{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9540481400437635,
  "eval_steps": 1.0,
  "global_step": 108,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "grad_norm": 6.122653587017292,
      "learning_rate": 0.0,
      "loss": 0.646,
      "step": 1
    },
    {
      "epoch": 0.05,
      "grad_norm": 4.716859801224774,
      "learning_rate": 5e-06,
      "loss": 0.5427,
      "step": 2
    },
    {
      "epoch": 0.08,
      "grad_norm": 5.019981578406485,
      "learning_rate": 7.924812503605782e-06,
      "loss": 0.549,
      "step": 3
    },
    {
      "epoch": 0.11,
      "grad_norm": 2.467488328063134,
      "learning_rate": 1e-05,
      "loss": 0.4574,
      "step": 4
    },
    {
      "epoch": 0.14,
      "grad_norm": 4.728992664952675,
      "learning_rate": 1e-05,
      "loss": 0.4952,
      "step": 5
    },
    {
      "epoch": 0.16,
      "grad_norm": 2.970029793004458,
      "learning_rate": 1e-05,
      "loss": 0.4373,
      "step": 6
    },
    {
      "epoch": 0.19,
      "grad_norm": 3.152396699018966,
      "learning_rate": 1e-05,
      "loss": 0.4491,
      "step": 7
    },
    {
      "epoch": 0.22,
      "grad_norm": 2.2898095220878196,
      "learning_rate": 1e-05,
      "loss": 0.421,
      "step": 8
    },
    {
      "epoch": 0.25,
      "grad_norm": 1.8312653886857113,
      "learning_rate": 1e-05,
      "loss": 0.3626,
      "step": 9
    },
    {
      "epoch": 0.27,
      "grad_norm": 1.8281463002677139,
      "learning_rate": 1e-05,
      "loss": 0.3628,
      "step": 10
    },
    {
      "epoch": 0.3,
      "grad_norm": 1.7367109671193546,
      "learning_rate": 1e-05,
      "loss": 0.3394,
      "step": 11
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.6255181971686055,
      "learning_rate": 1e-05,
      "loss": 0.333,
      "step": 12
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.6622699558938445,
      "learning_rate": 1e-05,
      "loss": 0.313,
      "step": 13
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.4241097114069272,
      "learning_rate": 1e-05,
      "loss": 0.2917,
      "step": 14
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.3259856150813727,
      "learning_rate": 1e-05,
      "loss": 0.2742,
      "step": 15
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.363385021609024,
      "learning_rate": 1e-05,
      "loss": 0.2931,
      "step": 16
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.612345679634704,
      "learning_rate": 1e-05,
      "loss": 0.2723,
      "step": 17
    },
    {
      "epoch": 0.49,
      "grad_norm": 1.394500889215576,
      "learning_rate": 1e-05,
      "loss": 0.2623,
      "step": 18
    },
    {
      "epoch": 0.52,
      "grad_norm": 1.2629355627249936,
      "learning_rate": 1e-05,
      "loss": 0.2719,
      "step": 19
    },
    {
      "epoch": 0.55,
      "grad_norm": 1.4121814773595716,
      "learning_rate": 1e-05,
      "loss": 0.2453,
      "step": 20
    },
    {
      "epoch": 0.57,
      "grad_norm": 1.2877132103814335,
      "learning_rate": 1e-05,
      "loss": 0.244,
      "step": 21
    },
    {
      "epoch": 0.6,
      "grad_norm": 1.2523378620255512,
      "learning_rate": 1e-05,
      "loss": 0.2349,
      "step": 22
    },
    {
      "epoch": 0.63,
      "grad_norm": 1.1902225037740863,
      "learning_rate": 1e-05,
      "loss": 0.2265,
      "step": 23
    },
    {
      "epoch": 0.66,
      "grad_norm": 1.1879613096031454,
      "learning_rate": 1e-05,
      "loss": 0.2144,
      "step": 24
    },
    {
      "epoch": 0.68,
      "grad_norm": 1.1464570881387057,
      "learning_rate": 1e-05,
      "loss": 0.2063,
      "step": 25
    },
    {
      "epoch": 0.71,
      "grad_norm": 1.2752585041011075,
      "learning_rate": 1e-05,
      "loss": 0.2157,
      "step": 26
    },
    {
      "epoch": 0.74,
      "grad_norm": 1.155543288413803,
      "learning_rate": 1e-05,
      "loss": 0.1916,
      "step": 27
    },
    {
