{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.10345,
  "eval_steps": 400,
  "global_step": 14000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5e-05,
      "loss": 2.4874,
      "step": 25
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.0001,
      "loss": 1.5625,
      "step": 50
    },
    {
      "epoch": 0.0,
      "learning_rate": 0.00015,
      "loss": 0.8084,
      "step": 75
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.181,
      "step": 100
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00025,
      "loss": 0.1378,
      "step": 125
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0003,
      "loss": 0.1267,
      "step": 150
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00035,
      "loss": 0.1118,
      "step": 175
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0004,
      "loss": 0.1189,
      "step": 200
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00045000000000000004,
      "loss": 0.11,
      "step": 225
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0005,
      "loss": 0.1131,
      "step": 250
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.00055,
      "loss": 0.1196,
      "step": 275
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0006,
      "loss": 0.0998,
      "step": 300
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0006500000000000001,
      "loss": 0.1087,
      "step": 325
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0007,
      "loss": 0.1098,
      "step": 350
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00075,
      "loss": 0.1138,
      "step": 375
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0008,
      "loss": 0.1212,
      "step": 400
    },
    {
      "epoch": 0.02,
      "eval_loss": 0.5537986159324646,
      "eval_runtime": 216.4127,
      "eval_samples_per_second": 4.621,
      "eval_steps_per_second": 0.291,
      "step": 400
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00085,
      "loss": 0.1228,
      "step": 425
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0009000000000000001,
      "loss": 0.1177,
      "step": 450
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00095,
      "loss": 0.1196,
      "step": 475
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.001,
      "loss": 0.1214,
      "step": 500
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0009987179487179487,
      "loss": 0.1415,
      "step": 525
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0009974358974358974,
      "loss": 0.1491,
      "step": 550
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0009961538461538463,
      "loss": 0.1467,
      "step": 575
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.000994871794871795,
      "loss": 0.1455,
      "step": 600
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0009935897435897436,
      "loss": 0.1393,
      "step": 625
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0009923076923076923,
      "loss": 0.1355,
      "step": 650
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0009910256410256412,
      "loss": 0.1299,
      "step": 675
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0009897435897435899,
      "loss": 0.124,
      "step": 700
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0009884615384615385,
      "loss": 0.1209,
      "step": 725
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0009871794871794872,
      "loss": 0.1247,
      "step": 750
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0009858974358974359,
      "loss": 0.1238,
      "step": 775
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0009846153846153848,
      "loss": 0.1211,
      "step": 800
    },
    {
      "epoch": 0.04,
      "eval_loss": 1.2810015678405762,
      "eval_runtime": 217.0046,
      "eval_samples_per_second": 4.608,
      "eval_steps_per_second": 0.29,
      "step": 800
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0009833333333333332,
      "loss": 0.1226,
      "step": 825
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0009820512820512821,
      "loss": 0.1267,
      "step": 850
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.000980820512820513,
      "loss": 0.1039,
      "step": 875
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0009795384615384616,
      "loss": 0.1142,
      "step": 900
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0009782564102564103,
      "loss": 0.1062,
      "step": 925
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.000976974358974359,
      "loss": 0.115,
      "step": 950
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0009756923076923077,
      "loss": 0.1131,
      "step": 975
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0009744102564102564,
      "loss": 0.1173,
      "step": 1000
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0009731282051282051,
      "loss": 0.1114,
      "step": 1025
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0009718461538461539,
      "loss": 0.1123,
      "step": 1050
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0009705641025641025,
      "loss": 0.1258,
      "step": 1075
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0009692820512820512,
      "loss": 0.1168,
      "step": 1100
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.000968,
      "loss": 0.1193,
      "step": 1125
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0009667179487179487,
      "loss": 0.1216,
      "step": 1150
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0009654358974358975,
      "loss": 0.1257,
      "step": 1175
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0009641538461538461,
      "loss": 0.1219,
      "step": 1200
    },
    {
      "epoch": 0.06,
      "eval_loss": 1.4006367921829224,
      "eval_runtime": 216.1116,
      "eval_samples_per_second": 4.627,
      "eval_steps_per_second": 0.292,
      "step": 1200
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0009628717948717949,
      "loss": 0.1096,
      "step": 1225
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0009615897435897436,
      "loss": 0.1234,
      "step": 1250
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0009603076923076923,
      "loss": 0.1406,
      "step": 1275
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0009590256410256411,
      "loss": 0.1189,
      "step": 1300
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0009577435897435897,
      "loss": 0.1211,
      "step": 1325
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0009564615384615385,
      "loss": 0.1153,
      "step": 1350
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0009551794871794872,
      "loss": 0.1161,
      "step": 1375
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.000953897435897436,
      "loss": 0.1179,
      "step": 1400
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0009526153846153847,
      "loss": 0.1102,
      "step": 1425
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0009513333333333334,
      "loss": 0.123,
      "step": 1450
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0009500512820512821,
      "loss": 0.1386,
      "step": 1475
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0009487692307692308,
      "loss": 0.0922,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0009474871794871796,
      "loss": 0.1243,
      "step": 1525
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0009462051282051282,
      "loss": 0.0956,
      "step": 1550
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.000944923076923077,
      "loss": 0.1027,
      "step": 1575
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0009436410256410256,
      "loss": 0.1142,
      "step": 1600
    },
    {
      "epoch": 0.08,
      "eval_loss": 0.10953158885240555,
      "eval_runtime": 216.1306,
      "eval_samples_per_second": 4.627,
      "eval_steps_per_second": 0.291,
      "step": 1600
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0009423589743589744,
      "loss": 0.117,
      "step": 1625
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0009410769230769231,
      "loss": 0.1146,
      "step": 1650
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0009397948717948717,
      "loss": 0.0974,
      "step": 1675
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0009385128205128205,
      "loss": 0.1041,
      "step": 1700
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0009372307692307692,
      "loss": 0.1048,
      "step": 1725
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.000935948717948718,
      "loss": 0.0934,
      "step": 1750
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0009346666666666667,
      "loss": 0.0982,
      "step": 1775
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0009333846153846154,
      "loss": 0.1014,
      "step": 1800
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0009321025641025641,
      "loss": 0.1304,
