{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.99825479930192,
  "global_step": 2860,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 12.9083,
      "step": 10
    },
    {
      "epoch": 0.07,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 12.5233,
      "step": 20
    },
    {
      "epoch": 0.1,
      "learning_rate": 8.699999999999999e-05,
      "loss": 12.1986,
      "step": 30
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00011099999999999999,
      "loss": 9.567,
      "step": 40
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.000138,
      "loss": 6.7421,
      "step": 50
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.000168,
      "loss": 4.3567,
      "step": 60
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.000198,
      "loss": 3.5698,
      "step": 70
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00022799999999999999,
      "loss": 3.2567,
      "step": 80
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.000258,
      "loss": 3.145,
      "step": 90
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00028799999999999995,
      "loss": 3.2802,
      "step": 100
    },
    {
      "epoch": 0.35,
      "eval_loss": 3.8881919384002686,
      "eval_runtime": 112.7654,
      "eval_samples_per_second": 12.867,
      "eval_steps_per_second": 0.807,
      "step": 100
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002993478260869565,
      "loss": 3.6105,
      "step": 110
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00029826086956521737,
      "loss": 3.1704,
      "step": 120
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00029717391304347825,
      "loss": 3.1124,
      "step": 130
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002960869565217391,
      "loss": 3.1266,
      "step": 140
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00029499999999999996,
      "loss": 3.2521,
      "step": 150
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00029391304347826084,
      "loss": 3.6024,
      "step": 160
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002928260869565217,
      "loss": 3.2086,
      "step": 170
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002917391304347826,
      "loss": 3.1616,
      "step": 180
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00029065217391304344,
      "loss": 3.1408,
      "step": 190
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002895652173913043,
      "loss": 3.2254,
      "step": 200
    },
    {
      "epoch": 0.7,
      "eval_loss": 3.885490894317627,
      "eval_runtime": 111.0614,
      "eval_samples_per_second": 13.065,
      "eval_steps_per_second": 0.819,
      "step": 200
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002884782608695652,
      "loss": 3.6835,
      "step": 210
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00028739130434782603,
      "loss": 3.1449,
      "step": 220
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002863043478260869,
      "loss": 3.2003,
      "step": 230
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0002852173913043478,
      "loss": 3.0836,
      "step": 240
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002841304347826087,
      "loss": 3.1128,
      "step": 250
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0002830434782608695,
      "loss": 3.3853,
      "step": 260
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002819565217391304,
      "loss": 3.102,
      "step": 270
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0002808695652173913,
      "loss": 3.0516,
      "step": 280
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0002797826086956521,
      "loss": 3.5681,
      "step": 290
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00027869565217391305,
      "loss": 3.1697,
      "step": 300
    },
    {
      "epoch": 1.05,
      "eval_loss": 3.419605255126953,
      "eval_runtime": 111.0117,
      "eval_samples_per_second": 13.071,
      "eval_steps_per_second": 0.82,
      "step": 300
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0002776086956521739,
      "loss": 3.0489,
      "step": 310
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00027652173913043476,
      "loss": 3.0347,
      "step": 320
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00027543478260869564,
      "loss": 3.0201,
      "step": 330
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00027434782608695647,
      "loss": 3.2079,
      "step": 340
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00027326086956521735,
      "loss": 3.0785,
      "step": 350
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00027217391304347824,
      "loss": 2.9747,
      "step": 360
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0002710869565217391,
      "loss": 2.9664,
      "step": 370
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00027,
      "loss": 2.9724,
      "step": 380
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00026891304347826083,
      "loss": 3.1082,
      "step": 390