      "epoch": 0.77,
      "grad_norm": 1.2042714749577323,
      "learning_rate": 1e-05,
      "loss": 0.2071,
      "step": 28
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.1346405455010145,
      "learning_rate": 1e-05,
      "loss": 0.1991,
      "step": 29
    },
    {
      "epoch": 0.82,
      "grad_norm": 1.0793677741286372,
      "learning_rate": 1e-05,
      "loss": 0.1895,
      "step": 30
    },
    {
      "epoch": 0.85,
      "grad_norm": 1.0641597541926833,
      "learning_rate": 1e-05,
      "loss": 0.1968,
      "step": 31
    },
    {
      "epoch": 0.88,
      "grad_norm": 1.1159083684301505,
      "learning_rate": 1e-05,
      "loss": 0.1846,
      "step": 32
    },
    {
      "epoch": 0.9,
      "grad_norm": 1.0614368255136861,
      "learning_rate": 1e-05,
      "loss": 0.1849,
      "step": 33
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.9659837421382899,
      "learning_rate": 1e-05,
      "loss": 0.1677,
      "step": 34
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.9869347069695258,
      "learning_rate": 1e-05,
      "loss": 0.1789,
      "step": 35
    },
    {
      "epoch": 0.98,
      "grad_norm": 1.0555806958429526,
      "learning_rate": 1e-05,
      "loss": 0.1901,
      "step": 36
    },
    {
      "epoch": 1.01,
      "grad_norm": 1.0003790636897225,
      "learning_rate": 1e-05,
      "loss": 0.1317,
      "step": 37
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.8755581193987241,
      "learning_rate": 1e-05,
      "loss": 0.109,
      "step": 38
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.9600336243215675,
      "learning_rate": 1e-05,
      "loss": 0.1225,
      "step": 39
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.983666193649008,
      "learning_rate": 1e-05,
      "loss": 0.1206,
      "step": 40
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.9984047080313273,
      "learning_rate": 1e-05,
      "loss": 0.114,
      "step": 41
    },
    {
      "epoch": 1.15,
      "grad_norm": 1.0701560459785802,
      "learning_rate": 1e-05,
      "loss": 0.1143,
      "step": 42
    },
    {
      "epoch": 1.18,
      "grad_norm": 1.0722426369355342,
      "learning_rate": 1e-05,
      "loss": 0.124,
      "step": 43
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.8817516131260538,
      "learning_rate": 1e-05,
      "loss": 0.1079,
      "step": 44
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.9625731291493045,
      "learning_rate": 1e-05,
      "loss": 0.1222,
      "step": 45
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.9812159017657305,
      "learning_rate": 1e-05,
      "loss": 0.1096,
      "step": 46
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.8745591684992073,
      "learning_rate": 1e-05,
      "loss": 0.1149,
      "step": 47
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.9507892298975904,
      "learning_rate": 1e-05,
      "loss": 0.1163,
      "step": 48
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.8611093549236812,
      "learning_rate": 1e-05,
      "loss": 0.11,
      "step": 49
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.8444613309525189,
      "learning_rate": 1e-05,
      "loss": 0.1054,
      "step": 50
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.9868965033294682,
      "learning_rate": 1e-05,
      "loss": 0.1206,
      "step": 51
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.7940733083936387,
      "learning_rate": 1e-05,
      "loss": 0.1072,
      "step": 52
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.8572147743881433,
      "learning_rate": 1e-05,
      "loss": 0.1044,
      "step": 53
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.8209526691122747,
      "learning_rate": 1e-05,
      "loss": 0.1009,
      "step": 54
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.7779846619307967,
      "learning_rate": 1e-05,
      "loss": 0.1018,
      "step": 55