      "step": 1825
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0009308205128205128,
      "loss": 0.0995,
      "step": 1850
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0009295384615384616,
      "loss": 0.0912,
      "step": 1875
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0009282564102564102,
      "loss": 0.0963,
      "step": 1900
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.000926974358974359,
      "loss": 0.1051,
      "step": 1925
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0009256923076923077,
      "loss": 0.0987,
      "step": 1950
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0009244102564102565,
      "loss": 0.0974,
      "step": 1975
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0009231282051282052,
      "loss": 0.1087,
      "step": 2000
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.10780761390924454,
      "eval_runtime": 216.1847,
      "eval_samples_per_second": 4.626,
      "eval_steps_per_second": 0.291,
      "step": 2000
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0009218461538461538,
      "loss": 0.1035,
      "step": 2025
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0009205641025641026,
      "loss": 0.1094,
      "step": 2050
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0009192820512820513,
      "loss": 0.0851,
      "step": 2075
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0009180000000000001,
      "loss": 0.0741,
      "step": 2100
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0009167179487179488,
      "loss": 0.0881,
      "step": 2125
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0009154358974358975,
      "loss": 0.0794,
      "step": 2150
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0009141538461538462,
      "loss": 0.1199,
      "step": 2175
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0009128717948717948,
      "loss": 0.1307,
      "step": 2200
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0009115897435897436,
      "loss": 0.108,
      "step": 2225
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0009103076923076923,
      "loss": 0.078,
      "step": 2250
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.000909025641025641,
      "loss": 0.0853,
      "step": 2275
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0009077435897435897,
      "loss": 0.0852,
      "step": 2300
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0009064615384615385,
      "loss": 0.0856,
      "step": 2325
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0009051794871794872,
      "loss": 0.0663,
      "step": 2350
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0009038974358974358,
      "loss": 0.0818,
      "step": 2375
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0009026153846153846,
      "loss": 0.0724,
      "step": 2400
    },
    {
      "epoch": 0.12,
      "eval_loss": 0.11282841861248016,
      "eval_runtime": 215.7164,
      "eval_samples_per_second": 4.636,
      "eval_steps_per_second": 0.292,
      "step": 2400
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0009013333333333333,
      "loss": 0.0824,
      "step": 2425
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0009000512820512821,
      "loss": 0.091,
      "step": 2450
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0008987692307692308,
      "loss": 0.0697,
      "step": 2475
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0008974871794871795,
      "loss": 0.076,
      "step": 2500
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0008962051282051282,
      "loss": 0.0769,
      "step": 2525
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0008949230769230769,
      "loss": 0.0724,
      "step": 2550
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0008936410256410257,
      "loss": 0.068,
      "step": 2575
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0008923589743589744,
      "loss": 0.0632,
      "step": 2600
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0008910769230769231,
      "loss": 0.0825,
      "step": 2625
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0008897948717948718,
      "loss": 0.0831,
      "step": 2650
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0008885128205128206,
      "loss": 0.0689,
      "step": 2675
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0008872307692307693,
      "loss": 0.0741,
      "step": 2700
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.000885948717948718,
      "loss": 0.0798,
      "step": 2725
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0008846666666666667,
      "loss": 0.0817,
      "step": 2750
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0008833846153846154,
      "loss": 0.081,
      "step": 2775
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0008821025641025642,
      "loss": 0.0867,
      "step": 2800
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.10396511852741241,
      "eval_runtime": 216.3083,
      "eval_samples_per_second": 4.623,
      "eval_steps_per_second": 0.291,
      "step": 2800
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0008808205128205128,
      "loss": 0.091,
      "step": 2825
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0008795384615384616,
      "loss": 0.0801,
      "step": 2850
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0008782564102564102,
      "loss": 0.1067,
      "step": 2875
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0008769743589743589,
      "loss": 0.0848,
      "step": 2900
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0008756923076923077,
      "loss": 0.0846,
      "step": 2925
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0008744102564102564,
      "loss": 0.076,
      "step": 2950
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0008731282051282051,
      "loss": 0.0729,
      "step": 2975
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0008718461538461538,
      "loss": 0.0994,
      "step": 3000
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0008705641025641026,
      "loss": 0.0901,
      "step": 3025
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0008692820512820513,
      "loss": 0.0735,
      "step": 3050
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0008680000000000001,
      "loss": 0.0796,
      "step": 3075
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0008667179487179487,
      "loss": 0.0888,
      "step": 3100
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0008654358974358974,
      "loss": 0.09,
      "step": 3125
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0008641538461538462,
      "loss": 0.0698,
      "step": 3150
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0008628717948717949,
      "loss": 0.074,
      "step": 3175
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0008615897435897437,
      "loss": 0.0644,
      "step": 3200
    },
    {
      "epoch": 0.16,
      "eval_loss": 0.09922421723604202,
      "eval_runtime": 218.0471,
      "eval_samples_per_second": 4.586,
      "eval_steps_per_second": 0.289,
      "step": 3200
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0008603076923076923,
      "loss": 0.0788,
      "step": 3225
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0008590256410256411,
      "loss": 0.0797,
      "step": 3250
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0008577948717948718,
      "loss": 0.11,
      "step": 3275
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0008565128205128206,
      "loss": 0.1003,
      "step": 3300
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0008552307692307693,
      "loss": 0.091,
      "step": 3325
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0008539487179487179,
      "loss": 0.1001,
      "step": 3350
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0008526666666666667,
      "loss": 0.087,
      "step": 3375
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0008513846153846154,
      "loss": 0.0857,
      "step": 3400
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0008501025641025642,
      "loss": 0.1035,
      "step": 3425
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0008488205128205129,
      "loss": 0.0913,
      "step": 3450
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0008475384615384616,
      "loss": 0.1002,
      "step": 3475
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0008462564102564103,
      "loss": 0.1144,
      "step": 3500
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.000844974358974359,
      "loss": 0.1348,
      "step": 3525
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0008436923076923078,
      "loss": 0.0958,
      "step": 3550
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0008424102564102565,
      "loss": 0.116,
      "step": 3575
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0008411282051282051,
