{
  "best_metric": 0.890625,
  "best_model_checkpoint": "/net/nfs/mosaic/seanw/ntptutorial/proofstep/EleutherAI/pythia-2.8b-deduped/checkpoint-5000",
  "epoch": 1.887504718761797,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 3.4478393790900113e-06,
      "loss": 2.7471,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.48574245242758e-06,
      "loss": 2.2453,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.092876829713298e-06,
      "loss": 2.1194,
      "step": 30
    },
    {
      "epoch": 0.02,
      "learning_rate": 5.523645525765149e-06,
      "loss": 1.9594,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 5.857775684842452e-06,
      "loss": 1.874,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 6.130779903050867e-06,
      "loss": 1.8366,
      "step": 60
    },
    {
      "epoch": 0.03,
      "learning_rate": 6.361601680642952e-06,
      "loss": 1.8257,
      "step": 70
    },
    {
      "epoch": 0.03,
      "learning_rate": 6.561548599102718e-06,
      "loss": 1.7581,
      "step": 80
    },
    {
      "epoch": 0.03,
      "learning_rate": 6.737914280336586e-06,
      "loss": 1.7984,
      "step": 90
    },
    {
      "epoch": 0.04,
      "learning_rate": 6.895678758180023e-06,
      "loss": 1.676,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 7.038394088074824e-06,
      "loss": 1.7201,
      "step": 110
    },
    {
      "epoch": 0.05,
      "learning_rate": 7.168682976388437e-06,
      "loss": 1.6249,
      "step": 120
    },
    {
      "epoch": 0.05,
      "learning_rate": 7.2885371352490585e-06,
      "loss": 1.5939,
      "step": 130
    },
    {
      "epoch": 0.05,
      "learning_rate": 7.39950475398052e-06,
      "loss": 1.5962,
      "step": 140
    },
    {
      "epoch": 0.06,
      "learning_rate": 7.502813135465739e-06,
      "loss": 1.6127,
      "step": 150
    },
    {
      "epoch": 0.06,
      "learning_rate": 7.599451672440288e-06,
      "loss": 1.6355,
      "step": 160
    },
    {
      "epoch": 0.06,
      "learning_rate": 7.690229624176853e-06,
      "loss": 1.598,
      "step": 170
    },
    {
      "epoch": 0.07,
      "learning_rate": 7.775817353674156e-06,
      "loss": 1.5307,
      "step": 180
    },
    {
      "epoch": 0.07,
      "learning_rate": 7.856776400608327e-06,
      "loss": 1.5646,
      "step": 190
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.93358183151759e-06,
      "loss": 1.5411,
      "step": 200
    },
    {
      "epoch": 0.08,
      "learning_rate": 8.006639131266238e-06,
      "loss": 1.5405,
      "step": 210
    },
    {
      "epoch": 0.08,
      "learning_rate": 8.076297161412393e-06,
      "loss": 1.5144,
      "step": 220
    },
    {
      "epoch": 0.09,
      "learning_rate": 8.142858235714492e-06,
      "loss": 1.485,
      "step": 230
    },
    {
      "epoch": 0.09,
      "learning_rate": 8.206586049726006e-06,
      "loss": 1.4476,
      "step": 240
    },
    {
      "epoch": 0.09,
      "learning_rate": 8.267711990594892e-06,
      "loss": 1.4511,
      "step": 250
    },
    {
      "epoch": 0.1,
      "learning_rate": 8.326440208586629e-06,
      "loss": 1.4577,
      "step": 260
    },
    {
      "epoch": 0.1,
      "learning_rate": 8.382951730959874e-06,
      "loss": 1.4848,
      "step": 270
    },
    {
      "epoch": 0.11,
      "learning_rate": 8.43740782731809e-06,
      "loss": 1.4021,
      "step": 280
    },
    {
      "epoch": 0.11,
      "learning_rate": 8.489952784148422e-06,
      "loss": 1.4138,
      "step": 290
    },
    {
      "epoch": 0.11,
      "learning_rate": 8.540716208803308e-06,
      "loss": 1.378,
      "step": 300
    },
    {
      "epoch": 0.12,
      "learning_rate": 8.589814955558195e-06,
      "loss": 1.3893,
      "step": 310
    },
    {
      "epoch": 0.12,
      "learning_rate": 8.637354745777857e-06,
      "loss": 1.3791,
      "step": 320
    },
    {
      "epoch": 0.12,
      "learning_rate": 8.683431538698111e-06,
      "loss": 1.4285,
      "step": 330
    },
    {
      "epoch": 0.13,
      "learning_rate": 8.728132697514422e-06,
      "loss": 1.4242,
      "step": 340
    },
    {
      "epoch": 0.13,
      "learning_rate": 8.77153798639539e-06,
      "loss": 1.3793,
      "step": 350
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.813720427011724e-06,
      "loss": 1.3997,
      "step": 360
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.854747037685042e-06,
      "loss": 1.4108,
      "step": 370
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.894679473945897e-06,
      "loss": 1.3924,
      "step": 380
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.933574585872348e-06,
      "loss": 1.3793,
      "step": 390
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.97148490485516e-06,
      "loss": 1.3523,
      "step": 400
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.008459070248975e-06,
      "loss": 1.3515,
      "step": 410
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.044542204603808e-06,
      "loss": 1.4118,
      "step": 420
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.07977624473865e-06,
      "loss": 1.3533,
      "step": 430
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.114200234749962e-06,
      "loss": 1.363,
      "step": 440
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.147850586089026e-06,
      "loss": 1.3651,
      "step": 450
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.180761309052062e-06,
      "loss": 1.3318,
      "step": 460
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.21296421937283e-06,
      "loss": 1.3013,
      "step": 470
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.244489123063576e-06,
      "loss": 1.3052,
      "step": 480
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.27536398219589e-06,
      "loss": 1.3055,
      "step": 490
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.305615063932462e-06,
      "loss": 1.3519,
      "step": 500
    },
    {
      "epoch": 0.19,
      "eval_loss": 1.3330078125,
      "eval_runtime": 65.706,
      "eval_samples_per_second": 61.684,
      "eval_steps_per_second": 1.933,
      "step": 500
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.335267074800138e-06,
      "loss": 1.2696,
      "step": 510
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.364343281924199e-06,
      "loss": 1.3356,
      "step": 520
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.392865622714283e-06,
      "loss": 1.2978,
      "step": 530
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.420854804297444e-06,
      "loss": 1.3221,
      "step": 540
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.448330393827263e-06,
      "loss": 1.252,
      "step": 550
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.475310900655658e-06,
      "loss": 1.2975,
      "step": 560
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.501813851231617e-06,
      "loss": 1.2992,
