{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.9983172065628945,
  "eval_steps": 500,
  "global_step": 2376,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 3.151260504201681e-08,
      "loss": 4.7778,
      "step": 15
    },
    {
      "epoch": 0.05,
      "learning_rate": 6.302521008403361e-08,
      "loss": 4.1934,
      "step": 30
    },
    {
      "epoch": 0.08,
      "learning_rate": 9.453781512605041e-08,
      "loss": 2.2385,
      "step": 45
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.2605042016806723e-07,
      "loss": 0.6474,
      "step": 60
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.5756302521008403e-07,
      "loss": 0.2815,
      "step": 75
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.8907563025210083e-07,
      "loss": 0.1927,
      "step": 90
    },
    {
      "epoch": 0.18,
      "learning_rate": 2.2058823529411763e-07,
      "loss": 0.1627,
      "step": 105
    },
    {
      "epoch": 0.2,
      "learning_rate": 2.5210084033613445e-07,
      "loss": 0.1106,
      "step": 120
    },
    {
      "epoch": 0.23,
      "learning_rate": 2.8361344537815123e-07,
      "loss": 0.0844,
      "step": 135
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.1512605042016805e-07,
      "loss": 0.075,
      "step": 150
    },
    {
      "epoch": 0.28,
      "learning_rate": 3.4663865546218483e-07,
      "loss": 0.07,
      "step": 165
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.7815126050420166e-07,
      "loss": 0.0633,
      "step": 180
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.096638655462185e-07,
      "loss": 0.0693,
      "step": 195
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.4117647058823526e-07,
      "loss": 0.0682,
      "step": 210
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.726890756302521e-07,
      "loss": 0.0639,
      "step": 225
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.99998920421976e-07,
      "loss": 0.0626,
      "step": 240
    },
    {
      "epoch": 0.43,
      "learning_rate": 4.999220044875031e-07,
      "loss": 0.0599,
      "step": 255
    },
    {
      "epoch": 0.45,
      "learning_rate": 4.997236787442598e-07,
      "loss": 0.0616,
      "step": 270
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.99404039537066e-07,
      "loss": 0.0562,
      "step": 285
    },
    {
      "epoch": 0.5,
      "learning_rate": 4.989632421437059e-07,
      "loss": 0.0554,
      "step": 300
    },
    {
      "epoch": 0.53,
      "learning_rate": 4.984015006994947e-07,
      "loss": 0.0514,
      "step": 315
    },
    {
      "epoch": 0.56,
      "learning_rate": 4.977190880932541e-07,
      "loss": 0.0561,
      "step": 330
    },
    {
      "epoch": 0.58,
      "learning_rate": 4.969163358347454e-07,
      "loss": 0.0552,
      "step": 345
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.959936338936248e-07,
      "loss": 0.0522,
      "step": 360
    },
    {
      "epoch": 0.63,
      "learning_rate": 4.949514305100001e-07,
      "loss": 0.0486,
      "step": 375
    },
    {
      "epoch": 0.66,
      "learning_rate": 4.937902319766796e-07,
      "loss": 0.06,
      "step": 390
    },
    {
      "epoch": 0.68,
      "learning_rate": 4.925106023932197e-07,
      "loss": 0.055,
      "step": 405
    },
    {
      "epoch": 0.71,
      "learning_rate": 4.911131633918905e-07,
      "loss": 0.0469,
      "step": 420
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.895985938356931e-07,
      "loss": 0.0522,
      "step": 435
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.879676294885731e-07,
      "loss": 0.0521,
      "step": 450
    },
    {
      "epoch": 0.78,
      "learning_rate": 4.862210626579943e-07,
      "loss": 0.0452,
      "step": 465
    },
    {
      "epoch": 0.81,
      "learning_rate": 4.843597418100424e-07,
      "loss": 0.0458,
      "step": 480
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.823845711572484e-07,
      "loss": 0.05,
      "step": 495
    },
    {
      "epoch": 0.86,
      "learning_rate": 4.802965102193303e-07,
