{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 399.0662251655629,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.07,
      "learning_rate": 2.4834437086092715e-05,
      "loss": 3.036,
      "step": 30
    },
    {
      "epoch": 5.07,
      "learning_rate": 4.966887417218543e-05,
      "loss": 2.727,
      "step": 60
    },
    {
      "epoch": 8.07,
      "learning_rate": 7.450331125827815e-05,
      "loss": 2.4733,
      "step": 90
    },
    {
      "epoch": 11.07,
      "learning_rate": 9.933774834437086e-05,
      "loss": 2.3326,
      "step": 120
    },
    {
      "epoch": 14.07,
      "learning_rate": 0.00012417218543046358,
      "loss": 2.2252,
      "step": 150
    },
    {
      "epoch": 17.07,
      "learning_rate": 0.0001490066225165563,
      "loss": 2.1346,
      "step": 180
    },
    {
      "epoch": 20.07,
      "learning_rate": 0.000173841059602649,
      "loss": 2.0513,
      "step": 210
    },
    {
      "epoch": 23.07,
      "learning_rate": 0.00019867549668874172,
      "loss": 1.9759,
      "step": 240
    },
    {
      "epoch": 26.07,
      "learning_rate": 0.00022350993377483444,
      "loss": 1.9071,
      "step": 270
    },
    {
      "epoch": 29.07,
      "learning_rate": 0.00024834437086092715,
      "loss": 1.838,
      "step": 300
    },
    {
      "epoch": 32.07,
      "learning_rate": 0.0002731788079470199,
      "loss": 1.7631,
      "step": 330
    },
    {
      "epoch": 35.07,
      "learning_rate": 0.0002980132450331126,
      "loss": 1.6968,
      "step": 360
    },
    {
      "epoch": 38.07,
      "learning_rate": 0.00032284768211920527,
      "loss": 1.6119,
      "step": 390
    },
    {
      "epoch": 41.07,
      "learning_rate": 0.000347682119205298,
      "loss": 1.5422,
      "step": 420
    },
    {
      "epoch": 44.07,
      "learning_rate": 0.00037251655629139076,
      "loss": 1.4734,
      "step": 450
    },
    {
      "epoch": 47.07,
      "learning_rate": 0.00039735099337748344,
      "loss": 1.3984,
      "step": 480
    },
    {
      "epoch": 50.07,
      "learning_rate": 0.0004221854304635762,
      "loss": 1.3118,
      "step": 510
    },
    {
      "epoch": 53.07,
      "learning_rate": 0.0004470198675496689,
      "loss": 1.2615,
      "step": 540
    },
    {
      "epoch": 56.07,
      "learning_rate": 0.00047185430463576156,
      "loss": 1.1864,
      "step": 570
    },
    {
      "epoch": 59.07,
      "learning_rate": 0.0004966887417218543,
      "loss": 1.1384,
      "step": 600
    },
    {
      "epoch": 62.07,
      "learning_rate": 0.000521523178807947,
      "loss": 1.0538,
      "step": 630
    },
    {
      "epoch": 65.07,
      "learning_rate": 0.0005463576158940398,
      "loss": 1.0015,
      "step": 660
    },
    {
      "epoch": 68.07,
      "learning_rate": 0.0005711920529801324,
      "loss": 0.939,
      "step": 690
    },
    {
      "epoch": 71.07,
      "learning_rate": 0.0005960264900662252,
      "loss": 0.8865,
      "step": 720
    },
    {
      "epoch": 74.07,
      "learning_rate": 0.0006208609271523179,
      "loss": 0.8369,
      "step": 750
    },
    {
      "epoch": 77.07,
      "learning_rate": 0.0006456953642384105,
      "loss": 0.7761,
      "step": 780
    },
    {
      "epoch": 80.07,
      "learning_rate": 0.0006705298013245033,
