{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 10,
  "global_step": 99,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010101010101010102,
      "grad_norm": 9.5,
      "learning_rate": 0.0,
      "loss": 2.0463,
      "step": 1
    },
    {
      "epoch": 0.020202020202020204,
      "grad_norm": 9.4375,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 2.2081,
      "step": 2
    },
    {
      "epoch": 0.030303030303030304,
      "grad_norm": 10.75,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.1588,
      "step": 3
    },
    {
      "epoch": 0.04040404040404041,
      "grad_norm": 9.4375,
      "learning_rate": 6e-06,
      "loss": 2.1989,
      "step": 4
    },
    {
      "epoch": 0.050505050505050504,
      "grad_norm": 6.8125,
      "learning_rate": 8.000000000000001e-06,
      "loss": 2.0533,
      "step": 5
    },
    {
      "epoch": 0.06060606060606061,
      "grad_norm": 5.84375,
      "learning_rate": 1e-05,
      "loss": 2.071,
      "step": 6
    },
    {
      "epoch": 0.0707070707070707,
      "grad_norm": 5.46875,
      "learning_rate": 9.893617021276596e-06,
      "loss": 2.0025,
      "step": 7
    },
    {
      "epoch": 0.08080808080808081,
      "grad_norm": 5.0625,
      "learning_rate": 9.787234042553192e-06,
      "loss": 1.7837,
      "step": 8
    },
    {
      "epoch": 0.09090909090909091,
      "grad_norm": 4.5625,
      "learning_rate": 9.680851063829787e-06,
      "loss": 1.8194,
      "step": 9
    },
    {
      "epoch": 0.10101010101010101,
      "grad_norm": 4.34375,
      "learning_rate": 9.574468085106385e-06,
      "loss": 1.7647,
      "step": 10
    },
    {
      "epoch": 0.10101010101010101,
      "eval_loss": 1.8501181602478027,
      "eval_model_preparation_time": 0.0192,
      "eval_runtime": 4.5153,
      "eval_samples_per_second": 31.891,
      "eval_steps_per_second": 15.946,
      "step": 10
    },
    {
      "epoch": 0.1111111111111111,
      "grad_norm": 4.1875,
      "learning_rate": 9.46808510638298e-06,
      "loss": 1.7479,
      "step": 11
    },
    {
      "epoch": 0.12121212121212122,
      "grad_norm": 4.6875,
      "learning_rate": 9.361702127659576e-06,
      "loss": 1.9077,
      "step": 12
    },
    {
      "epoch": 0.13131313131313133,
      "grad_norm": 3.984375,
      "learning_rate": 9.255319148936171e-06,
      "loss": 1.7011,
      "step": 13
    },
    {
      "epoch": 0.1414141414141414,
      "grad_norm": 3.921875,
      "learning_rate": 9.148936170212767e-06,
      "loss": 1.6132,
      "step": 14
    },
    {
      "epoch": 0.15151515151515152,
      "grad_norm": 3.484375,
      "learning_rate": 9.042553191489362e-06,
      "loss": 1.5589,
      "step": 15
    },
    {
      "epoch": 0.16161616161616163,
      "grad_norm": 4.28125,
      "learning_rate": 8.936170212765958e-06,
      "loss": 1.7689,
      "step": 16
    },
    {
      "epoch": 0.1717171717171717,
      "grad_norm": 3.90625,
      "learning_rate": 8.829787234042555e-06,
      "loss": 1.7021,
      "step": 17
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 3.90625,
      "learning_rate": 8.72340425531915e-06,
      "loss": 1.596,
      "step": 18
    },
    {
      "epoch": 0.1919191919191919,
      "grad_norm": 3.640625,
      "learning_rate": 8.617021276595746e-06,
      "loss": 1.5814,
      "step": 19
    },
    {
      "epoch": 0.20202020202020202,
      "grad_norm": 4.3125,
      "learning_rate": 8.510638297872341e-06,
      "loss": 1.6297,
      "step": 20
    },
    {
      "epoch": 0.20202020202020202,
      "eval_loss": 1.6177141666412354,
      "eval_model_preparation_time": 0.0192,
      "eval_runtime": 4.4588,