      "loss": 0.0437,
      "step": 510
    },
    {
      "epoch": 0.88,
      "learning_rate": 4.78096573357067e-07,
      "loss": 0.0509,
      "step": 525
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.757858292795316e-07,
      "loss": 0.041,
      "step": 540
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.7336540052492215e-07,
      "loss": 0.0416,
      "step": 555
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.7083646291524306e-07,
      "loss": 0.0381,
      "step": 570
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.6820024498510123e-07,
      "loss": 0.0427,
      "step": 585
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.05463867634534836,
      "eval_runtime": 53.6645,
      "eval_samples_per_second": 7.007,
      "eval_steps_per_second": 2.348,
      "step": 594
    },
    {
      "epoch": 1.01,
      "learning_rate": 4.6545802738489593e-07,
      "loss": 0.0407,
      "step": 600
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.6261114225868974e-07,
      "loss": 0.0275,
      "step": 615
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.5966097259706617e-07,
      "loss": 0.0489,
      "step": 630
    },
    {
      "epoch": 1.09,
      "learning_rate": 4.5660895156528477e-07,
      "loss": 0.0511,
      "step": 645
    },
    {
      "epoch": 1.11,
      "learning_rate": 4.5345656180706316e-07,
      "loss": 0.0453,
      "step": 660
    },
    {
      "epoch": 1.14,
      "learning_rate": 4.50205334724322e-07,
      "loss": 0.0421,
      "step": 675
    },
    {
      "epoch": 1.16,
      "learning_rate": 4.4685684973324447e-07,
      "loss": 0.0396,
      "step": 690
    },
    {
      "epoch": 1.19,
      "learning_rate": 4.434127334970099e-07,
      "loss": 0.0386,
      "step": 705
    },
    {
      "epoch": 1.21,
      "learning_rate": 4.398746591355763e-07,
      "loss": 0.0487,
      "step": 720
    },
    {
      "epoch": 1.24,
      "learning_rate": 4.3624434541289365e-07,
      "loss": 0.0394,
      "step": 735
    },
    {
      "epoch": 1.26,
      "learning_rate": 4.3252355590194444e-07,
      "loss": 0.041,
      "step": 750
    },
    {
      "epoch": 1.29,
      "learning_rate": 4.287140981280162e-07,
      "loss": 0.0414,
      "step": 765
    },
    {
      "epoch": 1.31,
      "learning_rate": 4.248178226906221e-07,
      "loss": 0.0398,
      "step": 780
    },
    {
      "epoch": 1.34,
      "learning_rate": 4.2083662236449694e-07,
      "loss": 0.0391,
      "step": 795
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.167724311801047e-07,
      "loss": 0.034,
      "step": 810
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.12627223484104e-07,
      "loss": 0.0466,
      "step": 825
    },
    {
      "epoch": 1.41,
      "learning_rate": 4.0840301298022904e-07,
      "loss": 0.0405,
      "step": 840
    },
    {
      "epoch": 1.44,
      "learning_rate": 4.0410185175105074e-07,
      "loss": 0.0408,
      "step": 855
    },
    {
      "epoch": 1.46,
      "learning_rate": 3.9972582926109373e-07,
      "loss": 0.0473,
      "step": 870
    },
    {
      "epoch": 1.49,
      "learning_rate": 3.95277071341794e-07,
      "loss": 0.0414,
      "step": 885
    },
    {
      "epoch": 1.51,
      "learning_rate": 3.907577391587894e-07,
      "loss": 0.0417,
      "step": 900
    },
    {
      "epoch": 1.54,
      "learning_rate": 3.861700281620455e-07,
      "loss": 0.0478,
      "step": 915
    },
    {
      "epoch": 1.56,
      "learning_rate": 3.8151616701932575e-07,
      "loss": 0.0356,
      "step": 930
    },
    {
      "epoch": 1.59,
      "learning_rate": 3.7679841653352563e-07,
      "loss": 0.0346,
      "step": 945
    },
    {
      "epoch": 1.62,
      "learning_rate": 3.7201906854439514e-07,
      "loss": 0.0434,
      "step": 960
    },
    {
      "epoch": 1.64,
      "learning_rate": 3.671804448151845e-07,
      "loss": 0.0511,
      "step": 975
    },
    {
      "epoch": 1.67,
      "learning_rate": 3.622848959047525e-07,
      "loss": 0.0413,