      "loss": 0.0895,
      "step": 3600
    },
    {
      "epoch": 0.18,
      "eval_loss": 0.1009831354022026,
      "eval_runtime": 216.8096,
      "eval_samples_per_second": 4.612,
      "eval_steps_per_second": 0.291,
      "step": 3600
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0008398461538461538,
      "loss": 0.0824,
      "step": 3625
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0008385641025641026,
      "loss": 0.1105,
      "step": 3650
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0008372820512820513,
      "loss": 0.1328,
      "step": 3675
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0008359999999999999,
      "loss": 0.1774,
      "step": 3700
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0008347179487179487,
      "loss": 0.1158,
      "step": 3725
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0008334358974358974,
      "loss": 0.1067,
      "step": 3750
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0008321538461538462,
      "loss": 0.0855,
      "step": 3775
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0008308717948717949,
      "loss": 0.0917,
      "step": 3800
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0008295897435897436,
      "loss": 0.0945,
      "step": 3825
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0008283076923076923,
      "loss": 0.0863,
      "step": 3850
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.000827025641025641,
      "loss": 0.1196,
      "step": 3875
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0008257435897435898,
      "loss": 0.0989,
      "step": 3900
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0008244615384615385,
      "loss": 0.1076,
      "step": 3925
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0008231794871794872,
      "loss": 0.1138,
      "step": 3950
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0008218974358974359,
      "loss": 0.1353,
      "step": 3975
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0008206153846153847,
      "loss": 0.236,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.9967941641807556,
      "eval_runtime": 215.9827,
      "eval_samples_per_second": 4.63,
      "eval_steps_per_second": 0.292,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0008193846153846154,
      "loss": 0.382,
      "step": 4025
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0008181025641025642,
      "loss": 0.3524,
      "step": 4050
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0008168205128205128,
      "loss": 0.1807,
      "step": 4075
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0008155384615384615,
      "loss": 0.1568,
      "step": 4100
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0008142564102564103,
      "loss": 0.1533,
      "step": 4125
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.000812974358974359,
      "loss": 0.1324,
      "step": 4150
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0008116923076923078,
      "loss": 0.1379,
      "step": 4175
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0008104102564102564,
      "loss": 0.1332,
      "step": 4200
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0008091282051282052,
      "loss": 0.1431,
      "step": 4225
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0008078461538461539,
      "loss": 0.1466,
      "step": 4250
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0008065641025641026,
      "loss": 0.1188,
      "step": 4275
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0008052820512820514,
      "loss": 0.1295,
      "step": 4300
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.000804,
      "loss": 0.1378,
      "step": 4325
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0008027179487179488,
      "loss": 0.1197,
      "step": 4350
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0008014358974358974,
      "loss": 0.1231,
      "step": 4375
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0008001538461538462,
      "loss": 0.1268,
      "step": 4400
    },
    {
      "epoch": 1.02,
      "eval_loss": 1.367616891860962,
      "eval_runtime": 215.6124,
      "eval_samples_per_second": 4.638,
      "eval_steps_per_second": 0.292,
      "step": 4400
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0007988717948717948,
      "loss": 0.1283,
      "step": 4425
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0007975897435897435,
      "loss": 0.112,
      "step": 4450
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0007963076923076923,
      "loss": 0.1102,
      "step": 4475
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.000795025641025641,
      "loss": 0.1225,
      "step": 4500
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0007937435897435898,
      "loss": 0.1155,
      "step": 4525
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0007924615384615384,
      "loss": 0.1176,
      "step": 4550
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0007911794871794872,
      "loss": 0.1049,
      "step": 4575
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0007898974358974359,
      "loss": 0.1204,
      "step": 4600
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0007886153846153847,
      "loss": 0.1112,
      "step": 4625
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.0007873333333333334,
      "loss": 0.1064,
      "step": 4650
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.000786051282051282,
      "loss": 0.1108,
      "step": 4675
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0007847692307692308,
      "loss": 0.1067,
      "step": 4700
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0007834871794871795,
      "loss": 0.0994,
      "step": 4725
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0007822051282051283,
      "loss": 0.105,
      "step": 4750
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.000780923076923077,
      "loss": 0.0964,
      "step": 4775
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0007796410256410257,
      "loss": 0.1118,
      "step": 4800
    },
    {
      "epoch": 1.04,
      "eval_loss": 1.3230507373809814,
      "eval_runtime": 217.4474,
      "eval_samples_per_second": 4.599,
      "eval_steps_per_second": 0.29,
      "step": 4800
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0007783589743589744,
      "loss": 0.1062,
      "step": 4825
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0007770769230769231,
      "loss": 0.0876,
      "step": 4850
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0007757948717948719,
      "loss": 0.0958,
      "step": 4875
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0007745128205128205,
      "loss": 0.0915,
      "step": 4900
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0007732307692307693,
      "loss": 0.0935,
      "step": 4925
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.000771948717948718,
      "loss": 0.0947,
      "step": 4950
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0007706666666666668,
      "loss": 0.0985,
      "step": 4975
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0007693846153846154,
      "loss": 0.0913,
      "step": 5000
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.000768102564102564,
      "loss": 0.0933,
      "step": 5025
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0007668205128205128,
      "loss": 0.0953,
      "step": 5050
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0007655384615384615,
      "loss": 0.0958,
      "step": 5075
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0007642564102564103,
      "loss": 0.0938,
      "step": 5100
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.000762974358974359,
      "loss": 0.0886,
      "step": 5125
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0007616923076923077,
      "loss": 0.0897,
      "step": 5150
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0007604102564102564,
      "loss": 0.0925,
      "step": 5175
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0007591282051282051,
      "loss": 0.0868,
      "step": 5200
    },
    {
      "epoch": 1.06,
      "eval_loss": 1.243537187576294,
      "eval_runtime": 216.704,
      "eval_samples_per_second": 4.615,
      "eval_steps_per_second": 0.291,
      "step": 5200
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0007578461538461539,
      "loss": 0.0964,
      "step": 5225
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0007565641025641025,
      "loss": 0.094,
      "step": 5250
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0007552820512820513,
      "loss": 0.0932,
      "step": 5275
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.000754,
      "loss": 0.0908,
      "step": 5300
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0007527179487179488,
      "loss": 0.0903,
      "step": 5325
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0007514358974358975,