      "eval_samples_per_second": 32.296,
      "eval_steps_per_second": 16.148,
      "step": 20
    },
    {
      "epoch": 0.21212121212121213,
      "grad_norm": 5.21875,
      "learning_rate": 8.404255319148937e-06,
      "loss": 1.4007,
      "step": 21
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 3.640625,
      "learning_rate": 8.297872340425532e-06,
      "loss": 1.5302,
      "step": 22
    },
    {
      "epoch": 0.23232323232323232,
      "grad_norm": 4.625,
      "learning_rate": 8.191489361702128e-06,
      "loss": 1.5728,
      "step": 23
    },
    {
      "epoch": 0.24242424242424243,
      "grad_norm": 4.71875,
      "learning_rate": 8.085106382978723e-06,
      "loss": 1.5623,
      "step": 24
    },
    {
      "epoch": 0.25252525252525254,
      "grad_norm": 4.875,
      "learning_rate": 7.97872340425532e-06,
      "loss": 1.6365,
      "step": 25
    },
    {
      "epoch": 0.26262626262626265,
      "grad_norm": 4.6875,
      "learning_rate": 7.872340425531916e-06,
      "loss": 1.5573,
      "step": 26
    },
    {
      "epoch": 0.2727272727272727,
      "grad_norm": 4.5625,
      "learning_rate": 7.765957446808511e-06,
      "loss": 1.5593,
      "step": 27
    },
    {
      "epoch": 0.2828282828282828,
      "grad_norm": 4.0625,
      "learning_rate": 7.659574468085107e-06,
      "loss": 1.5289,
      "step": 28
    },
    {
      "epoch": 0.29292929292929293,
      "grad_norm": 4.53125,
      "learning_rate": 7.553191489361703e-06,
      "loss": 1.4426,
      "step": 29
    },
    {
      "epoch": 0.30303030303030304,
      "grad_norm": 3.765625,
      "learning_rate": 7.446808510638298e-06,
      "loss": 1.5221,
      "step": 30
    },
    {
      "epoch": 0.30303030303030304,
      "eval_loss": 1.5113080739974976,
      "eval_model_preparation_time": 0.0192,
      "eval_runtime": 4.4717,
      "eval_samples_per_second": 32.202,
      "eval_steps_per_second": 16.101,
      "step": 30
    },
    {
      "epoch": 0.31313131313131315,
      "grad_norm": 3.640625,
      "learning_rate": 7.340425531914894e-06,
      "loss": 1.399,
      "step": 31
    },
    {
      "epoch": 0.32323232323232326,
      "grad_norm": 4.0625,
      "learning_rate": 7.234042553191491e-06,
      "loss": 1.5572,
      "step": 32
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 3.65625,
      "learning_rate": 7.127659574468085e-06,
      "loss": 1.3714,
      "step": 33
    },
    {
      "epoch": 0.3434343434343434,
      "grad_norm": 3.890625,
      "learning_rate": 7.021276595744682e-06,
      "loss": 1.4515,
      "step": 34
    },
    {
      "epoch": 0.35353535353535354,
      "grad_norm": 3.640625,
      "learning_rate": 6.914893617021278e-06,
      "loss": 1.4222,
      "step": 35
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 3.3125,
      "learning_rate": 6.808510638297873e-06,
      "loss": 1.375,
      "step": 36
    },
    {
      "epoch": 0.37373737373737376,
      "grad_norm": 3.625,
      "learning_rate": 6.702127659574469e-06,
      "loss": 1.4592,
      "step": 37
    },
    {
      "epoch": 0.3838383838383838,
      "grad_norm": 3.453125,
      "learning_rate": 6.595744680851064e-06,
      "loss": 1.3722,
      "step": 38
    },
    {
      "epoch": 0.3939393939393939,
      "grad_norm": 3.6875,
      "learning_rate": 6.48936170212766e-06,
      "loss": 1.4548,
      "step": 39
    },
    {
      "epoch": 0.40404040404040403,
      "grad_norm": 3.453125,
      "learning_rate": 6.382978723404256e-06,
      "loss": 1.5035,
      "step": 40
    },
    {
      "epoch": 0.40404040404040403,
      "eval_loss": 1.443503499031067,
      "eval_model_preparation_time": 0.0192,