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0002678260869565217,
      "loss": 3.0258,
      "step": 400
    },
    {
      "epoch": 1.4,
      "eval_loss": 3.195385217666626,
      "eval_runtime": 110.895,
      "eval_samples_per_second": 13.084,
      "eval_steps_per_second": 0.821,
      "step": 400
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002667391304347826,
      "loss": 2.9719,
      "step": 410
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00026565217391304343,
      "loss": 2.9884,
      "step": 420
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00026456521739130437,
      "loss": 2.9625,
      "step": 430
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0002634782608695652,
      "loss": 3.0735,
      "step": 440
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0002623913043478261,
      "loss": 2.9998,
      "step": 450
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00026130434782608696,
      "loss": 2.9621,
      "step": 460
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0002602173913043478,
      "loss": 2.9433,
      "step": 470
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002591304347826087,
      "loss": 2.9256,
      "step": 480
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00025804347826086956,
      "loss": 2.9996,
      "step": 490
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00025695652173913044,
      "loss": 2.8553,
      "step": 500
    },
    {
      "epoch": 1.75,
      "eval_loss": 3.017104387283325,
      "eval_runtime": 110.9455,
      "eval_samples_per_second": 13.078,
      "eval_steps_per_second": 0.82,
      "step": 500
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00025586956521739127,
      "loss": 2.7029,
      "step": 510
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00025478260869565215,
      "loss": 2.4464,
      "step": 520
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00025369565217391303,
      "loss": 2.2426,
      "step": 530
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00025260869565217386,
      "loss": 2.2914,
      "step": 540
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00025152173913043475,
      "loss": 1.9541,
      "step": 550
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00025043478260869563,
      "loss": 1.5822,
      "step": 560
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.0002493478260869565,
      "loss": 1.5446,
      "step": 570
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0002482608695652174,
      "loss": 2.0892,
      "step": 580
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0002471739130434782,
      "loss": 1.3977,
      "step": 590
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0002460869565217391,
      "loss": 1.2436,
      "step": 600
    },
    {
      "epoch": 2.1,
      "eval_loss": 2.155662775039673,
      "eval_runtime": 111.1811,
      "eval_samples_per_second": 13.051,
      "eval_steps_per_second": 0.818,
      "step": 600
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.000245,
      "loss": 1.1344,
      "step": 610
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.00024391304347826085,
      "loss": 1.1795,
      "step": 620
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0002428260869565217,
      "loss": 2.1046,
      "step": 630
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.00024173913043478261,
      "loss": 1.0792,
      "step": 640
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.00024065217391304347,
      "loss": 1.157,
      "step": 650
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.00023956521739130433,
      "loss": 1.0428,
      "step": 660
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.00023847826086956518,
      "loss": 1.1063,
      "step": 670
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.00023739130434782607,
      "loss": 1.6716,
      "step": 680
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.00023630434782608692,
      "loss": 0.9602,
      "step": 690
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.00023521739130434778,
      "loss": 0.9642,
      "step": 700
    },
    {
      "epoch": 2.45,
      "eval_loss": 1.5505317449569702,
      "eval_runtime": 111.9226,
      "eval_samples_per_second": 12.964,
      "eval_steps_per_second": 0.813,
      "step": 700
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0002341304347826087,
      "loss": 0.8874,
      "step": 710
    },
    {
      "epoch": 2.52,
      "learning_rate": 0.00023304347826086954,
      "loss": 0.9596,
      "step": 720
    },
    {
      "epoch": 2.55,
      "learning_rate": 0.00023195652173913043,
      "loss": 1.3932,
      "step": 730
    },
    {
      "epoch": 2.59,
      "learning_rate": 0.00023086956521739128,
      "loss": 0.8699,
      "step": 740
    },
    {
      "epoch": 2.62,
      "learning_rate": 0.00022978260869565214,
      "loss": 0.8153,
      "step": 750
    },
    {
      "epoch": 2.66,
      "learning_rate": 0.00022869565217391302,
