{
  "best_metric": 1.6449463367462158,
  "best_model_checkpoint": "lora_lr/meta-llama/Meta-Llama-3-8B-Instruct/unaligned/checkpoint-250",
  "epoch": 0.327575175943698,
  "eval_steps": 10,
  "global_step": 256,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 0.1787109375,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 2.2553,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.18359375,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.1481,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.15625,
      "learning_rate": 6e-06,
      "loss": 2.1875,
      "step": 3
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.1875,
      "learning_rate": 8.000000000000001e-06,
      "loss": 2.2196,
      "step": 4
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.1865234375,
      "learning_rate": 1e-05,
      "loss": 2.2149,
      "step": 5
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.1806640625,
      "learning_rate": 1.2e-05,
      "loss": 2.2424,
      "step": 6
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.1943359375,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 2.1425,
      "step": 7
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.1865234375,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 2.181,
      "step": 8
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.1884765625,
      "learning_rate": 1.8e-05,
      "loss": 2.2398,
      "step": 9
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.18359375,
      "learning_rate": 2e-05,
      "loss": 2.2213,
      "step": 10
    },
    {
      "epoch": 0.01,
      "eval_loss": 2.2066986560821533,
      "eval_runtime": 125.8714,
      "eval_samples_per_second": 39.723,
      "eval_steps_per_second": 1.247,
      "step": 10
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.169921875,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 2.1514,
      "step": 11
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.173828125,
      "learning_rate": 2.4e-05,
      "loss": 2.2107,
      "step": 12
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.2021484375,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 2.2672,
      "step": 13
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.1962890625,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 2.3302,
      "step": 14
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.1904296875,
      "learning_rate": 3e-05,
      "loss": 2.2157,
      "step": 15
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.1953125,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 2.227,
      "step": 16
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.22265625,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 2.1633,
      "step": 17
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.234375,
      "learning_rate": 3.6e-05,
      "loss": 2.2369,
      "step": 18
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.2314453125,
      "learning_rate": 3.8e-05,
      "loss": 2.2168,
      "step": 19
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.2490234375,
      "learning_rate": 4e-05,
      "loss": 2.1926,
      "step": 20
    },
    {
      "epoch": 0.03,
      "eval_loss": 2.1795005798339844,
      "eval_runtime": 126.2376,
      "eval_samples_per_second": 39.608,
      "eval_steps_per_second": 1.244,
      "step": 20
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.27734375,
      "learning_rate": 4.2e-05,
      "loss": 2.168,
      "step": 21
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.29296875,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 2.1932,
      "step": 22
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.34375,
      "learning_rate": 4.600000000000001e-05,
      "loss": 2.2095,
      "step": 23
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.345703125,
      "learning_rate": 4.8e-05,
      "loss": 2.2258,
      "step": 24
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.3671875,
      "learning_rate": 5e-05,
      "loss": 2.156,
      "step": 25
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.341796875,
      "learning_rate": 5.2000000000000004e-05,
      "loss": 2.2151,
      "step": 26
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.396484375,
      "learning_rate": 5.4000000000000005e-05,
      "loss": 2.118,
      "step": 27
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.439453125,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 2.1715,
      "step": 28
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.419921875,
      "learning_rate": 5.8e-05,
      "loss": 2.1332,
      "step": 29
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.400390625,
      "learning_rate": 6e-05,
      "loss": 2.0189,
      "step": 30
    },
    {
      "epoch": 0.04,
      "eval_loss": 2.0425634384155273,
      "eval_runtime": 126.2797,
      "eval_samples_per_second": 39.595,
      "eval_steps_per_second": 1.243,
      "step": 30
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.462890625,
      "learning_rate": 6.2e-05,
      "loss": 2.1203,
      "step": 31
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.41796875,
      "learning_rate": 6.400000000000001e-05,
      "loss": 2.0459,
      "step": 32
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.435546875,
      "learning_rate": 6.6e-05,
      "loss": 2.0343,
      "step": 33