      "loss": 0.0914,
      "step": 5350
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0007501538461538461,
      "loss": 0.0947,
      "step": 5375
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0007488717948717949,
      "loss": 0.0861,
      "step": 5400
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0007475897435897436,
      "loss": 0.087,
      "step": 5425
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0007463076923076924,
      "loss": 0.1056,
      "step": 5450
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0007450256410256411,
      "loss": 0.0759,
      "step": 5475
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0007437435897435898,
      "loss": 0.0963,
      "step": 5500
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0007424615384615385,
      "loss": 0.079,
      "step": 5525
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0007411794871794872,
      "loss": 0.0808,
      "step": 5550
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.000739897435897436,
      "loss": 0.0941,
      "step": 5575
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0007386153846153845,
      "loss": 0.0985,
      "step": 5600
    },
    {
      "epoch": 1.08,
      "eval_loss": 0.10466726869344711,
      "eval_runtime": 220.9516,
      "eval_samples_per_second": 4.526,
      "eval_steps_per_second": 0.285,
      "step": 5600
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0007373333333333333,
      "loss": 0.096,
      "step": 5625
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.000736051282051282,
      "loss": 0.0836,
      "step": 5650
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0007347692307692308,
      "loss": 0.0871,
      "step": 5675
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0007334871794871795,
      "loss": 0.0806,
      "step": 5700
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0007322051282051281,
      "loss": 0.0743,
      "step": 5725
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0007309230769230769,
      "loss": 0.0764,
      "step": 5750
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0007296410256410256,
      "loss": 0.0785,
      "step": 5775
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0007283589743589744,
      "loss": 0.1074,
      "step": 5800
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0007270769230769231,
      "loss": 0.085,
      "step": 5825
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0007257948717948718,
      "loss": 0.0737,
      "step": 5850
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0007245128205128205,
      "loss": 0.0804,
      "step": 5875
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0007232307692307692,
      "loss": 0.0889,
      "step": 5900
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.000721948717948718,
      "loss": 0.0962,
      "step": 5925
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0007206666666666667,
      "loss": 0.0869,
      "step": 5950
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0007193846153846154,
      "loss": 0.0913,
      "step": 5975
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0007181025641025641,
      "loss": 0.086,
      "step": 6000
    },
    {
      "epoch": 1.1,
      "eval_loss": 0.09825620800256729,
      "eval_runtime": 217.8898,
      "eval_samples_per_second": 4.589,
      "eval_steps_per_second": 0.289,
      "step": 6000
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0007168205128205129,
      "loss": 0.0859,
      "step": 6025
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0007155384615384616,
      "loss": 0.0757,
      "step": 6050
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0007142564102564103,
      "loss": 0.0643,
      "step": 6075
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.000712974358974359,
      "loss": 0.0771,
      "step": 6100
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0007116923076923077,
      "loss": 0.0666,
      "step": 6125
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0007104102564102565,
      "loss": 0.0963,
      "step": 6150
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0007091282051282052,
      "loss": 0.1062,
      "step": 6175
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0007078461538461538,
      "loss": 0.0881,
      "step": 6200
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0007065641025641025,
      "loss": 0.0587,
      "step": 6225
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0007052820512820512,
      "loss": 0.0673,
      "step": 6250
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.000704,
      "loss": 0.069,
      "step": 6275
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0007027179487179487,
      "loss": 0.0673,
      "step": 6300
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0007014358974358974,
      "loss": 0.0513,
      "step": 6325
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0007001538461538461,
      "loss": 0.0614,
      "step": 6350
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0006988717948717949,
      "loss": 0.0498,
      "step": 6375
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0006975897435897436,
      "loss": 0.0635,
      "step": 6400
    },
    {
      "epoch": 1.12,
      "eval_loss": 0.09686123579740524,
      "eval_runtime": 217.5581,
      "eval_samples_per_second": 4.596,
      "eval_steps_per_second": 0.29,
      "step": 6400
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0006963076923076924,
      "loss": 0.0656,
      "step": 6425
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.000695025641025641,
      "loss": 0.0574,
      "step": 6450
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0006937435897435897,
      "loss": 0.0583,
      "step": 6475
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0006924615384615385,
      "loss": 0.0594,
      "step": 6500
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0006911794871794872,
      "loss": 0.0607,
      "step": 6525
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.000689897435897436,
      "loss": 0.0544,
      "step": 6550
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0006886153846153846,
      "loss": 0.0466,
      "step": 6575
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0006873333333333334,
      "loss": 0.0574,
      "step": 6600
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0006860512820512821,
      "loss": 0.0655,
      "step": 6625
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0006847692307692308,
      "loss": 0.051,
      "step": 6650
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0006834871794871796,
      "loss": 0.0582,
      "step": 6675
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0006822051282051282,
      "loss": 0.0593,
      "step": 6700
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.000680923076923077,
      "loss": 0.0597,
      "step": 6725
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0006796410256410257,
      "loss": 0.0603,
      "step": 6750
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0006783589743589745,
      "loss": 0.0655,
      "step": 6775
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0006770769230769231,
      "loss": 0.0664,
      "step": 6800
    },
    {
      "epoch": 1.14,
      "eval_loss": 0.0915135070681572,
      "eval_runtime": 217.3938,
      "eval_samples_per_second": 4.6,
      "eval_steps_per_second": 0.29,
      "step": 6800
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0006757948717948717,
      "loss": 0.0604,
      "step": 6825
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0006745128205128205,
      "loss": 0.0765,
      "step": 6850
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0006732307692307692,
      "loss": 0.0622,
      "step": 6875
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.000671948717948718,
      "loss": 0.0572,
      "step": 6900
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0006706666666666666,
      "loss": 0.0423,
      "step": 6925
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0006693846153846154,
      "loss": 0.0504,
      "step": 6950
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0006681025641025641,
      "loss": 0.0747,
      "step": 6975
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0006668205128205128,
      "loss": 0.0651,
      "step": 7000
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0006655384615384616,
      "loss": 0.0549,
      "step": 7025
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0006642564102564102,
      "loss": 0.0512,
      "step": 7050
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.000662974358974359,
      "loss": 0.067,
      "step": 7075
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0006616923076923077,
      "loss": 0.0638,
      "step": 7100
