{
  "best_metric": 0.4666987955570221,
  "best_model_checkpoint": "checkpoints/1b_52k/checkpoint-6200",
  "epoch": 1.9935691318327975,
  "eval_steps": 200,
  "global_step": 6200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 2.6999999999999996e-05,
      "loss": 1.2981,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.1e-05,
      "loss": 1.2597,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 8.1e-05,
      "loss": 1.1419,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00011099999999999999,
      "loss": 0.8779,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00014099999999999998,
      "loss": 0.7661,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00017099999999999998,
      "loss": 0.7354,
      "step": 60
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.000201,
      "loss": 0.5858,
      "step": 70
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00023099999999999998,
      "loss": 0.5491,
      "step": 80
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.000261,
      "loss": 0.5445,
      "step": 90
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00029099999999999997,
      "loss": 0.5566,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002998640776699029,
      "loss": 0.5398,
      "step": 110
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002996699029126213,
      "loss": 0.528,
      "step": 120
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002994757281553398,
      "loss": 0.5818,
      "step": 130
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029928155339805826,
      "loss": 0.5048,
      "step": 140
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002990873786407767,
      "loss": 0.5234,
      "step": 150
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00029889320388349516,
      "loss": 0.5353,
      "step": 160
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002986990291262136,
      "loss": 0.5341,
      "step": 170
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.000298504854368932,
      "loss": 0.5185,
      "step": 180
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00029831067961165047,
      "loss": 0.5046,
      "step": 190
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002981165048543689,
      "loss": 0.4929,
      "step": 200
    },
    {
      "epoch": 0.06,
      "eval_loss": 0.5254431366920471,
      "eval_runtime": 503.7885,
      "eval_samples_per_second": 3.97,
      "eval_steps_per_second": 0.496,
      "step": 200
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029792233009708736,
      "loss": 0.5269,
      "step": 210
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002977281553398058,
      "loss": 0.5073,
      "step": 220
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00029753398058252425,
      "loss": 0.5534,
      "step": 230
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029733980582524267,
      "loss": 0.5466,
      "step": 240
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00029714563106796114,
      "loss": 0.5065,
      "step": 250
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002969514563106796,
      "loss": 0.5038,
      "step": 260
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00029675728155339803,
      "loss": 0.5398,
      "step": 270
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002965631067961165,
      "loss": 0.5068,
      "step": 280
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002963689320388349,
      "loss": 0.496,
      "step": 290
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002961747572815534,
      "loss": 0.5276,
      "step": 300
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002959805825242718,
      "loss": 0.5269,
      "step": 310
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00029578640776699023,
      "loss": 0.504,
      "step": 320
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002955922330097087,
      "loss": 0.5183,
      "step": 330
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002953980582524271,
      "loss": 0.4911,
      "step": 340
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002952038834951456,
      "loss": 0.4937,
      "step": 350
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029500970873786407,
      "loss": 0.4942,
      "step": 360
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002948155339805825,
      "loss": 0.4763,
      "step": 370
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00029462135922330096,
      "loss": 0.5261,
      "step": 380
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002944271844660194,
      "loss": 0.5031,
      "step": 390
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00029423300970873786,
      "loss": 0.5103,
      "step": 400
    },
    {
      "epoch": 0.13,
      "eval_loss": 0.5078505873680115,
      "eval_runtime": 523.4102,
      "eval_samples_per_second": 3.821,
      "eval_steps_per_second": 0.478,
      "step": 400
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002940388349514563,
      "loss": 0.5134,
      "step": 410
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029384466019417475,
      "loss": 0.5013,
      "step": 420
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029365048543689317,
      "loss": 0.5222,
      "step": 430
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029345631067961164,
      "loss": 0.5072,
      "step": 440
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00029326213592233006,
      "loss": 0.4943,
      "step": 450
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00029306796116504853,
      "loss": 0.4941,
      "step": 460
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00029287378640776695,
      "loss": 0.4767,
      "step": 470
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002926796116504854,
      "loss": 0.5166,
      "step": 480
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00029248543689320384,
      "loss": 0.487,
      "step": 490
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002922912621359223,
      "loss": 0.5165,
      "step": 500
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00029209708737864073,
      "loss": 0.4742,
      "step": 510
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002919029126213592,
      "loss": 0.496,
      "step": 520
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002917087378640776,
      "loss": 0.4934,
      "step": 530
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002915145631067961,
      "loss": 0.4915,
      "step": 540
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00029132038834951457,
      "loss": 0.4783,
      "step": 550
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.000291126213592233,
      "loss": 0.4869,
      "step": 560
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00029093203883495146,
      "loss": 0.4802,
      "step": 570
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002907378640776699,
      "loss": 0.4953,
      "step": 580
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00029054368932038835,
      "loss": 0.4859,
      "step": 590
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00029034951456310677,
      "loss": 0.5093,
      "step": 600
    },
    {
      "epoch": 0.19,
      "eval_loss": 0.4987016022205353,
      "eval_runtime": 527.0047,
      "eval_samples_per_second": 3.795,
      "eval_steps_per_second": 0.474,
      "step": 600
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002901553398058252,
      "loss": 0.4876,
      "step": 610
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00028996116504854366,
      "loss": 0.5086,
      "step": 620
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002897669902912621,
      "loss": 0.5089,
      "step": 630
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00028957281553398056,
      "loss": 0.5143,
      "step": 640
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.000289378640776699,
      "loss": 0.5274,
      "step": 650
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00028918446601941745,
      "loss": 0.4704,
      "step": 660
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002889902912621359,
      "loss": 0.4951,
      "step": 670
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00028879611650485434,
      "loss": 0.4949,
      "step": 680
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002886019417475728,
      "loss": 0.465,
      "step": 690
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00028840776699029123,
      "loss": 0.4942,
      "step": 700
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002882135922330097,
      "loss": 0.4861,
      "step": 710
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002880194174757281,
      "loss": 0.4591,
      "step": 720
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002878252427184466,
      "loss": 0.4983,
      "step": 730
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.000287631067961165,
      "loss": 0.4883,
      "step": 740
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00028743689320388343,
      "loss": 0.485,
      "step": 750
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002872427184466019,
      "loss": 0.4901,
      "step": 760
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002870485436893204,
      "loss": 0.5018,
      "step": 770
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002868543689320388,
      "loss": 0.4737,
      "step": 780
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00028666019417475727,
      "loss": 0.4885,
      "step": 790
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002864660194174757,
      "loss": 0.4958,
      "step": 800