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.408203125,
      "learning_rate": 6.800000000000001e-05,
      "loss": 1.9959,
      "step": 34
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.412109375,
      "learning_rate": 7e-05,
      "loss": 2.0212,
      "step": 35
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.376953125,
      "learning_rate": 7.2e-05,
      "loss": 1.9178,
      "step": 36
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.412109375,
      "learning_rate": 7.4e-05,
      "loss": 2.016,
      "step": 37
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.341796875,
      "learning_rate": 7.6e-05,
      "loss": 1.8999,
      "step": 38
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.296875,
      "learning_rate": 7.800000000000001e-05,
      "loss": 1.8407,
      "step": 39
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.2734375,
      "learning_rate": 8e-05,
      "loss": 1.8578,
      "step": 40
    },
    {
      "epoch": 0.05,
      "eval_loss": 1.8325164318084717,
      "eval_runtime": 126.1895,
      "eval_samples_per_second": 39.623,
      "eval_steps_per_second": 1.244,
      "step": 40
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.283203125,
      "learning_rate": 8.2e-05,
      "loss": 1.8992,
      "step": 41
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.322265625,
      "learning_rate": 8.4e-05,
      "loss": 1.8398,
      "step": 42
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.345703125,
      "learning_rate": 8.6e-05,
      "loss": 1.8156,
      "step": 43
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.3359375,
      "learning_rate": 8.800000000000001e-05,
      "loss": 1.7986,
      "step": 44
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.2890625,
      "learning_rate": 9e-05,
      "loss": 1.7476,
      "step": 45
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.251953125,
      "learning_rate": 9.200000000000001e-05,
      "loss": 1.7926,
      "step": 46
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.251953125,
      "learning_rate": 9.4e-05,
      "loss": 1.8071,
      "step": 47
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.251953125,
      "learning_rate": 9.6e-05,
      "loss": 1.8004,
      "step": 48
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.236328125,
      "learning_rate": 9.8e-05,
      "loss": 1.7274,
      "step": 49
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0001,
      "loss": 1.7916,
      "step": 50
    },
    {
      "epoch": 0.06,
      "eval_loss": 1.7592554092407227,
      "eval_runtime": 126.142,
      "eval_samples_per_second": 39.638,
      "eval_steps_per_second": 1.245,
      "step": 50
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.2021484375,
      "learning_rate": 0.00010200000000000001,
      "loss": 1.8609,
      "step": 51
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.2197265625,
      "learning_rate": 0.00010400000000000001,
      "loss": 1.7137,
      "step": 52
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.197265625,
      "learning_rate": 0.00010600000000000002,
      "loss": 1.7276,
      "step": 53
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.193359375,
      "learning_rate": 0.00010800000000000001,
      "loss": 1.7627,
      "step": 54
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.189453125,
      "learning_rate": 0.00011000000000000002,
      "loss": 1.6725,
      "step": 55
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.1796875,
      "learning_rate": 0.00011200000000000001,
      "loss": 1.7363,
      "step": 56
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.189453125,
      "learning_rate": 0.00011399999999999999,
      "loss": 1.7445,
      "step": 57
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.000116,
      "loss": 1.7819,
      "step": 58
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.162109375,
      "learning_rate": 0.000118,
      "loss": 1.7401,
      "step": 59
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.1728515625,
      "learning_rate": 0.00012,
      "loss": 1.7688,
      "step": 60
    },
    {
      "epoch": 0.08,
      "eval_loss": 1.72506582736969,
      "eval_runtime": 135.8454,
      "eval_samples_per_second": 36.807,
      "eval_steps_per_second": 1.156,
      "step": 60
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.166015625,
      "learning_rate": 0.000122,
      "loss": 1.7404,
      "step": 61
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.154296875,
      "learning_rate": 0.000124,
      "loss": 1.7477,
      "step": 62
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.189453125,
      "learning_rate": 0.000126,
      "loss": 1.7422,
      "step": 63
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.162109375,
      "learning_rate": 0.00012800000000000002,
      "loss": 1.6909,
      "step": 64
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.255859375,
      "learning_rate": 0.00013000000000000002,
      "loss": 1.7694,
      "step": 65
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.1669921875,
      "learning_rate": 0.000132,
      "loss": 1.8089,
      "step": 66
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.000134,
      "loss": 1.7403,
      "step": 67
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.158203125,
      "learning_rate": 0.00013600000000000003,
      "loss": 1.7219,
