{
  "best_metric": 0.8613905310630798,
  "best_model_checkpoint": "data/Mistral-7B_task-3_180-samples_config-2_full/checkpoint-59",
  "epoch": 14.0,
  "eval_steps": 500,
  "global_step": 119,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 1.021460771560669,
      "learning_rate": 2.5e-06,
      "loss": 1.264,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 1.0736006498336792,
      "learning_rate": 5e-06,
      "loss": 1.2818,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 0.9209890961647034,
      "learning_rate": 1e-05,
      "loss": 1.2809,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 0.9168510437011719,
      "learning_rate": 1.5e-05,
      "loss": 1.2115,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 0.730437695980072,
      "learning_rate": 2e-05,
      "loss": 1.224,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 1.1846206188201904,
      "eval_runtime": 39.0667,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.617757260799408,
      "learning_rate": 2.5e-05,
      "loss": 1.1954,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 0.43226438760757446,
      "learning_rate": 3e-05,
      "loss": 1.1241,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 0.34701934456825256,
      "learning_rate": 3.5e-05,
      "loss": 1.0914,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 0.3547401428222656,
      "learning_rate": 4e-05,
      "loss": 1.0894,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.0560972690582275,
      "eval_runtime": 39.0645,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.3320808708667755,
      "learning_rate": 4.5e-05,
      "loss": 1.0621,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.3414642810821533,
      "learning_rate": 5e-05,
      "loss": 1.0105,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.45840802788734436,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.983,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.4873791038990021,
      "learning_rate": 6e-05,
      "loss": 0.9794,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 0.9391953349113464,
      "eval_runtime": 39.062,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.4198255240917206,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.9151,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.23240754008293152,
      "learning_rate": 7e-05,
      "loss": 0.8902,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.2391039878129959,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.8502,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.24047456681728363,
      "learning_rate": 8e-05,
      "loss": 0.8868,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.22469373047351837,
      "learning_rate": 8.5e-05,
      "loss": 0.8318,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.8909127712249756,
      "eval_runtime": 39.073,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.19771155714988708,
      "learning_rate": 9e-05,
      "loss": 0.8409,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.2168087661266327,
      "learning_rate": 9.5e-05,
      "loss": 0.8314,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.2094077467918396,
      "learning_rate": 0.0001,
      "loss": 0.8331,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.23849743604660034,
      "learning_rate": 9.999238475781957e-05,
      "loss": 0.8075,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 0.8740689754486084,
      "eval_runtime": 39.0652,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.2399512231349945,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.7782,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.19294902682304382,
      "learning_rate": 9.99314767377287e-05,
      "loss": 0.81,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.20289131999015808,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.7839,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.21000699698925018,
      "learning_rate": 9.980973490458728e-05,
      "loss": 0.8096,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.8642433285713196,
      "eval_runtime": 39.0649,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.258810430765152,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.7543,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.2611052095890045,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.7355,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.279379278421402,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.7575,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.24716489017009735,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.7922,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 0.8613905310630798,
      "eval_runtime": 39.0644,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.28359732031822205,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.7442,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.3059193193912506,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.7241,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.5297988057136536,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.6951,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.39828604459762573,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.6883,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.4185999035835266,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.7089,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.8718762993812561,
      "eval_runtime": 39.0572,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.4456179440021515,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.6527,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.5093295574188232,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.6759,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.47449541091918945,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.6705,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 0.5402957201004028,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.6122,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 0.8933117985725403,
      "eval_runtime": 39.071,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.544838547706604,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.612,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.7277241349220276,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.6062,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.5708918571472168,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.5894,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.6501656174659729,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.569,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.9315780997276306,
      "eval_runtime": 39.0672,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.613765299320221,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.5322,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.7900691032409668,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.5371,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 0.7115367650985718,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.5023,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.889628529548645,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.508,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 0.9548236131668091,
      "eval_runtime": 39.0665,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.6886316537857056,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.5279,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 1.3477370738983154,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.4876,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 1.1403348445892334,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.4168,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 1.0651371479034424,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.4494,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.9667319059371948,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.4196,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.9916267395019531,
      "eval_runtime": 39.061,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 102
    },
    {
      "epoch": 12.235294117647058,
      "grad_norm": 0.8136600255966187,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.3845,
      "step": 104
    },
    {
      "epoch": 12.470588235294118,
      "grad_norm": 1.1212258338928223,
      "learning_rate": 9.193352839727121e-05,
      "loss": 0.3698,
      "step": 106
    },
    {
      "epoch": 12.705882352941176,
      "grad_norm": 0.991340160369873,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.384,
      "step": 108
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 0.9913235902786255,
      "learning_rate": 9.09576022144496e-05,
      "loss": 0.389,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "eval_loss": 1.0308579206466675,
      "eval_runtime": 39.0637,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 110
    },
    {
      "epoch": 13.176470588235293,
      "grad_norm": 0.8725358247756958,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.3553,
      "step": 112
    },
    {
      "epoch": 13.411764705882353,
      "grad_norm": 1.2270166873931885,
      "learning_rate": 8.993177550236464e-05,
      "loss": 0.304,
      "step": 114
    },
    {
      "epoch": 13.647058823529411,
      "grad_norm": 1.0568962097167969,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.3286,
      "step": 116
    },
    {
      "epoch": 13.882352941176471,
      "grad_norm": 1.1005165576934814,
      "learning_rate": 8.885729807284856e-05,
      "loss": 0.3283,
      "step": 118
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.128214955329895,
      "eval_runtime": 39.0643,
      "eval_samples_per_second": 0.922,
      "eval_steps_per_second": 0.922,
      "step": 119
    },
    {
      "epoch": 14.0,
      "step": 119,
      "total_flos": 1.8557653313231258e+17,
      "train_loss": 0.7292194727088223,
      "train_runtime": 6881.2832,
      "train_samples_per_second": 0.988,
      "train_steps_per_second": 0.058
    }
  ],
  "logging_steps": 2,
  "max_steps": 400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 25,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 7,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8557653313231258e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}