
    },
    {
      "epoch": 2.95,
      "learning_rate": 0.00012457328990228013,
      "loss": 0.415,
      "step": 4590
    },
    {
      "epoch": 2.96,
      "learning_rate": 0.00012418241042345276,
      "loss": 0.4247,
      "step": 4600
    },
    {
      "epoch": 2.96,
      "eval_loss": 0.44935256242752075,
      "eval_runtime": 220.1783,
      "eval_samples_per_second": 9.084,
      "eval_steps_per_second": 1.135,
      "step": 4600
    }
  ],
  "logging_steps": 10,
  "max_steps": 7775,
  "num_train_epochs": 5,
  "save_steps": 200,
  "total_flos": 2.3487809765824266e+18,
  "trial_name": null,
  "trial_params": null
}