      "step": 68
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.14453125,
      "learning_rate": 0.000138,
      "loss": 1.6921,
      "step": 69
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.142578125,
      "learning_rate": 0.00014,
      "loss": 1.6864,
      "step": 70
    },
    {
      "epoch": 0.09,
      "eval_loss": 1.7060683965682983,
      "eval_runtime": 141.5784,
      "eval_samples_per_second": 35.316,
      "eval_steps_per_second": 1.109,
      "step": 70
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.1484375,
      "learning_rate": 0.000142,
      "loss": 1.7129,
      "step": 71
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.1376953125,
      "learning_rate": 0.000144,
      "loss": 1.7353,
      "step": 72
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.12890625,
      "learning_rate": 0.000146,
      "loss": 1.678,
      "step": 73
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.1337890625,
      "learning_rate": 0.000148,
      "loss": 1.7835,
      "step": 74
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.142578125,
      "learning_rate": 0.00015000000000000001,
      "loss": 1.7583,
      "step": 75
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.162109375,
      "learning_rate": 0.000152,
      "loss": 1.6988,
      "step": 76
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.16015625,
      "learning_rate": 0.000154,
      "loss": 1.7599,
      "step": 77
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.2314453125,
      "learning_rate": 0.00015600000000000002,
      "loss": 1.7217,
      "step": 78
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.14453125,
      "learning_rate": 0.00015800000000000002,
      "loss": 1.6603,
      "step": 79
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.1435546875,
      "learning_rate": 0.00016,
      "loss": 1.6916,
      "step": 80
    },
    {
      "epoch": 0.1,
      "eval_loss": 1.6934484243392944,
      "eval_runtime": 126.2219,
      "eval_samples_per_second": 39.613,
      "eval_steps_per_second": 1.244,
      "step": 80
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.1796875,
      "learning_rate": 0.000162,
      "loss": 1.6855,
      "step": 81
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.1748046875,
      "learning_rate": 0.000164,
      "loss": 1.6786,
      "step": 82
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.1923828125,
      "learning_rate": 0.000166,
      "loss": 1.7373,
      "step": 83
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.1513671875,
      "learning_rate": 0.000168,
      "loss": 1.697,
      "step": 84
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.1298828125,
      "learning_rate": 0.00017,
      "loss": 1.7171,
      "step": 85
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.1650390625,
      "learning_rate": 0.000172,
      "loss": 1.6476,
      "step": 86
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.158203125,
      "learning_rate": 0.000174,
      "loss": 1.7017,
      "step": 87
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.1513671875,
      "learning_rate": 0.00017600000000000002,
      "loss": 1.7472,
      "step": 88
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.216796875,
      "learning_rate": 0.00017800000000000002,
      "loss": 1.6973,
      "step": 89
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.1787109375,
      "learning_rate": 0.00018,
      "loss": 1.6604,
      "step": 90
    },
    {
      "epoch": 0.12,
      "eval_loss": 1.6824843883514404,
      "eval_runtime": 126.172,
      "eval_samples_per_second": 39.628,
      "eval_steps_per_second": 1.244,
      "step": 90
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.150390625,
      "learning_rate": 0.000182,
      "loss": 1.7181,
      "step": 91
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.1669921875,
      "learning_rate": 0.00018400000000000003,
      "loss": 1.6215,
      "step": 92
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.1806640625,
      "learning_rate": 0.00018600000000000002,
      "loss": 1.7783,
      "step": 93
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.166015625,
      "learning_rate": 0.000188,
      "loss": 1.6328,
      "step": 94
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.208984375,
      "learning_rate": 0.00019,
      "loss": 1.6821,
      "step": 95
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.212890625,
      "learning_rate": 0.000192,
      "loss": 1.7397,
      "step": 96
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.1953125,
      "learning_rate": 0.000194,
      "loss": 1.6838,
      "step": 97
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.1474609375,
      "learning_rate": 0.000196,
      "loss": 1.6441,
      "step": 98
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.1962890625,
      "learning_rate": 0.00019800000000000002,
      "loss": 1.6274,
      "step": 99
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.2021484375,
      "learning_rate": 0.0002,
      "loss": 1.6126,
      "step": 100
    },
    {
      "epoch": 0.13,
      "eval_loss": 1.6749509572982788,
      "eval_runtime": 126.1716,
      "eval_samples_per_second": 39.629,
      "eval_steps_per_second": 1.244,
      "step": 100
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.1611328125,