      "loss": 0.7468,
      "step": 810
    },
    {
      "epoch": 83.07,
      "learning_rate": 0.000695364238410596,
      "loss": 0.6898,
      "step": 840
    },
    {
      "epoch": 86.07,
      "learning_rate": 0.0007201986754966887,
      "loss": 0.6711,
      "step": 870
    },
    {
      "epoch": 89.07,
      "learning_rate": 0.0007450331125827815,
      "loss": 0.6476,
      "step": 900
    },
    {
      "epoch": 92.07,
      "learning_rate": 0.0007698675496688743,
      "loss": 0.6337,
      "step": 930
    },
    {
      "epoch": 95.07,
      "learning_rate": 0.0007947019867549669,
      "loss": 0.5984,
      "step": 960
    },
    {
      "epoch": 98.07,
      "learning_rate": 0.0008195364238410596,
      "loss": 0.5906,
      "step": 990
    },
    {
      "epoch": 101.07,
      "learning_rate": 0.0008443708609271524,
      "loss": 0.5713,
      "step": 1020
    },
    {
      "epoch": 104.07,
      "learning_rate": 0.000869205298013245,
      "loss": 0.5323,
      "step": 1050
    },
    {
      "epoch": 107.07,
      "learning_rate": 0.0008940397350993377,
      "loss": 0.5379,
      "step": 1080
    },
    {
      "epoch": 110.07,
      "learning_rate": 0.0009188741721854305,
      "loss": 0.5165,
      "step": 1110
    },
    {
      "epoch": 113.07,
      "learning_rate": 0.0009437086092715231,
      "loss": 0.5117,
      "step": 1140
    },
    {
      "epoch": 116.07,
      "learning_rate": 0.0009685430463576159,
      "loss": 0.4923,
      "step": 1170
    },
    {
      "epoch": 119.07,
      "learning_rate": 0.0009933774834437086,
      "loss": 0.4754,
      "step": 1200
    },
    {
      "epoch": 122.07,
      "learning_rate": 0.0010182119205298015,
      "loss": 0.4723,
      "step": 1230
    },
    {
      "epoch": 125.07,
      "learning_rate": 0.001043046357615894,
      "loss": 0.4699,
      "step": 1260
    },
    {
      "epoch": 128.07,
      "learning_rate": 0.0010678807947019867,
      "loss": 0.4432,
      "step": 1290
    },
    {
      "epoch": 131.07,
      "learning_rate": 0.0010927152317880796,
      "loss": 0.4537,
      "step": 1320
    },
    {
      "epoch": 134.07,
      "learning_rate": 0.0011175496688741722,
      "loss": 0.4658,
      "step": 1350
    },
    {
      "epoch": 137.07,
      "learning_rate": 0.0011423841059602648,
      "loss": 0.4134,
      "step": 1380
    },
    {
      "epoch": 140.07,
      "learning_rate": 0.0011672185430463577,
      "loss": 0.4326,
      "step": 1410
    },
    {
      "epoch": 143.07,
      "learning_rate": 0.0011920529801324503,
      "loss": 0.452,
      "step": 1440
    },
    {
      "epoch": 146.07,
      "learning_rate": 0.001216887417218543,
      "loss": 0.4637,
      "step": 1470
    },
    {
      "epoch": 149.07,
      "learning_rate": 0.0012417218543046358,
      "loss": 0.4336,
      "step": 1500
    },
    {
      "epoch": 152.07,
      "learning_rate": 0.0012665562913907287,
      "loss": 0.4468,
      "step": 1530
    },
    {
      "epoch": 155.07,
      "learning_rate": 0.001291390728476821,
      "loss": 0.4415,
      "step": 1560
    },
    {
      "epoch": 158.07,
      "learning_rate": 0.001316225165562914,
      "loss": 0.4216,
      "step": 1590
    },
    {
      "epoch": 161.07,
      "learning_rate": 0.0013410596026490066,