      "loss": 0.9178,
      "step": 760
    },
    {
      "epoch": 2.69,
      "learning_rate": 0.00022760869565217388,
      "loss": 0.9647,
      "step": 770
    },
    {
      "epoch": 2.73,
      "learning_rate": 0.00022652173913043476,
      "loss": 1.0555,
      "step": 780
    },
    {
      "epoch": 2.76,
      "learning_rate": 0.00022543478260869565,
      "loss": 0.8834,
      "step": 790
    },
    {
      "epoch": 2.8,
      "learning_rate": 0.0002243478260869565,
      "loss": 0.752,
      "step": 800
    },
    {
      "epoch": 2.8,
      "eval_loss": 0.9008368253707886,
      "eval_runtime": 111.0482,
      "eval_samples_per_second": 13.066,
      "eval_steps_per_second": 0.819,
      "step": 800
    },
    {
      "epoch": 2.83,
      "learning_rate": 0.00022326086956521736,
      "loss": 0.8197,
      "step": 810
    },
    {
      "epoch": 2.87,
      "learning_rate": 0.00022217391304347824,
      "loss": 0.7771,
      "step": 820
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0002210869565217391,
      "loss": 0.9506,
      "step": 830
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.00021999999999999995,
      "loss": 0.7518,
      "step": 840
    },
    {
      "epoch": 2.97,
      "learning_rate": 0.00021891304347826086,
      "loss": 0.7297,
      "step": 850
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.00021782608695652172,
      "loss": 0.8188,
      "step": 860
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0002167391304347826,
      "loss": 0.6478,
      "step": 870
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.00021565217391304346,
      "loss": 0.6546,
      "step": 880
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.00021456521739130432,
      "loss": 0.7012,
      "step": 890
    },
    {
      "epoch": 3.15,
      "learning_rate": 0.0002134782608695652,
      "loss": 0.6164,
      "step": 900
    },
    {
      "epoch": 3.15,
      "eval_loss": 0.6567979454994202,
      "eval_runtime": 111.2282,
      "eval_samples_per_second": 13.045,
      "eval_steps_per_second": 0.818,
      "step": 900
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.00021239130434782605,
      "loss": 0.7296,
      "step": 910
    },
    {
      "epoch": 3.22,
      "learning_rate": 0.00021130434782608694,
      "loss": 0.6166,
      "step": 920
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.00021021739130434782,
      "loss": 0.6354,
      "step": 930
    },
    {
      "epoch": 3.29,
      "learning_rate": 0.00020913043478260868,
      "loss": 0.6388,
      "step": 940
    },
    {
      "epoch": 3.32,
      "learning_rate": 0.00020804347826086953,
      "loss": 0.6069,
      "step": 950
    },
    {
      "epoch": 3.36,
      "learning_rate": 0.00020695652173913042,
      "loss": 0.7107,
      "step": 960
    },
    {
      "epoch": 3.39,
      "learning_rate": 0.00020586956521739127,
      "loss": 0.6162,
      "step": 970
    },
    {
      "epoch": 3.43,
      "learning_rate": 0.00020478260869565213,
      "loss": 0.567,
      "step": 980
    },
    {
      "epoch": 3.46,
      "learning_rate": 0.00020369565217391304,
      "loss": 0.5679,
      "step": 990
    },
    {
      "epoch": 3.5,
      "learning_rate": 0.0002026086956521739,
      "loss": 0.6362,
      "step": 1000
    },
    {
      "epoch": 3.5,
      "eval_loss": 0.5460886359214783,
      "eval_runtime": 110.911,
      "eval_samples_per_second": 13.083,
      "eval_steps_per_second": 0.82,
      "step": 1000
    },
    {
      "epoch": 3.53,
      "learning_rate": 0.00020152173913043478,
      "loss": 0.6771,
      "step": 1010
    },
    {
      "epoch": 3.57,
      "learning_rate": 0.00020043478260869563,
      "loss": 0.6144,
      "step": 1020
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.0001993478260869565,
      "loss": 0.583,
      "step": 1030
    },
    {
      "epoch": 3.64,
      "learning_rate": 0.00019826086956521737,
      "loss": 0.5773,
      "step": 1040
    },
    {
      "epoch": 3.67,
      "learning_rate": 0.00019717391304347826,
      "loss": 0.5981,
      "step": 1050
    },
    {
      "epoch": 3.71,
      "learning_rate": 0.0001960869565217391,
      "loss": 0.6162,
      "step": 1060
    },
    {
      "epoch": 3.74,
      "learning_rate": 0.000195,
      "loss": 0.5387,
      "step": 1070
    },
    {
      "epoch": 3.77,
      "learning_rate": 0.00019391304347826085,
      "loss": 0.5651,
      "step": 1080
    },
    {
      "epoch": 3.81,
      "learning_rate": 0.0001928260869565217,
      "loss": 0.544,
      "step": 1090
    },
    {
      "epoch": 3.84,
      "learning_rate": 0.0001917391304347826,
      "loss": 0.5539,
      "step": 1100
    },
    {
      "epoch": 3.84,
      "eval_loss": 0.5009284615516663,
      "eval_runtime": 111.9745,
      "eval_samples_per_second": 12.958,
      "eval_steps_per_second": 0.813,
      "step": 1100
    },
    {
      "epoch": 3.88,
      "learning_rate": 0.00019065217391304345,
      "loss": 0.6894,
      "step": 1110
    },
    {
      "epoch": 3.91,