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0006604102564102565,
      "loss": 0.0512,
      "step": 7125
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0006591282051282051,
      "loss": 0.0557,
      "step": 7150
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0006578461538461538,
      "loss": 0.0497,
      "step": 7175
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0006565641025641026,
      "loss": 0.0605,
      "step": 7200
    },
    {
      "epoch": 1.16,
      "eval_loss": 0.09085912257432938,
      "eval_runtime": 217.4684,
      "eval_samples_per_second": 4.598,
      "eval_steps_per_second": 0.29,
      "step": 7200
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0006552820512820513,
      "loss": 0.0573,
      "step": 7225
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0006540000000000001,
      "loss": 0.0714,
      "step": 7250
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0006527179487179487,
      "loss": 0.0813,
      "step": 7275
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0006514358974358975,
      "loss": 0.0664,
      "step": 7300
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0006501538461538462,
      "loss": 0.0789,
      "step": 7325
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0006488717948717949,
      "loss": 0.0711,
      "step": 7350
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0006475897435897437,
      "loss": 0.0635,
      "step": 7375
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0006463076923076923,
      "loss": 0.0794,
      "step": 7400
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.000645025641025641,
      "loss": 0.0742,
      "step": 7425
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0006437435897435897,
      "loss": 0.0763,
      "step": 7450
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0006424615384615385,
      "loss": 0.0783,
      "step": 7475
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0006411794871794871,
      "loss": 0.1092,
      "step": 7500
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0006398974358974358,
      "loss": 0.0765,
      "step": 7525
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0006386153846153846,
      "loss": 0.0926,
      "step": 7550
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0006373333333333333,
      "loss": 0.0669,
      "step": 7575
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0006360512820512821,
      "loss": 0.0645,
      "step": 7600
    },
    {
      "epoch": 1.18,
      "eval_loss": 0.10231851041316986,
      "eval_runtime": 216.1204,
      "eval_samples_per_second": 4.627,
      "eval_steps_per_second": 0.292,
      "step": 7600
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0006347692307692307,
      "loss": 0.0898,
      "step": 7625
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0006334871794871795,
      "loss": 0.103,
      "step": 7650
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0006322051282051282,
      "loss": 0.1355,
      "step": 7675
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0006309230769230769,
      "loss": 0.0953,
      "step": 7700
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0006296410256410257,
      "loss": 0.0807,
      "step": 7725
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0006283589743589743,
      "loss": 0.069,
      "step": 7750
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0006271282051282051,
      "loss": 0.0823,
      "step": 7775
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0006258461538461538,
      "loss": 0.0687,
      "step": 7800
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0006245641025641026,
      "loss": 0.0701,
      "step": 7825
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0006232820512820513,
      "loss": 0.089,
      "step": 7850
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.000622,
      "loss": 0.0734,
      "step": 7875
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0006207179487179487,
      "loss": 0.0798,
      "step": 7900
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0006194358974358975,
      "loss": 0.0835,
      "step": 7925
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0006181538461538462,
      "loss": 0.1016,
      "step": 7950
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006168717948717949,
      "loss": 0.1292,
      "step": 7975
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006155897435897436,
      "loss": 0.1077,
      "step": 8000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8477774858474731,
      "eval_runtime": 216.383,
      "eval_samples_per_second": 4.621,
      "eval_steps_per_second": 0.291,
      "step": 8000
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006143076923076923,
      "loss": 0.109,
      "step": 8025
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0006130256410256411,
      "loss": 0.1169,
      "step": 8050
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006117435897435898,
      "loss": 0.1108,
      "step": 8075
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006104615384615386,
      "loss": 0.1014,
      "step": 8100
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006091794871794872,
      "loss": 0.0945,
      "step": 8125
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006078974358974359,
      "loss": 0.1005,
      "step": 8150
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006066153846153847,
      "loss": 0.0888,
      "step": 8175
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006053333333333333,
      "loss": 0.0959,
      "step": 8200
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.000604051282051282,
      "loss": 0.1038,
      "step": 8225
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0006027692307692307,
      "loss": 0.0909,
      "step": 8250
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006014871794871795,
      "loss": 0.0994,
      "step": 8275
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0006002051282051282,
      "loss": 0.0996,
      "step": 8300
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0005989230769230769,
      "loss": 0.0922,
      "step": 8325
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0005976410256410256,
      "loss": 0.099,
      "step": 8350
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0005963589743589743,
      "loss": 0.0924,
      "step": 8375
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0005950769230769231,
      "loss": 0.0891,
      "step": 8400
    },
    {
      "epoch": 2.02,
      "eval_loss": 0.8391351699829102,
      "eval_runtime": 216.0468,
      "eval_samples_per_second": 4.629,
      "eval_steps_per_second": 0.292,
      "step": 8400
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0005937948717948718,
      "loss": 0.0856,
      "step": 8425
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0005925128205128206,
      "loss": 0.082,
      "step": 8450
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0005912307692307692,
      "loss": 0.0887,
      "step": 8475
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0005899487179487179,
      "loss": 0.0836,
      "step": 8500
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0005886666666666667,
      "loss": 0.0841,
      "step": 8525
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0005873846153846154,
      "loss": 0.0814,
      "step": 8550
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0005861025641025642,
      "loss": 0.0903,
      "step": 8575
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0005848205128205128,
      "loss": 0.0846,
      "step": 8600
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0005835384615384616,
      "loss": 0.0759,
      "step": 8625
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.0005822564102564103,
      "loss": 0.0817,
      "step": 8650
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.000580974358974359,
      "loss": 0.0768,
      "step": 8675
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0005796923076923078,
      "loss": 0.0718,
      "step": 8700
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0005784102564102564,
      "loss": 0.0737,
      "step": 8725
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0005771282051282052,
      "loss": 0.0712,
      "step": 8750
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0005758461538461539,
      "loss": 0.0768,
      "step": 8775
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0005745641025641027,
      "loss": 0.0803,
      "step": 8800
    },
    {
      "epoch": 2.04,
      "eval_loss": 0.759874165058136,
      "eval_runtime": 215.1962,
      "eval_samples_per_second": 4.647,
      "eval_steps_per_second": 0.293,
      "step": 8800
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0005732820512820512,
      "loss": 0.068,
      "step": 8825
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0005719999999999999,