      "loss": 0.4268,
      "step": 1620
    },
    {
      "epoch": 164.07,
      "learning_rate": 0.0013658940397350994,
      "loss": 0.4177,
      "step": 1650
    },
    {
      "epoch": 167.07,
      "learning_rate": 0.001390728476821192,
      "loss": 0.4439,
      "step": 1680
    },
    {
      "epoch": 170.07,
      "learning_rate": 0.001415562913907285,
      "loss": 0.444,
      "step": 1710
    },
    {
      "epoch": 173.07,
      "learning_rate": 0.0014403973509933773,
      "loss": 0.4364,
      "step": 1740
    },
    {
      "epoch": 176.07,
      "learning_rate": 0.0014652317880794702,
      "loss": 0.4292,
      "step": 1770
    },
    {
      "epoch": 179.07,
      "learning_rate": 0.001490066225165563,
      "loss": 0.4343,
      "step": 1800
    },
    {
      "epoch": 182.07,
      "learning_rate": 0.0015149006622516557,
      "loss": 0.4513,
      "step": 1830
    },
    {
      "epoch": 185.07,
      "learning_rate": 0.0015397350993377485,
      "loss": 0.453,
      "step": 1860
    },
    {
      "epoch": 188.07,
      "learning_rate": 0.0015645695364238411,
      "loss": 0.4163,
      "step": 1890
    },
    {
      "epoch": 191.07,
      "learning_rate": 0.0015894039735099338,
      "loss": 0.441,
      "step": 1920
    },
    {
      "epoch": 194.07,
      "learning_rate": 0.0016142384105960264,
      "loss": 0.4564,
      "step": 1950
    },
    {
      "epoch": 197.07,
      "learning_rate": 0.0016390728476821193,
      "loss": 0.4733,
      "step": 1980
    },
    {
      "epoch": 200.07,
      "learning_rate": 0.001663907284768212,
      "loss": 0.4634,
      "step": 2010
    },
    {
      "epoch": 203.07,
      "learning_rate": 0.0016887417218543047,
      "loss": 0.4723,
      "step": 2040
    },
    {
      "epoch": 206.07,
      "learning_rate": 0.0017135761589403972,
      "loss": 0.4666,
      "step": 2070
    },
    {
      "epoch": 209.07,
      "learning_rate": 0.00173841059602649,
      "loss": 0.4924,
      "step": 2100
    },
    {
      "epoch": 212.07,
      "learning_rate": 0.0017632450331125829,
      "loss": 0.4821,
      "step": 2130
    },
    {
      "epoch": 215.07,
      "learning_rate": 0.0017880794701986755,
      "loss": 0.48,
      "step": 2160
    },
    {
      "epoch": 218.07,
      "learning_rate": 0.0018129139072847683,
      "loss": 0.5119,
      "step": 2190
    },
    {
      "epoch": 221.07,
      "learning_rate": 0.001837748344370861,
      "loss": 0.5396,
      "step": 2220
    },
    {
      "epoch": 224.07,
      "learning_rate": 0.0018625827814569538,
      "loss": 0.5597,
      "step": 2250
    },
    {
      "epoch": 227.07,
      "learning_rate": 0.0018874172185430462,
      "loss": 0.5535,
      "step": 2280
    },
    {
      "epoch": 230.07,
      "learning_rate": 0.001912251655629139,
      "loss": 0.5951,
      "step": 2310
    },
    {
      "epoch": 233.07,
      "learning_rate": 0.0019370860927152317,
      "loss": 0.553,
      "step": 2340
    },
    {
      "epoch": 236.07,
      "learning_rate": 0.0019619205298013246,
      "loss": 0.5303,
      "step": 2370
    },
    {
      "epoch": 239.07,
      "learning_rate": 0.001986754966887417,
      "loss": 0.5958,
      "step": 2400
    },
    {
      "epoch": 242.07,
      "learning_rate": 0.00201158940397351,
