{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.019438444924406,
  "eval_steps": 4,
  "global_step": 118,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.3477,
      "step": 1
    },
    {
      "epoch": 0.01,
      "eval_loss": 0.4987175166606903,
      "eval_runtime": 3.1166,
      "eval_samples_per_second": 6.417,
      "eval_steps_per_second": 3.209,
      "step": 1
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.3285,
      "step": 2
    },
    {
      "epoch": 0.03,
      "learning_rate": 6e-06,
      "loss": 0.3754,
      "step": 3
    },
    {
      "epoch": 0.03,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.3143,
      "step": 4
    },
    {
      "epoch": 0.03,
      "eval_loss": 0.4986831247806549,
      "eval_runtime": 3.101,
      "eval_samples_per_second": 6.45,
      "eval_steps_per_second": 3.225,
      "step": 4
    },
    {
      "epoch": 0.04,
      "learning_rate": 1e-05,
      "loss": 0.4661,
      "step": 5
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.2e-05,
      "loss": 0.3726,
      "step": 6
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 0.3515,
      "step": 7
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.2981,
      "step": 8
    },
    {
      "epoch": 0.07,
      "eval_loss": 0.4983077645301819,
      "eval_runtime": 3.1103,
      "eval_samples_per_second": 6.43,
      "eval_steps_per_second": 3.215,
      "step": 8
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.8e-05,
      "loss": 0.5434,
      "step": 9
    },
    {
      "epoch": 0.09,
      "learning_rate": 2e-05,
      "loss": 0.3998,
      "step": 10
    },
    {
      "epoch": 0.1,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 0.31,
      "step": 11
    },
    {
      "epoch": 0.1,
      "learning_rate": 2.4e-05,
      "loss": 0.4314,
      "step": 12
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.497723788022995,
      "eval_runtime": 3.1104,
      "eval_samples_per_second": 6.43,
      "eval_steps_per_second": 3.215,
      "step": 12
    },
    {
      "epoch": 0.11,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 0.3478,
      "step": 13
    },
    {
      "epoch": 0.12,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 0.3663,
      "step": 14
    },
    {
      "epoch": 0.13,
      "learning_rate": 3e-05,
      "loss": 0.3981,
      "step": 15
    },
    {
      "epoch": 0.14,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.3959,
      "step": 16
    },
    {
      "epoch": 0.14,
      "eval_loss": 0.49577316641807556,
      "eval_runtime": 3.1093,
      "eval_samples_per_second": 6.432,
      "eval_steps_per_second": 3.216,
      "step": 16
    },
    {
      "epoch": 0.15,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 0.4256,
      "step": 17
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.6e-05,
      "loss": 0.3393,
      "step": 18
    },
    {
      "epoch": 0.16,
      "learning_rate": 3.8e-05,
      "loss": 0.4055,
      "step": 19
    },
    {
      "epoch": 0.17,
      "learning_rate": 4e-05,
      "loss": 0.3353,
      "step": 20
    },
    {
      "epoch": 0.17,
      "eval_loss": 0.4915711581707001,
      "eval_runtime": 3.1139,
      "eval_samples_per_second": 6.423,
      "eval_steps_per_second": 3.211,
      "step": 20
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.2e-05,
      "loss": 0.3431,
      "step": 21
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.354,
      "step": 22
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.3472,
      "step": 23
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.8e-05,
      "loss": 0.4287,
      "step": 24
    },
    {
      "epoch": 0.21,
      "eval_loss": 0.4811338782310486,
      "eval_runtime": 3.1059,
      "eval_samples_per_second": 6.439,
      "eval_steps_per_second": 3.22,
      "step": 24
    },
    {
      "epoch": 0.22,
      "learning_rate": 5e-05,
      "loss": 0.3663,
      "step": 25
    },
    {
      "epoch": 0.22,
      "learning_rate": 5.2000000000000004e-05,
      "loss": 0.298,
      "step": 26
    },
    {
      "epoch": 0.23,
      "learning_rate": 5.4000000000000005e-05,
      "loss": 0.3649,
      "step": 27
    },
    {
      "epoch": 0.24,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 0.4018,
      "step": 28
    },
    {
      "epoch": 0.24,
      "eval_loss": 0.46300894021987915,
      "eval_runtime": 3.1124,
      "eval_samples_per_second": 6.426,
      "eval_steps_per_second": 3.213,
      "step": 28
    },
    {
      "epoch": 0.25,
      "learning_rate": 5.8e-05,
      "loss": 0.3116,
      "step": 29
    },
    {
      "epoch": 0.26,
      "learning_rate": 6e-05,
      "loss": 0.3147,