      "eval_runtime": 4.5368,
      "eval_samples_per_second": 31.74,
      "eval_steps_per_second": 15.87,
      "step": 40
    },
    {
      "epoch": 0.41414141414141414,
      "grad_norm": 3.1875,
      "learning_rate": 6.276595744680851e-06,
      "loss": 1.4053,
      "step": 41
    },
    {
      "epoch": 0.42424242424242425,
      "grad_norm": 3.453125,
      "learning_rate": 6.170212765957447e-06,
      "loss": 1.3533,
      "step": 42
    },
    {
      "epoch": 0.43434343434343436,
      "grad_norm": 4.09375,
      "learning_rate": 6.063829787234044e-06,
      "loss": 1.4969,
      "step": 43
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 3.8125,
      "learning_rate": 5.957446808510638e-06,
      "loss": 1.4565,
      "step": 44
    },
    {
      "epoch": 0.45454545454545453,
      "grad_norm": 4.0,
      "learning_rate": 5.851063829787235e-06,
      "loss": 1.389,
      "step": 45
    },
    {
      "epoch": 0.46464646464646464,
      "grad_norm": 4.625,
      "learning_rate": 5.744680851063831e-06,
      "loss": 1.4511,
      "step": 46
    },
    {
      "epoch": 0.47474747474747475,
      "grad_norm": 3.890625,
      "learning_rate": 5.638297872340426e-06,
      "loss": 1.4159,
      "step": 47
    },
    {
      "epoch": 0.48484848484848486,
      "grad_norm": 3.90625,
      "learning_rate": 5.531914893617022e-06,
      "loss": 1.3452,
      "step": 48
    },
    {
      "epoch": 0.494949494949495,
      "grad_norm": 4.375,
      "learning_rate": 5.425531914893617e-06,
      "loss": 1.3517,
      "step": 49
    },
    {
      "epoch": 0.5050505050505051,
      "grad_norm": 3.84375,
      "learning_rate": 5.319148936170213e-06,
      "loss": 1.3475,
      "step": 50
    },
    {
      "epoch": 0.5050505050505051,
      "eval_loss": 1.399296522140503,
      "eval_model_preparation_time": 0.0192,
      "eval_runtime": 4.4653,
      "eval_samples_per_second": 32.249,
      "eval_steps_per_second": 16.124,
      "step": 50
    },
    {
      "epoch": 0.5151515151515151,
      "grad_norm": 3.515625,
      "learning_rate": 5.212765957446809e-06,
      "loss": 1.3007,
      "step": 51
    },
    {
      "epoch": 0.5252525252525253,
      "grad_norm": 3.515625,
      "learning_rate": 5.106382978723404e-06,
      "loss": 1.3047,
      "step": 52
    },
    {
      "epoch": 0.5353535353535354,
      "grad_norm": 3.734375,
      "learning_rate": 5e-06,
      "loss": 1.3374,
      "step": 53
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 3.703125,
      "learning_rate": 4.893617021276596e-06,
      "loss": 1.2432,
      "step": 54
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 4.09375,
      "learning_rate": 4.787234042553192e-06,
      "loss": 1.4018,
      "step": 55
    },
    {
      "epoch": 0.5656565656565656,
      "grad_norm": 3.78125,
      "learning_rate": 4.680851063829788e-06,
      "loss": 1.4056,
      "step": 56
    },
    {
      "epoch": 0.5757575757575758,
      "grad_norm": 3.5625,
      "learning_rate": 4.574468085106383e-06,
      "loss": 1.3512,
      "step": 57
    },
    {
      "epoch": 0.5858585858585859,
      "grad_norm": 3.375,
      "learning_rate": 4.468085106382979e-06,
      "loss": 1.3737,
      "step": 58
    },
    {
      "epoch": 0.5959595959595959,
      "grad_norm": 3.5625,
      "learning_rate": 4.361702127659575e-06,
      "loss": 1.3453,
      "step": 59
    },
    {
      "epoch": 0.6060606060606061,
      "grad_norm": 3.25,
      "learning_rate": 4.255319148936171e-06,
      "loss": 1.3598,
      "step": 60
    },
    {
      "epoch": 0.6060606060606061,
      "eval_loss": 1.3689942359924316,