      "loss": 0.0679,
      "step": 8850
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0005707179487179487,
      "loss": 0.0687,
      "step": 8875
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0005694358974358974,
      "loss": 0.0675,
      "step": 8900
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0005681538461538462,
      "loss": 0.0736,
      "step": 8925
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0005668717948717948,
      "loss": 0.0737,
      "step": 8950
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0005655897435897436,
      "loss": 0.0737,
      "step": 8975
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0005643076923076923,
      "loss": 0.0674,
      "step": 9000
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.000563025641025641,
      "loss": 0.0703,
      "step": 9025
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0005617435897435898,
      "loss": 0.0698,
      "step": 9050
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0005604615384615384,
      "loss": 0.0723,
      "step": 9075
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0005591794871794872,
      "loss": 0.0688,
      "step": 9100
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0005578974358974359,
      "loss": 0.0664,
      "step": 9125
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0005566153846153847,
      "loss": 0.0677,
      "step": 9150
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0005553333333333334,
      "loss": 0.0624,
      "step": 9175
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.000554051282051282,
      "loss": 0.068,
      "step": 9200
    },
    {
      "epoch": 2.06,
      "eval_loss": 0.77803635597229,
      "eval_runtime": 216.146,
      "eval_samples_per_second": 4.627,
      "eval_steps_per_second": 0.291,
      "step": 9200
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0005527692307692308,
      "loss": 0.0717,
      "step": 9225
    },
    {
      "epoch": 2.06,
      "learning_rate": 0.0005514871794871795,
      "loss": 0.0692,
      "step": 9250
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0005502051282051283,
      "loss": 0.0657,
      "step": 9275
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.000548923076923077,
      "loss": 0.064,
      "step": 9300
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0005476410256410257,
      "loss": 0.0661,
      "step": 9325
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0005463589743589744,
      "loss": 0.0604,
      "step": 9350
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0005450769230769232,
      "loss": 0.066,
      "step": 9375
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0005437948717948719,
      "loss": 0.0617,
      "step": 9400
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0005425128205128204,
      "loss": 0.0622,
      "step": 9425
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.0005412307692307692,
      "loss": 0.056,
      "step": 9450
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0005399487179487179,
      "loss": 0.0621,
      "step": 9475
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0005386666666666667,
      "loss": 0.0578,
      "step": 9500
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0005373846153846154,
      "loss": 0.0526,
      "step": 9525
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0005361025641025641,
      "loss": 0.0703,
      "step": 9550
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0005348205128205128,
      "loss": 0.0629,
      "step": 9575
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0005335384615384615,
      "loss": 0.0723,
      "step": 9600
    },
    {
      "epoch": 2.08,
      "eval_loss": 0.09210965782403946,
      "eval_runtime": 216.3525,
      "eval_samples_per_second": 4.622,
      "eval_steps_per_second": 0.291,
      "step": 9600
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.0005322564102564103,
      "loss": 0.0606,
      "step": 9625
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.000530974358974359,
      "loss": 0.0577,
      "step": 9650
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0005296923076923077,
      "loss": 0.0595,
      "step": 9675
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0005284102564102564,
      "loss": 0.0561,
      "step": 9700
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0005271282051282052,
      "loss": 0.0558,
      "step": 9725
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0005258461538461539,
      "loss": 0.0562,
      "step": 9750
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0005245641025641025,
      "loss": 0.0756,
      "step": 9775
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0005232820512820513,
      "loss": 0.0624,
      "step": 9800
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.000522,
      "loss": 0.0532,
      "step": 9825
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0005207179487179488,
      "loss": 0.0546,
      "step": 9850
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0005194358974358975,
      "loss": 0.0603,
      "step": 9875
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0005181538461538462,
      "loss": 0.0633,
      "step": 9900
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0005168717948717949,
      "loss": 0.0598,
      "step": 9925
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0005155897435897436,
      "loss": 0.0635,
      "step": 9950
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0005143076923076924,
      "loss": 0.0587,
      "step": 9975
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0005130256410256411,
      "loss": 0.0615,
      "step": 10000
    },
    {
      "epoch": 2.1,
      "eval_loss": 0.09180190414190292,
      "eval_runtime": 215.773,
      "eval_samples_per_second": 4.635,
      "eval_steps_per_second": 0.292,
      "step": 10000
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0005117435897435897,
      "loss": 0.0519,
      "step": 10025
    },
    {
      "epoch": 2.1,
      "learning_rate": 0.0005104615384615384,
      "loss": 0.0445,
      "step": 10050
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0005091794871794872,
      "loss": 0.0528,
      "step": 10075
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0005078974358974359,
      "loss": 0.0458,
      "step": 10100
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0005066153846153845,
      "loss": 0.0629,
      "step": 10125
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0005053333333333333,
      "loss": 0.08,
      "step": 10150
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.000504051282051282,
      "loss": 0.0623,
      "step": 10175
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0005027692307692308,
      "loss": 0.0479,
      "step": 10200
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0005014871794871795,
      "loss": 0.052,
      "step": 10225
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0005002051282051282,
      "loss": 0.0546,
      "step": 10250
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0004989230769230769,
      "loss": 0.0495,
      "step": 10275
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0004976410256410256,
      "loss": 0.0403,
      "step": 10300
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0004963589743589744,
      "loss": 0.0468,
      "step": 10325
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0004950769230769231,
      "loss": 0.0393,
      "step": 10350
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0004937948717948718,
      "loss": 0.0454,
      "step": 10375
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0004925128205128205,
      "loss": 0.0493,
      "step": 10400
    },
    {
      "epoch": 2.12,
      "eval_loss": 0.09357196092605591,
      "eval_runtime": 218.7372,
      "eval_samples_per_second": 4.572,
      "eval_steps_per_second": 0.288,
      "step": 10400
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0004912307692307693,
      "loss": 0.0402,
      "step": 10425
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.000489948717948718,
      "loss": 0.0379,
      "step": 10450
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0004886666666666667,
      "loss": 0.0403,
      "step": 10475
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0004873846153846154,
      "loss": 0.0404,
      "step": 10500
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0004861025641025641,
      "loss": 0.0376,
      "step": 10525
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0004848205128205128,
      "loss": 0.0336,
      "step": 10550
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0004835384615384615,
      "loss": 0.039,
      "step": 10575
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.00048225641025641025,
      "loss": 0.0459,
      "step": 10600