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.4947653114795685,
      "eval_runtime": 537.1211,
      "eval_samples_per_second": 3.724,
      "eval_steps_per_second": 0.465,
      "step": 800
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00028627184466019416,
      "loss": 0.4811,
      "step": 810
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002860776699029126,
      "loss": 0.5244,
      "step": 820
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00028588349514563105,
      "loss": 0.4735,
      "step": 830
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00028568932038834947,
      "loss": 0.5116,
      "step": 840
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00028549514563106794,
      "loss": 0.5015,
      "step": 850
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002853009708737864,
      "loss": 0.4914,
      "step": 860
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00028510679611650484,
      "loss": 0.5336,
      "step": 870
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002849126213592233,
      "loss": 0.4975,
      "step": 880
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00028471844660194173,
      "loss": 0.4693,
      "step": 890
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00028452427184466015,
      "loss": 0.4915,
      "step": 900
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002843300970873786,
      "loss": 0.4595,
      "step": 910
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00028413592233009704,
      "loss": 0.5101,
      "step": 920
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002839417475728155,
      "loss": 0.4835,
      "step": 930
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00028374757281553393,
      "loss": 0.4749,
      "step": 940
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002835533980582524,
      "loss": 0.4943,
      "step": 950
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002833592233009709,
      "loss": 0.5031,
      "step": 960
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002831650485436893,
      "loss": 0.472,
      "step": 970
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00028297087378640777,
      "loss": 0.5117,
      "step": 980
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002827766990291262,
      "loss": 0.4917,
      "step": 990
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00028258252427184466,
      "loss": 0.4764,
      "step": 1000
    },
    {
      "epoch": 0.32,
      "eval_loss": 0.49110785126686096,
      "eval_runtime": 534.8721,
      "eval_samples_per_second": 3.739,
      "eval_steps_per_second": 0.467,
      "step": 1000
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002823883495145631,
      "loss": 0.4696,
      "step": 1010
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00028219417475728155,
      "loss": 0.4927,
      "step": 1020
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00028199999999999997,
      "loss": 0.4699,
      "step": 1030
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002818058252427184,
      "loss": 0.4746,
      "step": 1040
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00028161165048543686,
      "loss": 0.4837,
      "step": 1050
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002814174757281553,
      "loss": 0.4655,
      "step": 1060
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00028122330097087375,
      "loss": 0.4977,
      "step": 1070
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002810291262135922,
      "loss": 0.4806,
      "step": 1080
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00028083495145631064,
      "loss": 0.4728,
      "step": 1090
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002806407766990291,
      "loss": 0.4704,
      "step": 1100
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00028044660194174754,
      "loss": 0.4583,
      "step": 1110
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.000280252427184466,
      "loss": 0.4733,
      "step": 1120
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00028005825242718443,
      "loss": 0.4797,
      "step": 1130
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002798640776699029,
      "loss": 0.4971,
      "step": 1140
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002796699029126213,
      "loss": 0.4556,
      "step": 1150
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002794757281553398,
      "loss": 0.4836,
      "step": 1160
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00027928155339805827,
      "loss": 0.4931,
      "step": 1170
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002790873786407767,
      "loss": 0.4741,
      "step": 1180
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002788932038834951,
      "loss": 0.4926,
      "step": 1190
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002786990291262136,
      "loss": 0.4791,
      "step": 1200
    },
    {
      "epoch": 0.39,
      "eval_loss": 0.48850855231285095,
      "eval_runtime": 532.9556,
      "eval_samples_per_second": 3.753,
      "eval_steps_per_second": 0.469,
      "step": 1200
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.000278504854368932,
      "loss": 0.4925,
      "step": 1210
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00027831067961165047,
      "loss": 0.4947,
      "step": 1220
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002781165048543689,
      "loss": 0.4804,
      "step": 1230
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00027792233009708736,
      "loss": 0.5022,
      "step": 1240
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002777281553398058,
      "loss": 0.5123,
      "step": 1250
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00027753398058252425,
      "loss": 0.5035,
      "step": 1260
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002773398058252427,
      "loss": 0.4622,
      "step": 1270
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00027714563106796114,
      "loss": 0.4921,
      "step": 1280
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002769514563106796,
      "loss": 0.4908,
      "step": 1290
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00027675728155339803,
      "loss": 0.4937,
      "step": 1300
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002765631067961165,
      "loss": 0.4787,
      "step": 1310
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002763689320388349,
      "loss": 0.4844,
      "step": 1320
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00027617475728155334,
      "loss": 0.4835,
      "step": 1330
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0002759805825242718,
      "loss": 0.4707,
      "step": 1340
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00027578640776699024,
      "loss": 0.4695,
      "step": 1350
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002755922330097087,
      "loss": 0.4692,
      "step": 1360
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002753980582524272,
      "loss": 0.4865,
      "step": 1370
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002752038834951456,
      "loss": 0.4816,
      "step": 1380
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002750097087378641,
      "loss": 0.4665,
      "step": 1390
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002748155339805825,
      "loss": 0.5001,
      "step": 1400
    },
    {
      "epoch": 0.45,
      "eval_loss": 0.4845764636993408,
      "eval_runtime": 528.8954,
      "eval_samples_per_second": 3.781,
      "eval_steps_per_second": 0.473,
      "step": 1400
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00027462135922330097,
      "loss": 0.4748,
      "step": 1410
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002744271844660194,
      "loss": 0.4855,
      "step": 1420
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00027423300970873786,
      "loss": 0.4921,
      "step": 1430
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002740388349514563,
      "loss": 0.4978,
      "step": 1440
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00027384466019417475,
      "loss": 0.4807,
      "step": 1450
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002736504854368932,
      "loss": 0.49,
      "step": 1460
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002734563106796116,
      "loss": 0.4812,
      "step": 1470
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00027326213592233006,
      "loss": 0.4823,
      "step": 1480
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00027306796116504853,
      "loss": 0.4581,
      "step": 1490
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00027287378640776695,
      "loss": 0.4975,
      "step": 1500
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002726796116504854,
      "loss": 0.4859,
      "step": 1510
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00027248543689320384,
      "loss": 0.4661,
      "step": 1520
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002722912621359223,
      "loss": 0.4699,
      "step": 1530
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00027209708737864073,
      "loss": 0.4768,
      "step": 1540
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002719029126213592,
      "loss": 0.5059,
      "step": 1550
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002717087378640776,
      "loss": 0.49,
      "step": 1560
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002715145631067961,
      "loss": 0.495,
      "step": 1570
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00027132038834951457,
      "loss": 0.4662,
      "step": 1580
