{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1386,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 4.875,
      "learning_rate": 0.0002,
      "loss": 18.0666,
      "step": 10
    },
    {
      "epoch": 0.04,
      "grad_norm": 8.625,
      "learning_rate": 0.0002,
      "loss": 11.9963,
      "step": 20
    },
    {
      "epoch": 0.06,
      "grad_norm": 3.84375,
      "learning_rate": 0.0002,
      "loss": 6.0667,
      "step": 30
    },
    {
      "epoch": 0.09,
      "grad_norm": 2.375,
      "learning_rate": 0.0002,
      "loss": 2.4769,
      "step": 40
    },
    {
      "epoch": 0.11,
      "grad_norm": 2.109375,
      "learning_rate": 0.0002,
      "loss": 2.2233,
      "step": 50
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0002,
      "loss": 2.1739,
      "step": 60
    },
    {
      "epoch": 0.15,
      "grad_norm": 3.375,
      "learning_rate": 0.0002,
      "loss": 2.0811,
      "step": 70
    },
    {
      "epoch": 0.17,
      "grad_norm": 4.0,
      "learning_rate": 0.0002,
      "loss": 1.9928,
      "step": 80
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002,
      "loss": 1.9649,
      "step": 90
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0002,
      "loss": 1.9645,
      "step": 100
    },
    {
      "epoch": 0.24,
      "grad_norm": 2.171875,
      "learning_rate": 0.0002,
      "loss": 1.9812,
      "step": 110
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002,
      "loss": 1.9106,
      "step": 120
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.984375,
      "learning_rate": 0.0002,
      "loss": 1.8943,
      "step": 130
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002,
      "loss": 1.9205,
      "step": 140
    },
    {
      "epoch": 0.32,
      "grad_norm": 2.234375,
      "learning_rate": 0.0002,
      "loss": 1.9445,
      "step": 150
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002,
      "loss": 1.8882,
      "step": 160
    },
    {
      "epoch": 0.37,
      "grad_norm": 2.984375,
      "learning_rate": 0.0002,
      "loss": 1.8806,
      "step": 170
    },
    {
      "epoch": 0.39,
      "grad_norm": 2.46875,
      "learning_rate": 0.0002,
      "loss": 1.8692,
      "step": 180
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002,
      "loss": 1.8291,
      "step": 190
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002,
      "loss": 1.8069,
      "step": 200
    },
    {
      "epoch": 0.45,
      "grad_norm": 2.671875,
      "learning_rate": 0.0002,
      "loss": 1.8406,
      "step": 210
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.625,
      "learning_rate": 0.0002,
      "loss": 1.8694,
      "step": 220
    },
    {
      "epoch": 0.5,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0002,
      "loss": 1.821,
      "step": 230
    },
    {
      "epoch": 0.52,
      "grad_norm": 2.03125,
      "learning_rate": 0.0002,
      "loss": 1.8243,
      "step": 240
    },
    {
      "epoch": 0.54,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002,
      "loss": 1.7678,
      "step": 250
    },
    {
      "epoch": 0.56,
      "grad_norm": 2.0,
      "learning_rate": 0.0002,
      "loss": 1.8301,
      "step": 260
    },
    {
      "epoch": 0.58,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002,
      "loss": 1.8845,
      "step": 270
    },
    {
      "epoch": 0.61,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002,
      "loss": 1.8901,
      "step": 280
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002,
      "loss": 1.8198,
      "step": 290
    },
    {
      "epoch": 0.65,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002,
      "loss": 1.7615,
      "step": 300
    },
    {
      "epoch": 0.67,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002,
      "loss": 1.8717,
      "step": 310
    },
    {
      "epoch": 0.69,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002,
      "loss": 1.7862,
      "step": 320
    },
    {
      "epoch": 0.71,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0002,
      "loss": 1.8674,
      "step": 330
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002,
      "loss": 1.822,
      "step": 340
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.921875,
      "learning_rate": 0.0002,
      "loss": 1.7834,
      "step": 350
    },
    {
      "epoch": 0.78,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002,
      "loss": 1.8157,
      "step": 360
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002,
      "loss": 1.7164,
      "step": 370
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.875,
      "learning_rate": 0.0002,
      "loss": 1.835,
      "step": 380
    },
    {
      "epoch": 0.84,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002,
      "loss": 1.7927,
      "step": 390
    },
    {
      "epoch": 0.87,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0002,
      "loss": 1.7038,
      "step": 400
    },
    {
      "epoch": 0.89,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002,
      "loss": 1.7914,
      "step": 410
    },
    {
      "epoch": 0.91,
      "grad_norm": 2.921875,
      "learning_rate": 0.0002,
      "loss": 1.7892,
      "step": 420
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0002,
      "loss": 1.7335,
      "step": 430
    },
    {
      "epoch": 0.95,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002,
      "loss": 1.8013,
      "step": 440
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002,
      "loss": 1.7118,
      "step": 450
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002,
      "loss": 1.8202,
      "step": 460
    },
    {
      "epoch": 1.02,
      "grad_norm": 2.265625,
      "learning_rate": 0.0002,
      "loss": 1.7011,
      "step": 470
    },
    {
      "epoch": 1.04,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002,
      "loss": 1.6769,
      "step": 480