      "step": 990
    },
    {
      "epoch": 1.69,
      "learning_rate": 3.57334800025687e-07,
      "loss": 0.0355,
      "step": 1005
    },
    {
      "epoch": 1.72,
      "learning_rate": 3.523325618889909e-07,
      "loss": 0.033,
      "step": 1020
    },
    {
      "epoch": 1.74,
      "learning_rate": 3.4728061153589595e-07,
      "loss": 0.0316,
      "step": 1035
    },
    {
      "epoch": 1.77,
      "learning_rate": 3.4218140315737075e-07,
      "loss": 0.0415,
      "step": 1050
    },
    {
      "epoch": 1.79,
      "learning_rate": 3.3703741390189765e-07,
      "loss": 0.0401,
      "step": 1065
    },
    {
      "epoch": 1.82,
      "learning_rate": 3.318511426720967e-07,
      "loss": 0.0333,
      "step": 1080
    },
    {
      "epoch": 1.84,
      "learning_rate": 3.2662510891078175e-07,
      "loss": 0.0347,
      "step": 1095
    },
    {
      "epoch": 1.87,
      "learning_rate": 3.2136185137703856e-07,
      "loss": 0.0389,
      "step": 1110
    },
    {
      "epoch": 1.89,
      "learning_rate": 3.160639269129186e-07,
      "loss": 0.0406,
      "step": 1125
    },
    {
      "epoch": 1.92,
      "learning_rate": 3.10733909201349e-07,
      "loss": 0.0354,
      "step": 1140
    },
    {
      "epoch": 1.94,
      "learning_rate": 3.053743875158611e-07,
      "loss": 0.0334,
      "step": 1155
    },
    {
      "epoch": 1.97,
      "learning_rate": 2.9998796546274496e-07,
      "loss": 0.0337,
      "step": 1170
    },
    {
      "epoch": 1.99,
      "learning_rate": 2.9457725971624173e-07,
      "loss": 0.0473,
      "step": 1185
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.0474083386361599,
      "eval_runtime": 53.4669,
      "eval_samples_per_second": 7.032,
      "eval_steps_per_second": 2.357,
      "step": 1188
    },
    {
      "epoch": 2.02,
      "learning_rate": 2.891448987473873e-07,
      "loss": 0.0316,
      "step": 1200
    },
    {
      "epoch": 2.04,
      "learning_rate": 2.8369352154712493e-07,
      "loss": 0.0316,
      "step": 1215
    },
    {
      "epoch": 2.07,
      "learning_rate": 2.782257763443082e-07,
      "loss": 0.0194,
      "step": 1230
    },
    {
      "epoch": 2.1,
      "learning_rate": 2.7274431931921563e-07,
      "loss": 0.0172,
      "step": 1245
    },
    {
      "epoch": 2.12,
      "learning_rate": 2.6725181331320274e-07,
      "loss": 0.0236,
      "step": 1260
    },
    {
      "epoch": 2.15,
      "learning_rate": 2.617509265351181e-07,
      "loss": 0.0281,
      "step": 1275
    },
    {
      "epoch": 2.17,
      "learning_rate": 2.5624433126511254e-07,
      "loss": 0.0228,
      "step": 1290
    },
    {
      "epoch": 2.2,
      "learning_rate": 2.507347025564693e-07,
      "loss": 0.0296,
      "step": 1305
    },
    {
      "epoch": 2.22,
      "learning_rate": 2.452247169360885e-07,
      "loss": 0.0332,
      "step": 1320
    },
    {
      "epoch": 2.25,
      "learning_rate": 2.3971705110425454e-07,
      "loss": 0.0218,
      "step": 1335
    },
    {
      "epoch": 2.27,
      "learning_rate": 2.3421438063431996e-07,
      "loss": 0.0255,
      "step": 1350
    },
    {
      "epoch": 2.3,
      "learning_rate": 2.2871937867293638e-07,
      "loss": 0.0272,
      "step": 1365
    },
    {
      "epoch": 2.32,
      "learning_rate": 2.2323471464146477e-07,
      "loss": 0.0253,
      "step": 1380
    },
    {
      "epoch": 2.35,
      "learning_rate": 2.1776305293919473e-07,
      "loss": 0.0319,
      "step": 1395
    },
    {
      "epoch": 2.37,
      "learning_rate": 2.1230705164900408e-07,
      "loss": 0.0279,
      "step": 1410
    },
    {
      "epoch": 2.4,
      "learning_rate": 2.0686936124608682e-07,
      "loss": 0.0243,
      "step": 1425
    },
    {
      "epoch": 2.42,
      "learning_rate": 2.0145262331037628e-07,
      "loss": 0.0266,
      "step": 1440
    },
    {
      "epoch": 2.45,
      "learning_rate": 1.960594692432901e-07,
      "loss": 0.0277,
      "step": 1455
    },
    {
      "epoch": 2.47,
      "learning_rate": 1.9069251898941995e-07,