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.000480974358974359,
      "loss": 0.036,
      "step": 10625
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.0004796923076923077,
      "loss": 0.0392,
      "step": 10650
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.00047841025641025644,
      "loss": 0.0443,
      "step": 10675
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.00047712820512820517,
      "loss": 0.0413,
      "step": 10700
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0004758461538461539,
      "loss": 0.0412,
      "step": 10725
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0004745641025641026,
      "loss": 0.0454,
      "step": 10750
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0004732820512820513,
      "loss": 0.0439,
      "step": 10775
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.000472,
      "loss": 0.0412,
      "step": 10800
    },
    {
      "epoch": 2.14,
      "eval_loss": 0.08619751781225204,
      "eval_runtime": 217.221,
      "eval_samples_per_second": 4.604,
      "eval_steps_per_second": 0.29,
      "step": 10800
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0004707179487179487,
      "loss": 0.0537,
      "step": 10825
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.00046943589743589744,
      "loss": 0.0436,
      "step": 10850
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.00046815384615384617,
      "loss": 0.0404,
      "step": 10875
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0004668717948717949,
      "loss": 0.0303,
      "step": 10900
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0004655897435897436,
      "loss": 0.0332,
      "step": 10925
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0004643076923076923,
      "loss": 0.0501,
      "step": 10950
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.00046302564102564104,
      "loss": 0.0483,
      "step": 10975
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.00046174358974358977,
      "loss": 0.038,
      "step": 11000
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0004604615384615385,
      "loss": 0.034,
      "step": 11025
    },
    {
      "epoch": 2.15,
      "learning_rate": 0.0004591794871794872,
      "loss": 0.0446,
      "step": 11050
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.00045789743589743595,
      "loss": 0.0476,
      "step": 11075
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0004566153846153846,
      "loss": 0.0368,
      "step": 11100
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0004553333333333333,
      "loss": 0.0411,
      "step": 11125
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.00045405128205128204,
      "loss": 0.0349,
      "step": 11150
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.00045276923076923077,
      "loss": 0.0417,
      "step": 11175
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0004514871794871795,
      "loss": 0.0402,
      "step": 11200
    },
    {
      "epoch": 2.16,
      "eval_loss": 0.08717386424541473,
      "eval_runtime": 217.4573,
      "eval_samples_per_second": 4.599,
      "eval_steps_per_second": 0.29,
      "step": 11200
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.0004502051282051282,
      "loss": 0.0464,
      "step": 11225
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.00044892307692307696,
      "loss": 0.0563,
      "step": 11250
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.00044764102564102563,
      "loss": 0.0448,
      "step": 11275
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.00044635897435897436,
      "loss": 0.0525,
      "step": 11300
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0004450769230769231,
      "loss": 0.0475,
      "step": 11325
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0004437948717948718,
      "loss": 0.0429,
      "step": 11350
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.00044251282051282055,
      "loss": 0.0523,
      "step": 11375
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0004412307692307692,
      "loss": 0.0494,
      "step": 11400
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.00043994871794871796,
      "loss": 0.0507,
      "step": 11425
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.00043866666666666663,
      "loss": 0.0486,
      "step": 11450
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.00043738461538461536,
      "loss": 0.0738,
      "step": 11475
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0004361025641025641,
      "loss": 0.0443,
      "step": 11500
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0004348205128205128,
      "loss": 0.06,
      "step": 11525
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.00043353846153846155,
      "loss": 0.0438,
      "step": 11550
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0004322564102564103,
      "loss": 0.0393,
      "step": 11575
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.000430974358974359,
      "loss": 0.0561,
      "step": 11600
    },
    {
      "epoch": 2.18,
      "eval_loss": 0.08517900854349136,
      "eval_runtime": 215.8861,
      "eval_samples_per_second": 4.632,
      "eval_steps_per_second": 0.292,
      "step": 11600
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.00042969230769230774,
      "loss": 0.0633,
      "step": 11625
    },
    {
      "epoch": 2.18,
      "learning_rate": 0.0004284102564102564,
      "loss": 0.0937,
      "step": 11650
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.00042712820512820515,
      "loss": 0.0647,
      "step": 11675
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0004258461538461538,
      "loss": 0.0517,
      "step": 11700
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.00042456410256410255,
      "loss": 0.044,
      "step": 11725
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0004232820512820513,
      "loss": 0.048,
      "step": 11750
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.000422,
      "loss": 0.0429,
      "step": 11775
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.00042071794871794874,
      "loss": 0.0452,
      "step": 11800
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.0004194358974358974,
      "loss": 0.0535,
      "step": 11825
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.00041815384615384615,
      "loss": 0.049,
      "step": 11850
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0004168717948717949,
      "loss": 0.0464,
      "step": 11875
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0004155897435897436,
      "loss": 0.0564,
      "step": 11900
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.00041430769230769234,
      "loss": 0.0673,
      "step": 11925
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00041302564102564107,
      "loss": 0.0679,
      "step": 11950
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0004117435897435898,
      "loss": 0.0642,
      "step": 11975
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00041046153846153847,
      "loss": 0.0664,
      "step": 12000
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.10094589740037918,
      "eval_runtime": 219.2666,
      "eval_samples_per_second": 4.561,
      "eval_steps_per_second": 0.287,
      "step": 12000
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.00040917948717948715,
      "loss": 0.0683,
      "step": 12025
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0004078974358974359,
      "loss": 0.0699,
      "step": 12050
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0004066153846153846,
      "loss": 0.0645,
      "step": 12075
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.00040533333333333334,
      "loss": 0.0548,
      "step": 12100
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.00040405128205128207,
      "loss": 0.0603,
      "step": 12125
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0004027692307692308,
      "loss": 0.0552,
      "step": 12150
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0004014871794871795,
      "loss": 0.0555,
      "step": 12175
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0004002051282051282,
      "loss": 0.0632,
      "step": 12200
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.00039892307692307693,
      "loss": 0.0572,
      "step": 12225
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.00039764102564102566,
      "loss": 0.0557,
      "step": 12250
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0003963589743589744,
      "loss": 0.0599,
      "step": 12275
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0003950769230769231,
      "loss": 0.056,
      "step": 12300
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0003937948717948718,
      "loss": 0.0583,
      "step": 12325
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.00039251282051282053,
      "loss": 0.0564,