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.000271126213592233,
      "loss": 0.4845,
      "step": 1590
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00027093203883495146,
      "loss": 0.4706,
      "step": 1600
    },
    {
      "epoch": 0.51,
      "eval_loss": 0.48333096504211426,
      "eval_runtime": 527.3194,
      "eval_samples_per_second": 3.793,
      "eval_steps_per_second": 0.474,
      "step": 1600
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002707378640776699,
      "loss": 0.4668,
      "step": 1610
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002705436893203883,
      "loss": 0.4956,
      "step": 1620
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0002703495145631068,
      "loss": 0.4624,
      "step": 1630
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002701553398058252,
      "loss": 0.5262,
      "step": 1640
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00026996116504854367,
      "loss": 0.4741,
      "step": 1650
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002697669902912621,
      "loss": 0.4764,
      "step": 1660
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00026957281553398056,
      "loss": 0.4791,
      "step": 1670
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00026937864077669903,
      "loss": 0.4867,
      "step": 1680
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00026918446601941745,
      "loss": 0.4743,
      "step": 1690
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002689902912621359,
      "loss": 0.4764,
      "step": 1700
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00026879611650485434,
      "loss": 0.4837,
      "step": 1710
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0002686019417475728,
      "loss": 0.4765,
      "step": 1720
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00026840776699029123,
      "loss": 0.4898,
      "step": 1730
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002682135922330097,
      "loss": 0.4708,
      "step": 1740
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002680194174757281,
      "loss": 0.5113,
      "step": 1750
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00026782524271844654,
      "loss": 0.4852,
      "step": 1760
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.000267631067961165,
      "loss": 0.4643,
      "step": 1770
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002674368932038835,
      "loss": 0.4666,
      "step": 1780
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002672427184466019,
      "loss": 0.4927,
      "step": 1790
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002670485436893204,
      "loss": 0.4778,
      "step": 1800
    },
    {
      "epoch": 0.58,
      "eval_loss": 0.4814455807209015,
      "eval_runtime": 528.5641,
      "eval_samples_per_second": 3.784,
      "eval_steps_per_second": 0.473,
      "step": 1800
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002668543689320388,
      "loss": 0.4695,
      "step": 1810
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00026666019417475727,
      "loss": 0.468,
      "step": 1820
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002664660194174757,
      "loss": 0.4492,
      "step": 1830
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00026627184466019416,
      "loss": 0.4654,
      "step": 1840
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002660776699029126,
      "loss": 0.4909,
      "step": 1850
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00026588349514563106,
      "loss": 0.4786,
      "step": 1860
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00026568932038834953,
      "loss": 0.4633,
      "step": 1870
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00026549514563106795,
      "loss": 0.4826,
      "step": 1880
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002653009708737864,
      "loss": 0.4644,
      "step": 1890
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00026510679611650484,
      "loss": 0.4573,
      "step": 1900
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00026491262135922326,
      "loss": 0.4402,
      "step": 1910
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00026471844660194173,
      "loss": 0.5006,
      "step": 1920
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00026452427184466015,
      "loss": 0.5296,
      "step": 1930
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0002643300970873786,
      "loss": 0.4689,
      "step": 1940
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00026413592233009704,
      "loss": 0.4728,
      "step": 1950
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002639417475728155,
      "loss": 0.476,
      "step": 1960
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00026374757281553393,
      "loss": 0.4545,
      "step": 1970
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002635533980582524,
      "loss": 0.47,
      "step": 1980
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002633592233009709,
      "loss": 0.5014,
      "step": 1990
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0002631650485436893,
      "loss": 0.4726,
      "step": 2000
    },
    {
      "epoch": 0.64,
      "eval_loss": 0.4796612560749054,
      "eval_runtime": 521.7692,
      "eval_samples_per_second": 3.833,
      "eval_steps_per_second": 0.479,
      "step": 2000
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00026297087378640777,
      "loss": 0.4673,
      "step": 2010
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002627766990291262,
      "loss": 0.4892,
      "step": 2020
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00026258252427184466,
      "loss": 0.47,
      "step": 2030
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002623883495145631,
      "loss": 0.4532,
      "step": 2040
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002621941747572815,
      "loss": 0.4949,
      "step": 2050
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00026199999999999997,
      "loss": 0.493,
      "step": 2060
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0002618058252427184,
      "loss": 0.4609,
      "step": 2070
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00026161165048543686,
      "loss": 0.5041,
      "step": 2080
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00026141747572815534,
      "loss": 0.4871,
      "step": 2090
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00026122330097087375,
      "loss": 0.4476,
      "step": 2100
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00026102912621359223,
      "loss": 0.4874,
      "step": 2110
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00026083495145631065,
      "loss": 0.478,
      "step": 2120
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002606407766990291,
      "loss": 0.5009,
      "step": 2130
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00026044660194174754,
      "loss": 0.4905,
      "step": 2140
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.000260252427184466,
      "loss": 0.4664,
      "step": 2150
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00026005825242718443,
      "loss": 0.4743,
      "step": 2160
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002598640776699029,
      "loss": 0.4867,
      "step": 2170
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002596699029126214,
      "loss": 0.4851,
      "step": 2180
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002594757281553398,
      "loss": 0.4743,
      "step": 2190
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002592815533980582,
      "loss": 0.5097,
      "step": 2200
    },
    {
      "epoch": 0.71,
      "eval_loss": 0.47833874821662903,
      "eval_runtime": 519.0074,
      "eval_samples_per_second": 3.854,
      "eval_steps_per_second": 0.482,
      "step": 2200
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002590873786407767,
      "loss": 0.4953,
      "step": 2210
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002588932038834951,
      "loss": 0.4722,
      "step": 2220
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0002586990291262136,
      "loss": 0.4592,
      "step": 2230
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.000258504854368932,
      "loss": 0.4625,
      "step": 2240
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00025831067961165047,
      "loss": 0.4712,
      "step": 2250
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002581165048543689,
      "loss": 0.4728,
      "step": 2260
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00025792233009708736,
      "loss": 0.4849,
      "step": 2270
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00025772815533980583,
      "loss": 0.4907,
      "step": 2280
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00025753398058252425,
      "loss": 0.4499,
      "step": 2290
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0002573398058252427,
      "loss": 0.4648,
      "step": 2300
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00025714563106796114,
      "loss": 0.4616,
      "step": 2310
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0002569514563106796,
      "loss": 0.4872,
      "step": 2320
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00025675728155339804,
      "loss": 0.4986,
      "step": 2330
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00025656310679611645,
      "loss": 0.4899,
      "step": 2340
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00025636893203883493,
      "loss": 0.4848,
      "step": 2350
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00025617475728155335,