      "learning_rate": 0.00018956521739130436,
      "loss": 0.5409,
      "step": 1120
    },
    {
      "epoch": 3.95,
      "learning_rate": 0.00018847826086956521,
      "loss": 0.5676,
      "step": 1130
    },
    {
      "epoch": 3.98,
      "learning_rate": 0.00018739130434782607,
      "loss": 0.5543,
      "step": 1140
    },
    {
      "epoch": 4.02,
      "learning_rate": 0.00018630434782608695,
      "loss": 0.6413,
      "step": 1150
    },
    {
      "epoch": 4.06,
      "learning_rate": 0.0001852173913043478,
      "loss": 0.4664,
      "step": 1160
    },
    {
      "epoch": 4.09,
      "learning_rate": 0.00018413043478260867,
      "loss": 0.5098,
      "step": 1170
    },
    {
      "epoch": 4.13,
      "learning_rate": 0.00018304347826086955,
      "loss": 0.4832,
      "step": 1180
    },
    {
      "epoch": 4.16,
      "learning_rate": 0.00018195652173913043,
      "loss": 0.4859,
      "step": 1190
    },
    {
      "epoch": 4.2,
      "learning_rate": 0.0001808695652173913,
      "loss": 0.5144,
      "step": 1200
    },
    {
      "epoch": 4.2,
      "eval_loss": 0.4911543130874634,
      "eval_runtime": 111.7063,
      "eval_samples_per_second": 12.989,
      "eval_steps_per_second": 0.815,
      "step": 1200
    },
    {
      "epoch": 4.23,
      "learning_rate": 0.00017978260869565217,
      "loss": 0.4853,
      "step": 1210
    },
    {
      "epoch": 4.27,
      "learning_rate": 0.00017869565217391303,
      "loss": 0.5149,
      "step": 1220
    },
    {
      "epoch": 4.3,
      "learning_rate": 0.00017760869565217388,
      "loss": 0.4608,
      "step": 1230
    },
    {
      "epoch": 4.34,
      "learning_rate": 0.00017652173913043477,
      "loss": 0.4917,
      "step": 1240
    },
    {
      "epoch": 4.37,
      "learning_rate": 0.00017543478260869562,
      "loss": 0.4615,
      "step": 1250
    },
    {
      "epoch": 4.4,
      "learning_rate": 0.00017434782608695653,
      "loss": 0.4175,
      "step": 1260
    },
    {
      "epoch": 4.44,
      "learning_rate": 0.0001732608695652174,
      "loss": 0.4752,
      "step": 1270
    },
    {
      "epoch": 4.47,
      "learning_rate": 0.00017217391304347825,
      "loss": 0.4979,
      "step": 1280
    },
    {
      "epoch": 4.51,
      "learning_rate": 0.00017108695652173913,
      "loss": 0.4272,
      "step": 1290
    },
    {
      "epoch": 4.54,
      "learning_rate": 0.00016999999999999999,
      "loss": 0.5067,
      "step": 1300
    },
    {
      "epoch": 4.54,
      "eval_loss": 0.4602111279964447,
      "eval_runtime": 111.5487,
      "eval_samples_per_second": 13.008,
      "eval_steps_per_second": 0.816,
      "step": 1300
    },
    {
      "epoch": 4.58,
      "learning_rate": 0.00016891304347826084,
      "loss": 0.483,
      "step": 1310
    },
    {
      "epoch": 4.61,
      "learning_rate": 0.0001678260869565217,
      "loss": 0.4671,
      "step": 1320
    },
    {
      "epoch": 4.65,
      "learning_rate": 0.0001667391304347826,
      "loss": 0.4806,
      "step": 1330
    },
    {
      "epoch": 4.68,
      "learning_rate": 0.00016565217391304346,
      "loss": 0.4798,
      "step": 1340
    },
    {
      "epoch": 4.72,
      "learning_rate": 0.00016456521739130435,
      "loss": 0.5021,
      "step": 1350
    },
    {
      "epoch": 4.75,
      "learning_rate": 0.0001634782608695652,
      "loss": 0.4445,
      "step": 1360
    },
    {
      "epoch": 4.79,
      "learning_rate": 0.00016239130434782606,
      "loss": 0.4462,
      "step": 1370
    },
    {
      "epoch": 4.82,
      "learning_rate": 0.00016130434782608694,
      "loss": 0.5033,
      "step": 1380
    },
    {
      "epoch": 4.86,
      "learning_rate": 0.0001602173913043478,
      "loss": 0.4577,
      "step": 1390
    },
    {
      "epoch": 4.89,
      "learning_rate": 0.0001591304347826087,
      "loss": 0.4999,
      "step": 1400
    },
    {
      "epoch": 4.89,
      "eval_loss": 0.4281094968318939,
      "eval_runtime": 111.6556,
      "eval_samples_per_second": 12.995,
      "eval_steps_per_second": 0.815,
      "step": 1400
    },
    {
      "epoch": 4.93,
      "learning_rate": 0.00015804347826086956,
      "loss": 0.4366,
      "step": 1410
    },
    {
      "epoch": 4.96,
      "learning_rate": 0.00015695652173913042,
      "loss": 0.4696,
      "step": 1420
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.00015586956521739128,
      "loss": 0.491,
      "step": 1430
    },
    {
      "epoch": 5.03,
      "learning_rate": 0.00015478260869565216,
      "loss": 0.4202,
      "step": 1440
    },
    {
      "epoch": 5.07,
      "learning_rate": 0.00015369565217391302,
      "loss": 0.4307,
      "step": 1450
    },
    {
      "epoch": 5.1,
      "learning_rate": 0.00015260869565217387,
      "loss": 0.4103,
      "step": 1460
    },
    {
      "epoch": 5.14,
      "learning_rate": 0.00015152173913043478,
      "loss": 0.4157,
      "step": 1470
    },
    {
      "epoch": 5.17,
      "learning_rate": 0.00015043478260869564,