    },
    {
      "epoch": 1.06,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002,
      "loss": 1.6393,
      "step": 490
    },
    {
      "epoch": 1.08,
      "grad_norm": 1.25,
      "learning_rate": 0.0002,
      "loss": 1.6685,
      "step": 500
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0002,
      "loss": 1.665,
      "step": 510
    },
    {
      "epoch": 1.13,
      "grad_norm": 2.046875,
      "learning_rate": 0.0002,
      "loss": 1.6446,
      "step": 520
    },
    {
      "epoch": 1.15,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002,
      "loss": 1.5743,
      "step": 530
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002,
      "loss": 1.6622,
      "step": 540
    },
    {
      "epoch": 1.19,
      "grad_norm": 1.125,
      "learning_rate": 0.0002,
      "loss": 1.5741,
      "step": 550
    },
    {
      "epoch": 1.21,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002,
      "loss": 1.6434,
      "step": 560
    },
    {
      "epoch": 1.23,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002,
      "loss": 1.6343,
      "step": 570
    },
    {
      "epoch": 1.26,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002,
      "loss": 1.662,
      "step": 580
    },
    {
      "epoch": 1.28,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002,
      "loss": 1.6246,
      "step": 590
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.96875,
      "learning_rate": 0.0002,
      "loss": 1.614,
      "step": 600
    },
    {
      "epoch": 1.32,
      "grad_norm": 2.703125,
      "learning_rate": 0.0002,
      "loss": 1.6817,
      "step": 610
    },
    {
      "epoch": 1.34,
      "grad_norm": 2.203125,
      "learning_rate": 0.0002,
      "loss": 1.6343,
      "step": 620
    },
    {
      "epoch": 1.36,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002,
      "loss": 1.6237,
      "step": 630
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0002,
      "loss": 1.6367,
      "step": 640
    },
    {
      "epoch": 1.41,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002,
      "loss": 1.6304,
      "step": 650
    },
    {
      "epoch": 1.43,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002,
      "loss": 1.5319,
      "step": 660
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.921875,
      "learning_rate": 0.0002,
      "loss": 1.643,
      "step": 670
    },
    {
      "epoch": 1.47,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002,
      "loss": 1.6247,
      "step": 680
    },
    {
      "epoch": 1.49,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0002,
      "loss": 1.5913,
      "step": 690
    },
    {
      "epoch": 1.52,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002,
      "loss": 1.6415,
      "step": 700
    },
    {
      "epoch": 1.54,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002,
      "loss": 1.636,
      "step": 710
    },
    {
      "epoch": 1.56,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002,
      "loss": 1.6304,
      "step": 720
    },
    {
      "epoch": 1.58,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002,
      "loss": 1.6156,
      "step": 730
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002,
      "loss": 1.5906,
      "step": 740
    },
    {
      "epoch": 1.62,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002,
      "loss": 1.5717,
      "step": 750
    },
    {
      "epoch": 1.65,
      "grad_norm": 1.25,
      "learning_rate": 0.0002,
      "loss": 1.6747,
      "step": 760
    },
    {
      "epoch": 1.67,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002,
      "loss": 1.6095,
      "step": 770
    },
    {
      "epoch": 1.69,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002,
      "loss": 1.6571,
      "step": 780
    },
    {
      "epoch": 1.71,
      "grad_norm": 1.5,
      "learning_rate": 0.0002,
      "loss": 1.5913,
      "step": 790
    },
    {
      "epoch": 1.73,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002,
      "loss": 1.592,
      "step": 800
    },
    {
      "epoch": 1.75,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0002,
      "loss": 1.5279,
      "step": 810
    },
    {
      "epoch": 1.77,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002,
      "loss": 1.6599,
      "step": 820
    },
    {
      "epoch": 1.8,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002,
      "loss": 1.6947,
      "step": 830
    },
    {
      "epoch": 1.82,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002,
      "loss": 1.608,
      "step": 840
    },
    {
      "epoch": 1.84,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002,
      "loss": 1.6716,
      "step": 850
    },
    {
      "epoch": 1.86,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002,
      "loss": 1.6847,
      "step": 860
    },
    {
      "epoch": 1.88,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002,
      "loss": 1.5283,
      "step": 870
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0002,
      "loss": 1.6639,
      "step": 880
    },
    {
      "epoch": 1.93,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002,
      "loss": 1.6154,
      "step": 890
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0002,
      "loss": 1.6248,
      "step": 900
    },
    {
      "epoch": 1.97,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002,
      "loss": 1.5525,
      "step": 910
    },
    {
      "epoch": 1.99,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0002,
      "loss": 1.5069,
      "step": 920
    },
    {
      "epoch": 2.01,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002,
      "loss": 1.5104,
      "step": 930
    },
    {
      "epoch": 2.03,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002,
      "loss": 1.3058,
      "step": 940
    },
    {
      "epoch": 2.06,
      "grad_norm": 1.25,