      "eval_model_preparation_time": 0.0192,
      "eval_runtime": 4.891,
      "eval_samples_per_second": 29.442,
      "eval_steps_per_second": 14.721,
      "step": 60
    },
    {
      "epoch": 0.6161616161616161,
      "grad_norm": 3.90625,
      "learning_rate": 4.148936170212766e-06,
      "loss": 1.4119,
      "step": 61
    },
    {
      "epoch": 0.6262626262626263,
      "grad_norm": 3.734375,
      "learning_rate": 4.042553191489362e-06,
      "loss": 1.3028,
      "step": 62
    },
    {
      "epoch": 0.6363636363636364,
      "grad_norm": 3.53125,
      "learning_rate": 3.936170212765958e-06,
      "loss": 1.2088,
      "step": 63
    },
    {
      "epoch": 0.6464646464646465,
      "grad_norm": 3.5,
      "learning_rate": 3.8297872340425535e-06,
      "loss": 1.3143,
      "step": 64
    },
    {
      "epoch": 0.6565656565656566,
      "grad_norm": 3.65625,
      "learning_rate": 3.723404255319149e-06,
      "loss": 1.3123,
      "step": 65
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 4.25,
      "learning_rate": 3.6170212765957453e-06,
      "loss": 1.356,
      "step": 66
    },
    {
      "epoch": 0.6767676767676768,
      "grad_norm": 3.796875,
      "learning_rate": 3.510638297872341e-06,
      "loss": 1.4043,
      "step": 67
    },
    {
      "epoch": 0.6868686868686869,
      "grad_norm": 3.9375,
      "learning_rate": 3.4042553191489363e-06,
      "loss": 1.3052,
      "step": 68
    },
    {
      "epoch": 0.696969696969697,
      "grad_norm": 3.578125,
      "learning_rate": 3.297872340425532e-06,
      "loss": 1.3282,
      "step": 69
    },
    {
      "epoch": 0.7070707070707071,
      "grad_norm": 4.1875,
      "learning_rate": 3.191489361702128e-06,
      "loss": 1.4262,
      "step": 70
    },
    {
      "epoch": 0.7070707070707071,
      "eval_loss": 1.3482567071914673,
      "eval_model_preparation_time": 0.0192,
      "eval_runtime": 4.9944,
      "eval_samples_per_second": 28.832,
      "eval_steps_per_second": 14.416,
      "step": 70
    },
    {
      "epoch": 0.7171717171717171,
      "grad_norm": 3.984375,
      "learning_rate": 3.0851063829787237e-06,
      "loss": 1.4252,
      "step": 71
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 4.1875,
      "learning_rate": 2.978723404255319e-06,
      "loss": 1.3866,
      "step": 72
    },
    {
      "epoch": 0.7373737373737373,
      "grad_norm": 4.03125,
      "learning_rate": 2.8723404255319155e-06,
      "loss": 1.3893,
      "step": 73
    },
    {
      "epoch": 0.7474747474747475,
      "grad_norm": 3.828125,
      "learning_rate": 2.765957446808511e-06,
      "loss": 1.3093,
      "step": 74
    },
    {
      "epoch": 0.7575757575757576,
      "grad_norm": 3.6875,
      "learning_rate": 2.6595744680851065e-06,
      "loss": 1.2562,
      "step": 75
    },
    {
      "epoch": 0.7676767676767676,
      "grad_norm": 3.8125,
      "learning_rate": 2.553191489361702e-06,
      "loss": 1.2896,
      "step": 76
    },
    {
      "epoch": 0.7777777777777778,
      "grad_norm": 3.546875,
      "learning_rate": 2.446808510638298e-06,
      "loss": 1.3052,
      "step": 77
    },
    {
      "epoch": 0.7878787878787878,
      "grad_norm": 4.28125,
      "learning_rate": 2.340425531914894e-06,
      "loss": 1.2805,
      "step": 78
    },
    {
      "epoch": 0.797979797979798,
      "grad_norm": 3.796875,
      "learning_rate": 2.2340425531914894e-06,
      "loss": 1.2598,
      "step": 79
    },
    {
      "epoch": 0.8080808080808081,
      "grad_norm": 3.734375,
      "learning_rate": 2.1276595744680853e-06,
      "loss": 1.3316,
      "step": 80
    },
    {
      "epoch": 0.8080808080808081,