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.8291999953627118,
      "learning_rate": 1e-05,
      "loss": 0.1025,
      "step": 56
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.8321877301816655,
      "learning_rate": 1e-05,
      "loss": 0.1094,
      "step": 57
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.8542389871893485,
      "learning_rate": 1e-05,
      "loss": 0.1079,
      "step": 58
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.7737670400014411,
      "learning_rate": 1e-05,
      "loss": 0.0958,
      "step": 59
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.8129322360573784,
      "learning_rate": 1e-05,
      "loss": 0.093,
      "step": 60
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.8293838232530079,
      "learning_rate": 1e-05,
      "loss": 0.1054,
      "step": 61
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.6810818138246434,
      "learning_rate": 1e-05,
      "loss": 0.0906,
      "step": 62
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.7937807744835117,
      "learning_rate": 1e-05,
      "loss": 0.098,
      "step": 63
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.8224807756832562,
      "learning_rate": 1e-05,
      "loss": 0.1101,
      "step": 64
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.7304387601530952,
      "learning_rate": 1e-05,
      "loss": 0.1,
      "step": 65
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.8142026342771219,
      "learning_rate": 1e-05,
      "loss": 0.0964,
      "step": 66
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.7431609339195293,
      "learning_rate": 1e-05,
      "loss": 0.0936,
      "step": 67
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.7512520528680077,
      "learning_rate": 1e-05,
      "loss": 0.0949,
      "step": 68
    },
    {
      "epoch": 1.89,
      "grad_norm": 0.7538760164866836,
      "learning_rate": 1e-05,
      "loss": 0.0989,
      "step": 69
    },
    {
      "epoch": 1.91,
      "grad_norm": 0.8103341693726498,
      "learning_rate": 1e-05,
      "loss": 0.1028,
      "step": 70
    },
    {
      "epoch": 1.94,
      "grad_norm": 0.8357385002863533,
      "learning_rate": 1e-05,
      "loss": 0.0966,
      "step": 71
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.7944109386823767,
      "learning_rate": 1e-05,
      "loss": 0.1032,
      "step": 72
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8167924040143067,
      "learning_rate": 1e-05,
      "loss": 0.1003,
      "step": 73
    },
    {
      "epoch": 2.02,
      "grad_norm": 0.6224097456559627,
      "learning_rate": 1e-05,
      "loss": 0.0604,
      "step": 74
    },
    {
      "epoch": 2.05,
      "grad_norm": 0.6419960808589802,
      "learning_rate": 1e-05,
      "loss": 0.0652,
      "step": 75
    },
    {
      "epoch": 2.08,
      "grad_norm": 0.7787247503593108,
      "learning_rate": 1e-05,
      "loss": 0.0709,
      "step": 76
    },
    {
      "epoch": 2.11,
      "grad_norm": 0.7252667545531377,
      "learning_rate": 1e-05,
      "loss": 0.0571,
      "step": 77
    },
    {
      "epoch": 2.13,
      "grad_norm": 0.7217097658566882,
      "learning_rate": 1e-05,
      "loss": 0.0656,
      "step": 78
    },
    {
      "epoch": 2.16,
      "grad_norm": 0.7168372530155407,
      "learning_rate": 1e-05,
      "loss": 0.0527,
      "step": 79
    },
    {
      "epoch": 2.19,
      "grad_norm": 0.7418408777634922,
      "learning_rate": 1e-05,
      "loss": 0.0585,
      "step": 80
    },
    {
      "epoch": 2.22,
      "grad_norm": 0.7101096286468248,
      "learning_rate": 1e-05,
      "loss": 0.0509,
      "step": 81
    },
    {
      "epoch": 2.24,
      "grad_norm": 0.7006749099813174,
      "learning_rate": 1e-05,
      "loss": 0.0576,
      "step": 82
    },
    {
      "epoch": 2.27,
      "grad_norm": 0.7944497077223811,
      "learning_rate": 1e-05,
      "loss": 0.0545,
      "step": 83
    },
    {
      "epoch": 2.3,