      "step": 30
    },
    {
      "epoch": 0.27,
      "learning_rate": 6.2e-05,
      "loss": 0.2776,
      "step": 31
    },
    {
      "epoch": 0.28,
      "learning_rate": 6.400000000000001e-05,
      "loss": 0.3278,
      "step": 32
    },
    {
      "epoch": 0.28,
      "eval_loss": 0.4446893334388733,
      "eval_runtime": 3.1075,
      "eval_samples_per_second": 6.436,
      "eval_steps_per_second": 3.218,
      "step": 32
    },
    {
      "epoch": 0.29,
      "learning_rate": 6.6e-05,
      "loss": 0.2848,
      "step": 33
    },
    {
      "epoch": 0.29,
      "learning_rate": 6.800000000000001e-05,
      "loss": 0.2776,
      "step": 34
    },
    {
      "epoch": 0.3,
      "learning_rate": 7e-05,
      "loss": 0.3078,
      "step": 35
    },
    {
      "epoch": 0.31,
      "learning_rate": 7.2e-05,
      "loss": 0.2566,
      "step": 36
    },
    {
      "epoch": 0.31,
      "eval_loss": 0.4287605881690979,
      "eval_runtime": 3.1063,
      "eval_samples_per_second": 6.438,
      "eval_steps_per_second": 3.219,
      "step": 36
    },
    {
      "epoch": 0.32,
      "learning_rate": 7.4e-05,
      "loss": 0.2715,
      "step": 37
    },
    {
      "epoch": 0.33,
      "learning_rate": 7.6e-05,
      "loss": 0.3393,
      "step": 38
    },
    {
      "epoch": 0.34,
      "learning_rate": 7.800000000000001e-05,
      "loss": 0.2669,
      "step": 39
    },
    {
      "epoch": 0.35,
      "learning_rate": 8e-05,
      "loss": 0.3797,
      "step": 40
    },
    {
      "epoch": 0.35,
      "eval_loss": 0.4169308543205261,
      "eval_runtime": 3.1074,
      "eval_samples_per_second": 6.436,
      "eval_steps_per_second": 3.218,
      "step": 40
    },
    {
      "epoch": 0.35,
      "learning_rate": 8.2e-05,
      "loss": 0.2702,
      "step": 41
    },
    {
      "epoch": 0.36,
      "learning_rate": 8.4e-05,
      "loss": 0.3824,
      "step": 42
    },
    {
      "epoch": 0.37,
      "learning_rate": 8.6e-05,
      "loss": 0.274,
      "step": 43
    },
    {
      "epoch": 0.38,
      "learning_rate": 8.800000000000001e-05,
      "loss": 0.2966,
      "step": 44
    },
    {
      "epoch": 0.38,
      "eval_loss": 0.41074520349502563,
      "eval_runtime": 3.1076,
      "eval_samples_per_second": 6.436,
      "eval_steps_per_second": 3.218,
      "step": 44
    },
    {
      "epoch": 0.39,
      "learning_rate": 9e-05,
      "loss": 0.485,
      "step": 45
    },
    {
      "epoch": 0.4,
      "learning_rate": 9.200000000000001e-05,
      "loss": 0.2395,
      "step": 46
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.4e-05,
      "loss": 0.3023,
      "step": 47
    },
    {
      "epoch": 0.41,
      "learning_rate": 9.6e-05,
      "loss": 0.2676,
      "step": 48
    },
    {
      "epoch": 0.41,
      "eval_loss": 0.4023456573486328,
      "eval_runtime": 3.1109,
      "eval_samples_per_second": 6.429,
      "eval_steps_per_second": 3.215,
      "step": 48
    },
    {
      "epoch": 0.42,
      "learning_rate": 9.8e-05,
      "loss": 0.2378,
      "step": 49
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001,
      "loss": 0.2607,
      "step": 50
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00010200000000000001,
      "loss": 0.4538,
      "step": 51
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00010400000000000001,
      "loss": 0.2456,
      "step": 52
    },
    {
      "epoch": 0.45,
      "eval_loss": 0.3957594335079193,
      "eval_runtime": 3.1123,
      "eval_samples_per_second": 6.426,
      "eval_steps_per_second": 3.213,
      "step": 52
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00010600000000000002,
      "loss": 0.3435,
      "step": 53
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00010800000000000001,
      "loss": 0.3035,
      "step": 54
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00011000000000000002,
      "loss": 0.3564,
      "step": 55
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00011200000000000001,
      "loss": 0.3267,
      "step": 56
    },
    {
      "epoch": 0.48,
      "eval_loss": 0.39143189787864685,
      "eval_runtime": 3.1134,
      "eval_samples_per_second": 6.424,
      "eval_steps_per_second": 3.212,
      "step": 56
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00011399999999999999,
      "loss": 0.2641,
      "step": 57
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.000116,
      "loss": 0.2805,
      "step": 58
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.000118,
      "loss": 0.2784,
      "step": 59
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00012,
      "loss": 0.2901,
      "step": 60
    },