      "loss": 0.6002,
      "step": 2430
    },
    {
      "epoch": 245.07,
      "learning_rate": 0.002036423841059603,
      "loss": 0.6858,
      "step": 2460
    },
    {
      "epoch": 248.07,
      "learning_rate": 0.0020612582781456956,
      "loss": 0.6152,
      "step": 2490
    },
    {
      "epoch": 251.07,
      "learning_rate": 0.002086092715231788,
      "loss": 0.6171,
      "step": 2520
    },
    {
      "epoch": 254.07,
      "learning_rate": 0.002110927152317881,
      "loss": 0.6383,
      "step": 2550
    },
    {
      "epoch": 257.07,
      "learning_rate": 0.0021357615894039735,
      "loss": 0.7459,
      "step": 2580
    },
    {
      "epoch": 260.07,
      "learning_rate": 0.002160596026490066,
      "loss": 0.7926,
      "step": 2610
    },
    {
      "epoch": 263.07,
      "learning_rate": 0.002185430463576159,
      "loss": 0.8945,
      "step": 2640
    },
    {
      "epoch": 266.07,
      "learning_rate": 0.002210264900662252,
      "loss": 0.8908,
      "step": 2670
    },
    {
      "epoch": 269.07,
      "learning_rate": 0.0022350993377483444,
      "loss": 0.9268,
      "step": 2700
    },
    {
      "epoch": 272.07,
      "learning_rate": 0.002259933774834437,
      "loss": 0.8545,
      "step": 2730
    },
    {
      "epoch": 275.07,
      "learning_rate": 0.0022847682119205297,
      "loss": 0.8369,
      "step": 2760
    },
    {
      "epoch": 278.07,
      "learning_rate": 0.0023096026490066228,
      "loss": 0.8851,
      "step": 2790
    },
    {
      "epoch": 281.07,
      "learning_rate": 0.0023344370860927154,
      "loss": 0.8496,
      "step": 2820
    },
    {
      "epoch": 284.07,
      "learning_rate": 0.002359271523178808,
      "loss": 0.8121,
      "step": 2850
    },
    {
      "epoch": 287.07,
      "learning_rate": 0.0023841059602649007,
      "loss": 0.76,
      "step": 2880
    },
    {
      "epoch": 290.07,
      "learning_rate": 0.0024089403973509937,
      "loss": 0.8552,
      "step": 2910
    },
    {
      "epoch": 293.07,
      "learning_rate": 0.002433774834437086,
      "loss": 0.9308,
      "step": 2940
    },
    {
      "epoch": 296.07,
      "learning_rate": 0.002458609271523179,
      "loss": 0.9471,
      "step": 2970
    },
    {
      "epoch": 299.07,
      "learning_rate": 0.0024834437086092716,
      "loss": 0.9702,
      "step": 3000
    },
    {
      "epoch": 302.07,
      "learning_rate": 0.0025082781456953643,
      "loss": 1.0656,
      "step": 3030
    },
    {
      "epoch": 305.07,
      "learning_rate": 0.0025331125827814573,
      "loss": 1.0608,
      "step": 3060
    },
    {
      "epoch": 308.07,
      "learning_rate": 0.0025579470198675495,
      "loss": 1.0397,
      "step": 3090
    },
    {
      "epoch": 311.07,
      "learning_rate": 0.002582781456953642,
      "loss": 1.0805,
      "step": 3120
    },
    {
      "epoch": 314.07,
      "learning_rate": 0.0026076158940397352,
      "loss": 1.1374,
      "step": 3150
    },
    {
      "epoch": 317.07,
      "learning_rate": 0.002632450331125828,
      "loss": 1.4001,
      "step": 3180
    },
    {
      "epoch": 320.07,
      "learning_rate": 0.0026572847682119205,
      "loss": 1.5648,
      "step": 3210
    },
    {
      "epoch": 323.07,
      "learning_rate": 0.002682119205298013,