      "step": 570
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.527855857485992e-06,
      "loss": 1.2637,
      "step": 580
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.553452679370556e-06,
      "loss": 1.2805,
      "step": 590
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.578619282140878e-06,
      "loss": 1.2689,
      "step": 600
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.603369888904405e-06,
      "loss": 1.2505,
      "step": 610
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.627718028895765e-06,
      "loss": 1.2542,
      "step": 620
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.651676581889526e-06,
      "loss": 1.2741,
      "step": 630
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.675257819115427e-06,
      "loss": 1.2702,
      "step": 640
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.6984734410015e-06,
      "loss": 1.2626,
      "step": 650
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.721334612035681e-06,
      "loss": 1.2431,
      "step": 660
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.743851993005941e-06,
      "loss": 1.2712,
      "step": 670
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.76603577085199e-06,
      "loss": 1.2467,
      "step": 680
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.787895686337778e-06,
      "loss": 1.2577,
      "step": 690
    },
    {
      "epoch": 0.26,
      "learning_rate": 9.80944105973296e-06,
      "loss": 1.2069,
      "step": 700
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.830680814672785e-06,
      "loss": 1.1961,
      "step": 710
    },
    {
      "epoch": 0.27,
      "learning_rate": 9.851623500349294e-06,
      "loss": 1.2356,
      "step": 720
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.872277312171945e-06,
      "loss": 1.2155,
      "step": 730
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.892650111022612e-06,
      "loss": 1.233,
      "step": 740
    },
    {
      "epoch": 0.28,
      "learning_rate": 9.91274944121818e-06,
      "loss": 1.247,
      "step": 750
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.932582547283467e-06,
      "loss": 1.2464,
      "step": 760
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.952156389627764e-06,
      "loss": 1.2211,
      "step": 770
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.971477659209916e-06,
      "loss": 1.2357,
      "step": 780
    },
    {
      "epoch": 0.3,
      "learning_rate": 9.990552791269229e-06,
      "loss": 1.2205,
      "step": 790
    },
    {
      "epoch": 0.3,
      "learning_rate": 1e-05,
      "loss": 1.2111,
      "step": 800
    },
    {
      "epoch": 0.31,
      "learning_rate": 1e-05,
      "loss": 1.1852,
      "step": 810
    },
    {
      "epoch": 0.31,
      "learning_rate": 1e-05,
      "loss": 1.2291,
      "step": 820
    },
    {
      "epoch": 0.31,
      "learning_rate": 1e-05,
      "loss": 1.217,
      "step": 830
    },
    {
      "epoch": 0.32,
      "learning_rate": 1e-05,
      "loss": 1.184,
      "step": 840
    },
    {
      "epoch": 0.32,
      "learning_rate": 1e-05,
      "loss": 1.1567,
      "step": 850
    },
    {
      "epoch": 0.32,
      "learning_rate": 1e-05,
      "loss": 1.1987,
      "step": 860
    },
    {
      "epoch": 0.33,
      "learning_rate": 1e-05,
      "loss": 1.2121,
      "step": 870
    },
    {
      "epoch": 0.33,
      "learning_rate": 1e-05,
      "loss": 1.191,
      "step": 880
    },
    {
      "epoch": 0.34,
      "learning_rate": 1e-05,
      "loss": 1.2109,
      "step": 890
    },
    {
      "epoch": 0.34,
      "learning_rate": 1e-05,
      "loss": 1.1724,
      "step": 900
    },
    {
      "epoch": 0.34,
      "learning_rate": 1e-05,
      "loss": 1.1727,
      "step": 910
    },
    {
      "epoch": 0.35,
      "learning_rate": 1e-05,
      "loss": 1.154,
      "step": 920
    },
    {
      "epoch": 0.35,
      "learning_rate": 1e-05,
      "loss": 1.168,
      "step": 930
    },
    {
      "epoch": 0.35,
      "learning_rate": 1e-05,
      "loss": 1.1514,
      "step": 940
    },
    {
      "epoch": 0.36,
      "learning_rate": 1e-05,
      "loss": 1.1677,
      "step": 950
    },
    {
      "epoch": 0.36,
      "learning_rate": 1e-05,
      "loss": 1.1848,
      "step": 960
    },
    {
      "epoch": 0.37,
      "learning_rate": 1e-05,
      "loss": 1.1555,
      "step": 970
    },
    {
      "epoch": 0.37,
      "learning_rate": 1e-05,
      "loss": 1.1451,
      "step": 980
    },
    {
      "epoch": 0.37,
      "learning_rate": 1e-05,
      "loss": 1.1323,
      "step": 990
    },
    {
      "epoch": 0.38,
      "learning_rate": 1e-05,
      "loss": 1.1172,
      "step": 1000
    },
    {
      "epoch": 0.38,
      "eval_loss": 1.17578125,
      "eval_runtime": 63.361,
      "eval_samples_per_second": 63.967,
      "eval_steps_per_second": 2.004,
      "step": 1000
    },
    {
      "epoch": 0.38,
      "learning_rate": 1e-05,
      "loss": 1.1784,
      "step": 1010
    },
    {
      "epoch": 0.39,
      "learning_rate": 1e-05,
      "loss": 1.1424,
      "step": 1020
    },
    {
      "epoch": 0.39,
      "learning_rate": 1e-05,
      "loss": 1.1858,
      "step": 1030
    },
    {
      "epoch": 0.39,
      "learning_rate": 1e-05,
      "loss": 1.1484,
      "step": 1040
    },
    {
      "epoch": 0.4,
      "learning_rate": 1e-05,
      "loss": 1.1682,
      "step": 1050
    },
    {
      "epoch": 0.4,
      "learning_rate": 1e-05,
      "loss": 1.1212,
      "step": 1060
    },
    {
      "epoch": 0.4,
      "learning_rate": 1e-05,
      "loss": 1.1324,
      "step": 1070
    },
    {
      "epoch": 0.41,
      "learning_rate": 1e-05,
      "loss": 1.1367,
      "step": 1080
    },
    {
      "epoch": 0.41,
      "learning_rate": 1e-05,
      "loss": 1.1443,
      "step": 1090
    },
    {
      "epoch": 0.42,
      "learning_rate": 1e-05,
      "loss": 1.1407,
      "step": 1100
    },
    {
      "epoch": 0.42,
      "learning_rate": 1e-05,
      "loss": 1.1193,
      "step": 1110
    },
    {
      "epoch": 0.42,
      "learning_rate": 1e-05,
      "loss": 1.1317,
      "step": 1120
    },
    {
      "epoch": 0.43,
      "learning_rate": 1e-05,
      "loss": 1.1556,
      "step": 1130
    },
    {
      "epoch": 0.43,
      "learning_rate": 1e-05,
      "loss": 1.1482,
      "step": 1140
    },
    {
      "epoch": 0.43,
      "learning_rate": 1e-05,
      "loss": 1.1255,
      "step": 1150
    },
    {
      "epoch": 0.44,
      "learning_rate": 1e-05,
      "loss": 1.1249,
      "step": 1160
    },
    {
      "epoch": 0.44,
      "learning_rate": 1e-05,
      "loss": 1.1379,
      "step": 1170
    },
    {
      "epoch": 0.45,
      "learning_rate": 1e-05,
      "loss": 1.1077,
      "step": 1180
    },
    {
      "epoch": 0.45,
      "learning_rate": 1e-05,
      "loss": 1.115,
      "step": 1190
    },