      "grad_norm": 0.6722848545285588,
      "learning_rate": 1e-05,
      "loss": 0.0563,
      "step": 84
    },
    {
      "epoch": 2.32,
      "grad_norm": 0.7120414843684311,
      "learning_rate": 1e-05,
      "loss": 0.058,
      "step": 85
    },
    {
      "epoch": 2.35,
      "grad_norm": 0.7279580151325783,
      "learning_rate": 1e-05,
      "loss": 0.0552,
      "step": 86
    },
    {
      "epoch": 2.38,
      "grad_norm": 0.7029506482473885,
      "learning_rate": 1e-05,
      "loss": 0.0577,
      "step": 87
    },
    {
      "epoch": 2.41,
      "grad_norm": 0.6189514282541002,
      "learning_rate": 1e-05,
      "loss": 0.0547,
      "step": 88
    },
    {
      "epoch": 2.43,
      "grad_norm": 0.6229759439930223,
      "learning_rate": 1e-05,
      "loss": 0.0513,
      "step": 89
    },
    {
      "epoch": 2.46,
      "grad_norm": 0.6863028991803624,
      "learning_rate": 1e-05,
      "loss": 0.0613,
      "step": 90
    },
    {
      "epoch": 2.49,
      "grad_norm": 0.7364535429784711,
      "learning_rate": 1e-05,
      "loss": 0.0652,
      "step": 91
    },
    {
      "epoch": 2.52,
      "grad_norm": 0.7387032302781582,
      "learning_rate": 1e-05,
      "loss": 0.0638,
      "step": 92
    },
    {
      "epoch": 2.54,
      "grad_norm": 0.6757297547267043,
      "learning_rate": 1e-05,
      "loss": 0.0586,
      "step": 93
    },
    {
      "epoch": 2.57,
      "grad_norm": 0.6588502217604668,
      "learning_rate": 1e-05,
      "loss": 0.05,
      "step": 94
    },
    {
      "epoch": 2.6,
      "grad_norm": 0.6612243757810015,
      "learning_rate": 1e-05,
      "loss": 0.0565,
      "step": 95
    },
    {
      "epoch": 2.63,
      "grad_norm": 0.6510872422256165,
      "learning_rate": 1e-05,
      "loss": 0.0564,
      "step": 96
    },
    {
      "epoch": 2.65,
      "grad_norm": 0.6599878520531972,
      "learning_rate": 1e-05,
      "loss": 0.0584,
      "step": 97
    },
    {
      "epoch": 2.68,
      "grad_norm": 0.6723176479777001,
      "learning_rate": 1e-05,
      "loss": 0.0596,
      "step": 98
    },
    {
      "epoch": 2.71,
      "grad_norm": 0.6738851793824463,
      "learning_rate": 1e-05,
      "loss": 0.0568,
      "step": 99
    },
    {
      "epoch": 2.74,
      "grad_norm": 0.6730157693288188,
      "learning_rate": 1e-05,
      "loss": 0.0567,
      "step": 100
    },
    {
      "epoch": 2.76,
      "grad_norm": 0.6025834169032148,
      "learning_rate": 1e-05,
      "loss": 0.0543,
      "step": 101
    },
    {
      "epoch": 2.79,
      "grad_norm": 0.5662111947365751,
      "learning_rate": 1e-05,
      "loss": 0.0521,
      "step": 102
    },
    {
      "epoch": 2.82,
      "grad_norm": 0.6744169703896066,
      "learning_rate": 1e-05,
      "loss": 0.0589,
      "step": 103
    },
    {
      "epoch": 2.84,
      "grad_norm": 0.6312659616633817,
      "learning_rate": 1e-05,
      "loss": 0.0544,
      "step": 104
    },
    {
      "epoch": 2.87,
      "grad_norm": 0.6011739294981976,
      "learning_rate": 1e-05,
      "loss": 0.055,
      "step": 105
    },
    {
      "epoch": 2.9,
      "grad_norm": 0.6427838412250556,
      "learning_rate": 1e-05,
      "loss": 0.0582,
      "step": 106
    },
    {
      "epoch": 2.93,
      "grad_norm": 0.6537825081243189,
      "learning_rate": 1e-05,
      "loss": 0.0579,
      "step": 107
    },
    {
      "epoch": 2.95,
      "grad_norm": 0.6762138754041659,
      "learning_rate": 1e-05,
      "loss": 0.0588,
      "step": 108
    },
    {
      "epoch": 2.95,
      "step": 108,
      "total_flos": 90174812381184.0,
      "train_loss": 0.15581304762788392,
      "train_runtime": 6548.7943,
      "train_samples_per_second": 2.513,
      "train_steps_per_second": 0.016
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 108,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1.0,
  "total_flos": 90174812381184.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}