      "step": 12350
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.0003912307692307692,
      "loss": 0.0538,
      "step": 12375
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.00038994871794871793,
      "loss": 0.0525,
      "step": 12400
    },
    {
      "epoch": 3.02,
      "eval_loss": 0.13796697556972504,
      "eval_runtime": 216.1893,
      "eval_samples_per_second": 4.626,
      "eval_steps_per_second": 0.291,
      "step": 12400
    },
    {
      "epoch": 3.02,
      "learning_rate": 0.00038866666666666666,
      "loss": 0.0508,
      "step": 12425
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0003873846153846154,
      "loss": 0.0562,
      "step": 12450
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0003861025641025641,
      "loss": 0.0489,
      "step": 12475
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.00038482051282051285,
      "loss": 0.0565,
      "step": 12500
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0003835384615384616,
      "loss": 0.0522,
      "step": 12525
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.00038225641025641026,
      "loss": 0.0534,
      "step": 12550
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.000380974358974359,
      "loss": 0.0539,
      "step": 12575
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0003796923076923077,
      "loss": 0.0457,
      "step": 12600
    },
    {
      "epoch": 3.03,
      "learning_rate": 0.0003784102564102564,
      "loss": 0.0509,
      "step": 12625
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0003771282051282051,
      "loss": 0.0512,
      "step": 12650
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.00037584615384615385,
      "loss": 0.043,
      "step": 12675
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0003745641025641026,
      "loss": 0.0513,
      "step": 12700
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.00037328205128205126,
      "loss": 0.0423,
      "step": 12725
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.000372,
      "loss": 0.0482,
      "step": 12750
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0003707179487179487,
      "loss": 0.0517,
      "step": 12775
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.00036943589743589745,
      "loss": 0.0414,
      "step": 12800
    },
    {
      "epoch": 3.04,
      "eval_loss": 0.1226244568824768,
      "eval_runtime": 216.9336,
      "eval_samples_per_second": 4.61,
      "eval_steps_per_second": 0.29,
      "step": 12800
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.0003681538461538462,
      "loss": 0.0451,
      "step": 12825
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0003668717948717949,
      "loss": 0.0432,
      "step": 12850
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.00036558974358974364,
      "loss": 0.0428,
      "step": 12875
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0003643076923076923,
      "loss": 0.0486,
      "step": 12900
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.000363025641025641,
      "loss": 0.0459,
      "step": 12925
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0003617435897435897,
      "loss": 0.0465,
      "step": 12950
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.00036046153846153845,
      "loss": 0.0417,
      "step": 12975
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.0003591794871794872,
      "loss": 0.0417,
      "step": 13000
    },
    {
      "epoch": 3.05,
      "learning_rate": 0.00035794871794871797,
      "loss": 0.0433,
      "step": 13025
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0003566666666666667,
      "loss": 0.0494,
      "step": 13050
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.00035538461538461543,
      "loss": 0.0407,
      "step": 13075
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0003541025641025641,
      "loss": 0.0372,
      "step": 13100
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0003528205128205128,
      "loss": 0.0455,
      "step": 13125
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0003515384615384615,
      "loss": 0.0375,
      "step": 13150
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.00035025641025641024,
      "loss": 0.0435,
      "step": 13175
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.00034897435897435897,
      "loss": 0.0437,
      "step": 13200
    },
    {
      "epoch": 3.06,
      "eval_loss": 0.1415424644947052,
      "eval_runtime": 215.4136,
      "eval_samples_per_second": 4.642,
      "eval_steps_per_second": 0.292,
      "step": 13200
    },
    {
      "epoch": 3.06,
      "learning_rate": 0.0003476923076923077,
      "loss": 0.0432,
      "step": 13225
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.00034641025641025643,
      "loss": 0.0393,
      "step": 13250
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.00034512820512820516,
      "loss": 0.0446,
      "step": 13275
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0003438461538461539,
      "loss": 0.0417,
      "step": 13300
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.00034256410256410256,
      "loss": 0.0386,
      "step": 13325
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0003412820512820513,
      "loss": 0.0395,
      "step": 13350
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.00034,
      "loss": 0.0394,
      "step": 13375
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.0003387179487179487,
      "loss": 0.0364,
      "step": 13400
    },
    {
      "epoch": 3.07,
      "learning_rate": 0.00033743589743589743,
      "loss": 0.0334,
      "step": 13425
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.00033615384615384616,
      "loss": 0.0374,
      "step": 13450
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0003348717948717949,
      "loss": 0.0352,
      "step": 13475
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.00033358974358974357,
      "loss": 0.0317,
      "step": 13500
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0003323076923076923,
      "loss": 0.0427,
      "step": 13525
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.000331025641025641,
      "loss": 0.038,
      "step": 13550
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.00032974358974358976,
      "loss": 0.0414,
      "step": 13575
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0003284615384615385,
      "loss": 0.0368,
      "step": 13600
    },
    {
      "epoch": 3.08,
      "eval_loss": 0.08724867552518845,
      "eval_runtime": 216.1216,
      "eval_samples_per_second": 4.627,
      "eval_steps_per_second": 0.292,
      "step": 13600
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.0003271794871794872,
      "loss": 0.0357,
      "step": 13625
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.00032589743589743594,
      "loss": 0.0337,
      "step": 13650
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0003246153846153846,
      "loss": 0.0337,
      "step": 13675
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0003233333333333333,
      "loss": 0.033,
      "step": 13700
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.000322051282051282,
      "loss": 0.0326,
      "step": 13725
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.00032076923076923076,
      "loss": 0.0434,
      "step": 13750
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0003194871794871795,
      "loss": 0.0365,
      "step": 13775
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.0003182051282051282,
      "loss": 0.0327,
      "step": 13800
    },
    {
      "epoch": 3.09,
      "learning_rate": 0.00031692307692307695,
      "loss": 0.0324,
      "step": 13825
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0003156410256410256,
      "loss": 0.038,
      "step": 13850
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.00031435897435897435,
      "loss": 0.0379,
      "step": 13875
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0003130769230769231,
      "loss": 0.0409,
      "step": 13900
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.0003117948717948718,
      "loss": 0.0391,
      "step": 13925
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.00031051282051282054,
      "loss": 0.036,
      "step": 13950
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.00030923076923076927,
      "loss": 0.038,
      "step": 13975
    },
    {
      "epoch": 3.1,
      "learning_rate": 0.000307948717948718,
      "loss": 0.0311,
      "step": 14000
    },
    {
      "epoch": 3.1,
      "eval_loss": 0.08812109380960464,
      "eval_runtime": 216.1669,
      "eval_samples_per_second": 4.626,
      "eval_steps_per_second": 0.291,
      "step": 14000
    }
  ],
  "logging_steps": 25,
  "max_steps": 20000,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 400,
  "total_flos": 1.441894654420992e+21,
  "trial_name": null,
  "trial_params": null
}