    {
      "epoch": 0.45,
      "learning_rate": 1e-05,
      "loss": 1.0913,
      "step": 1200
    },
    {
      "epoch": 0.46,
      "learning_rate": 1e-05,
      "loss": 1.0948,
      "step": 1210
    },
    {
      "epoch": 0.46,
      "learning_rate": 1e-05,
      "loss": 1.1593,
      "step": 1220
    },
    {
      "epoch": 0.46,
      "learning_rate": 1e-05,
      "loss": 1.1208,
      "step": 1230
    },
    {
      "epoch": 0.47,
      "learning_rate": 1e-05,
      "loss": 1.1261,
      "step": 1240
    },
    {
      "epoch": 0.47,
      "learning_rate": 1e-05,
      "loss": 1.0816,
      "step": 1250
    },
    {
      "epoch": 0.48,
      "learning_rate": 1e-05,
      "loss": 1.0907,
      "step": 1260
    },
    {
      "epoch": 0.48,
      "learning_rate": 1e-05,
      "loss": 1.0547,
      "step": 1270
    },
    {
      "epoch": 0.48,
      "learning_rate": 1e-05,
      "loss": 1.1173,
      "step": 1280
    },
    {
      "epoch": 0.49,
      "learning_rate": 1e-05,
      "loss": 1.1133,
      "step": 1290
    },
    {
      "epoch": 0.49,
      "learning_rate": 1e-05,
      "loss": 1.1072,
      "step": 1300
    },
    {
      "epoch": 0.49,
      "learning_rate": 1e-05,
      "loss": 1.0862,
      "step": 1310
    },
    {
      "epoch": 0.5,
      "learning_rate": 1e-05,
      "loss": 1.1178,
      "step": 1320
    },
    {
      "epoch": 0.5,
      "learning_rate": 1e-05,
      "loss": 1.099,
      "step": 1330
    },
    {
      "epoch": 0.51,
      "learning_rate": 1e-05,
      "loss": 1.0853,
      "step": 1340
    },
    {
      "epoch": 0.51,
      "learning_rate": 1e-05,
      "loss": 1.0652,
      "step": 1350
    },
    {
      "epoch": 0.51,
      "learning_rate": 1e-05,
      "loss": 1.0959,
      "step": 1360
    },
    {
      "epoch": 0.52,
      "learning_rate": 1e-05,
      "loss": 1.0585,
      "step": 1370
    },
    {
      "epoch": 0.52,
      "learning_rate": 1e-05,
      "loss": 1.0644,
      "step": 1380
    },
    {
      "epoch": 0.52,
      "learning_rate": 1e-05,
      "loss": 1.018,
      "step": 1390
    },
    {
      "epoch": 0.53,
      "learning_rate": 1e-05,
      "loss": 1.0694,
      "step": 1400
    },
    {
      "epoch": 0.53,
      "learning_rate": 1e-05,
      "loss": 1.0861,
      "step": 1410
    },
    {
      "epoch": 0.54,
      "learning_rate": 1e-05,
      "loss": 1.0846,
      "step": 1420
    },
    {
      "epoch": 0.54,
      "learning_rate": 1e-05,
      "loss": 1.1364,
      "step": 1430
    },
    {
      "epoch": 0.54,
      "learning_rate": 1e-05,
      "loss": 1.0504,
      "step": 1440
    },
    {
      "epoch": 0.55,
      "learning_rate": 1e-05,
      "loss": 1.0912,
      "step": 1450
    },
    {
      "epoch": 0.55,
      "learning_rate": 1e-05,
      "loss": 1.0495,
      "step": 1460
    },
    {
      "epoch": 0.55,
      "learning_rate": 1e-05,
      "loss": 1.0541,
      "step": 1470
    },
    {
      "epoch": 0.56,
      "learning_rate": 1e-05,
      "loss": 1.0622,
      "step": 1480
    },
    {
      "epoch": 0.56,
      "learning_rate": 1e-05,
      "loss": 1.0501,
      "step": 1490
    },
    {
      "epoch": 0.57,
      "learning_rate": 1e-05,
      "loss": 1.06,
      "step": 1500
    },
    {
      "epoch": 0.57,
      "eval_loss": 1.0966796875,
      "eval_runtime": 63.326,
      "eval_samples_per_second": 64.002,
      "eval_steps_per_second": 2.005,
      "step": 1500
    },
    {
      "epoch": 0.57,
      "learning_rate": 1e-05,
      "loss": 1.0511,
      "step": 1510
    },
    {
      "epoch": 0.57,
      "learning_rate": 1e-05,
      "loss": 1.025,
      "step": 1520
    },
    {
      "epoch": 0.58,
      "learning_rate": 1e-05,
      "loss": 1.0679,
      "step": 1530
    },
    {
      "epoch": 0.58,
      "learning_rate": 1e-05,
      "loss": 1.0676,
      "step": 1540
    },
    {
      "epoch": 0.59,
      "learning_rate": 1e-05,
      "loss": 1.0678,
      "step": 1550
    },
    {
      "epoch": 0.59,
      "learning_rate": 1e-05,
      "loss": 1.0666,
      "step": 1560
    },
    {
      "epoch": 0.59,
      "learning_rate": 1e-05,
      "loss": 1.1123,
      "step": 1570
    },
    {
      "epoch": 0.6,
      "learning_rate": 1e-05,
      "loss": 1.045,
      "step": 1580
    },
    {
      "epoch": 0.6,
      "learning_rate": 1e-05,
      "loss": 1.0311,
      "step": 1590
    },
    {
      "epoch": 0.6,
      "learning_rate": 1e-05,
      "loss": 1.0675,
      "step": 1600
    },
    {
      "epoch": 0.61,
      "learning_rate": 1e-05,
      "loss": 1.0417,
      "step": 1610
    },
    {
      "epoch": 0.61,
      "learning_rate": 1e-05,
      "loss": 1.0181,
      "step": 1620
    },
    {
      "epoch": 0.62,
      "learning_rate": 1e-05,
      "loss": 1.0364,
      "step": 1630
    },
    {
      "epoch": 0.62,
      "learning_rate": 1e-05,
      "loss": 1.023,
      "step": 1640
    },
    {
      "epoch": 0.62,
      "learning_rate": 1e-05,
      "loss": 1.0502,
      "step": 1650
    },
    {
      "epoch": 0.63,
      "learning_rate": 1e-05,
      "loss": 1.0789,
      "step": 1660
    },
    {
      "epoch": 0.63,
      "learning_rate": 1e-05,
      "loss": 1.062,
      "step": 1670
    },
    {
      "epoch": 0.63,
      "learning_rate": 1e-05,
      "loss": 1.0341,
      "step": 1680
    },
    {
      "epoch": 0.64,
      "learning_rate": 1e-05,
      "loss": 1.0126,
      "step": 1690
    },
    {
      "epoch": 0.64,
      "learning_rate": 1e-05,
      "loss": 1.0094,
      "step": 1700
    },
    {
      "epoch": 0.65,
      "learning_rate": 1e-05,
      "loss": 1.0378,
      "step": 1710
    },
    {
      "epoch": 0.65,
      "learning_rate": 1e-05,
      "loss": 1.0002,
      "step": 1720
    },
    {
      "epoch": 0.65,
      "learning_rate": 1e-05,
      "loss": 1.0389,
      "step": 1730
    },
    {
      "epoch": 0.66,
      "learning_rate": 1e-05,
      "loss": 1.0541,
      "step": 1740
    },
    {
      "epoch": 0.66,
      "learning_rate": 1e-05,
      "loss": 1.0776,
      "step": 1750
    },
    {
      "epoch": 0.66,
      "learning_rate": 1e-05,
      "loss": 1.0088,
      "step": 1760
    },
    {
      "epoch": 0.67,
      "learning_rate": 1e-05,
      "loss": 1.0139,
      "step": 1770
    },
    {
      "epoch": 0.67,
      "learning_rate": 1e-05,
      "loss": 0.9772,
      "step": 1780
    },
    {
      "epoch": 0.68,
      "learning_rate": 1e-05,
      "loss": 1.0363,
      "step": 1790
    },
    {
      "epoch": 0.68,
      "learning_rate": 1e-05,
      "loss": 1.0499,
      "step": 1800
    },
    {
      "epoch": 0.68,
      "learning_rate": 1e-05,
      "loss": 1.0171,
      "step": 1810
    },
    {
      "epoch": 0.69,
      "learning_rate": 1e-05,
      "loss": 1.0138,
      "step": 1820
    },
    {
      "epoch": 0.69,
      "learning_rate": 1e-05,
      "loss": 0.973,
      "step": 1830
    },
    {
      "epoch": 0.69,