      "loss": 0.4586,
      "step": 2360
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002559805825242718,
      "loss": 0.4641,
      "step": 2370
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00025578640776699024,
      "loss": 0.4726,
      "step": 2380
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002555922330097087,
      "loss": 0.4858,
      "step": 2390
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002553980582524272,
      "loss": 0.4631,
      "step": 2400
    },
    {
      "epoch": 0.77,
      "eval_loss": 0.47692033648490906,
      "eval_runtime": 523.0687,
      "eval_samples_per_second": 3.824,
      "eval_steps_per_second": 0.478,
      "step": 2400
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0002552038834951456,
      "loss": 0.4537,
      "step": 2410
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002550097087378641,
      "loss": 0.5043,
      "step": 2420
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002548155339805825,
      "loss": 0.4889,
      "step": 2430
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00025462135922330097,
      "loss": 0.4713,
      "step": 2440
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002544271844660194,
      "loss": 0.4593,
      "step": 2450
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00025423300970873786,
      "loss": 0.4617,
      "step": 2460
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0002540388349514563,
      "loss": 0.4587,
      "step": 2470
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002538446601941747,
      "loss": 0.4471,
      "step": 2480
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00025365048543689317,
      "loss": 0.4732,
      "step": 2490
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00025345631067961164,
      "loss": 0.4999,
      "step": 2500
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00025326213592233006,
      "loss": 0.4736,
      "step": 2510
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00025306796116504853,
      "loss": 0.4893,
      "step": 2520
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00025287378640776695,
      "loss": 0.476,
      "step": 2530
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002526796116504854,
      "loss": 0.4713,
      "step": 2540
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00025248543689320384,
      "loss": 0.4982,
      "step": 2550
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002522912621359223,
      "loss": 0.4694,
      "step": 2560
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00025209708737864074,
      "loss": 0.4822,
      "step": 2570
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0002519029126213592,
      "loss": 0.4657,
      "step": 2580
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0002517087378640777,
      "loss": 0.4796,
      "step": 2590
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002515145631067961,
      "loss": 0.4606,
      "step": 2600
    },
    {
      "epoch": 0.84,
      "eval_loss": 0.4760785698890686,
      "eval_runtime": 515.4107,
      "eval_samples_per_second": 3.88,
      "eval_steps_per_second": 0.485,
      "step": 2600
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002513203883495146,
      "loss": 0.4735,
      "step": 2610
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.000251126213592233,
      "loss": 0.4597,
      "step": 2620
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002509320388349514,
      "loss": 0.4722,
      "step": 2630
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002507378640776699,
      "loss": 0.4378,
      "step": 2640
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002505436893203883,
      "loss": 0.438,
      "step": 2650
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002503495145631068,
      "loss": 0.4754,
      "step": 2660
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002501553398058252,
      "loss": 0.4753,
      "step": 2670
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00024996116504854367,
      "loss": 0.4605,
      "step": 2680
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00024976699029126214,
      "loss": 0.4981,
      "step": 2690
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00024957281553398056,
      "loss": 0.4848,
      "step": 2700
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00024937864077669903,
      "loss": 0.4493,
      "step": 2710
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00024918446601941745,
      "loss": 0.4865,
      "step": 2720
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002489902912621359,
      "loss": 0.4728,
      "step": 2730
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00024879611650485434,
      "loss": 0.4769,
      "step": 2740
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002486019417475728,
      "loss": 0.4817,
      "step": 2750
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00024840776699029123,
      "loss": 0.4685,
      "step": 2760
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00024821359223300965,
      "loss": 0.4718,
      "step": 2770
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002480194174757281,
      "loss": 0.4627,
      "step": 2780
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00024782524271844654,
      "loss": 0.4779,
      "step": 2790
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.000247631067961165,
      "loss": 0.4673,
      "step": 2800
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.47458600997924805,
      "eval_runtime": 513.2712,
      "eval_samples_per_second": 3.897,
      "eval_steps_per_second": 0.487,
      "step": 2800
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002474368932038835,
      "loss": 0.4714,
      "step": 2810
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002472427184466019,
      "loss": 0.5071,
      "step": 2820
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002470485436893204,
      "loss": 0.478,
      "step": 2830
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002468543689320388,
      "loss": 0.4769,
      "step": 2840
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002466601941747573,
      "loss": 0.4871,
      "step": 2850
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002464660194174757,
      "loss": 0.5094,
      "step": 2860
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00024627184466019417,
      "loss": 0.4551,
      "step": 2870
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002460776699029126,
      "loss": 0.4891,
      "step": 2880
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00024588349514563106,
      "loss": 0.4588,
      "step": 2890
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00024568932038834953,
      "loss": 0.4609,
      "step": 2900
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00024549514563106795,
      "loss": 0.4792,
      "step": 2910
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00024530097087378637,
      "loss": 0.4758,
      "step": 2920
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00024510679611650484,
      "loss": 0.5096,
      "step": 2930
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00024491262135922326,
      "loss": 0.4798,
      "step": 2940
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00024471844660194173,
      "loss": 0.4777,
      "step": 2950
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00024452427184466015,
      "loss": 0.4827,
      "step": 2960
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002443300970873786,
      "loss": 0.4713,
      "step": 2970
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00024413592233009707,
      "loss": 0.4716,
      "step": 2980
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00024394174757281552,
      "loss": 0.4644,
      "step": 2990
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00024374757281553396,
      "loss": 0.4695,
      "step": 3000
    },
    {
      "epoch": 0.96,
      "eval_loss": 0.4740288257598877,
      "eval_runtime": 514.5469,
      "eval_samples_per_second": 3.887,
      "eval_steps_per_second": 0.486,
      "step": 3000
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0002435533980582524,
      "loss": 0.4763,
      "step": 3010
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00024335922330097085,
      "loss": 0.4761,
      "step": 3020
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0002431650485436893,
      "loss": 0.4926,
      "step": 3030
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00024297087378640777,
      "loss": 0.4571,
      "step": 3040
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00024277669902912616,
      "loss": 0.4774,
      "step": 3050
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00024258252427184464,
      "loss": 0.4651,
      "step": 3060
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00024238834951456308,
      "loss": 0.4816,
      "step": 3070
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00024219417475728153,
      "loss": 0.4713,
      "step": 3080
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00024199999999999997,
      "loss": 0.503,
      "step": 3090
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00024180582524271842,
      "loss": 0.5096,
      "step": 3100
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00024161165048543687,
      "loss": 0.4786,
      "step": 3110
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002414174757281553,
      "loss": 0.4194,
      "step": 3120
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00024122330097087376,
      "loss": 0.4373,
      "step": 3130