      "loss": 0.0265,
      "step": 1470
    },
    {
      "epoch": 2.5,
      "learning_rate": 1.8535437976378608e-07,
      "loss": 0.0239,
      "step": 1485
    },
    {
      "epoch": 2.52,
      "learning_rate": 1.8004764478527641e-07,
      "loss": 0.0182,
      "step": 1500
    },
    {
      "epoch": 2.55,
      "learning_rate": 1.7477489201688505e-07,
      "loss": 0.0235,
      "step": 1515
    },
    {
      "epoch": 2.57,
      "learning_rate": 1.6953868291336103e-07,
      "loss": 0.0241,
      "step": 1530
    },
    {
      "epoch": 2.6,
      "learning_rate": 1.6434156117687747e-07,
      "loss": 0.0248,
      "step": 1545
    },
    {
      "epoch": 2.63,
      "learning_rate": 1.591860515213247e-07,
      "loss": 0.0257,
      "step": 1560
    },
    {
      "epoch": 2.65,
      "learning_rate": 1.540746584458269e-07,
      "loss": 0.0296,
      "step": 1575
    },
    {
      "epoch": 2.68,
      "learning_rate": 1.4900986501808e-07,
      "loss": 0.024,
      "step": 1590
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.4399413166810027e-07,
      "loss": 0.0289,
      "step": 1605
    },
    {
      "epoch": 2.73,
      "learning_rate": 1.3902989499296953e-07,
      "loss": 0.0245,
      "step": 1620
    },
    {
      "epoch": 2.75,
      "learning_rate": 1.3411956657315935e-07,
      "loss": 0.0219,
      "step": 1635
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.2926553180100714e-07,
      "loss": 0.0213,
      "step": 1650
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.244701487219146e-07,
      "loss": 0.0179,
      "step": 1665
    },
    {
      "epoch": 2.83,
      "learning_rate": 1.1973574688883075e-07,
      "loss": 0.0239,
      "step": 1680
    },
    {
      "epoch": 2.85,
      "learning_rate": 1.1506462623057698e-07,
      "loss": 0.0227,
      "step": 1695
    },
    {
      "epoch": 2.88,
      "learning_rate": 1.1045905593456254e-07,
      "loss": 0.0228,
      "step": 1710
    },
    {
      "epoch": 2.9,
      "learning_rate": 1.0592127334443398e-07,
      "loss": 0.0212,
      "step": 1725
    },
    {
      "epoch": 2.93,
      "learning_rate": 1.0145348287319471e-07,
      "loss": 0.019,
      "step": 1740
    },
    {
      "epoch": 2.95,
      "learning_rate": 9.705785493232125e-08,
      "loss": 0.0243,
      "step": 1755
    },
    {
      "epoch": 2.98,
      "learning_rate": 9.27365248773969e-08,
      "loss": 0.0211,
      "step": 1770
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.05001327767968178,
      "eval_runtime": 53.5686,
      "eval_samples_per_second": 7.019,
      "eval_steps_per_second": 2.352,
      "step": 1782
    },
    {
      "epoch": 3.0,
      "learning_rate": 8.849159197077633e-08,
      "loss": 0.0222,
      "step": 1785
    },
    {
      "epoch": 3.03,
      "learning_rate": 8.432511836178294e-08,
      "loss": 0.0098,
      "step": 1800
    },
    {
      "epoch": 3.05,
      "learning_rate": 8.02391280849355e-08,
      "loss": 0.0086,
      "step": 1815
    },
    {
      "epoch": 3.08,
      "learning_rate": 7.623560607669117e-08,
      "loss": 0.0112,
      "step": 1830
    },
    {
      "epoch": 3.1,
      "learning_rate": 7.231649721118163e-08,
      "loss": 0.0106,
      "step": 1845
    },
    {
      "epoch": 3.13,
      "learning_rate": 6.848370535541068e-08,
      "loss": 0.0125,
      "step": 1860
    },
    {
      "epoch": 3.16,
      "learning_rate": 6.473909244437376e-08,
      "loss": 0.008,
      "step": 1875
    },
    {
      "epoch": 3.18,
      "learning_rate": 6.108447757654678e-08,
      "loss": 0.0102,
      "step": 1890
    },
    {
      "epoch": 3.21,
      "learning_rate": 5.752163613018471e-08,
      "loss": 0.0074,
      "step": 1905
    },
    {
      "epoch": 3.23,
      "learning_rate": 5.405229890085949e-08,
      "loss": 0.006,
      "step": 1920
    },
    {
      "epoch": 3.26,
      "learning_rate": 5.0678151260655335e-08,
      "loss": 0.0084,
      "step": 1935
    },
    {
      "epoch": 3.28,
      "learning_rate": 4.740083233943021e-08,