      "learning_rate": 1e-05,
      "loss": 1.0022,
      "step": 1840
    },
    {
      "epoch": 0.7,
      "learning_rate": 1e-05,
      "loss": 1.049,
      "step": 1850
    },
    {
      "epoch": 0.7,
      "learning_rate": 1e-05,
      "loss": 1.0375,
      "step": 1860
    },
    {
      "epoch": 0.71,
      "learning_rate": 1e-05,
      "loss": 1.0273,
      "step": 1870
    },
    {
      "epoch": 0.71,
      "learning_rate": 1e-05,
      "loss": 1.0266,
      "step": 1880
    },
    {
      "epoch": 0.71,
      "learning_rate": 1e-05,
      "loss": 0.9753,
      "step": 1890
    },
    {
      "epoch": 0.72,
      "learning_rate": 1e-05,
      "loss": 1.032,
      "step": 1900
    },
    {
      "epoch": 0.72,
      "learning_rate": 1e-05,
      "loss": 1.0195,
      "step": 1910
    },
    {
      "epoch": 0.72,
      "learning_rate": 1e-05,
      "loss": 1.0094,
      "step": 1920
    },
    {
      "epoch": 0.73,
      "learning_rate": 1e-05,
      "loss": 0.9572,
      "step": 1930
    },
    {
      "epoch": 0.73,
      "learning_rate": 1e-05,
      "loss": 1.0336,
      "step": 1940
    },
    {
      "epoch": 0.74,
      "learning_rate": 1e-05,
      "loss": 1.0516,
      "step": 1950
    },
    {
      "epoch": 0.74,
      "learning_rate": 1e-05,
      "loss": 1.0169,
      "step": 1960
    },
    {
      "epoch": 0.74,
      "learning_rate": 1e-05,
      "loss": 0.9965,
      "step": 1970
    },
    {
      "epoch": 0.75,
      "learning_rate": 1e-05,
      "loss": 0.9932,
      "step": 1980
    },
    {
      "epoch": 0.75,
      "learning_rate": 1e-05,
      "loss": 0.9958,
      "step": 1990
    },
    {
      "epoch": 0.76,
      "learning_rate": 1e-05,
      "loss": 1.066,
      "step": 2000
    },
    {
      "epoch": 0.76,
      "eval_loss": 1.0322265625,
      "eval_runtime": 63.4242,
      "eval_samples_per_second": 63.903,
      "eval_steps_per_second": 2.002,
      "step": 2000
    },
    {
      "epoch": 0.76,
      "learning_rate": 1e-05,
      "loss": 1.0399,
      "step": 2010
    },
    {
      "epoch": 0.76,
      "learning_rate": 1e-05,
      "loss": 1.0053,
      "step": 2020
    },
    {
      "epoch": 0.77,
      "learning_rate": 1e-05,
      "loss": 0.9866,
      "step": 2030
    },
    {
      "epoch": 0.77,
      "learning_rate": 1e-05,
      "loss": 0.9962,
      "step": 2040
    },
    {
      "epoch": 0.77,
      "learning_rate": 1e-05,
      "loss": 1.003,
      "step": 2050
    },
    {
      "epoch": 0.78,
      "learning_rate": 1e-05,
      "loss": 0.9685,
      "step": 2060
    },
    {
      "epoch": 0.78,
      "learning_rate": 1e-05,
      "loss": 0.9901,
      "step": 2070
    },
    {
      "epoch": 0.79,
      "learning_rate": 1e-05,
      "loss": 1.0062,
      "step": 2080
    },
    {
      "epoch": 0.79,
      "learning_rate": 1e-05,
      "loss": 0.9667,
      "step": 2090
    },
    {
      "epoch": 0.79,
      "learning_rate": 1e-05,
      "loss": 1.0048,
      "step": 2100
    },
    {
      "epoch": 0.8,
      "learning_rate": 1e-05,
      "loss": 0.9898,
      "step": 2110
    },
    {
      "epoch": 0.8,
      "learning_rate": 1e-05,
      "loss": 0.989,
      "step": 2120
    },
    {
      "epoch": 0.8,
      "learning_rate": 1e-05,
      "loss": 0.9959,
      "step": 2130
    },
    {
      "epoch": 0.81,
      "learning_rate": 1e-05,
      "loss": 0.9936,
      "step": 2140
    },
    {
      "epoch": 0.81,
      "learning_rate": 1e-05,
      "loss": 0.9809,
      "step": 2150
    },
    {
      "epoch": 0.82,
      "learning_rate": 1e-05,
      "loss": 0.9878,
      "step": 2160
    },
    {
      "epoch": 0.82,
      "learning_rate": 1e-05,
      "loss": 0.9786,
      "step": 2170
    },
    {
      "epoch": 0.82,
      "learning_rate": 1e-05,
      "loss": 0.9745,
      "step": 2180
    },
    {
      "epoch": 0.83,
      "learning_rate": 1e-05,
      "loss": 0.9598,
      "step": 2190
    },
    {
      "epoch": 0.83,
      "learning_rate": 1e-05,
      "loss": 0.9861,
      "step": 2200
    },
    {
      "epoch": 0.83,
      "learning_rate": 1e-05,
      "loss": 1.0198,
      "step": 2210
    },
    {
      "epoch": 0.84,
      "learning_rate": 1e-05,
      "loss": 0.958,
      "step": 2220
    },
    {
      "epoch": 0.84,
      "learning_rate": 1e-05,
      "loss": 1.0112,
      "step": 2230
    },
    {
      "epoch": 0.85,
      "learning_rate": 1e-05,
      "loss": 0.9824,
      "step": 2240
    },
    {
      "epoch": 0.85,
      "learning_rate": 1e-05,
      "loss": 1.016,
      "step": 2250
    },
    {
      "epoch": 0.85,
      "learning_rate": 1e-05,
      "loss": 0.9787,
      "step": 2260
    },
    {
      "epoch": 0.86,
      "learning_rate": 1e-05,
      "loss": 0.9343,
      "step": 2270
    },
    {
      "epoch": 0.86,
      "learning_rate": 1e-05,
      "loss": 0.985,
      "step": 2280
    },
    {
      "epoch": 0.86,
      "learning_rate": 1e-05,
      "loss": 1.0047,
      "step": 2290
    },
    {
      "epoch": 0.87,
      "learning_rate": 1e-05,
      "loss": 0.9547,
      "step": 2300
    },
    {
      "epoch": 0.87,
      "learning_rate": 1e-05,
      "loss": 0.9782,
      "step": 2310
    },
    {
      "epoch": 0.88,
      "learning_rate": 1e-05,
      "loss": 0.9645,
      "step": 2320
    },
    {
      "epoch": 0.88,
      "learning_rate": 1e-05,
      "loss": 0.9736,
      "step": 2330
    },
    {
      "epoch": 0.88,
      "learning_rate": 1e-05,
      "loss": 0.98,
      "step": 2340
    },
    {
      "epoch": 0.89,
      "learning_rate": 1e-05,
      "loss": 0.9407,
      "step": 2350
    },
    {
      "epoch": 0.89,
      "learning_rate": 1e-05,
      "loss": 0.9727,
      "step": 2360
    },
    {
      "epoch": 0.89,
      "learning_rate": 1e-05,
      "loss": 0.9996,
      "step": 2370
    },
    {
      "epoch": 0.9,
      "learning_rate": 1e-05,
      "loss": 0.9139,
      "step": 2380
    },
    {
      "epoch": 0.9,
      "learning_rate": 1e-05,
      "loss": 0.9455,
      "step": 2390
    },
    {
      "epoch": 0.91,
      "learning_rate": 1e-05,
      "loss": 0.9754,
      "step": 2400
    },
    {
      "epoch": 0.91,
      "learning_rate": 1e-05,
      "loss": 0.9237,
      "step": 2410
    },
    {
      "epoch": 0.91,
      "learning_rate": 1e-05,
      "loss": 0.9409,
      "step": 2420
    },
    {
      "epoch": 0.92,
      "learning_rate": 1e-05,
      "loss": 0.9335,
      "step": 2430
    },
    {
      "epoch": 0.92,
      "learning_rate": 1e-05,
      "loss": 0.97,
      "step": 2440
    },
    {
      "epoch": 0.92,
      "learning_rate": 1e-05,
      "loss": 0.9355,
      "step": 2450
    },
    {
      "epoch": 0.93,
      "learning_rate": 1e-05,
      "loss": 0.9547,
      "step": 2460
    },
    {
      "epoch": 0.93,
      "learning_rate": 1e-05,
      "loss": 0.978,
      "step": 2470
    },
    {
      "epoch": 0.94,
      "learning_rate": 1e-05,
      "loss": 0.9312,