      "learning_rate": 0.00019871794871794874,
      "loss": 1.6243,
      "step": 101
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.1552734375,
      "learning_rate": 0.00019743589743589744,
      "loss": 1.6823,
      "step": 102
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.1689453125,
      "learning_rate": 0.00019615384615384615,
      "loss": 1.7351,
      "step": 103
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.158203125,
      "learning_rate": 0.00019487179487179487,
      "loss": 1.6981,
      "step": 104
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.158203125,
      "learning_rate": 0.0001935897435897436,
      "loss": 1.6748,
      "step": 105
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.1484375,
      "learning_rate": 0.00019230769230769233,
      "loss": 1.6985,
      "step": 106
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.14453125,
      "learning_rate": 0.00019102564102564104,
      "loss": 1.6849,
      "step": 107
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.158203125,
      "learning_rate": 0.00018974358974358974,
      "loss": 1.6547,
      "step": 108
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.00018846153846153847,
      "loss": 1.6749,
      "step": 109
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.14453125,
      "learning_rate": 0.0001871794871794872,
      "loss": 1.7316,
      "step": 110
    },
    {
      "epoch": 0.14,
      "eval_loss": 1.6681605577468872,
      "eval_runtime": 126.374,
      "eval_samples_per_second": 39.565,
      "eval_steps_per_second": 1.242,
      "step": 110
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.1865234375,
      "learning_rate": 0.0001858974358974359,
      "loss": 1.6359,
      "step": 111
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.169921875,
      "learning_rate": 0.00018461538461538463,
      "loss": 1.6691,
      "step": 112
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.146484375,
      "learning_rate": 0.00018333333333333334,
      "loss": 1.6756,
      "step": 113
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.1533203125,
      "learning_rate": 0.00018205128205128207,
      "loss": 1.6792,
      "step": 114
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.16796875,
      "learning_rate": 0.00018076923076923077,
      "loss": 1.664,
      "step": 115
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.1767578125,
      "learning_rate": 0.0001794871794871795,
      "loss": 1.6874,
      "step": 116
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.1826171875,
      "learning_rate": 0.00017820512820512823,
      "loss": 1.7538,
      "step": 117
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.1767578125,
      "learning_rate": 0.00017692307692307693,
      "loss": 1.6786,
      "step": 118
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.185546875,
      "learning_rate": 0.00017564102564102566,
      "loss": 1.6342,
      "step": 119
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.158203125,
      "learning_rate": 0.00017435897435897436,
      "loss": 1.7201,
      "step": 120
    },
    {
      "epoch": 0.15,
      "eval_loss": 1.663940668106079,
      "eval_runtime": 126.2162,
      "eval_samples_per_second": 39.615,
      "eval_steps_per_second": 1.244,
      "step": 120
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.1748046875,
      "learning_rate": 0.0001730769230769231,
      "loss": 1.7054,
      "step": 121
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.181640625,
      "learning_rate": 0.0001717948717948718,
      "loss": 1.6278,
      "step": 122
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.154296875,
      "learning_rate": 0.00017051282051282053,
      "loss": 1.6609,
      "step": 123
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.00016923076923076923,
      "loss": 1.6739,
      "step": 124
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.1865234375,
      "learning_rate": 0.00016794871794871796,
      "loss": 1.6719,
      "step": 125
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.1708984375,
      "learning_rate": 0.0001666666666666667,
      "loss": 1.6738,
      "step": 126
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.1533203125,
      "learning_rate": 0.0001653846153846154,
      "loss": 1.6973,
      "step": 127
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.1650390625,
      "learning_rate": 0.0001641025641025641,
      "loss": 1.6434,
      "step": 128
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.1845703125,
      "learning_rate": 0.00016282051282051282,
      "loss": 1.719,
      "step": 129
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.17578125,
      "learning_rate": 0.00016153846153846155,
      "loss": 1.7266,
      "step": 130
    },
    {
      "epoch": 0.17,
      "eval_loss": 1.6603440046310425,
      "eval_runtime": 126.1026,
      "eval_samples_per_second": 39.65,
      "eval_steps_per_second": 1.245,
      "step": 130
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.1806640625,
      "learning_rate": 0.00016025641025641028,
      "loss": 1.6324,
      "step": 131
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.1708984375,
      "learning_rate": 0.00015897435897435896,
      "loss": 1.7067,
      "step": 132
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.1806640625,
      "learning_rate": 0.0001576923076923077,