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00024102912621359223,
      "loss": 0.4531,
      "step": 3140
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00024083495145631068,
      "loss": 0.4706,
      "step": 3150
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00024064077669902912,
      "loss": 0.4283,
      "step": 3160
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00024044660194174757,
      "loss": 0.4589,
      "step": 3170
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.000240252427184466,
      "loss": 0.4506,
      "step": 3180
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00024005825242718446,
      "loss": 0.4305,
      "step": 3190
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00023986407766990288,
      "loss": 0.451,
      "step": 3200
    },
    {
      "epoch": 1.03,
      "eval_loss": 0.4739780128002167,
      "eval_runtime": 520.0007,
      "eval_samples_per_second": 3.846,
      "eval_steps_per_second": 0.481,
      "step": 3200
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00023966990291262132,
      "loss": 0.4354,
      "step": 3210
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00023947572815533977,
      "loss": 0.4396,
      "step": 3220
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00023928155339805821,
      "loss": 0.423,
      "step": 3230
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00023908737864077666,
      "loss": 0.4737,
      "step": 3240
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00023889320388349513,
      "loss": 0.4466,
      "step": 3250
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00023869902912621358,
      "loss": 0.4624,
      "step": 3260
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00023850485436893203,
      "loss": 0.455,
      "step": 3270
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00023831067961165047,
      "loss": 0.4699,
      "step": 3280
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00023811650485436892,
      "loss": 0.4218,
      "step": 3290
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00023792233009708736,
      "loss": 0.4495,
      "step": 3300
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0002377281553398058,
      "loss": 0.4738,
      "step": 3310
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00023753398058252425,
      "loss": 0.4574,
      "step": 3320
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002373398058252427,
      "loss": 0.4699,
      "step": 3330
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00023714563106796112,
      "loss": 0.454,
      "step": 3340
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00023695145631067956,
      "loss": 0.463,
      "step": 3350
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00023675728155339804,
      "loss": 0.4533,
      "step": 3360
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00023656310679611648,
      "loss": 0.4525,
      "step": 3370
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00023636893203883493,
      "loss": 0.4391,
      "step": 3380
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00023617475728155338,
      "loss": 0.4542,
      "step": 3390
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00023598058252427182,
      "loss": 0.4255,
      "step": 3400
    },
    {
      "epoch": 1.09,
      "eval_loss": 0.47360363602638245,
      "eval_runtime": 513.4848,
      "eval_samples_per_second": 3.895,
      "eval_steps_per_second": 0.487,
      "step": 3400
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00023578640776699027,
      "loss": 0.4341,
      "step": 3410
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002355922330097087,
      "loss": 0.4439,
      "step": 3420
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00023539805825242716,
      "loss": 0.4512,
      "step": 3430
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0002352038834951456,
      "loss": 0.4454,
      "step": 3440
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00023500970873786408,
      "loss": 0.4621,
      "step": 3450
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00023481553398058252,
      "loss": 0.4597,
      "step": 3460
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00023462135922330097,
      "loss": 0.4546,
      "step": 3470
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00023442718446601941,
      "loss": 0.4488,
      "step": 3480
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00023423300970873783,
      "loss": 0.4553,
      "step": 3490
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00023403883495145628,
      "loss": 0.4334,
      "step": 3500
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00023384466019417473,
      "loss": 0.4609,
      "step": 3510
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00023365048543689317,
      "loss": 0.4517,
      "step": 3520
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00023345631067961162,
      "loss": 0.4627,
      "step": 3530
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00023326213592233006,
      "loss": 0.4355,
      "step": 3540
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00023306796116504854,
      "loss": 0.4451,
      "step": 3550
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00023287378640776698,
      "loss": 0.4565,
      "step": 3560
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00023267961165048543,
      "loss": 0.4652,
      "step": 3570
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00023248543689320387,
      "loss": 0.4422,
      "step": 3580
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00023229126213592232,
      "loss": 0.4361,
      "step": 3590
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00023209708737864076,
      "loss": 0.4279,
      "step": 3600
    },
    {
      "epoch": 1.16,
      "eval_loss": 0.4739069640636444,
      "eval_runtime": 517.674,
      "eval_samples_per_second": 3.863,
      "eval_steps_per_second": 0.483,
      "step": 3600
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0002319029126213592,
      "loss": 0.4432,
      "step": 3610
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00023170873786407766,
      "loss": 0.4719,
      "step": 3620
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00023151456310679608,
      "loss": 0.4603,
      "step": 3630
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00023132038834951452,
      "loss": 0.4878,
      "step": 3640
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00023112621359223297,
      "loss": 0.4439,
      "step": 3650
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00023093203883495144,
      "loss": 0.4301,
      "step": 3660
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00023073786407766989,
      "loss": 0.4362,
      "step": 3670
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00023054368932038833,
      "loss": 0.4612,
      "step": 3680
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00023034951456310678,
      "loss": 0.4681,
      "step": 3690
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00023015533980582522,
      "loss": 0.4303,
      "step": 3700
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00022996116504854367,
      "loss": 0.4663,
      "step": 3710
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00022976699029126211,
      "loss": 0.4523,
      "step": 3720
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00022957281553398056,
      "loss": 0.4478,
      "step": 3730
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.000229378640776699,
      "loss": 0.4428,
      "step": 3740
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00022918446601941748,
      "loss": 0.4427,
      "step": 3750
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00022899029126213593,
      "loss": 0.4247,
      "step": 3760
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00022879611650485437,
      "loss": 0.4431,
      "step": 3770
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0002286019417475728,
      "loss": 0.466,
      "step": 3780
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00022840776699029124,
      "loss": 0.4704,
      "step": 3790
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00022821359223300968,
      "loss": 0.4698,
      "step": 3800
    },
    {
      "epoch": 1.22,
      "eval_loss": 0.4726303517818451,
      "eval_runtime": 519.452,
      "eval_samples_per_second": 3.85,
      "eval_steps_per_second": 0.481,
      "step": 3800
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00022801941747572813,
      "loss": 0.4643,
      "step": 3810
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00022782524271844657,
      "loss": 0.4357,
      "step": 3820
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00022763106796116502,
      "loss": 0.4749,
      "step": 3830
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00022743689320388346,
      "loss": 0.428,
      "step": 3840
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002272427184466019,
      "loss": 0.4705,
      "step": 3850
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00022704854368932038,
      "loss": 0.4448,
      "step": 3860
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00022685436893203883,
      "loss": 0.4613,
      "step": 3870
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00022666019417475728,
      "loss": 0.4643,
      "step": 3880
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00022646601941747572,
      "loss": 0.4416,
      "step": 3890
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00022627184466019417,
      "loss": 0.4494,
      "step": 3900