      "step": 2480
    },
    {
      "epoch": 0.94,
      "learning_rate": 1e-05,
      "loss": 0.9677,
      "step": 2490
    },
    {
      "epoch": 0.94,
      "learning_rate": 1e-05,
      "loss": 0.9637,
      "step": 2500
    },
    {
      "epoch": 0.94,
      "eval_loss": 0.986328125,
      "eval_runtime": 63.3465,
      "eval_samples_per_second": 63.981,
      "eval_steps_per_second": 2.005,
      "step": 2500
    },
    {
      "epoch": 0.95,
      "learning_rate": 1e-05,
      "loss": 0.9618,
      "step": 2510
    },
    {
      "epoch": 0.95,
      "learning_rate": 1e-05,
      "loss": 0.9371,
      "step": 2520
    },
    {
      "epoch": 0.96,
      "learning_rate": 1e-05,
      "loss": 0.9065,
      "step": 2530
    },
    {
      "epoch": 0.96,
      "learning_rate": 1e-05,
      "loss": 0.9718,
      "step": 2540
    },
    {
      "epoch": 0.96,
      "learning_rate": 1e-05,
      "loss": 0.9376,
      "step": 2550
    },
    {
      "epoch": 0.97,
      "learning_rate": 1e-05,
      "loss": 0.9367,
      "step": 2560
    },
    {
      "epoch": 0.97,
      "learning_rate": 1e-05,
      "loss": 0.9576,
      "step": 2570
    },
    {
      "epoch": 0.97,
      "learning_rate": 1e-05,
      "loss": 0.9476,
      "step": 2580
    },
    {
      "epoch": 0.98,
      "learning_rate": 1e-05,
      "loss": 0.941,
      "step": 2590
    },
    {
      "epoch": 0.98,
      "learning_rate": 1e-05,
      "loss": 0.9305,
      "step": 2600
    },
    {
      "epoch": 0.99,
      "learning_rate": 1e-05,
      "loss": 0.9501,
      "step": 2610
    },
    {
      "epoch": 0.99,
      "learning_rate": 1e-05,
      "loss": 0.9287,
      "step": 2620
    },
    {
      "epoch": 0.99,
      "learning_rate": 1e-05,
      "loss": 0.9217,
      "step": 2630
    },
    {
      "epoch": 1.0,
      "learning_rate": 1e-05,
      "loss": 0.9236,
      "step": 2640
    },
    {
      "epoch": 1.0,
      "learning_rate": 1e-05,
      "loss": 0.9539,
      "step": 2650
    },
    {
      "epoch": 1.0,
      "learning_rate": 1e-05,
      "loss": 0.7743,
      "step": 2660
    },
    {
      "epoch": 1.01,
      "learning_rate": 1e-05,
      "loss": 0.7539,
      "step": 2670
    },
    {
      "epoch": 1.01,
      "learning_rate": 1e-05,
      "loss": 0.7568,
      "step": 2680
    },
    {
      "epoch": 1.02,
      "learning_rate": 1e-05,
      "loss": 0.7495,
      "step": 2690
    },
    {
      "epoch": 1.02,
      "learning_rate": 1e-05,
      "loss": 0.7418,
      "step": 2700
    },
    {
      "epoch": 1.02,
      "learning_rate": 1e-05,
      "loss": 0.7482,
      "step": 2710
    },
    {
      "epoch": 1.03,
      "learning_rate": 1e-05,
      "loss": 0.7582,
      "step": 2720
    },
    {
      "epoch": 1.03,
      "learning_rate": 1e-05,
      "loss": 0.7262,
      "step": 2730
    },
    {
      "epoch": 1.03,
      "learning_rate": 1e-05,
      "loss": 0.7819,
      "step": 2740
    },
    {
      "epoch": 1.04,
      "learning_rate": 1e-05,
      "loss": 0.7121,
      "step": 2750
    },
    {
      "epoch": 1.04,
      "learning_rate": 1e-05,
      "loss": 0.7488,
      "step": 2760
    },
    {
      "epoch": 1.05,
      "learning_rate": 1e-05,
      "loss": 0.7519,
      "step": 2770
    },
    {
      "epoch": 1.05,
      "learning_rate": 1e-05,
      "loss": 0.749,
      "step": 2780
    },
    {
      "epoch": 1.05,
      "learning_rate": 1e-05,
      "loss": 0.7465,
      "step": 2790
    },
    {
      "epoch": 1.06,
      "learning_rate": 1e-05,
      "loss": 0.763,
      "step": 2800
    },
    {
      "epoch": 1.06,
      "learning_rate": 1e-05,
      "loss": 0.7538,
      "step": 2810
    },
    {
      "epoch": 1.06,
      "learning_rate": 1e-05,
      "loss": 0.767,
      "step": 2820
    },
    {
      "epoch": 1.07,
      "learning_rate": 1e-05,
      "loss": 0.7657,
      "step": 2830
    },
    {
      "epoch": 1.07,
      "learning_rate": 1e-05,
      "loss": 0.7488,
      "step": 2840
    },
    {
      "epoch": 1.08,
      "learning_rate": 1e-05,
      "loss": 0.7788,
      "step": 2850
    },
    {
      "epoch": 1.08,
      "learning_rate": 1e-05,
      "loss": 0.7324,
      "step": 2860
    },
    {
      "epoch": 1.08,
      "learning_rate": 1e-05,
      "loss": 0.7477,
      "step": 2870
    },
    {
      "epoch": 1.09,
      "learning_rate": 1e-05,
      "loss": 0.7515,
      "step": 2880
    },
    {
      "epoch": 1.09,
      "learning_rate": 1e-05,
      "loss": 0.7256,
      "step": 2890
    },
    {
      "epoch": 1.09,
      "learning_rate": 1e-05,
      "loss": 0.762,
      "step": 2900
    },
    {
      "epoch": 1.1,
      "learning_rate": 1e-05,
      "loss": 0.7495,
      "step": 2910
    },
    {
      "epoch": 1.1,
      "learning_rate": 1e-05,
      "loss": 0.756,
      "step": 2920
    },
    {
      "epoch": 1.11,
      "learning_rate": 1e-05,
      "loss": 0.7124,
      "step": 2930
    },
    {
      "epoch": 1.11,
      "learning_rate": 1e-05,
      "loss": 0.7457,
      "step": 2940
    },
    {
      "epoch": 1.11,
      "learning_rate": 1e-05,
      "loss": 0.7372,
      "step": 2950
    },
    {
      "epoch": 1.12,
      "learning_rate": 1e-05,
      "loss": 0.7532,
      "step": 2960
    },
    {
      "epoch": 1.12,
      "learning_rate": 1e-05,
      "loss": 0.7486,
      "step": 2970
    },
    {
      "epoch": 1.12,
      "learning_rate": 1e-05,
      "loss": 0.7631,
      "step": 2980
    },
    {
      "epoch": 1.13,
      "learning_rate": 1e-05,
      "loss": 0.7308,
      "step": 2990
    },
    {
      "epoch": 1.13,
      "learning_rate": 1e-05,
      "loss": 0.7603,
      "step": 3000
    },
    {
      "epoch": 1.13,
      "eval_loss": 0.984375,
      "eval_runtime": 63.3332,
      "eval_samples_per_second": 63.995,
      "eval_steps_per_second": 2.005,
      "step": 3000
    },
    {
      "epoch": 1.14,
      "learning_rate": 1e-05,
      "loss": 0.731,
      "step": 3010
    },
    {
      "epoch": 1.14,
      "learning_rate": 1e-05,
      "loss": 0.7598,
      "step": 3020
    },
    {
      "epoch": 1.14,
      "learning_rate": 1e-05,
      "loss": 0.7506,
      "step": 3030
    },
    {
      "epoch": 1.15,
      "learning_rate": 1e-05,
      "loss": 0.772,
      "step": 3040
    },
    {
      "epoch": 1.15,
      "learning_rate": 1e-05,
      "loss": 0.7408,
      "step": 3050
    },
    {
      "epoch": 1.16,
      "learning_rate": 1e-05,
      "loss": 0.7467,
      "step": 3060
    },
    {
      "epoch": 1.16,
      "learning_rate": 1e-05,
      "loss": 0.7571,
      "step": 3070
    },
    {
      "epoch": 1.16,
      "learning_rate": 1e-05,
      "loss": 0.7578,
      "step": 3080
    },
    {
      "epoch": 1.17,
      "learning_rate": 1e-05,
      "loss": 0.7561,
      "step": 3090
    },
    {
      "epoch": 1.17,
      "learning_rate": 1e-05,
      "loss": 0.7461,
      "step": 3100
    },
    {
      "epoch": 1.17,
