{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.7241379310344827,
  "eval_steps": 9,
  "global_step": 75,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.022988505747126436,
      "grad_norm": 8.76852035522461,
      "learning_rate": 1e-05,
      "loss": 20.8384,
      "step": 1
    },
    {
      "epoch": 0.022988505747126436,
      "eval_loss": 21.21566390991211,
      "eval_runtime": 2.179,
      "eval_samples_per_second": 67.004,
      "eval_steps_per_second": 2.295,
      "step": 1
    },
    {
      "epoch": 0.04597701149425287,
      "grad_norm": 7.297959327697754,
      "learning_rate": 2e-05,
      "loss": 20.9799,
      "step": 2
    },
    {
      "epoch": 0.06896551724137931,
      "grad_norm": 8.222094535827637,
      "learning_rate": 3e-05,
      "loss": 20.9579,
      "step": 3
    },
    {
      "epoch": 0.09195402298850575,
      "grad_norm": 8.73771858215332,
      "learning_rate": 4e-05,
      "loss": 21.4096,
      "step": 4
    },
    {
      "epoch": 0.11494252873563218,
      "grad_norm": 9.340126037597656,
      "learning_rate": 5e-05,
      "loss": 20.8177,
      "step": 5
    },
    {
      "epoch": 0.13793103448275862,
      "grad_norm": 10.907496452331543,
      "learning_rate": 6e-05,
      "loss": 20.0583,
      "step": 6
    },
    {
      "epoch": 0.16091954022988506,
      "grad_norm": 13.2294340133667,
      "learning_rate": 7e-05,
      "loss": 19.5326,
      "step": 7
    },
    {
      "epoch": 0.1839080459770115,
      "grad_norm": 16.79814910888672,
      "learning_rate": 8e-05,
      "loss": 17.6944,
      "step": 8
    },
    {
      "epoch": 0.20689655172413793,
      "grad_norm": 20.354450225830078,
      "learning_rate": 9e-05,
      "loss": 15.8224,
      "step": 9
    },
    {
      "epoch": 0.20689655172413793,
      "eval_loss": 13.134273529052734,
      "eval_runtime": 2.1879,
      "eval_samples_per_second": 66.732,
      "eval_steps_per_second": 2.285,
      "step": 9
    },
    {
      "epoch": 0.22988505747126436,
      "grad_norm": 28.363195419311523,
      "learning_rate": 0.0001,
      "loss": 13.1636,
      "step": 10
    },
    {
      "epoch": 0.25287356321839083,
      "grad_norm": 20.680286407470703,
      "learning_rate": 9.99695413509548e-05,
      "loss": 10.6378,
      "step": 11
    },
    {
      "epoch": 0.27586206896551724,
      "grad_norm": 18.535959243774414,
      "learning_rate": 9.987820251299122e-05,
      "loss": 9.0366,
      "step": 12
    },
    {
      "epoch": 0.2988505747126437,
      "grad_norm": 13.819395065307617,
      "learning_rate": 9.972609476841367e-05,
      "loss": 6.7707,
      "step": 13
    },
    {
      "epoch": 0.3218390804597701,
      "grad_norm": 10.581539154052734,
      "learning_rate": 9.951340343707852e-05,
      "loss": 6.3245,
      "step": 14
    },
    {
      "epoch": 0.3448275862068966,
      "grad_norm": 9.617717742919922,
      "learning_rate": 9.924038765061042e-05,
      "loss": 5.1628,
      "step": 15
    },
    {
      "epoch": 0.367816091954023,
      "grad_norm": 6.453233242034912,
      "learning_rate": 9.890738003669029e-05,
      "loss": 4.81,
      "step": 16
    },
    {
      "epoch": 0.39080459770114945,
      "grad_norm": 4.609827041625977,
      "learning_rate": 9.851478631379982e-05,
      "loss": 4.5561,
      "step": 17
    },
    {
      "epoch": 0.41379310344827586,
      "grad_norm": 4.55856466293335,
      "learning_rate": 9.806308479691595e-05,
      "loss": 4.1678,
      "step": 18
    },
    {
      "epoch": 0.41379310344827586,
      "eval_loss": 4.0816650390625,
      "eval_runtime": 2.1607,
      "eval_samples_per_second": 67.571,
      "eval_steps_per_second": 2.314,