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0002260776699029126,
      "loss": 0.4343,
      "step": 3910
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00022588349514563103,
      "loss": 0.4322,
      "step": 3920
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00022568932038834948,
      "loss": 0.454,
      "step": 3930
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00022549514563106792,
      "loss": 0.4432,
      "step": 3940
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00022530097087378637,
      "loss": 0.4393,
      "step": 3950
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00022510679611650484,
      "loss": 0.4438,
      "step": 3960
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0002249126213592233,
      "loss": 0.4144,
      "step": 3970
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00022471844660194173,
      "loss": 0.4304,
      "step": 3980
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00022452427184466018,
      "loss": 0.454,
      "step": 3990
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00022433009708737863,
      "loss": 0.4663,
      "step": 4000
    },
    {
      "epoch": 1.29,
      "eval_loss": 0.4728265702724457,
      "eval_runtime": 520.4674,
      "eval_samples_per_second": 3.843,
      "eval_steps_per_second": 0.48,
      "step": 4000
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00022413592233009707,
      "loss": 0.4803,
      "step": 4010
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00022394174757281552,
      "loss": 0.4528,
      "step": 4020
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00022374757281553396,
      "loss": 0.4333,
      "step": 4030
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0002235533980582524,
      "loss": 0.4548,
      "step": 4040
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00022335922330097088,
      "loss": 0.4533,
      "step": 4050
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00022316504854368927,
      "loss": 0.4269,
      "step": 4060
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00022297087378640775,
      "loss": 0.4489,
      "step": 4070
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002227766990291262,
      "loss": 0.4581,
      "step": 4080
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00022258252427184464,
      "loss": 0.4474,
      "step": 4090
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00022238834951456308,
      "loss": 0.4498,
      "step": 4100
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00022219417475728153,
      "loss": 0.4737,
      "step": 4110
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00022199999999999998,
      "loss": 0.4412,
      "step": 4120
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00022180582524271842,
      "loss": 0.4365,
      "step": 4130
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00022161165048543687,
      "loss": 0.4266,
      "step": 4140
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0002214174757281553,
      "loss": 0.4492,
      "step": 4150
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00022122330097087379,
      "loss": 0.4472,
      "step": 4160
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00022102912621359223,
      "loss": 0.4137,
      "step": 4170
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00022083495145631068,
      "loss": 0.4497,
      "step": 4180
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00022064077669902912,
      "loss": 0.4324,
      "step": 4190
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00022044660194174757,
      "loss": 0.4332,
      "step": 4200
    },
    {
      "epoch": 1.35,
      "eval_loss": 0.471797376871109,
      "eval_runtime": 518.6631,
      "eval_samples_per_second": 3.856,
      "eval_steps_per_second": 0.482,
      "step": 4200
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.000220252427184466,
      "loss": 0.452,
      "step": 4210
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00022005825242718443,
      "loss": 0.4372,
      "step": 4220
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00021986407766990288,
      "loss": 0.4346,
      "step": 4230
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00021966990291262133,
      "loss": 0.4511,
      "step": 4240
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00021947572815533977,
      "loss": 0.4594,
      "step": 4250
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00021928155339805822,
      "loss": 0.4655,
      "step": 4260
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0002190873786407767,
      "loss": 0.4613,
      "step": 4270
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00021889320388349514,
      "loss": 0.4436,
      "step": 4280
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00021869902912621358,
      "loss": 0.4434,
      "step": 4290
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00021850485436893203,
      "loss": 0.4554,
      "step": 4300
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00021831067961165047,
      "loss": 0.4838,
      "step": 4310
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00021811650485436892,
      "loss": 0.4451,
      "step": 4320
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00021792233009708736,
      "loss": 0.4606,
      "step": 4330
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0002177281553398058,
      "loss": 0.4239,
      "step": 4340
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00021753398058252423,
      "loss": 0.4328,
      "step": 4350
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00021733980582524268,
      "loss": 0.4712,
      "step": 4360
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00021714563106796112,
      "loss": 0.4412,
      "step": 4370
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0002169514563106796,
      "loss": 0.4629,
      "step": 4380
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00021675728155339804,
      "loss": 0.4536,
      "step": 4390
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00021656310679611649,
      "loss": 0.4208,
      "step": 4400
    },
    {
      "epoch": 1.41,
      "eval_loss": 0.4719524681568146,
      "eval_runtime": 524.6738,
      "eval_samples_per_second": 3.812,
      "eval_steps_per_second": 0.476,
      "step": 4400
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00021636893203883493,
      "loss": 0.438,
      "step": 4410
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00021617475728155338,
      "loss": 0.4399,
      "step": 4420
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00021598058252427182,
      "loss": 0.4568,
      "step": 4430
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00021578640776699027,
      "loss": 0.4695,
      "step": 4440
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00021559223300970871,
      "loss": 0.4387,
      "step": 4450
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002153980582524272,
      "loss": 0.4466,
      "step": 4460
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00021520388349514563,
      "loss": 0.4652,
      "step": 4470
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00021500970873786408,
      "loss": 0.4155,
      "step": 4480
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00021481553398058253,
      "loss": 0.4457,
      "step": 4490
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00021462135922330094,
      "loss": 0.4349,
      "step": 4500
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0002144271844660194,
      "loss": 0.4304,
      "step": 4510
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00021423300970873784,
      "loss": 0.4222,
      "step": 4520
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00021403883495145628,
      "loss": 0.4373,
      "step": 4530
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00021384466019417473,
      "loss": 0.4526,
      "step": 4540
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00021365048543689317,
      "loss": 0.4804,
      "step": 4550
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00021345631067961162,
      "loss": 0.468,
      "step": 4560
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0002132621359223301,
      "loss": 0.4664,
      "step": 4570
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00021306796116504854,
      "loss": 0.4688,
      "step": 4580
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00021287378640776698,
      "loss": 0.4262,
      "step": 4590
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00021267961165048543,
      "loss": 0.4431,
      "step": 4600
    },
    {
      "epoch": 1.48,
      "eval_loss": 0.47054779529571533,
      "eval_runtime": 526.5834,
      "eval_samples_per_second": 3.798,
      "eval_steps_per_second": 0.475,
      "step": 4600
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00021248543689320387,
      "loss": 0.4371,
      "step": 4610
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00021229126213592232,
      "loss": 0.4555,
      "step": 4620
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00021209708737864077,
      "loss": 0.4538,
      "step": 4630
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00021190291262135919,
      "loss": 0.4501,
      "step": 4640
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00021170873786407763,
      "loss": 0.4201,
      "step": 4650
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00021151456310679608,
      "loss": 0.4346,
      "step": 4660
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00021132038834951452,
      "loss": 0.447,
      "step": 4670