      "learning_rate": 1e-05,
      "loss": 0.7609,
      "step": 3110
    },
    {
      "epoch": 1.18,
      "learning_rate": 1e-05,
      "loss": 0.7579,
      "step": 3120
    },
    {
      "epoch": 1.18,
      "learning_rate": 1e-05,
      "loss": 0.7346,
      "step": 3130
    },
    {
      "epoch": 1.19,
      "learning_rate": 1e-05,
      "loss": 0.7582,
      "step": 3140
    },
    {
      "epoch": 1.19,
      "learning_rate": 1e-05,
      "loss": 0.7515,
      "step": 3150
    },
    {
      "epoch": 1.19,
      "learning_rate": 1e-05,
      "loss": 0.7084,
      "step": 3160
    },
    {
      "epoch": 1.2,
      "learning_rate": 1e-05,
      "loss": 0.7534,
      "step": 3170
    },
    {
      "epoch": 1.2,
      "learning_rate": 1e-05,
      "loss": 0.722,
      "step": 3180
    },
    {
      "epoch": 1.2,
      "learning_rate": 1e-05,
      "loss": 0.7267,
      "step": 3190
    },
    {
      "epoch": 1.21,
      "learning_rate": 1e-05,
      "loss": 0.7094,
      "step": 3200
    },
    {
      "epoch": 1.21,
      "learning_rate": 1e-05,
      "loss": 0.7181,
      "step": 3210
    },
    {
      "epoch": 1.22,
      "learning_rate": 1e-05,
      "loss": 0.7269,
      "step": 3220
    },
    {
      "epoch": 1.22,
      "learning_rate": 1e-05,
      "loss": 0.7156,
      "step": 3230
    },
    {
      "epoch": 1.22,
      "learning_rate": 1e-05,
      "loss": 0.7522,
      "step": 3240
    },
    {
      "epoch": 1.23,
      "learning_rate": 1e-05,
      "loss": 0.7427,
      "step": 3250
    },
    {
      "epoch": 1.23,
      "learning_rate": 1e-05,
      "loss": 0.7311,
      "step": 3260
    },
    {
      "epoch": 1.23,
      "learning_rate": 1e-05,
      "loss": 0.7845,
      "step": 3270
    },
    {
      "epoch": 1.24,
      "learning_rate": 1e-05,
      "loss": 0.7514,
      "step": 3280
    },
    {
      "epoch": 1.24,
      "learning_rate": 1e-05,
      "loss": 0.7294,
      "step": 3290
    },
    {
      "epoch": 1.25,
      "learning_rate": 1e-05,
      "loss": 0.7567,
      "step": 3300
    },
    {
      "epoch": 1.25,
      "learning_rate": 1e-05,
      "loss": 0.7449,
      "step": 3310
    },
    {
      "epoch": 1.25,
      "learning_rate": 1e-05,
      "loss": 0.7431,
      "step": 3320
    },
    {
      "epoch": 1.26,
      "learning_rate": 1e-05,
      "loss": 0.7834,
      "step": 3330
    },
    {
      "epoch": 1.26,
      "learning_rate": 1e-05,
      "loss": 0.7338,
      "step": 3340
    },
    {
      "epoch": 1.26,
      "learning_rate": 1e-05,
      "loss": 0.7162,
      "step": 3350
    },
    {
      "epoch": 1.27,
      "learning_rate": 1e-05,
      "loss": 0.7502,
      "step": 3360
    },
    {
      "epoch": 1.27,
      "learning_rate": 1e-05,
      "loss": 0.7363,
      "step": 3370
    },
    {
      "epoch": 1.28,
      "learning_rate": 1e-05,
      "loss": 0.7367,
      "step": 3380
    },
    {
      "epoch": 1.28,
      "learning_rate": 1e-05,
      "loss": 0.7237,
      "step": 3390
    },
    {
      "epoch": 1.28,
      "learning_rate": 1e-05,
      "loss": 0.7386,
      "step": 3400
    },
    {
      "epoch": 1.29,
      "learning_rate": 1e-05,
      "loss": 0.7276,
      "step": 3410
    },
    {
      "epoch": 1.29,
      "learning_rate": 1e-05,
      "loss": 0.7408,
      "step": 3420
    },
    {
      "epoch": 1.29,
      "learning_rate": 1e-05,
      "loss": 0.7531,
      "step": 3430
    },
    {
      "epoch": 1.3,
      "learning_rate": 1e-05,
      "loss": 0.754,
      "step": 3440
    },
    {
      "epoch": 1.3,
      "learning_rate": 1e-05,
      "loss": 0.7348,
      "step": 3450
    },
    {
      "epoch": 1.31,
      "learning_rate": 1e-05,
      "loss": 0.7325,
      "step": 3460
    },
    {
      "epoch": 1.31,
      "learning_rate": 1e-05,
      "loss": 0.7191,
      "step": 3470
    },
    {
      "epoch": 1.31,
      "learning_rate": 1e-05,
      "loss": 0.7066,
      "step": 3480
    },
    {
      "epoch": 1.32,
      "learning_rate": 1e-05,
      "loss": 0.7459,
      "step": 3490
    },
    {
      "epoch": 1.32,
      "learning_rate": 1e-05,
      "loss": 0.7565,
      "step": 3500
    },
    {
      "epoch": 1.32,
      "eval_loss": 0.955078125,
      "eval_runtime": 63.3357,
      "eval_samples_per_second": 63.992,
      "eval_steps_per_second": 2.005,
      "step": 3500
    },
    {
      "epoch": 1.33,
      "learning_rate": 1e-05,
      "loss": 0.7043,
      "step": 3510
    },
    {
      "epoch": 1.33,
      "learning_rate": 1e-05,
      "loss": 0.7165,
      "step": 3520
    },
    {
      "epoch": 1.33,
      "learning_rate": 1e-05,
      "loss": 0.7536,
      "step": 3530
    },
    {
      "epoch": 1.34,
      "learning_rate": 1e-05,
      "loss": 0.7346,
      "step": 3540
    },
    {
      "epoch": 1.34,
      "learning_rate": 1e-05,
      "loss": 0.7011,
      "step": 3550
    },
    {
      "epoch": 1.34,
      "learning_rate": 1e-05,
      "loss": 0.7522,
      "step": 3560
    },
    {
      "epoch": 1.35,
      "learning_rate": 1e-05,
      "loss": 0.7458,
      "step": 3570
    },
    {
      "epoch": 1.35,
      "learning_rate": 1e-05,
      "loss": 0.7432,
      "step": 3580
    },
    {
      "epoch": 1.36,
      "learning_rate": 1e-05,
      "loss": 0.7417,
      "step": 3590
    },
    {
      "epoch": 1.36,
      "learning_rate": 1e-05,
      "loss": 0.7519,
      "step": 3600
    },
    {
      "epoch": 1.36,
      "learning_rate": 1e-05,
      "loss": 0.7754,
      "step": 3610
    },
    {
      "epoch": 1.37,
      "learning_rate": 1e-05,
      "loss": 0.7256,
      "step": 3620
    },
    {
      "epoch": 1.37,
      "learning_rate": 1e-05,
      "loss": 0.7529,
      "step": 3630
    },
    {
      "epoch": 1.37,
      "learning_rate": 1e-05,
      "loss": 0.7545,
      "step": 3640
    },
    {
      "epoch": 1.38,
      "learning_rate": 1e-05,
      "loss": 0.7778,
      "step": 3650
    },
    {
      "epoch": 1.38,
      "learning_rate": 1e-05,
      "loss": 0.6999,
      "step": 3660
    },
    {
      "epoch": 1.39,
      "learning_rate": 1e-05,
      "loss": 0.741,
      "step": 3670
    },
    {
      "epoch": 1.39,
      "learning_rate": 1e-05,
      "loss": 0.7464,
      "step": 3680
    },
    {
      "epoch": 1.39,
      "learning_rate": 1e-05,
      "loss": 0.7568,
      "step": 3690
    },
    {
      "epoch": 1.4,
      "learning_rate": 1e-05,
      "loss": 0.7132,
      "step": 3700
    },
    {
      "epoch": 1.4,
      "learning_rate": 1e-05,
      "loss": 0.7391,
      "step": 3710
    },
    {
      "epoch": 1.4,
      "learning_rate": 1e-05,
      "loss": 0.7498,
      "step": 3720
    },
    {
      "epoch": 1.41,
      "learning_rate": 1e-05,
      "loss": 0.7508,
      "step": 3730
    },
    {
      "epoch": 1.41,
      "learning_rate": 1e-05,
      "loss": 0.7193,
      "step": 3740
    },
    {
      "epoch": 1.42,
      "learning_rate": 1e-05,