      "step": 18
    },
    {
      "epoch": 0.4367816091954023,
      "grad_norm": 5.335641384124756,
      "learning_rate": 9.755282581475769e-05,
      "loss": 4.1428,
      "step": 19
    },
    {
      "epoch": 0.45977011494252873,
      "grad_norm": 5.87034273147583,
      "learning_rate": 9.698463103929542e-05,
      "loss": 4.1433,
      "step": 20
    },
    {
      "epoch": 0.4827586206896552,
      "grad_norm": 10.226850509643555,
      "learning_rate": 9.635919272833938e-05,
      "loss": 3.8563,
      "step": 21
    },
    {
      "epoch": 0.5057471264367817,
      "grad_norm": 5.1733317375183105,
      "learning_rate": 9.567727288213005e-05,
      "loss": 3.7241,
      "step": 22
    },
    {
      "epoch": 0.5287356321839081,
      "grad_norm": 7.562571048736572,
      "learning_rate": 9.493970231495835e-05,
      "loss": 3.6946,
      "step": 23
    },
    {
      "epoch": 0.5517241379310345,
      "grad_norm": 8.14027214050293,
      "learning_rate": 9.414737964294636e-05,
      "loss": 3.7754,
      "step": 24
    },
    {
      "epoch": 0.5747126436781609,
      "grad_norm": 10.570993423461914,
      "learning_rate": 9.330127018922194e-05,
      "loss": 3.5848,
      "step": 25
    },
    {
      "epoch": 0.5977011494252874,
      "grad_norm": 7.933102130889893,
      "learning_rate": 9.24024048078213e-05,
      "loss": 3.4477,
      "step": 26
    },
    {
      "epoch": 0.6206896551724138,
      "grad_norm": 2.8181567192077637,
      "learning_rate": 9.145187862775209e-05,
      "loss": 3.5206,
      "step": 27
    },
    {
      "epoch": 0.6206896551724138,
      "eval_loss": 3.504189968109131,
      "eval_runtime": 2.1676,
      "eval_samples_per_second": 67.357,
      "eval_steps_per_second": 2.307,
      "step": 27
    },
    {
      "epoch": 0.6436781609195402,
      "grad_norm": 3.624206781387329,
      "learning_rate": 9.045084971874738e-05,
      "loss": 3.5185,
      "step": 28
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 3.64929461479187,
      "learning_rate": 8.940053768033609e-05,
      "loss": 3.4151,
      "step": 29
    },
    {
      "epoch": 0.6896551724137931,
      "grad_norm": 6.643653392791748,
      "learning_rate": 8.83022221559489e-05,
      "loss": 3.6092,
      "step": 30
    },
    {
      "epoch": 0.7126436781609196,
      "grad_norm": 7.6192240715026855,
      "learning_rate": 8.715724127386972e-05,
      "loss": 3.5512,
      "step": 31
    },
    {
      "epoch": 0.735632183908046,
      "grad_norm": 6.675508499145508,
      "learning_rate": 8.596699001693255e-05,
      "loss": 3.4623,
      "step": 32
    },
    {
      "epoch": 0.7586206896551724,
      "grad_norm": 3.877248525619507,
      "learning_rate": 8.473291852294987e-05,
      "loss": 3.4185,
      "step": 33
    },
    {
      "epoch": 0.7816091954022989,
      "grad_norm": 4.328342914581299,
      "learning_rate": 8.345653031794292e-05,
      "loss": 3.419,
      "step": 34
    },
    {
      "epoch": 0.8045977011494253,
      "grad_norm": 5.183217525482178,
      "learning_rate": 8.213938048432697e-05,
      "loss": 3.4315,
      "step": 35
    },
    {
      "epoch": 0.8275862068965517,
      "grad_norm": 2.042294979095459,
      "learning_rate": 8.07830737662829e-05,
      "loss": 3.3882,
      "step": 36
    },
    {
      "epoch": 0.8275862068965517,
      "eval_loss": 3.3608548641204834,
      "eval_runtime": 2.1534,
      "eval_samples_per_second": 67.799,
      "eval_steps_per_second": 2.322,
      "step": 36
    },
    {
      "epoch": 0.8505747126436781,
      "grad_norm": 1.6554501056671143,
      "learning_rate": 7.938926261462366e-05,
      "loss": 3.347,
      "step": 37
    },