      "loss": 0.4326,
      "step": 1480
    },
    {
      "epoch": 5.21,
      "learning_rate": 0.00014934782608695652,
      "loss": 0.3842,
      "step": 1490
    },
    {
      "epoch": 5.24,
      "learning_rate": 0.00014826086956521738,
      "loss": 0.4072,
      "step": 1500
    },
    {
      "epoch": 5.24,
      "eval_loss": 0.416363924741745,
      "eval_runtime": 111.6811,
      "eval_samples_per_second": 12.992,
      "eval_steps_per_second": 0.815,
      "step": 1500
    },
    {
      "epoch": 5.28,
      "learning_rate": 0.00014717391304347823,
      "loss": 0.4093,
      "step": 1510
    },
    {
      "epoch": 5.31,
      "learning_rate": 0.00014608695652173912,
      "loss": 0.4013,
      "step": 1520
    },
    {
      "epoch": 5.35,
      "learning_rate": 0.000145,
      "loss": 0.4534,
      "step": 1530
    },
    {
      "epoch": 5.38,
      "learning_rate": 0.00014391304347826086,
      "loss": 0.3509,
      "step": 1540
    },
    {
      "epoch": 5.42,
      "learning_rate": 0.00014282608695652174,
      "loss": 0.4058,
      "step": 1550
    },
    {
      "epoch": 5.45,
      "learning_rate": 0.0001417391304347826,
      "loss": 0.3956,
      "step": 1560
    },
    {
      "epoch": 5.49,
      "learning_rate": 0.00014065217391304345,
      "loss": 0.415,
      "step": 1570
    },
    {
      "epoch": 5.52,
      "learning_rate": 0.00013956521739130434,
      "loss": 0.4255,
      "step": 1580
    },
    {
      "epoch": 5.56,
      "learning_rate": 0.00013847826086956522,
      "loss": 0.3925,
      "step": 1590
    },
    {
      "epoch": 5.59,
      "learning_rate": 0.00013739130434782607,
      "loss": 0.3855,
      "step": 1600
    },
    {
      "epoch": 5.59,
      "eval_loss": 0.43422192335128784,
      "eval_runtime": 111.3707,
      "eval_samples_per_second": 13.029,
      "eval_steps_per_second": 0.817,
      "step": 1600
    },
    {
      "epoch": 5.63,
      "learning_rate": 0.00013630434782608693,
      "loss": 0.4364,
      "step": 1610
    },
    {
      "epoch": 5.66,
      "learning_rate": 0.00013521739130434781,
      "loss": 0.3908,
      "step": 1620
    },
    {
      "epoch": 5.7,
      "learning_rate": 0.0001341304347826087,
      "loss": 0.4419,
      "step": 1630
    },
    {
      "epoch": 5.73,
      "learning_rate": 0.00013304347826086955,
      "loss": 0.3667,
      "step": 1640
    },
    {
      "epoch": 5.77,
      "learning_rate": 0.0001319565217391304,
      "loss": 0.3971,
      "step": 1650
    },
    {
      "epoch": 5.8,
      "learning_rate": 0.0001308695652173913,
      "loss": 0.4102,
      "step": 1660
    },
    {
      "epoch": 5.84,
      "learning_rate": 0.00012978260869565218,
      "loss": 0.4063,
      "step": 1670
    },
    {
      "epoch": 5.87,
      "learning_rate": 0.00012869565217391303,
      "loss": 0.4576,
      "step": 1680
    },
    {
      "epoch": 5.91,
      "learning_rate": 0.00012760869565217392,
      "loss": 0.3524,
      "step": 1690
    },
    {
      "epoch": 5.94,
      "learning_rate": 0.00012652173913043477,
      "loss": 0.4075,
      "step": 1700
    },
    {
      "epoch": 5.94,
      "eval_loss": 0.40860703587532043,
      "eval_runtime": 112.0658,
      "eval_samples_per_second": 12.948,
      "eval_steps_per_second": 0.812,
      "step": 1700
    },
    {
      "epoch": 5.98,
      "learning_rate": 0.00012543478260869563,
      "loss": 0.3744,
      "step": 1710
    },
    {
      "epoch": 6.01,
      "learning_rate": 0.0001243478260869565,
      "loss": 0.3863,
      "step": 1720
    },
    {
      "epoch": 6.05,
      "learning_rate": 0.0001232608695652174,
      "loss": 0.338,
      "step": 1730
    },
    {
      "epoch": 6.08,
      "learning_rate": 0.00012217391304347825,
      "loss": 0.3789,
      "step": 1740
    },
    {
      "epoch": 6.12,
      "learning_rate": 0.00012108695652173912,
      "loss": 0.3253,
      "step": 1750
    },
    {
      "epoch": 6.15,
      "learning_rate": 0.00011999999999999999,
      "loss": 0.3878,
      "step": 1760
    },
    {
      "epoch": 6.19,
      "learning_rate": 0.00011891304347826086,
      "loss": 0.3732,
      "step": 1770
    },
    {
      "epoch": 6.22,
      "learning_rate": 0.00011782608695652173,
      "loss": 0.3494,
      "step": 1780
    },
    {
      "epoch": 6.26,
      "learning_rate": 0.00011673913043478258,
      "loss": 0.3701,
      "step": 1790
    },
    {
      "epoch": 6.29,
      "learning_rate": 0.00011565217391304347,
      "loss": 0.347,
      "step": 1800
    },
    {
      "epoch": 6.29,
      "eval_loss": 0.42197009921073914,
      "eval_runtime": 111.9078,
      "eval_samples_per_second": 12.966,
      "eval_steps_per_second": 0.813,
      "step": 1800
    },
    {
      "epoch": 6.33,
      "learning_rate": 0.00011456521739130434,
      "loss": 0.3484,
      "step": 1810
    },
    {
      "epoch": 6.36,
      "learning_rate": 0.00011347826086956521,
      "loss": 0.3328,
      "step": 1820
    },
    {
      "epoch": 6.4,
      "learning_rate": 0.00011239130434782608,