      "learning_rate": 0.0002,
      "loss": 1.4888,
      "step": 950
    },
    {
      "epoch": 2.08,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002,
      "loss": 1.3972,
      "step": 960
    },
    {
      "epoch": 2.1,
      "grad_norm": 1.375,
      "learning_rate": 0.0002,
      "loss": 1.4524,
      "step": 970
    },
    {
      "epoch": 2.12,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002,
      "loss": 1.4534,
      "step": 980
    },
    {
      "epoch": 2.14,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0002,
      "loss": 1.4601,
      "step": 990
    },
    {
      "epoch": 2.16,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002,
      "loss": 1.4352,
      "step": 1000
    },
    {
      "epoch": 2.19,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002,
      "loss": 1.4183,
      "step": 1010
    },
    {
      "epoch": 2.21,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002,
      "loss": 1.3654,
      "step": 1020
    },
    {
      "epoch": 2.23,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002,
      "loss": 1.4216,
      "step": 1030
    },
    {
      "epoch": 2.25,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002,
      "loss": 1.3554,
      "step": 1040
    },
    {
      "epoch": 2.27,
      "grad_norm": 2.5,
      "learning_rate": 0.0002,
      "loss": 1.3716,
      "step": 1050
    },
    {
      "epoch": 2.29,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002,
      "loss": 1.426,
      "step": 1060
    },
    {
      "epoch": 2.32,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002,
      "loss": 1.4755,
      "step": 1070
    },
    {
      "epoch": 2.34,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0002,
      "loss": 1.3976,
      "step": 1080
    },
    {
      "epoch": 2.36,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002,
      "loss": 1.472,
      "step": 1090
    },
    {
      "epoch": 2.38,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002,
      "loss": 1.461,
      "step": 1100
    },
    {
      "epoch": 2.4,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002,
      "loss": 1.4307,
      "step": 1110
    },
    {
      "epoch": 2.42,
      "grad_norm": 1.625,
      "learning_rate": 0.0002,
      "loss": 1.4603,
      "step": 1120
    },
    {
      "epoch": 2.45,
      "grad_norm": 2.03125,
      "learning_rate": 0.0002,
      "loss": 1.573,
      "step": 1130
    },
    {
      "epoch": 2.47,
      "grad_norm": 1.984375,
      "learning_rate": 0.0002,
      "loss": 1.3899,
      "step": 1140
    },
    {
      "epoch": 2.49,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002,
      "loss": 1.463,
      "step": 1150
    },
    {
      "epoch": 2.51,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002,
      "loss": 1.3866,
      "step": 1160
    },
    {
      "epoch": 2.53,
      "grad_norm": 1.125,
      "learning_rate": 0.0002,
      "loss": 1.3927,
      "step": 1170
    },
    {
      "epoch": 2.55,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002,
      "loss": 1.3956,
      "step": 1180
    },
    {
      "epoch": 2.58,
      "grad_norm": 8.75,
      "learning_rate": 0.0002,
      "loss": 1.335,
      "step": 1190
    },
    {
      "epoch": 2.6,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002,
      "loss": 1.4117,
      "step": 1200
    },
    {
      "epoch": 2.62,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002,
      "loss": 1.385,
      "step": 1210
    },
    {
      "epoch": 2.64,
      "grad_norm": 1.25,
      "learning_rate": 0.0002,
      "loss": 1.414,
      "step": 1220
    },
    {
      "epoch": 2.66,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002,
      "loss": 1.3831,
      "step": 1230
    },
    {
      "epoch": 2.68,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002,
      "loss": 1.5191,
      "step": 1240
    },
    {
      "epoch": 2.71,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002,
      "loss": 1.3568,
      "step": 1250
    },
    {
      "epoch": 2.73,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002,
      "loss": 1.4097,
      "step": 1260
    },
    {
      "epoch": 2.75,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002,
      "loss": 1.4798,
      "step": 1270
    },
    {
      "epoch": 2.77,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002,
      "loss": 1.3602,
      "step": 1280
    },
    {
      "epoch": 2.79,
      "grad_norm": 1.671875,
      "learning_rate": 0.0002,
      "loss": 1.4344,
      "step": 1290
    },
    {
      "epoch": 2.81,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002,
      "loss": 1.4734,
      "step": 1300
    },
    {
      "epoch": 2.84,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002,
      "loss": 1.4536,
      "step": 1310
    },
    {
      "epoch": 2.86,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002,
      "loss": 1.3822,
      "step": 1320
    },
    {
      "epoch": 2.88,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002,
      "loss": 1.4099,
      "step": 1330
    },
    {
      "epoch": 2.9,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002,
      "loss": 1.4668,
      "step": 1340
    },
    {
      "epoch": 2.92,
      "grad_norm": 1.75,
      "learning_rate": 0.0002,
      "loss": 1.4382,
      "step": 1350
    },
    {
      "epoch": 2.94,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0002,
      "loss": 1.4847,
      "step": 1360
    },
    {
      "epoch": 2.97,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002,
      "loss": 1.4274,
      "step": 1370
    },
    {
      "epoch": 2.99,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0002,
      "loss": 1.4039,
      "step": 1380
    }
  ],
  "logging_steps": 10,
  "max_steps": 1386,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 3.907934746733445e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}