{
  "best_metric": 0.40765851736068726,
  "best_model_checkpoint": "data/Mistral-7B_task-3_180-samples_config-2/checkpoint-42",
  "epoch": 12.0,
  "eval_steps": 500,
  "global_step": 102,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 14.504705429077148,
      "learning_rate": 2.5e-06,
      "loss": 3.3326,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 20.7569580078125,
      "learning_rate": 5e-06,
      "loss": 4.3434,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 15.900101661682129,
      "learning_rate": 1e-05,
      "loss": 3.6601,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 11.783173561096191,
      "learning_rate": 1.5e-05,
      "loss": 3.1182,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 9.379878997802734,
      "learning_rate": 2e-05,
      "loss": 2.7267,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 2.1989693641662598,
      "eval_runtime": 39.0777,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 7.800622940063477,
      "learning_rate": 2.5e-05,
      "loss": 2.0051,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 7.2994065284729,
      "learning_rate": 3e-05,
      "loss": 1.5395,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 4.580909729003906,
      "learning_rate": 3.5e-05,
      "loss": 1.0346,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 2.24992036819458,
      "learning_rate": 4e-05,
      "loss": 0.8147,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6679273843765259,
      "eval_runtime": 39.074,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 1.516501545906067,
      "learning_rate": 4.5e-05,
      "loss": 0.6609,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.9916050434112549,
      "learning_rate": 5e-05,
      "loss": 0.607,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.7745568752288818,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.4275,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.7072432637214661,
      "learning_rate": 6e-05,
      "loss": 0.4601,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 0.47325241565704346,
      "eval_runtime": 39.0692,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.7457287907600403,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.3543,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.8342892527580261,
      "learning_rate": 7e-05,
      "loss": 0.4855,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.42753827571868896,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.3148,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 3.2698042392730713,
      "learning_rate": 8e-05,
      "loss": 0.38,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.6027395725250244,
      "learning_rate": 8.5e-05,
      "loss": 0.244,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.4196014404296875,
      "eval_runtime": 39.0757,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.6179048418998718,
      "learning_rate": 9e-05,
      "loss": 0.3033,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 3.261169195175171,
      "learning_rate": 9.5e-05,
      "loss": 0.3497,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.5675766468048096,
      "learning_rate": 0.0001,
      "loss": 0.2755,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.48287370800971985,
      "learning_rate": 9.999238475781957e-05,
      "loss": 0.2392,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 0.40765851736068726,
      "eval_runtime": 39.0794,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.5523759722709656,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.2591,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.31405553221702576,
      "learning_rate": 9.99314767377287e-05,
      "loss": 0.1407,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.8077965974807739,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.1793,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.6415297389030457,
      "learning_rate": 9.980973490458728e-05,
      "loss": 0.2333,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.4224318265914917,
      "eval_runtime": 39.0789,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.4493468701839447,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.1872,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.593795895576477,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.1529,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.7886911630630493,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.0964,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 1.127954363822937,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.1101,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 0.4902816414833069,
      "eval_runtime": 39.0736,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.7894813418388367,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.1401,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.5589519739151001,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.0716,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 1.3564684391021729,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.0646,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.9714311361312866,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.0327,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.2530443668365479,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.0532,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.5853617191314697,
      "eval_runtime": 39.0872,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.4537618160247803,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.0194,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 1.3883719444274902,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.0309,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.31875643134117126,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.0204,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 1.1900216341018677,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0266,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 0.664294958114624,
      "eval_runtime": 39.0752,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.5460513234138489,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.0099,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.4250204563140869,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.011,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.5882348418235779,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.0084,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.15863734483718872,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.009,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.7163164019584656,
      "eval_runtime": 39.0742,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.20348787307739258,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.007,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.13509777188301086,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0025,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 0.6861823797225952,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.008,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.32561418414115906,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.0074,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 0.7387914657592773,
      "eval_runtime": 39.0731,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.1485217809677124,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.0028,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 0.41832828521728516,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0047,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 0.09504914283752441,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.0014,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.6861616969108582,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0072,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.15043994784355164,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.0055,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.7272784113883972,
      "eval_runtime": 39.0768,
      "eval_samples_per_second": 0.921,
      "eval_steps_per_second": 0.921,
      "step": 102
    },
    {
      "epoch": 12.0,
      "step": 102,
      "total_flos": 1.5906559946339123e+17,
      "train_loss": 0.5047409891552639,
      "train_runtime": 5898.6129,
      "train_samples_per_second": 1.153,
      "train_steps_per_second": 0.068
    }
  ],
  "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": 1.5906559946339123e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}