      "eval_loss": 1.3349997997283936,
      "eval_model_preparation_time": 0.0192,
      "eval_runtime": 4.7579,
      "eval_samples_per_second": 30.265,
      "eval_steps_per_second": 15.133,
      "step": 80
    },
    {
      "epoch": 0.8181818181818182,
      "grad_norm": 3.578125,
      "learning_rate": 2.021276595744681e-06,
      "loss": 1.2956,
      "step": 81
    },
    {
      "epoch": 0.8282828282828283,
      "grad_norm": 3.53125,
      "learning_rate": 1.9148936170212767e-06,
      "loss": 1.317,
      "step": 82
    },
    {
      "epoch": 0.8383838383838383,
      "grad_norm": 3.71875,
      "learning_rate": 1.8085106382978727e-06,
      "loss": 1.2709,
      "step": 83
    },
    {
      "epoch": 0.8484848484848485,
      "grad_norm": 3.5625,
      "learning_rate": 1.7021276595744682e-06,
      "loss": 1.3115,
      "step": 84
    },
    {
      "epoch": 0.8585858585858586,
      "grad_norm": 3.796875,
      "learning_rate": 1.595744680851064e-06,
      "loss": 1.3131,
      "step": 85
    },
    {
      "epoch": 0.8686868686868687,
      "grad_norm": 3.375,
      "learning_rate": 1.4893617021276596e-06,
      "loss": 1.2993,
      "step": 86
    },
    {
      "epoch": 0.8787878787878788,
      "grad_norm": 3.828125,
      "learning_rate": 1.3829787234042555e-06,
      "loss": 1.2934,
      "step": 87
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 3.890625,
      "learning_rate": 1.276595744680851e-06,
      "loss": 1.3148,
      "step": 88
    },
    {
      "epoch": 0.898989898989899,
      "grad_norm": 4.125,
      "learning_rate": 1.170212765957447e-06,
      "loss": 1.4243,
      "step": 89
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 3.53125,
      "learning_rate": 1.0638297872340427e-06,
      "loss": 1.2164,
      "step": 90
    },
    {
      "epoch": 0.9090909090909091,
      "eval_loss": 1.3271279335021973,
      "eval_model_preparation_time": 0.0192,
      "eval_runtime": 4.7066,
      "eval_samples_per_second": 30.595,
      "eval_steps_per_second": 15.298,
      "step": 90
    },
    {
      "epoch": 0.9191919191919192,
      "grad_norm": 4.125,
      "learning_rate": 9.574468085106384e-07,
      "loss": 1.4069,
      "step": 91
    },
    {
      "epoch": 0.9292929292929293,
      "grad_norm": 3.921875,
      "learning_rate": 8.510638297872341e-07,
      "loss": 1.3035,
      "step": 92
    },
    {
      "epoch": 0.9393939393939394,
      "grad_norm": 3.75,
      "learning_rate": 7.446808510638298e-07,
      "loss": 1.2968,
      "step": 93
    },
    {
      "epoch": 0.9494949494949495,
      "grad_norm": 3.609375,
      "learning_rate": 6.382978723404255e-07,
      "loss": 1.3289,
      "step": 94
    },
    {
      "epoch": 0.9595959595959596,
      "grad_norm": 3.78125,
      "learning_rate": 5.319148936170213e-07,
      "loss": 1.3311,
      "step": 95
    },
    {
      "epoch": 0.9696969696969697,
      "grad_norm": 3.75,
      "learning_rate": 4.2553191489361704e-07,
      "loss": 1.3339,
      "step": 96
    },
    {
      "epoch": 0.9797979797979798,
      "grad_norm": 3.796875,
      "learning_rate": 3.1914893617021275e-07,
      "loss": 1.3035,
      "step": 97
    },
    {
      "epoch": 0.98989898989899,
      "grad_norm": 3.3125,
      "learning_rate": 2.1276595744680852e-07,
      "loss": 1.2145,
      "step": 98
    },
    {
      "epoch": 1.0,
      "grad_norm": 3.640625,
      "learning_rate": 1.0638297872340426e-07,
      "loss": 1.2494,
      "step": 99
    }
  ],
  "logging_steps": 1,
  "max_steps": 99,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.698740109213696e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}