      "loss": 0.74,
      "step": 3750
    },
    {
      "epoch": 1.42,
      "learning_rate": 1e-05,
      "loss": 0.7091,
      "step": 3760
    },
    {
      "epoch": 1.42,
      "learning_rate": 1e-05,
      "loss": 0.7419,
      "step": 3770
    },
    {
      "epoch": 1.43,
      "learning_rate": 1e-05,
      "loss": 0.7241,
      "step": 3780
    },
    {
      "epoch": 1.43,
      "learning_rate": 1e-05,
      "loss": 0.7318,
      "step": 3790
    },
    {
      "epoch": 1.43,
      "learning_rate": 1e-05,
      "loss": 0.716,
      "step": 3800
    },
    {
      "epoch": 1.44,
      "learning_rate": 1e-05,
      "loss": 0.7247,
      "step": 3810
    },
    {
      "epoch": 1.44,
      "learning_rate": 1e-05,
      "loss": 0.7435,
      "step": 3820
    },
    {
      "epoch": 1.45,
      "learning_rate": 1e-05,
      "loss": 0.7483,
      "step": 3830
    },
    {
      "epoch": 1.45,
      "learning_rate": 1e-05,
      "loss": 0.7394,
      "step": 3840
    },
    {
      "epoch": 1.45,
      "learning_rate": 1e-05,
      "loss": 0.7231,
      "step": 3850
    },
    {
      "epoch": 1.46,
      "learning_rate": 1e-05,
      "loss": 0.7005,
      "step": 3860
    },
    {
      "epoch": 1.46,
      "learning_rate": 1e-05,
      "loss": 0.7177,
      "step": 3870
    },
    {
      "epoch": 1.46,
      "learning_rate": 1e-05,
      "loss": 0.7449,
      "step": 3880
    },
    {
      "epoch": 1.47,
      "learning_rate": 1e-05,
      "loss": 0.7093,
      "step": 3890
    },
    {
      "epoch": 1.47,
      "learning_rate": 1e-05,
      "loss": 0.7262,
      "step": 3900
    },
    {
      "epoch": 1.48,
      "learning_rate": 1e-05,
      "loss": 0.7039,
      "step": 3910
    },
    {
      "epoch": 1.48,
      "learning_rate": 1e-05,
      "loss": 0.7563,
      "step": 3920
    },
    {
      "epoch": 1.48,
      "learning_rate": 1e-05,
      "loss": 0.7385,
      "step": 3930
    },
    {
      "epoch": 1.49,
      "learning_rate": 1e-05,
      "loss": 0.7416,
      "step": 3940
    },
    {
      "epoch": 1.49,
      "learning_rate": 1e-05,
      "loss": 0.725,
      "step": 3950
    },
    {
      "epoch": 1.49,
      "learning_rate": 1e-05,
      "loss": 0.7311,
      "step": 3960
    },
    {
      "epoch": 1.5,
      "learning_rate": 1e-05,
      "loss": 0.7245,
      "step": 3970
    },
    {
      "epoch": 1.5,
      "learning_rate": 1e-05,
      "loss": 0.7307,
      "step": 3980
    },
    {
      "epoch": 1.51,
      "learning_rate": 1e-05,
      "loss": 0.7105,
      "step": 3990
    },
    {
      "epoch": 1.51,
      "learning_rate": 1e-05,
      "loss": 0.7363,
      "step": 4000
    },
    {
      "epoch": 1.51,
      "eval_loss": 0.93310546875,
      "eval_runtime": 63.3035,
      "eval_samples_per_second": 64.025,
      "eval_steps_per_second": 2.006,
      "step": 4000
    },
    {
      "epoch": 1.51,
      "learning_rate": 1e-05,
      "loss": 0.726,
      "step": 4010
    },
    {
      "epoch": 1.52,
      "learning_rate": 1e-05,
      "loss": 0.7045,
      "step": 4020
    },
    {
      "epoch": 1.52,
      "learning_rate": 1e-05,
      "loss": 0.7705,
      "step": 4030
    },
    {
      "epoch": 1.53,
      "learning_rate": 1e-05,
      "loss": 0.7275,
      "step": 4040
    },
    {
      "epoch": 1.53,
      "learning_rate": 1e-05,
      "loss": 0.7203,
      "step": 4050
    },
    {
      "epoch": 1.53,
      "learning_rate": 1e-05,
      "loss": 0.7247,
      "step": 4060
    },
    {
      "epoch": 1.54,
      "learning_rate": 1e-05,
      "loss": 0.7406,
      "step": 4070
    },
    {
      "epoch": 1.54,
      "learning_rate": 1e-05,
      "loss": 0.7232,
      "step": 4080
    },
    {
      "epoch": 1.54,
      "learning_rate": 1e-05,
      "loss": 0.7175,
      "step": 4090
    },
    {
      "epoch": 1.55,
      "learning_rate": 1e-05,
      "loss": 0.7253,
      "step": 4100
    },
    {
      "epoch": 1.55,
      "learning_rate": 1e-05,
      "loss": 0.711,
      "step": 4110
    },
    {
      "epoch": 1.56,
      "learning_rate": 1e-05,
      "loss": 0.7011,
      "step": 4120
    },
    {
      "epoch": 1.56,
      "learning_rate": 1e-05,
      "loss": 0.698,
      "step": 4130
    },
    {
      "epoch": 1.56,
      "learning_rate": 1e-05,
      "loss": 0.7018,
      "step": 4140
    },
    {
      "epoch": 1.57,
      "learning_rate": 1e-05,
      "loss": 0.7196,
      "step": 4150
    },
    {
      "epoch": 1.57,
      "learning_rate": 1e-05,
      "loss": 0.7241,
      "step": 4160
    },
    {
      "epoch": 1.57,
      "learning_rate": 1e-05,
      "loss": 0.7109,
      "step": 4170
    },
    {
      "epoch": 1.58,
      "learning_rate": 1e-05,
      "loss": 0.7066,
      "step": 4180
    },
    {
      "epoch": 1.58,
      "learning_rate": 1e-05,
      "loss": 0.7301,
      "step": 4190
    },
    {
      "epoch": 1.59,
      "learning_rate": 1e-05,
      "loss": 0.7094,
      "step": 4200
    },
    {
      "epoch": 1.59,
      "learning_rate": 1e-05,
      "loss": 0.7125,
      "step": 4210
    },
    {
      "epoch": 1.59,
      "learning_rate": 1e-05,
      "loss": 0.7067,
      "step": 4220
    },
    {
      "epoch": 1.6,
      "learning_rate": 1e-05,
      "loss": 0.7244,
      "step": 4230
    },
    {
      "epoch": 1.6,
      "learning_rate": 1e-05,
      "loss": 0.7316,
      "step": 4240
    },
    {
      "epoch": 1.6,
      "learning_rate": 1e-05,
      "loss": 0.7565,
      "step": 4250
    },
    {
      "epoch": 1.61,
      "learning_rate": 1e-05,
      "loss": 0.7009,
      "step": 4260
    },
    {
      "epoch": 1.61,
      "learning_rate": 1e-05,
      "loss": 0.71,
      "step": 4270
    },
    {
      "epoch": 1.62,
      "learning_rate": 1e-05,
      "loss": 0.706,
      "step": 4280
    },
    {
      "epoch": 1.62,
      "learning_rate": 1e-05,
      "loss": 0.688,
      "step": 4290
    },
    {
      "epoch": 1.62,
      "learning_rate": 1e-05,
      "loss": 0.7051,
      "step": 4300
    },
    {
      "epoch": 1.63,
      "learning_rate": 1e-05,
      "loss": 0.7253,
      "step": 4310
    },
    {
      "epoch": 1.63,
      "learning_rate": 1e-05,
      "loss": 0.7152,
      "step": 4320
    },
    {
      "epoch": 1.63,
      "learning_rate": 1e-05,
      "loss": 0.7489,
      "step": 4330
    },
    {
      "epoch": 1.64,
      "learning_rate": 1e-05,
      "loss": 0.7165,
      "step": 4340
    },
    {
      "epoch": 1.64,
      "learning_rate": 1e-05,
      "loss": 0.7563,
      "step": 4350
    },
    {
      "epoch": 1.65,
      "learning_rate": 1e-05,
      "loss": 0.7121,
      "step": 4360
    },
    {
      "epoch": 1.65,
      "learning_rate": 1e-05,
      "loss": 0.6982,
      "step": 4370
    },
    {
      "epoch": 1.65,
      "learning_rate": 1e-05,
      "loss": 0.7037,
      "step": 4380
    },
    {
      "epoch": 1.66,
      "learning_rate": 1e-05,
      "loss": 0.7059,