      "loss": 0.3444,
      "step": 1830
    },
    {
      "epoch": 6.43,
      "learning_rate": 0.00011130434782608695,
      "loss": 0.321,
      "step": 1840
    },
    {
      "epoch": 6.47,
      "learning_rate": 0.00011021739130434782,
      "loss": 0.4089,
      "step": 1850
    },
    {
      "epoch": 6.5,
      "learning_rate": 0.00010913043478260867,
      "loss": 0.335,
      "step": 1860
    },
    {
      "epoch": 6.54,
      "learning_rate": 0.00010804347826086956,
      "loss": 0.376,
      "step": 1870
    },
    {
      "epoch": 6.57,
      "learning_rate": 0.00010695652173913043,
      "loss": 0.3324,
      "step": 1880
    },
    {
      "epoch": 6.61,
      "learning_rate": 0.0001058695652173913,
      "loss": 0.3359,
      "step": 1890
    },
    {
      "epoch": 6.64,
      "learning_rate": 0.00010478260869565216,
      "loss": 0.3708,
      "step": 1900
    },
    {
      "epoch": 6.64,
      "eval_loss": 0.4211980998516083,
      "eval_runtime": 113.1099,
      "eval_samples_per_second": 12.828,
      "eval_steps_per_second": 0.805,
      "step": 1900
    },
    {
      "epoch": 6.68,
      "learning_rate": 0.00010369565217391303,
      "loss": 0.3512,
      "step": 1910
    },
    {
      "epoch": 6.71,
      "learning_rate": 0.0001026086956521739,
      "loss": 0.3381,
      "step": 1920
    },
    {
      "epoch": 6.75,
      "learning_rate": 0.00010152173913043479,
      "loss": 0.2925,
      "step": 1930
    },
    {
      "epoch": 6.78,
      "learning_rate": 0.00010043478260869564,
      "loss": 0.3805,
      "step": 1940
    },
    {
      "epoch": 6.82,
      "learning_rate": 9.934782608695651e-05,
      "loss": 0.3803,
      "step": 1950
    },
    {
      "epoch": 6.85,
      "learning_rate": 9.826086956521738e-05,
      "loss": 0.3499,
      "step": 1960
    },
    {
      "epoch": 6.89,
      "learning_rate": 9.717391304347825e-05,
      "loss": 0.3419,
      "step": 1970
    },
    {
      "epoch": 6.92,
      "learning_rate": 9.608695652173912e-05,
      "loss": 0.2918,
      "step": 1980
    },
    {
      "epoch": 6.96,
      "learning_rate": 9.499999999999999e-05,
      "loss": 0.3602,
      "step": 1990
    },
    {
      "epoch": 6.99,
      "learning_rate": 9.391304347826087e-05,
      "loss": 0.3788,
      "step": 2000
    },
    {
      "epoch": 6.99,
      "eval_loss": 0.419994056224823,
      "eval_runtime": 111.5298,
      "eval_samples_per_second": 13.01,
      "eval_steps_per_second": 0.816,
      "step": 2000
    },
    {
      "epoch": 7.03,
      "learning_rate": 9.282608695652173e-05,
      "loss": 0.3368,
      "step": 2010
    },
    {
      "epoch": 7.06,
      "learning_rate": 9.17391304347826e-05,
      "loss": 0.3186,
      "step": 2020
    },
    {
      "epoch": 7.1,
      "learning_rate": 9.065217391304346e-05,
      "loss": 0.3134,
      "step": 2030
    },
    {
      "epoch": 7.13,
      "learning_rate": 8.956521739130434e-05,
      "loss": 0.3021,
      "step": 2040
    },
    {
      "epoch": 7.17,
      "learning_rate": 8.847826086956521e-05,
      "loss": 0.3168,
      "step": 2050
    },
    {
      "epoch": 7.2,
      "learning_rate": 8.739130434782608e-05,
      "loss": 0.2573,
      "step": 2060
    },
    {
      "epoch": 7.24,
      "learning_rate": 8.630434782608696e-05,
      "loss": 0.3107,
      "step": 2070
    },
    {
      "epoch": 7.27,
      "learning_rate": 8.521739130434782e-05,
      "loss": 0.2961,
      "step": 2080
    },
    {
      "epoch": 7.31,
      "learning_rate": 8.413043478260869e-05,
      "loss": 0.3487,
      "step": 2090
    },
    {
      "epoch": 7.34,
      "learning_rate": 8.304347826086954e-05,
      "loss": 0.337,
      "step": 2100
    },
    {
      "epoch": 7.34,
      "eval_loss": 0.41391661763191223,
      "eval_runtime": 111.525,
      "eval_samples_per_second": 13.011,
      "eval_steps_per_second": 0.816,
      "step": 2100
    },
    {
      "epoch": 7.38,
      "learning_rate": 8.195652173913043e-05,
      "loss": 0.3379,
      "step": 2110
    },
    {
      "epoch": 7.41,
      "learning_rate": 8.08695652173913e-05,
      "loss": 0.3168,
      "step": 2120
    },
    {
      "epoch": 7.45,
      "learning_rate": 7.978260869565217e-05,
      "loss": 0.3339,
      "step": 2130
    },
    {
      "epoch": 7.48,
      "learning_rate": 7.869565217391304e-05,
      "loss": 0.3712,
      "step": 2140
    },
    {
      "epoch": 7.52,
      "learning_rate": 7.76086956521739e-05,
      "loss": 0.2807,
      "step": 2150
    },
    {
      "epoch": 7.55,
      "learning_rate": 7.652173913043478e-05,
      "loss": 0.3018,
      "step": 2160
    },
    {
      "epoch": 7.59,
      "learning_rate": 7.543478260869563e-05,
      "loss": 0.311,
      "step": 2170
    },
    {
      "epoch": 7.62,
      "learning_rate": 7.434782608695652e-05,
      "loss": 0.309,
      "step": 2180
    },
    {
      "epoch": 7.66,
      "learning_rate": 7.326086956521738e-05,
      "loss": 0.3175,
      "step": 2190
    },
    {
      "epoch": 7.69,