      "loss": 1.7019,
      "step": 133
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.16796875,
      "learning_rate": 0.00015641025641025642,
      "loss": 1.7202,
      "step": 134
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.1728515625,
      "learning_rate": 0.00015512820512820515,
      "loss": 1.6778,
      "step": 135
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.19140625,
      "learning_rate": 0.00015384615384615385,
      "loss": 1.6928,
      "step": 136
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.1982421875,
      "learning_rate": 0.00015256410256410255,
      "loss": 1.6346,
      "step": 137
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.17578125,
      "learning_rate": 0.00015128205128205128,
      "loss": 1.6308,
      "step": 138
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.22265625,
      "learning_rate": 0.00015000000000000001,
      "loss": 1.7073,
      "step": 139
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.197265625,
      "learning_rate": 0.00014871794871794872,
      "loss": 1.6467,
      "step": 140
    },
    {
      "epoch": 0.18,
      "eval_loss": 1.6569123268127441,
      "eval_runtime": 126.2301,
      "eval_samples_per_second": 39.61,
      "eval_steps_per_second": 1.244,
      "step": 140
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.1875,
      "learning_rate": 0.00014743589743589745,
      "loss": 1.6789,
      "step": 141
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.1611328125,
      "learning_rate": 0.00014615384615384615,
      "loss": 1.6766,
      "step": 142
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.177734375,
      "learning_rate": 0.00014487179487179488,
      "loss": 1.6749,
      "step": 143
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.1630859375,
      "learning_rate": 0.0001435897435897436,
      "loss": 1.6991,
      "step": 144
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.0001423076923076923,
      "loss": 1.6876,
      "step": 145
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.1728515625,
      "learning_rate": 0.00014102564102564104,
      "loss": 1.707,
      "step": 146
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.16796875,
      "learning_rate": 0.00013974358974358974,
      "loss": 1.662,
      "step": 147
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.1552734375,
      "learning_rate": 0.00013846153846153847,
      "loss": 1.5826,
      "step": 148
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.1591796875,
      "learning_rate": 0.00013717948717948718,
      "loss": 1.7056,
      "step": 149
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.1591796875,
      "learning_rate": 0.0001358974358974359,
      "loss": 1.711,
      "step": 150
    },
    {
      "epoch": 0.19,
      "eval_loss": 1.6547765731811523,
      "eval_runtime": 126.2019,
      "eval_samples_per_second": 39.619,
      "eval_steps_per_second": 1.244,
      "step": 150
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.19140625,
      "learning_rate": 0.00013461538461538464,
      "loss": 1.6004,
      "step": 151
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.169921875,
      "learning_rate": 0.00013333333333333334,
      "loss": 1.7267,
      "step": 152
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.16015625,
      "learning_rate": 0.00013205128205128204,
      "loss": 1.6589,
      "step": 153
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.1787109375,
      "learning_rate": 0.00013076923076923077,
      "loss": 1.6143,
      "step": 154
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.166015625,
      "learning_rate": 0.0001294871794871795,
      "loss": 1.6815,
      "step": 155
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.158203125,
      "learning_rate": 0.00012820512820512823,
      "loss": 1.6267,
      "step": 156
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.1630859375,
      "learning_rate": 0.00012692307692307693,
      "loss": 1.6705,
      "step": 157
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.1796875,
      "learning_rate": 0.00012564102564102564,
      "loss": 1.6853,
      "step": 158
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.16015625,
      "learning_rate": 0.00012435897435897437,
      "loss": 1.6755,
      "step": 159
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.1572265625,
      "learning_rate": 0.0001230769230769231,
      "loss": 1.6165,
      "step": 160
    },
    {
      "epoch": 0.2,
      "eval_loss": 1.6526222229003906,
      "eval_runtime": 126.3096,
      "eval_samples_per_second": 39.585,
      "eval_steps_per_second": 1.243,
      "step": 160
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.166015625,
      "learning_rate": 0.00012179487179487179,
      "loss": 1.6913,
      "step": 161
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.162109375,
      "learning_rate": 0.00012051282051282052,
      "loss": 1.6403,
      "step": 162
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.162109375,
      "learning_rate": 0.00011923076923076923,
      "loss": 1.6667,
      "step": 163
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.1748046875,
      "learning_rate": 0.00011794871794871796,
      "loss": 1.616,
      "step": 164
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.177734375,