    {
      "epoch": 0.52,
      "eval_loss": 0.38809266686439514,
      "eval_runtime": 3.1079,
      "eval_samples_per_second": 6.435,
      "eval_steps_per_second": 3.218,
      "step": 60
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.000122,
      "loss": 0.3039,
      "step": 61
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.000124,
      "loss": 0.2035,
      "step": 62
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.000126,
      "loss": 0.326,
      "step": 63
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00012800000000000002,
      "loss": 0.1843,
      "step": 64
    },
    {
      "epoch": 0.55,
      "eval_loss": 0.3881237506866455,
      "eval_runtime": 3.1107,
      "eval_samples_per_second": 6.429,
      "eval_steps_per_second": 3.215,
      "step": 64
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00013000000000000002,
      "loss": 0.296,
      "step": 65
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.000132,
      "loss": 0.2766,
      "step": 66
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.000134,
      "loss": 0.304,
      "step": 67
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00013600000000000003,
      "loss": 0.2585,
      "step": 68
    },
    {
      "epoch": 0.59,
      "eval_loss": 0.3869483470916748,
      "eval_runtime": 3.1091,
      "eval_samples_per_second": 6.433,
      "eval_steps_per_second": 3.216,
      "step": 68
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.000138,
      "loss": 0.2639,
      "step": 69
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00014,
      "loss": 0.3013,
      "step": 70
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.000142,
      "loss": 0.4263,
      "step": 71
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.000144,
      "loss": 0.3108,
      "step": 72
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.3839200437068939,
      "eval_runtime": 3.1098,
      "eval_samples_per_second": 6.431,
      "eval_steps_per_second": 3.216,
      "step": 72
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.000146,
      "loss": 0.4283,
      "step": 73
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.000148,
      "loss": 0.2582,
      "step": 74
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.2703,
      "step": 75
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.000152,
      "loss": 0.3799,
      "step": 76
    },
    {
      "epoch": 0.66,
      "eval_loss": 0.3816559314727783,
      "eval_runtime": 3.1096,
      "eval_samples_per_second": 6.432,
      "eval_steps_per_second": 3.216,
      "step": 76
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.000154,
      "loss": 0.252,
      "step": 77
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00015600000000000002,
      "loss": 0.3297,
      "step": 78
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00015800000000000002,
      "loss": 0.2607,
      "step": 79
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00016,
      "loss": 0.2064,
      "step": 80
    },
    {
      "epoch": 0.69,
      "eval_loss": 0.37915611267089844,
      "eval_runtime": 3.1123,
      "eval_samples_per_second": 6.426,
      "eval_steps_per_second": 3.213,
      "step": 80
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.000162,
      "loss": 0.3341,
      "step": 81
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.000164,
      "loss": 0.3159,
      "step": 82
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.000166,
      "loss": 0.2457,
      "step": 83
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.000168,
      "loss": 0.3176,
      "step": 84
    },
    {
      "epoch": 0.73,
      "eval_loss": 0.3792770504951477,
      "eval_runtime": 3.13,
      "eval_samples_per_second": 6.39,
      "eval_steps_per_second": 3.195,
      "step": 84
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00017,
      "loss": 0.3043,
      "step": 85
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.000172,
      "loss": 0.2346,
      "step": 86
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.000174,
      "loss": 0.2872,
      "step": 87
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00017600000000000002,
      "loss": 0.2307,
      "step": 88
    },
    {
      "epoch": 0.76,
      "eval_loss": 0.3768787384033203,
      "eval_runtime": 3.1088,
      "eval_samples_per_second": 6.433,
      "eval_steps_per_second": 3.217,
      "step": 88
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00017800000000000002,
      "loss": 0.2661,
      "step": 89
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00018,
      "loss": 0.3068,
      "step": 90