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.000211126213592233,
      "loss": 0.451,
      "step": 4680
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00021093203883495144,
      "loss": 0.4548,
      "step": 4690
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002107378640776699,
      "loss": 0.4386,
      "step": 4700
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00021054368932038833,
      "loss": 0.4408,
      "step": 4710
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00021034951456310678,
      "loss": 0.467,
      "step": 4720
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00021015533980582522,
      "loss": 0.4447,
      "step": 4730
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00020996116504854367,
      "loss": 0.4566,
      "step": 4740
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00020976699029126212,
      "loss": 0.4391,
      "step": 4750
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00020957281553398056,
      "loss": 0.4547,
      "step": 4760
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00020937864077669904,
      "loss": 0.4466,
      "step": 4770
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00020918446601941748,
      "loss": 0.4368,
      "step": 4780
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0002089902912621359,
      "loss": 0.446,
      "step": 4790
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00020879611650485435,
      "loss": 0.4228,
      "step": 4800
    },
    {
      "epoch": 1.54,
      "eval_loss": 0.4704694151878357,
      "eval_runtime": 522.9734,
      "eval_samples_per_second": 3.824,
      "eval_steps_per_second": 0.478,
      "step": 4800
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0002086019417475728,
      "loss": 0.4241,
      "step": 4810
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00020840776699029124,
      "loss": 0.4635,
      "step": 4820
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00020821359223300968,
      "loss": 0.4618,
      "step": 4830
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00020801941747572813,
      "loss": 0.4364,
      "step": 4840
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00020782524271844657,
      "loss": 0.4532,
      "step": 4850
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00020763106796116502,
      "loss": 0.4522,
      "step": 4860
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0002074368932038835,
      "loss": 0.4459,
      "step": 4870
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00020724271844660194,
      "loss": 0.4599,
      "step": 4880
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00020704854368932039,
      "loss": 0.4526,
      "step": 4890
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00020685436893203883,
      "loss": 0.4497,
      "step": 4900
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00020666019417475728,
      "loss": 0.4357,
      "step": 4910
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00020646601941747572,
      "loss": 0.4353,
      "step": 4920
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00020627184466019414,
      "loss": 0.4383,
      "step": 4930
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0002060776699029126,
      "loss": 0.4654,
      "step": 4940
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00020588349514563103,
      "loss": 0.455,
      "step": 4950
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00020568932038834948,
      "loss": 0.4544,
      "step": 4960
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00020549514563106792,
      "loss": 0.4531,
      "step": 4970
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002053009708737864,
      "loss": 0.4586,
      "step": 4980
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00020510679611650484,
      "loss": 0.4648,
      "step": 4990
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0002049126213592233,
      "loss": 0.4527,
      "step": 5000
    },
    {
      "epoch": 1.61,
      "eval_loss": 0.47035008668899536,
      "eval_runtime": 520.9438,
      "eval_samples_per_second": 3.839,
      "eval_steps_per_second": 0.48,
      "step": 5000
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00020471844660194174,
      "loss": 0.4497,
      "step": 5010
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00020452427184466018,
      "loss": 0.4321,
      "step": 5020
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00020433009708737863,
      "loss": 0.4548,
      "step": 5030
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00020413592233009707,
      "loss": 0.4687,
      "step": 5040
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00020394174757281552,
      "loss": 0.4773,
      "step": 5050
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00020374757281553396,
      "loss": 0.4692,
      "step": 5060
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00020355339805825244,
      "loss": 0.4443,
      "step": 5070
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00020335922330097083,
      "loss": 0.4751,
      "step": 5080
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0002031650485436893,
      "loss": 0.4473,
      "step": 5090
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00020297087378640775,
      "loss": 0.4551,
      "step": 5100
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0002027766990291262,
      "loss": 0.4753,
      "step": 5110
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00020258252427184464,
      "loss": 0.4325,
      "step": 5120
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00020238834951456309,
      "loss": 0.4563,
      "step": 5130
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00020219417475728153,
      "loss": 0.456,
      "step": 5140
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00020199999999999998,
      "loss": 0.4598,
      "step": 5150
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00020180582524271842,
      "loss": 0.4407,
      "step": 5160
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00020161165048543687,
      "loss": 0.4424,
      "step": 5170
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00020141747572815534,
      "loss": 0.4291,
      "step": 5180
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0002012233009708738,
      "loss": 0.4544,
      "step": 5190
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00020102912621359223,
      "loss": 0.4351,
      "step": 5200
    },
    {
      "epoch": 1.67,
      "eval_loss": 0.46933242678642273,
      "eval_runtime": 521.6481,
      "eval_samples_per_second": 3.834,
      "eval_steps_per_second": 0.479,
      "step": 5200
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00020083495145631068,
      "loss": 0.4503,
      "step": 5210
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002006407766990291,
      "loss": 0.4352,
      "step": 5220
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00020044660194174754,
      "loss": 0.4509,
      "step": 5230
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.000200252427184466,
      "loss": 0.4656,
      "step": 5240
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00020005825242718444,
      "loss": 0.4578,
      "step": 5250
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00019986407766990288,
      "loss": 0.4338,
      "step": 5260
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00019966990291262133,
      "loss": 0.4285,
      "step": 5270
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001994757281553398,
      "loss": 0.4219,
      "step": 5280
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00019928155339805825,
      "loss": 0.4709,
      "step": 5290
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001990873786407767,
      "loss": 0.4335,
      "step": 5300
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00019889320388349514,
      "loss": 0.421,
      "step": 5310
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00019869902912621358,
      "loss": 0.4594,
      "step": 5320
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00019850485436893203,
      "loss": 0.4361,
      "step": 5330
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00019831067961165047,
      "loss": 0.4673,
      "step": 5340
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00019811650485436892,
      "loss": 0.4499,
      "step": 5350
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00019792233009708734,
      "loss": 0.4634,
      "step": 5360
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00019772815533980579,
      "loss": 0.4267,
      "step": 5370
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00019753398058252423,
      "loss": 0.453,
      "step": 5380
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0001973398058252427,
      "loss": 0.4406,
      "step": 5390
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00019714563106796115,
      "loss": 0.4516,
      "step": 5400
    },
    {
      "epoch": 1.74,
      "eval_loss": 0.4691580832004547,
      "eval_runtime": 518.7603,
      "eval_samples_per_second": 3.855,
      "eval_steps_per_second": 0.482,
      "step": 5400
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.0001969514563106796,
      "loss": 0.4744,
      "step": 5410
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00019675728155339804,
      "loss": 0.4744,
      "step": 5420
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0001965631067961165,
      "loss": 0.4546,
      "step": 5430
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00019636893203883493,
      "loss": 0.4553,
      "step": 5440
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00019617475728155338,