      "learning_rate": 7.217391304347825e-05,
      "loss": 0.3045,
      "step": 2200
    },
    {
      "epoch": 7.69,
      "eval_loss": 0.4140127897262573,
      "eval_runtime": 111.986,
      "eval_samples_per_second": 12.957,
      "eval_steps_per_second": 0.813,
      "step": 2200
    },
    {
      "epoch": 7.73,
      "learning_rate": 7.108695652173912e-05,
      "loss": 0.3065,
      "step": 2210
    },
    {
      "epoch": 7.76,
      "learning_rate": 7e-05,
      "loss": 0.2784,
      "step": 2220
    },
    {
      "epoch": 7.8,
      "learning_rate": 6.891304347826086e-05,
      "loss": 0.3367,
      "step": 2230
    },
    {
      "epoch": 7.83,
      "learning_rate": 6.782608695652173e-05,
      "loss": 0.3176,
      "step": 2240
    },
    {
      "epoch": 7.87,
      "learning_rate": 6.67391304347826e-05,
      "loss": 0.3126,
      "step": 2250
    },
    {
      "epoch": 7.9,
      "learning_rate": 6.565217391304347e-05,
      "loss": 0.3586,
      "step": 2260
    },
    {
      "epoch": 7.94,
      "learning_rate": 6.456521739130434e-05,
      "loss": 0.3199,
      "step": 2270
    },
    {
      "epoch": 7.97,
      "learning_rate": 6.347826086956521e-05,
      "loss": 0.3302,
      "step": 2280
    },
    {
      "epoch": 8.01,
      "learning_rate": 6.239130434782608e-05,
      "loss": 0.346,
      "step": 2290
    },
    {
      "epoch": 8.04,
      "learning_rate": 6.130434782608695e-05,
      "loss": 0.2547,
      "step": 2300
    },
    {
      "epoch": 8.04,
      "eval_loss": 0.41661086678504944,
      "eval_runtime": 111.4943,
      "eval_samples_per_second": 13.014,
      "eval_steps_per_second": 0.816,
      "step": 2300
    },
    {
      "epoch": 8.08,
      "learning_rate": 6.021739130434782e-05,
      "loss": 0.2923,
      "step": 2310
    },
    {
      "epoch": 8.11,
      "learning_rate": 5.913043478260869e-05,
      "loss": 0.2889,
      "step": 2320
    },
    {
      "epoch": 8.15,
      "learning_rate": 5.804347826086956e-05,
      "loss": 0.2858,
      "step": 2330
    },
    {
      "epoch": 8.18,
      "learning_rate": 5.695652173913043e-05,
      "loss": 0.3082,
      "step": 2340
    },
    {
      "epoch": 8.22,
      "learning_rate": 5.58695652173913e-05,
      "loss": 0.2721,
      "step": 2350
    },
    {
      "epoch": 8.25,
      "learning_rate": 5.478260869565217e-05,
      "loss": 0.2911,
      "step": 2360
    },
    {
      "epoch": 8.29,
      "learning_rate": 5.3695652173913046e-05,
      "loss": 0.3015,
      "step": 2370
    },
    {
      "epoch": 8.32,
      "learning_rate": 5.260869565217391e-05,
      "loss": 0.3471,
      "step": 2380
    },
    {
      "epoch": 8.36,
      "learning_rate": 5.152173913043478e-05,
      "loss": 0.3053,
      "step": 2390
    },
    {
      "epoch": 8.39,
      "learning_rate": 5.043478260869565e-05,
      "loss": 0.2584,
      "step": 2400
    },
    {
      "epoch": 8.39,
      "eval_loss": 0.408309668302536,
      "eval_runtime": 111.5385,
      "eval_samples_per_second": 13.009,
      "eval_steps_per_second": 0.816,
      "step": 2400
    },
    {
      "epoch": 8.43,
      "learning_rate": 4.934782608695652e-05,
      "loss": 0.3029,
      "step": 2410
    },
    {
      "epoch": 8.46,
      "learning_rate": 4.826086956521738e-05,
      "loss": 0.2933,
      "step": 2420
    },
    {
      "epoch": 8.5,
      "learning_rate": 4.717391304347826e-05,
      "loss": 0.2807,
      "step": 2430
    },
    {
      "epoch": 8.53,
      "learning_rate": 4.6086956521739126e-05,
      "loss": 0.2899,
      "step": 2440
    },
    {
      "epoch": 8.57,
      "learning_rate": 4.4999999999999996e-05,
      "loss": 0.229,
      "step": 2450
    },
    {
      "epoch": 8.6,
      "learning_rate": 4.3913043478260866e-05,
      "loss": 0.2919,
      "step": 2460
    },
    {
      "epoch": 8.64,
      "learning_rate": 4.2826086956521735e-05,
      "loss": 0.2989,
      "step": 2470
    },
    {
      "epoch": 8.67,
      "learning_rate": 4.1739130434782605e-05,
      "loss": 0.2919,
      "step": 2480
    },
    {
      "epoch": 8.71,
      "learning_rate": 4.065217391304348e-05,
      "loss": 0.2988,
      "step": 2490
    },
    {
      "epoch": 8.74,
      "learning_rate": 3.9565217391304344e-05,
      "loss": 0.2673,
      "step": 2500
    },
    {
      "epoch": 8.74,
      "eval_loss": 0.39945822954177856,
      "eval_runtime": 111.4334,
      "eval_samples_per_second": 13.021,
      "eval_steps_per_second": 0.817,
      "step": 2500
    },
    {
      "epoch": 8.77,
      "learning_rate": 3.8478260869565214e-05,
      "loss": 0.2668,
      "step": 2510
    },
    {
      "epoch": 8.81,
      "learning_rate": 3.7391304347826084e-05,
      "loss": 0.3159,
      "step": 2520
    },
    {
      "epoch": 8.84,
      "learning_rate": 3.630434782608695e-05,
      "loss": 0.3037,
      "step": 2530
    },
    {
      "epoch": 8.88,
      "learning_rate": 3.521739130434782e-05,
      "loss": 0.2869,
      "step": 2540
    },
    {
      "epoch": 8.91,