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.000182,
      "loss": 0.2866,
      "step": 91
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00018400000000000003,
      "loss": 0.2826,
      "step": 92
    },
    {
      "epoch": 0.79,
      "eval_loss": 0.3746011555194855,
      "eval_runtime": 3.1104,
      "eval_samples_per_second": 6.43,
      "eval_steps_per_second": 3.215,
      "step": 92
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00018600000000000002,
      "loss": 0.4888,
      "step": 93
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.000188,
      "loss": 0.1667,
      "step": 94
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00019,
      "loss": 0.2683,
      "step": 95
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.000192,
      "loss": 0.2718,
      "step": 96
    },
    {
      "epoch": 0.83,
      "eval_loss": 0.373685747385025,
      "eval_runtime": 3.1134,
      "eval_samples_per_second": 6.424,
      "eval_steps_per_second": 3.212,
      "step": 96
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.000194,
      "loss": 0.2193,
      "step": 97
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.000196,
      "loss": 0.378,
      "step": 98
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00019800000000000002,
      "loss": 0.2917,
      "step": 99
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0002,
      "loss": 0.2945,
      "step": 100
    },
    {
      "epoch": 0.86,
      "eval_loss": 0.3727183938026428,
      "eval_runtime": 3.1085,
      "eval_samples_per_second": 6.434,
      "eval_steps_per_second": 3.217,
      "step": 100
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00019999177886783194,
      "loss": 0.2355,
      "step": 101
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.000199967116823068,
      "loss": 0.2656,
      "step": 102
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00019992601792070679,
      "loss": 0.2803,
      "step": 103
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00019986848891833845,
      "loss": 0.2313,
      "step": 104
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.371459424495697,
      "eval_runtime": 3.1114,
      "eval_samples_per_second": 6.428,
      "eval_steps_per_second": 3.214,
      "step": 104
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00019979453927503364,
      "loss": 0.2983,
      "step": 105
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0001997041811497882,
      "loss": 0.2547,
      "step": 106
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00019959742939952392,
      "loss": 0.2463,
      "step": 107
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00019947430157664576,
      "loss": 0.2519,
      "step": 108
    },
    {
      "epoch": 0.93,
      "eval_loss": 0.37180376052856445,
      "eval_runtime": 3.1113,
      "eval_samples_per_second": 6.428,
      "eval_steps_per_second": 3.214,
      "step": 108
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00019933481792615583,
      "loss": 0.2032,
      "step": 109
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001991790013823246,
      "loss": 0.2868,
      "step": 110
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.0001990068775649202,
      "loss": 0.2653,
      "step": 111
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00019881847477499557,
      "loss": 0.2468,
      "step": 112
    },
    {
      "epoch": 0.97,
      "eval_loss": 0.37166160345077515,
      "eval_runtime": 3.1194,
      "eval_samples_per_second": 6.412,
      "eval_steps_per_second": 3.206,
      "step": 112
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001986138239902355,
      "loss": 0.3997,
      "step": 113
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00019839295885986296,
      "loss": 0.172,
      "step": 114
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00019815591569910654,
      "loss": 0.2713,
      "step": 115
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001979027334832293,
      "loss": 0.2034,
      "step": 116
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.37163203954696655,
      "eval_runtime": 3.1126,
      "eval_samples_per_second": 6.425,
      "eval_steps_per_second": 3.213,
      "step": 116
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00019763345384112043,
      "loss": 0.1824,
      "step": 117
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00019734812104845047,
      "loss": 0.3669,
      "step": 118
    }
  ],
  "logging_steps": 1,
  "max_steps": 345,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 5.999495556523622e+16,
  "trial_name": null,
  "trial_params": null
}