      "loss": 0.0099,
      "step": 1950
    },
    {
      "epoch": 3.31,
      "learning_rate": 4.4221934228542144e-08,
      "loss": 0.0051,
      "step": 1965
    },
    {
      "epoch": 3.33,
      "learning_rate": 4.114300120742564e-08,
      "loss": 0.0124,
      "step": 1980
    },
    {
      "epoch": 3.36,
      "learning_rate": 3.8165528993394756e-08,
      "loss": 0.0032,
      "step": 1995
    },
    {
      "epoch": 3.38,
      "learning_rate": 3.5290964015037836e-08,
      "loss": 0.0091,
      "step": 2010
    },
    {
      "epoch": 3.41,
      "learning_rate": 3.2520702709555564e-08,
      "loss": 0.0044,
      "step": 2025
    },
    {
      "epoch": 3.43,
      "learning_rate": 2.985609084438442e-08,
      "loss": 0.0152,
      "step": 2040
    },
    {
      "epoch": 3.46,
      "learning_rate": 2.729842286343581e-08,
      "loss": 0.0053,
      "step": 2055
    },
    {
      "epoch": 3.48,
      "learning_rate": 2.4848941258266643e-08,
      "loss": 0.0053,
      "step": 2070
    },
    {
      "epoch": 3.51,
      "learning_rate": 2.2508835964488465e-08,
      "loss": 0.0059,
      "step": 2085
    },
    {
      "epoch": 3.53,
      "learning_rate": 2.0279243783707904e-08,
      "loss": 0.0046,
      "step": 2100
    },
    {
      "epoch": 3.56,
      "learning_rate": 1.8161247831278785e-08,
      "loss": 0.0063,
      "step": 2115
    },
    {
      "epoch": 3.58,
      "learning_rate": 1.6155877010134488e-08,
      "loss": 0.0067,
      "step": 2130
    },
    {
      "epoch": 3.61,
      "learning_rate": 1.426410551095672e-08,
      "loss": 0.003,
      "step": 2145
    },
    {
      "epoch": 3.63,
      "learning_rate": 1.2486852338922438e-08,
      "loss": 0.011,
      "step": 2160
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.0824980867259526e-08,
      "loss": 0.0068,
      "step": 2175
    },
    {
      "epoch": 3.69,
      "learning_rate": 9.279298417828263e-09,
      "loss": 0.0113,
      "step": 2190
    },
    {
      "epoch": 3.71,
      "learning_rate": 7.85055586893163e-09,
      "loss": 0.0069,
      "step": 2205
    },
    {
      "epoch": 3.74,
      "learning_rate": 6.539447290545563e-09,
      "loss": 0.0105,
      "step": 2220
    },
    {
      "epoch": 3.76,
      "learning_rate": 5.346609607146141e-09,
      "loss": 0.0093,
      "step": 2235
    },
    {
      "epoch": 3.79,
      "learning_rate": 4.272622288297695e-09,
      "loss": 0.004,
      "step": 2250
    },
    {
      "epoch": 3.81,
      "learning_rate": 3.3180070671515736e-09,
      "loss": 0.0082,
      "step": 2265
    },
    {
      "epoch": 3.84,
      "learning_rate": 2.4832276869931056e-09,
      "loss": 0.0065,
      "step": 2280
    },
    {
      "epoch": 3.86,
      "learning_rate": 1.7686896759595282e-09,
      "loss": 0.0063,
      "step": 2295
    },
    {
      "epoch": 3.89,
      "learning_rate": 1.174740150037823e-09,
      "loss": 0.0123,
      "step": 2310
    },
    {
      "epoch": 3.91,
      "learning_rate": 7.016676444392966e-10,
      "loss": 0.0083,
      "step": 2325
    },
    {
      "epoch": 3.94,
      "learning_rate": 3.497019734316209e-10,
      "loss": 0.005,
      "step": 2340
    },
    {
      "epoch": 3.96,
      "learning_rate": 1.1901411869724708e-10,
      "loss": 0.0076,
      "step": 2355
    },
    {
      "epoch": 3.99,
      "learning_rate": 9.716146272098492e-12,
      "loss": 0.0072,
      "step": 2370
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.07049984484910965,
      "eval_runtime": 53.5111,
      "eval_samples_per_second": 7.027,
      "eval_steps_per_second": 2.355,
      "step": 2376
    },
    {
      "epoch": 4.0,
      "step": 2376,
      "total_flos": 134540166758400.0,
      "train_loss": 0.10876847104865568,
      "train_runtime": 16521.7456,
      "train_samples_per_second": 1.726,
      "train_steps_per_second": 0.144
    }
  ],
  "logging_steps": 15,
  "max_steps": 2376,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 134540166758400.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}