    {
      "epoch": 0.8735632183908046,
      "grad_norm": 4.561530590057373,
      "learning_rate": 7.795964517353735e-05,
      "loss": 3.2378,
      "step": 38
    },
    {
      "epoch": 0.896551724137931,
      "grad_norm": 1.5960023403167725,
      "learning_rate": 7.649596321166024e-05,
      "loss": 3.3006,
      "step": 39
    },
    {
      "epoch": 0.9195402298850575,
      "grad_norm": 1.9813002347946167,
      "learning_rate": 7.500000000000001e-05,
      "loss": 3.3413,
      "step": 40
    },
    {
      "epoch": 0.9425287356321839,
      "grad_norm": 1.86942458152771,
      "learning_rate": 7.347357813929454e-05,
      "loss": 3.3407,
      "step": 41
    },
    {
      "epoch": 0.9655172413793104,
      "grad_norm": 1.8890225887298584,
      "learning_rate": 7.191855733945387e-05,
      "loss": 3.3447,
      "step": 42
    },
    {
      "epoch": 0.9885057471264368,
      "grad_norm": 2.1058900356292725,
      "learning_rate": 7.033683215379002e-05,
      "loss": 3.2648,
      "step": 43
    },
    {
      "epoch": 1.0114942528735633,
      "grad_norm": 4.351009368896484,
      "learning_rate": 6.873032967079561e-05,
      "loss": 4.8492,
      "step": 44
    },
    {
      "epoch": 1.0344827586206897,
      "grad_norm": 2.976430892944336,
      "learning_rate": 6.710100716628344e-05,
      "loss": 3.3652,
      "step": 45
    },
    {
      "epoch": 1.0344827586206897,
      "eval_loss": 3.286372661590576,
      "eval_runtime": 2.1582,
      "eval_samples_per_second": 67.647,
      "eval_steps_per_second": 2.317,
      "step": 45
    },
    {
      "epoch": 1.0574712643678161,
      "grad_norm": 2.045285224914551,
      "learning_rate": 6.545084971874738e-05,
      "loss": 3.2875,
      "step": 46
    },
    {
      "epoch": 1.0804597701149425,
      "grad_norm": 2.3181397914886475,
      "learning_rate": 6.378186779084995e-05,
      "loss": 3.0245,
      "step": 47
    },
    {
      "epoch": 1.103448275862069,
      "grad_norm": 2.984717607498169,
      "learning_rate": 6.209609477998338e-05,
      "loss": 3.4331,
      "step": 48
    },
    {
      "epoch": 1.1264367816091954,
      "grad_norm": 1.0449074506759644,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 3.3764,
      "step": 49
    },
    {
      "epoch": 1.1494252873563218,
      "grad_norm": 0.9436253905296326,
      "learning_rate": 5.868240888334653e-05,
      "loss": 2.9765,
      "step": 50
    },
    {
      "epoch": 1.1724137931034484,
      "grad_norm": 4.141868591308594,
      "learning_rate": 5.695865504800327e-05,
      "loss": 3.597,
      "step": 51
    },
    {
      "epoch": 1.1954022988505748,
      "grad_norm": 1.9662575721740723,
      "learning_rate": 5.522642316338268e-05,
      "loss": 3.136,
      "step": 52
    },
    {
      "epoch": 1.2183908045977012,
      "grad_norm": 1.9346809387207031,
      "learning_rate": 5.348782368720626e-05,
      "loss": 2.8733,
      "step": 53
    },
    {
      "epoch": 1.2413793103448276,
      "grad_norm": 5.7119364738464355,
      "learning_rate": 5.174497483512506e-05,
      "loss": 3.6832,
      "step": 54
    },
    {
      "epoch": 1.2413793103448276,
      "eval_loss": 3.386746644973755,
      "eval_runtime": 2.1621,
      "eval_samples_per_second": 67.528,
      "eval_steps_per_second": 2.313,
      "step": 54
    },
    {
      "epoch": 1.264367816091954,
      "grad_norm": 7.614490985870361,
      "learning_rate": 5e-05,
      "loss": 3.3329,
      "step": 55
    },
    {
      "epoch": 1.2873563218390804,
      "grad_norm": 8.576265335083008,
      "learning_rate": 4.825502516487497e-05,
      "loss": 3.1484,
      "step": 56
    },
    {
      "epoch": 1.3103448275862069,