      "learning_rate": 0.00011666666666666668,
      "loss": 1.5773,
      "step": 165
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.1884765625,
      "learning_rate": 0.00011538461538461538,
      "loss": 1.6668,
      "step": 166
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.1708984375,
      "learning_rate": 0.0001141025641025641,
      "loss": 1.6904,
      "step": 167
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.177734375,
      "learning_rate": 0.00011282051282051283,
      "loss": 1.6935,
      "step": 168
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.1640625,
      "learning_rate": 0.00011153846153846154,
      "loss": 1.6615,
      "step": 169
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.185546875,
      "learning_rate": 0.00011025641025641027,
      "loss": 1.7506,
      "step": 170
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.650907278060913,
      "eval_runtime": 125.9533,
      "eval_samples_per_second": 39.697,
      "eval_steps_per_second": 1.246,
      "step": 170
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.2158203125,
      "learning_rate": 0.00010897435897435896,
      "loss": 1.6593,
      "step": 171
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.166015625,
      "learning_rate": 0.0001076923076923077,
      "loss": 1.6011,
      "step": 172
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.171875,
      "learning_rate": 0.00010641025641025641,
      "loss": 1.675,
      "step": 173
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.177734375,
      "learning_rate": 0.00010512820512820514,
      "loss": 1.6491,
      "step": 174
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.169921875,
      "learning_rate": 0.00010384615384615386,
      "loss": 1.7251,
      "step": 175
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.1669921875,
      "learning_rate": 0.00010256410256410256,
      "loss": 1.6881,
      "step": 176
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.177734375,
      "learning_rate": 0.00010128205128205129,
      "loss": 1.6465,
      "step": 177
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.1943359375,
      "learning_rate": 0.0001,
      "loss": 1.6203,
      "step": 178
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.18359375,
      "learning_rate": 9.871794871794872e-05,
      "loss": 1.6638,
      "step": 179
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.1904296875,
      "learning_rate": 9.743589743589744e-05,
      "loss": 1.6636,
      "step": 180
    },
    {
      "epoch": 0.23,
      "eval_loss": 1.6491334438323975,
      "eval_runtime": 126.2676,
      "eval_samples_per_second": 39.598,
      "eval_steps_per_second": 1.243,
      "step": 180
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.1767578125,
      "learning_rate": 9.615384615384617e-05,
      "loss": 1.6722,
      "step": 181
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.158203125,
      "learning_rate": 9.487179487179487e-05,
      "loss": 1.6301,
      "step": 182
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.1728515625,
      "learning_rate": 9.35897435897436e-05,
      "loss": 1.6467,
      "step": 183
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.1904296875,
      "learning_rate": 9.230769230769232e-05,
      "loss": 1.6395,
      "step": 184
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.1689453125,
      "learning_rate": 9.102564102564103e-05,
      "loss": 1.7357,
      "step": 185
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.169921875,
      "learning_rate": 8.974358974358975e-05,
      "loss": 1.6724,
      "step": 186
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.185546875,
      "learning_rate": 8.846153846153847e-05,
      "loss": 1.7003,
      "step": 187
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.1787109375,
      "learning_rate": 8.717948717948718e-05,
      "loss": 1.6373,
      "step": 188
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.185546875,
      "learning_rate": 8.58974358974359e-05,
      "loss": 1.699,
      "step": 189
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.1748046875,
      "learning_rate": 8.461538461538461e-05,
      "loss": 1.6927,
      "step": 190
    },
    {
      "epoch": 0.24,
      "eval_loss": 1.64810311794281,
      "eval_runtime": 126.1264,
      "eval_samples_per_second": 39.643,
      "eval_steps_per_second": 1.245,
      "step": 190
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.177734375,
      "learning_rate": 8.333333333333334e-05,
      "loss": 1.6843,
      "step": 191
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.18359375,
      "learning_rate": 8.205128205128205e-05,
      "loss": 1.7133,
      "step": 192
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.1669921875,
      "learning_rate": 8.076923076923078e-05,
      "loss": 1.5758,
      "step": 193
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.1630859375,
      "learning_rate": 7.948717948717948e-05,
      "loss": 1.7406,
      "step": 194
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.169921875,
      "learning_rate": 7.820512820512821e-05,
      "loss": 1.6312,
      "step": 195
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.1669921875,
      "learning_rate": 7.692307692307693e-05,
      "loss": 1.6398,
      "step": 196
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.1767578125,