      "loss": 0.4562,
      "step": 5450
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00019598058252427182,
      "loss": 0.4426,
      "step": 5460
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00019578640776699027,
      "loss": 0.4425,
      "step": 5470
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00019559223300970874,
      "loss": 0.4288,
      "step": 5480
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0001953980582524272,
      "loss": 0.4413,
      "step": 5490
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00019520388349514564,
      "loss": 0.4488,
      "step": 5500
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00019500970873786405,
      "loss": 0.4461,
      "step": 5510
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0001948155339805825,
      "loss": 0.4361,
      "step": 5520
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00019462135922330095,
      "loss": 0.4146,
      "step": 5530
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0001944271844660194,
      "loss": 0.4378,
      "step": 5540
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00019423300970873784,
      "loss": 0.4375,
      "step": 5550
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00019403883495145628,
      "loss": 0.4541,
      "step": 5560
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00019384466019417473,
      "loss": 0.4228,
      "step": 5570
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00019365048543689317,
      "loss": 0.4349,
      "step": 5580
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00019345631067961165,
      "loss": 0.4401,
      "step": 5590
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0001932621359223301,
      "loss": 0.4494,
      "step": 5600
    },
    {
      "epoch": 1.8,
      "eval_loss": 0.46821147203445435,
      "eval_runtime": 516.2558,
      "eval_samples_per_second": 3.874,
      "eval_steps_per_second": 0.484,
      "step": 5600
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00019306796116504854,
      "loss": 0.4666,
      "step": 5610
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00019287378640776699,
      "loss": 0.4385,
      "step": 5620
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00019267961165048543,
      "loss": 0.4202,
      "step": 5630
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00019248543689320388,
      "loss": 0.4564,
      "step": 5640
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0001922912621359223,
      "loss": 0.4315,
      "step": 5650
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00019209708737864074,
      "loss": 0.4549,
      "step": 5660
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0001919029126213592,
      "loss": 0.4515,
      "step": 5670
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00019170873786407763,
      "loss": 0.4478,
      "step": 5680
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0001915145631067961,
      "loss": 0.4541,
      "step": 5690
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00019132038834951455,
      "loss": 0.4282,
      "step": 5700
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.000191126213592233,
      "loss": 0.4333,
      "step": 5710
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00019093203883495144,
      "loss": 0.4587,
      "step": 5720
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0001907378640776699,
      "loss": 0.4671,
      "step": 5730
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00019054368932038834,
      "loss": 0.4618,
      "step": 5740
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00019034951456310678,
      "loss": 0.4576,
      "step": 5750
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00019015533980582523,
      "loss": 0.4435,
      "step": 5760
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00018996116504854367,
      "loss": 0.4311,
      "step": 5770
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00018976699029126215,
      "loss": 0.4414,
      "step": 5780
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0001895728155339806,
      "loss": 0.455,
      "step": 5790
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.000189378640776699,
      "loss": 0.47,
      "step": 5800
    },
    {
      "epoch": 1.86,
      "eval_loss": 0.46795403957366943,
      "eval_runtime": 528.6661,
      "eval_samples_per_second": 3.783,
      "eval_steps_per_second": 0.473,
      "step": 5800
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00018918446601941746,
      "loss": 0.4624,
      "step": 5810
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0001889902912621359,
      "loss": 0.4606,
      "step": 5820
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00018879611650485435,
      "loss": 0.4533,
      "step": 5830
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0001886019417475728,
      "loss": 0.4458,
      "step": 5840
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00018840776699029124,
      "loss": 0.4455,
      "step": 5850
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00018821359223300968,
      "loss": 0.4416,
      "step": 5860
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00018801941747572813,
      "loss": 0.4364,
      "step": 5870
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00018782524271844658,
      "loss": 0.4537,
      "step": 5880
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00018763106796116505,
      "loss": 0.4368,
      "step": 5890
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0001874368932038835,
      "loss": 0.4618,
      "step": 5900
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00018724271844660194,
      "loss": 0.4295,
      "step": 5910
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0001870485436893204,
      "loss": 0.437,
      "step": 5920
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00018685436893203883,
      "loss": 0.4455,
      "step": 5930
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00018666019417475725,
      "loss": 0.4536,
      "step": 5940
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.0001864660194174757,
      "loss": 0.4268,
      "step": 5950
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00018627184466019414,
      "loss": 0.4446,
      "step": 5960
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0001860776699029126,
      "loss": 0.4402,
      "step": 5970
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00018588349514563103,
      "loss": 0.4461,
      "step": 5980
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00018568932038834948,
      "loss": 0.4428,
      "step": 5990
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00018549514563106795,
      "loss": 0.4511,
      "step": 6000
    },
    {
      "epoch": 1.93,
      "eval_loss": 0.46734848618507385,
      "eval_runtime": 541.2095,
      "eval_samples_per_second": 3.695,
      "eval_steps_per_second": 0.462,
      "step": 6000
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0001853009708737864,
      "loss": 0.4351,
      "step": 6010
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00018510679611650485,
      "loss": 0.4829,
      "step": 6020
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0001849126213592233,
      "loss": 0.4493,
      "step": 6030
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00018471844660194174,
      "loss": 0.4435,
      "step": 6040
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00018452427184466018,
      "loss": 0.4295,
      "step": 6050
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00018433009708737863,
      "loss": 0.4416,
      "step": 6060
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00018413592233009707,
      "loss": 0.4233,
      "step": 6070
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00018394174757281555,
      "loss": 0.4569,
      "step": 6080
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00018374757281553394,
      "loss": 0.4402,
      "step": 6090
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0001835533980582524,
      "loss": 0.4358,
      "step": 6100
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00018335922330097086,
      "loss": 0.4552,
      "step": 6110
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0001831650485436893,
      "loss": 0.4475,
      "step": 6120
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00018297087378640775,
      "loss": 0.4479,
      "step": 6130
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0001827766990291262,
      "loss": 0.4538,
      "step": 6140
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00018258252427184464,
      "loss": 0.4635,
      "step": 6150
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0001823883495145631,
      "loss": 0.4366,
      "step": 6160
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00018219417475728153,
      "loss": 0.4194,
      "step": 6170
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00018199999999999998,
      "loss": 0.4795,
      "step": 6180
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00018180582524271845,
      "loss": 0.4484,
      "step": 6190
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.0001816116504854369,
      "loss": 0.4451,
      "step": 6200
    },
    {
      "epoch": 1.99,
      "eval_loss": 0.4666987955570221,
      "eval_runtime": 560.4767,
      "eval_samples_per_second": 3.568,
      "eval_steps_per_second": 0.446,
      "step": 6200
    }
  ],
  "logging_steps": 10,
  "max_steps": 15550,
  "num_train_epochs": 5,
  "save_steps": 200,
  "total_flos": 1.5245364281459343e+18,
  "trial_name": null,
  "trial_params": null
}