      "step": 4390
    },
    {
      "epoch": 1.66,
      "learning_rate": 1e-05,
      "loss": 0.7096,
      "step": 4400
    },
    {
      "epoch": 1.66,
      "learning_rate": 1e-05,
      "loss": 0.7103,
      "step": 4410
    },
    {
      "epoch": 1.67,
      "learning_rate": 1e-05,
      "loss": 0.7088,
      "step": 4420
    },
    {
      "epoch": 1.67,
      "learning_rate": 1e-05,
      "loss": 0.7573,
      "step": 4430
    },
    {
      "epoch": 1.68,
      "learning_rate": 1e-05,
      "loss": 0.7192,
      "step": 4440
    },
    {
      "epoch": 1.68,
      "learning_rate": 1e-05,
      "loss": 0.7296,
      "step": 4450
    },
    {
      "epoch": 1.68,
      "learning_rate": 1e-05,
      "loss": 0.7115,
      "step": 4460
    },
    {
      "epoch": 1.69,
      "learning_rate": 1e-05,
      "loss": 0.7153,
      "step": 4470
    },
    {
      "epoch": 1.69,
      "learning_rate": 1e-05,
      "loss": 0.6649,
      "step": 4480
    },
    {
      "epoch": 1.69,
      "learning_rate": 1e-05,
      "loss": 0.6822,
      "step": 4490
    },
    {
      "epoch": 1.7,
      "learning_rate": 1e-05,
      "loss": 0.7234,
      "step": 4500
    },
    {
      "epoch": 1.7,
      "eval_loss": 0.908203125,
      "eval_runtime": 63.3469,
      "eval_samples_per_second": 63.981,
      "eval_steps_per_second": 2.005,
      "step": 4500
    },
    {
      "epoch": 1.7,
      "learning_rate": 1e-05,
      "loss": 0.7053,
      "step": 4510
    },
    {
      "epoch": 1.71,
      "learning_rate": 1e-05,
      "loss": 0.7222,
      "step": 4520
    },
    {
      "epoch": 1.71,
      "learning_rate": 1e-05,
      "loss": 0.6748,
      "step": 4530
    },
    {
      "epoch": 1.71,
      "learning_rate": 1e-05,
      "loss": 0.7166,
      "step": 4540
    },
    {
      "epoch": 1.72,
      "learning_rate": 1e-05,
      "loss": 0.7074,
      "step": 4550
    },
    {
      "epoch": 1.72,
      "learning_rate": 1e-05,
      "loss": 0.7214,
      "step": 4560
    },
    {
      "epoch": 1.73,
      "learning_rate": 1e-05,
      "loss": 0.6938,
      "step": 4570
    },
    {
      "epoch": 1.73,
      "learning_rate": 1e-05,
      "loss": 0.7237,
      "step": 4580
    },
    {
      "epoch": 1.73,
      "learning_rate": 1e-05,
      "loss": 0.6958,
      "step": 4590
    },
    {
      "epoch": 1.74,
      "learning_rate": 1e-05,
      "loss": 0.6849,
      "step": 4600
    },
    {
      "epoch": 1.74,
      "learning_rate": 1e-05,
      "loss": 0.6959,
      "step": 4610
    },
    {
      "epoch": 1.74,
      "learning_rate": 1e-05,
      "loss": 0.7126,
      "step": 4620
    },
    {
      "epoch": 1.75,
      "learning_rate": 1e-05,
      "loss": 0.7266,
      "step": 4630
    },
    {
      "epoch": 1.75,
      "learning_rate": 1e-05,
      "loss": 0.6683,
      "step": 4640
    },
    {
      "epoch": 1.76,
      "learning_rate": 1e-05,
      "loss": 0.7248,
      "step": 4650
    },
    {
      "epoch": 1.76,
      "learning_rate": 1e-05,
      "loss": 0.7117,
      "step": 4660
    },
    {
      "epoch": 1.76,
      "learning_rate": 1e-05,
      "loss": 0.6998,
      "step": 4670
    },
    {
      "epoch": 1.77,
      "learning_rate": 1e-05,
      "loss": 0.6986,
      "step": 4680
    },
    {
      "epoch": 1.77,
      "learning_rate": 1e-05,
      "loss": 0.6631,
      "step": 4690
    },
    {
      "epoch": 1.77,
      "learning_rate": 1e-05,
      "loss": 0.6857,
      "step": 4700
    },
    {
      "epoch": 1.78,
      "learning_rate": 1e-05,
      "loss": 0.6649,
      "step": 4710
    },
    {
      "epoch": 1.78,
      "learning_rate": 1e-05,
      "loss": 0.699,
      "step": 4720
    },
    {
      "epoch": 1.79,
      "learning_rate": 1e-05,
      "loss": 0.6991,
      "step": 4730
    },
    {
      "epoch": 1.79,
      "learning_rate": 1e-05,
      "loss": 0.703,
      "step": 4740
    },
    {
      "epoch": 1.79,
      "learning_rate": 1e-05,
      "loss": 0.7259,
      "step": 4750
    },
    {
      "epoch": 1.8,
      "learning_rate": 1e-05,
      "loss": 0.7023,
      "step": 4760
    },
    {
      "epoch": 1.8,
      "learning_rate": 1e-05,
      "loss": 0.7281,
      "step": 4770
    },
    {
      "epoch": 1.8,
      "learning_rate": 1e-05,
      "loss": 0.7213,
      "step": 4780
    },
    {
      "epoch": 1.81,
      "learning_rate": 1e-05,
      "loss": 0.6808,
      "step": 4790
    },
    {
      "epoch": 1.81,
      "learning_rate": 1e-05,
      "loss": 0.7253,
      "step": 4800
    },
    {
      "epoch": 1.82,
      "learning_rate": 1e-05,
      "loss": 0.7199,
      "step": 4810
    },
    {
      "epoch": 1.82,
      "learning_rate": 1e-05,
      "loss": 0.745,
      "step": 4820
    },
    {
      "epoch": 1.82,
      "learning_rate": 1e-05,
      "loss": 0.6928,
      "step": 4830
    },
    {
      "epoch": 1.83,
      "learning_rate": 1e-05,
      "loss": 0.6942,
      "step": 4840
    },
    {
      "epoch": 1.83,
      "learning_rate": 1e-05,
      "loss": 0.6963,
      "step": 4850
    },
    {
      "epoch": 1.83,
      "learning_rate": 1e-05,
      "loss": 0.7025,
      "step": 4860
    },
    {
      "epoch": 1.84,
      "learning_rate": 1e-05,
      "loss": 0.6947,
      "step": 4870
    },
    {
      "epoch": 1.84,
      "learning_rate": 1e-05,
      "loss": 0.6959,
      "step": 4880
    },
    {
      "epoch": 1.85,
      "learning_rate": 1e-05,
      "loss": 0.6984,
      "step": 4890
    },
    {
      "epoch": 1.85,
      "learning_rate": 1e-05,
      "loss": 0.7351,
      "step": 4900
    },
    {
      "epoch": 1.85,
      "learning_rate": 1e-05,
      "loss": 0.7095,
      "step": 4910
    },
    {
      "epoch": 1.86,
      "learning_rate": 1e-05,
      "loss": 0.6915,
      "step": 4920
    },
    {
      "epoch": 1.86,
      "learning_rate": 1e-05,
      "loss": 0.694,
      "step": 4930
    },
    {
      "epoch": 1.86,
      "learning_rate": 1e-05,
      "loss": 0.6907,
      "step": 4940
    },
    {
      "epoch": 1.87,
      "learning_rate": 1e-05,
      "loss": 0.7157,
      "step": 4950
    },
    {
      "epoch": 1.87,
      "learning_rate": 1e-05,
      "loss": 0.6906,
      "step": 4960
    },
    {
      "epoch": 1.88,
      "learning_rate": 1e-05,
      "loss": 0.675,
      "step": 4970
    },
    {
      "epoch": 1.88,
      "learning_rate": 1e-05,
      "loss": 0.6764,
      "step": 4980
    },
    {
      "epoch": 1.88,
      "learning_rate": 1e-05,
      "loss": 0.6821,
      "step": 4990
    },
    {
      "epoch": 1.89,
      "learning_rate": 1e-05,
      "loss": 0.7015,
      "step": 5000
    },
    {
      "epoch": 1.89,
      "eval_loss": 0.890625,
      "eval_runtime": 63.2179,
      "eval_samples_per_second": 64.112,
      "eval_steps_per_second": 2.009,
      "step": 5000
    }
  ],
  "max_steps": 26490,
  "num_train_epochs": 10,
  "total_flos": 851208285061120.0,
  "trial_name": null,
  "trial_params": null
}