      "learning_rate": 7.564102564102564e-05,
      "loss": 1.6418,
      "step": 197
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.1591796875,
      "learning_rate": 7.435897435897436e-05,
      "loss": 1.6438,
      "step": 198
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.169921875,
      "learning_rate": 7.307692307692307e-05,
      "loss": 1.6514,
      "step": 199
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.1943359375,
      "learning_rate": 7.17948717948718e-05,
      "loss": 1.631,
      "step": 200
    },
    {
      "epoch": 0.26,
      "eval_loss": 1.6470773220062256,
      "eval_runtime": 126.2841,
      "eval_samples_per_second": 39.593,
      "eval_steps_per_second": 1.243,
      "step": 200
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.1904296875,
      "learning_rate": 7.051282051282052e-05,
      "loss": 1.6412,
      "step": 201
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.1728515625,
      "learning_rate": 6.923076923076924e-05,
      "loss": 1.7023,
      "step": 202
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.1708984375,
      "learning_rate": 6.794871794871795e-05,
      "loss": 1.7129,
      "step": 203
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.17578125,
      "learning_rate": 6.666666666666667e-05,
      "loss": 1.6855,
      "step": 204
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.18359375,
      "learning_rate": 6.538461538461539e-05,
      "loss": 1.6017,
      "step": 205
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.1787109375,
      "learning_rate": 6.410256410256412e-05,
      "loss": 1.6791,
      "step": 206
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.1767578125,
      "learning_rate": 6.282051282051282e-05,
      "loss": 1.5874,
      "step": 207
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.166015625,
      "learning_rate": 6.153846153846155e-05,
      "loss": 1.6805,
      "step": 208
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.2041015625,
      "learning_rate": 6.025641025641026e-05,
      "loss": 1.6416,
      "step": 209
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.1650390625,
      "learning_rate": 5.897435897435898e-05,
      "loss": 1.7009,
      "step": 210
    },
    {
      "epoch": 0.27,
      "eval_loss": 1.64640212059021,
      "eval_runtime": 126.2026,
      "eval_samples_per_second": 39.619,
      "eval_steps_per_second": 1.244,
      "step": 210
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.1787109375,
      "learning_rate": 5.769230769230769e-05,
      "loss": 1.6571,
      "step": 211
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.1875,
      "learning_rate": 5.6410256410256414e-05,
      "loss": 1.6447,
      "step": 212
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.173828125,
      "learning_rate": 5.512820512820514e-05,
      "loss": 1.6619,
      "step": 213
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.1728515625,
      "learning_rate": 5.384615384615385e-05,
      "loss": 1.6299,
      "step": 214
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.185546875,
      "learning_rate": 5.256410256410257e-05,
      "loss": 1.59,
      "step": 215
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.1826171875,
      "learning_rate": 5.128205128205128e-05,
      "loss": 1.6238,
      "step": 216
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.2041015625,
      "learning_rate": 5e-05,
      "loss": 1.673,
      "step": 217
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.1806640625,
      "learning_rate": 4.871794871794872e-05,
      "loss": 1.6962,
      "step": 218
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.1796875,
      "learning_rate": 4.7435897435897435e-05,
      "loss": 1.6256,
      "step": 219
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.177734375,
      "learning_rate": 4.615384615384616e-05,
      "loss": 1.6825,
      "step": 220
    },
    {
      "epoch": 0.28,
      "eval_loss": 1.645533561706543,
      "eval_runtime": 126.1441,
      "eval_samples_per_second": 39.637,
      "eval_steps_per_second": 1.245,
      "step": 220
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.2021484375,
      "learning_rate": 4.4871794871794874e-05,
      "loss": 1.6856,
      "step": 221
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.1728515625,
      "learning_rate": 4.358974358974359e-05,
      "loss": 1.6845,
      "step": 222
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.1728515625,
      "learning_rate": 4.230769230769231e-05,
      "loss": 1.6687,
      "step": 223
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.1650390625,
      "learning_rate": 4.1025641025641023e-05,
      "loss": 1.6538,
      "step": 224
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.2109375,
      "learning_rate": 3.974358974358974e-05,
      "loss": 1.6264,
      "step": 225
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.185546875,
      "learning_rate": 3.846153846153846e-05,
      "loss": 1.5731,
      "step": 226
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.162109375,
      "learning_rate": 3.717948717948718e-05,
      "loss": 1.6814,
      "step": 227
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.18359375,
      "learning_rate": 3.58974358974359e-05,
      "loss": 1.662,
      "step": 228