      "learning_rate": 3.413043478260869e-05,
      "loss": 0.2792,
      "step": 2550
    },
    {
      "epoch": 8.95,
      "learning_rate": 3.304347826086956e-05,
      "loss": 0.2939,
      "step": 2560
    },
    {
      "epoch": 8.98,
      "learning_rate": 3.195652173913043e-05,
      "loss": 0.2781,
      "step": 2570
    },
    {
      "epoch": 9.02,
      "learning_rate": 3.08695652173913e-05,
      "loss": 0.3154,
      "step": 2580
    },
    {
      "epoch": 9.06,
      "learning_rate": 2.978260869565217e-05,
      "loss": 0.301,
      "step": 2590
    },
    {
      "epoch": 9.09,
      "learning_rate": 2.869565217391304e-05,
      "loss": 0.2765,
      "step": 2600
    },
    {
      "epoch": 9.09,
      "eval_loss": 0.4026023745536804,
      "eval_runtime": 111.3595,
      "eval_samples_per_second": 13.03,
      "eval_steps_per_second": 0.817,
      "step": 2600
    },
    {
      "epoch": 9.13,
      "learning_rate": 2.760869565217391e-05,
      "loss": 0.2801,
      "step": 2610
    },
    {
      "epoch": 9.16,
      "learning_rate": 2.652173913043478e-05,
      "loss": 0.263,
      "step": 2620
    },
    {
      "epoch": 9.2,
      "learning_rate": 2.543478260869565e-05,
      "loss": 0.2586,
      "step": 2630
    },
    {
      "epoch": 9.23,
      "learning_rate": 2.4347826086956516e-05,
      "loss": 0.2496,
      "step": 2640
    },
    {
      "epoch": 9.27,
      "learning_rate": 2.3260869565217393e-05,
      "loss": 0.269,
      "step": 2650
    },
    {
      "epoch": 9.3,
      "learning_rate": 2.217391304347826e-05,
      "loss": 0.296,
      "step": 2660
    },
    {
      "epoch": 9.34,
      "learning_rate": 2.108695652173913e-05,
      "loss": 0.2531,
      "step": 2670
    },
    {
      "epoch": 9.37,
      "learning_rate": 1.9999999999999998e-05,
      "loss": 0.2871,
      "step": 2680
    },
    {
      "epoch": 9.4,
      "learning_rate": 1.8913043478260868e-05,
      "loss": 0.2616,
      "step": 2690
    },
    {
      "epoch": 9.44,
      "learning_rate": 1.7826086956521738e-05,
      "loss": 0.2453,
      "step": 2700
    },
    {
      "epoch": 9.44,
      "eval_loss": 0.4079287052154541,
      "eval_runtime": 111.4001,
      "eval_samples_per_second": 13.025,
      "eval_steps_per_second": 0.817,
      "step": 2700
    },
    {
      "epoch": 9.47,
      "learning_rate": 1.6739130434782607e-05,
      "loss": 0.3043,
      "step": 2710
    },
    {
      "epoch": 9.51,
      "learning_rate": 1.5652173913043477e-05,
      "loss": 0.2297,
      "step": 2720
    },
    {
      "epoch": 9.54,
      "learning_rate": 1.4565217391304347e-05,
      "loss": 0.2844,
      "step": 2730
    },
    {
      "epoch": 9.58,
      "learning_rate": 1.3478260869565216e-05,
      "loss": 0.2485,
      "step": 2740
    },
    {
      "epoch": 9.61,
      "learning_rate": 1.2391304347826086e-05,
      "loss": 0.2695,
      "step": 2750
    },
    {
      "epoch": 9.65,
      "learning_rate": 1.1304347826086956e-05,
      "loss": 0.3031,
      "step": 2760
    },
    {
      "epoch": 9.68,
      "learning_rate": 1.0217391304347825e-05,
      "loss": 0.2716,
      "step": 2770
    },
    {
      "epoch": 9.72,
      "learning_rate": 9.130434782608695e-06,
      "loss": 0.2424,
      "step": 2780
    },
    {
      "epoch": 9.75,
      "learning_rate": 8.043478260869565e-06,
      "loss": 0.2652,
      "step": 2790
    },
    {
      "epoch": 9.79,
      "learning_rate": 6.956521739130434e-06,
      "loss": 0.2883,
      "step": 2800
    },
    {
      "epoch": 9.79,
      "eval_loss": 0.4038572311401367,
      "eval_runtime": 111.7946,
      "eval_samples_per_second": 12.979,
      "eval_steps_per_second": 0.814,
      "step": 2800
    },
    {
      "epoch": 9.82,
      "learning_rate": 5.869565217391305e-06,
      "loss": 0.2743,
      "step": 2810
    },
    {
      "epoch": 9.86,
      "learning_rate": 4.7826086956521735e-06,
      "loss": 0.2823,
      "step": 2820
    },
    {
      "epoch": 9.89,
      "learning_rate": 3.695652173913043e-06,
      "loss": 0.2766,
      "step": 2830
    },
    {
      "epoch": 9.93,
      "learning_rate": 2.608695652173913e-06,
      "loss": 0.2881,
      "step": 2840
    },
    {
      "epoch": 9.96,
      "learning_rate": 1.5217391304347823e-06,
      "loss": 0.3219,
      "step": 2850
    },
    {
      "epoch": 10.0,
      "learning_rate": 4.3478260869565214e-07,
      "loss": 0.294,
      "step": 2860
    },
    {
      "epoch": 10.0,
      "step": 2860,
      "total_flos": 6.827433558323335e+18,
      "train_loss": 1.1227069218675574,
      "train_runtime": 9275.3367,
      "train_samples_per_second": 4.936,
      "train_steps_per_second": 0.308
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.40599533915519714,
      "eval_runtime": 111.6061,
      "eval_samples_per_second": 13.001,
      "eval_steps_per_second": 0.815,
      "step": 2860
    }
  ],
  "max_steps": 2860,
  "num_train_epochs": 10,
  "total_flos": 6.827433558323335e+18,
  "trial_name": null,
  "trial_params": null
}