      "grad_norm": 2.174159049987793,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 3.041,
      "step": 57
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 1.6589335203170776,
      "learning_rate": 4.477357683661734e-05,
      "loss": 3.3295,
      "step": 58
    },
    {
      "epoch": 1.3563218390804597,
      "grad_norm": 5.215062141418457,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 3.2889,
      "step": 59
    },
    {
      "epoch": 1.3793103448275863,
      "grad_norm": 2.2243685722351074,
      "learning_rate": 4.131759111665349e-05,
      "loss": 3.2102,
      "step": 60
    },
    {
      "epoch": 1.4022988505747127,
      "grad_norm": 6.214241981506348,
      "learning_rate": 3.960441545911204e-05,
      "loss": 3.3818,
      "step": 61
    },
    {
      "epoch": 1.4252873563218391,
      "grad_norm": 6.380526542663574,
      "learning_rate": 3.790390522001662e-05,
      "loss": 3.3575,
      "step": 62
    },
    {
      "epoch": 1.4482758620689655,
      "grad_norm": 4.525617599487305,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 3.2245,
      "step": 63
    },
    {
      "epoch": 1.4482758620689655,
      "eval_loss": 3.2319209575653076,
      "eval_runtime": 2.1603,
      "eval_samples_per_second": 67.582,
      "eval_steps_per_second": 2.314,
      "step": 63
    },
    {
      "epoch": 1.471264367816092,
      "grad_norm": 1.7523378133773804,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 3.2592,
      "step": 64
    },
    {
      "epoch": 1.4942528735632183,
      "grad_norm": 3.5593461990356445,
      "learning_rate": 3.289899283371657e-05,
      "loss": 3.1112,
      "step": 65
    },
    {
      "epoch": 1.5172413793103448,
      "grad_norm": 3.8328521251678467,
      "learning_rate": 3.12696703292044e-05,
      "loss": 3.354,
      "step": 66
    },
    {
      "epoch": 1.5402298850574714,
      "grad_norm": 2.212510824203491,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 3.2925,
      "step": 67
    },
    {
      "epoch": 1.5632183908045976,
      "grad_norm": 2.8463003635406494,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 3.0859,
      "step": 68
    },
    {
      "epoch": 1.5862068965517242,
      "grad_norm": 2.9076786041259766,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 3.1054,
      "step": 69
    },
    {
      "epoch": 1.6091954022988506,
      "grad_norm": 2.6354947090148926,
      "learning_rate": 2.500000000000001e-05,
      "loss": 3.1639,
      "step": 70
    },
    {
      "epoch": 1.632183908045977,
      "grad_norm": 1.233817219734192,
      "learning_rate": 2.350403678833976e-05,
      "loss": 3.1697,
      "step": 71
    },
    {
      "epoch": 1.6551724137931034,
      "grad_norm": 3.946552038192749,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 3.3657,
      "step": 72
    },
    {
      "epoch": 1.6551724137931034,
      "eval_loss": 3.222416400909424,
      "eval_runtime": 2.1573,
      "eval_samples_per_second": 67.678,
      "eval_steps_per_second": 2.318,
      "step": 72
    },
    {
      "epoch": 1.6781609195402298,
      "grad_norm": 3.0761539936065674,
      "learning_rate": 2.061073738537635e-05,
      "loss": 3.1671,
      "step": 73
    },
    {
      "epoch": 1.7011494252873565,
      "grad_norm": 1.3344730138778687,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 3.1811,
      "step": 74
    },
    {
      "epoch": 1.7241379310344827,
      "grad_norm": 4.6756062507629395,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 3.2284,
      "step": 75
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.225661850681344e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}