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.1884765625,
      "learning_rate": 3.461538461538462e-05,
      "loss": 1.5692,
      "step": 229
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.1767578125,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.6712,
      "step": 230
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.6453105211257935,
      "eval_runtime": 125.1792,
      "eval_samples_per_second": 39.943,
      "eval_steps_per_second": 1.254,
      "step": 230
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.2021484375,
      "learning_rate": 3.205128205128206e-05,
      "loss": 1.67,
      "step": 231
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.1796875,
      "learning_rate": 3.0769230769230774e-05,
      "loss": 1.5761,
      "step": 232
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.1875,
      "learning_rate": 2.948717948717949e-05,
      "loss": 1.6017,
      "step": 233
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.1982421875,
      "learning_rate": 2.8205128205128207e-05,
      "loss": 1.684,
      "step": 234
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.18359375,
      "learning_rate": 2.6923076923076923e-05,
      "loss": 1.6708,
      "step": 235
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.1865234375,
      "learning_rate": 2.564102564102564e-05,
      "loss": 1.6748,
      "step": 236
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.17578125,
      "learning_rate": 2.435897435897436e-05,
      "loss": 1.6203,
      "step": 237
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.1904296875,
      "learning_rate": 2.307692307692308e-05,
      "loss": 1.6658,
      "step": 238
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.173828125,
      "learning_rate": 2.1794871794871795e-05,
      "loss": 1.7107,
      "step": 239
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.1953125,
      "learning_rate": 2.0512820512820512e-05,
      "loss": 1.6987,
      "step": 240
    },
    {
      "epoch": 0.31,
      "eval_loss": 1.6449843645095825,
      "eval_runtime": 125.1567,
      "eval_samples_per_second": 39.95,
      "eval_steps_per_second": 1.254,
      "step": 240
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.1806640625,
      "learning_rate": 1.923076923076923e-05,
      "loss": 1.6973,
      "step": 241
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.1875,
      "learning_rate": 1.794871794871795e-05,
      "loss": 1.5889,
      "step": 242
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.1845703125,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.6322,
      "step": 243
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.1953125,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 1.611,
      "step": 244
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.169921875,
      "learning_rate": 1.4102564102564104e-05,
      "loss": 1.7048,
      "step": 245
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.1796875,
      "learning_rate": 1.282051282051282e-05,
      "loss": 1.6283,
      "step": 246
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.1748046875,
      "learning_rate": 1.153846153846154e-05,
      "loss": 1.6868,
      "step": 247
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.1943359375,
      "learning_rate": 1.0256410256410256e-05,
      "loss": 1.7102,
      "step": 248
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.193359375,
      "learning_rate": 8.974358974358976e-06,
      "loss": 1.6453,
      "step": 249
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.1796875,
      "learning_rate": 7.692307692307694e-06,
      "loss": 1.6217,
      "step": 250
    },
    {
      "epoch": 0.32,
      "eval_loss": 1.6449463367462158,
      "eval_runtime": 126.0264,
      "eval_samples_per_second": 39.674,
      "eval_steps_per_second": 1.246,
      "step": 250
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.189453125,
      "learning_rate": 6.41025641025641e-06,
      "loss": 1.7208,
      "step": 251
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.181640625,
      "learning_rate": 5.128205128205128e-06,
      "loss": 1.6421,
      "step": 252
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.1982421875,
      "learning_rate": 3.846153846153847e-06,
      "loss": 1.5272,
      "step": 253
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.1708984375,
      "learning_rate": 2.564102564102564e-06,
      "loss": 1.5856,
      "step": 254
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.1748046875,
      "learning_rate": 1.282051282051282e-06,
      "loss": 1.6374,
      "step": 255
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.1767578125,
      "learning_rate": 0.0,
      "loss": 1.6896,
      "step": 256
    },
    {
      "epoch": 0.33,
      "step": 256,
      "total_flos": 1.813217341042852e+17,
      "train_loss": 1.7538025458343327,
      "train_runtime": 4604.2872,
      "train_samples_per_second": 3.558,
      "train_steps_per_second": 0.056
    },
    {
      "epoch": 0.33,
      "eval_loss": 1.6449463367462158,
      "eval_runtime": 125.933,
      "eval_samples_per_second": 39.704,
      "eval_steps_per_second": 1.247,
      "step": 256
    }
  ],
  "logging_steps": 1,
  "max_steps": 256,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "total_flos": 1.813217341042852e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}