      "loss": 1.7286,
      "step": 3240
    },
    {
      "epoch": 326.07,
      "learning_rate": 0.002706953642384106,
      "loss": 1.5511,
      "step": 3270
    },
    {
      "epoch": 329.07,
      "learning_rate": 0.002731788079470199,
      "loss": 1.3367,
      "step": 3300
    },
    {
      "epoch": 332.07,
      "learning_rate": 0.0027566225165562915,
      "loss": 1.2894,
      "step": 3330
    },
    {
      "epoch": 335.07,
      "learning_rate": 0.002781456953642384,
      "loss": 1.3288,
      "step": 3360
    },
    {
      "epoch": 338.07,
      "learning_rate": 0.002806291390728477,
      "loss": 1.2965,
      "step": 3390
    },
    {
      "epoch": 341.07,
      "learning_rate": 0.00283112582781457,
      "loss": 1.1965,
      "step": 3420
    },
    {
      "epoch": 344.07,
      "learning_rate": 0.002855960264900662,
      "loss": 1.1625,
      "step": 3450
    },
    {
      "epoch": 347.07,
      "learning_rate": 0.0028807947019867546,
      "loss": 1.2772,
      "step": 3480
    },
    {
      "epoch": 350.07,
      "learning_rate": 0.0029056291390728477,
      "loss": 1.2967,
      "step": 3510
    },
    {
      "epoch": 353.07,
      "learning_rate": 0.0029304635761589403,
      "loss": 1.5922,
      "step": 3540
    },
    {
      "epoch": 356.07,
      "learning_rate": 0.002955298013245033,
      "loss": 1.5614,
      "step": 3570
    },
    {
      "epoch": 359.07,
      "learning_rate": 0.002980132450331126,
      "loss": 1.9061,
      "step": 3600
    },
    {
      "epoch": 362.07,
      "learning_rate": 0.0030049668874172187,
      "loss": 2.0664,
      "step": 3630
    },
    {
      "epoch": 365.07,
      "learning_rate": 0.0030298013245033113,
      "loss": 1.8159,
      "step": 3660
    },
    {
      "epoch": 368.07,
      "learning_rate": 0.003054635761589404,
      "loss": 1.6505,
      "step": 3690
    },
    {
      "epoch": 371.07,
      "learning_rate": 0.003079470198675497,
      "loss": 1.4858,
      "step": 3720
    },
    {
      "epoch": 374.07,
      "learning_rate": 0.0031043046357615896,
      "loss": 1.4008,
      "step": 3750
    },
    {
      "epoch": 377.07,
      "learning_rate": 0.0031291390728476823,
      "loss": 1.2804,
      "step": 3780
    },
    {
      "epoch": 380.07,
      "learning_rate": 0.0031539735099337745,
      "loss": 1.3041,
      "step": 3810
    },
    {
      "epoch": 383.07,
      "learning_rate": 0.0031788079470198675,
      "loss": 1.3115,
      "step": 3840
    },
    {
      "epoch": 386.07,
      "learning_rate": 0.00320364238410596,
      "loss": 1.3194,
      "step": 3870
    },
    {
      "epoch": 389.07,
      "learning_rate": 0.003228476821192053,
      "loss": 1.3681,
      "step": 3900
    },
    {
      "epoch": 392.07,
      "learning_rate": 0.003253311258278146,
      "loss": 1.3676,
      "step": 3930
    },
    {
      "epoch": 395.07,
      "learning_rate": 0.0032781456953642385,
      "loss": 1.702,
      "step": 3960
    },
    {
      "epoch": 398.07,
      "learning_rate": 0.003302980132450331,
      "loss": 2.7594,
      "step": 3990
    }
  ],
  "logging_steps": 30,
  "max_steps": 60400,
  "num_train_epochs": 400,
  "save_steps": 500,
  "total_flos": 2204652011520000.0,
  "trial_name": null,
  "trial_params": null
}