{
  "best_metric": 0.4528728723526001,
  "best_model_checkpoint": "data/Gemma-2-2B_task-3_180-samples_config-2/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": 5.613330841064453,
      "learning_rate": 2.5e-06,
      "loss": 3.087,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 8.413951873779297,
      "learning_rate": 5e-06,
      "loss": 3.8565,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 6.310426712036133,
      "learning_rate": 1e-05,
      "loss": 3.3773,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 6.2464518547058105,
      "learning_rate": 1.5e-05,
      "loss": 3.1725,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 5.46867561340332,
      "learning_rate": 2e-05,
      "loss": 3.2784,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 3.045137643814087,
      "eval_runtime": 11.3253,
      "eval_samples_per_second": 3.179,
      "eval_steps_per_second": 3.179,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 6.5437703132629395,
      "learning_rate": 2.5e-05,
      "loss": 2.9201,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 7.045949935913086,
      "learning_rate": 3e-05,
      "loss": 2.6453,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 4.619212627410889,
      "learning_rate": 3.5e-05,
      "loss": 2.2159,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 4.102094650268555,
      "learning_rate": 4e-05,
      "loss": 1.6125,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.3315627574920654,
      "eval_runtime": 11.3228,
      "eval_samples_per_second": 3.179,
      "eval_steps_per_second": 3.179,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 2.700011730194092,
      "learning_rate": 4.5e-05,
      "loss": 1.3726,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 1.972001314163208,
      "learning_rate": 5e-05,
      "loss": 1.1323,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 1.5637657642364502,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.8894,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.854964017868042,
      "learning_rate": 6e-05,
      "loss": 0.7031,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 0.6672905087471008,
      "eval_runtime": 11.2959,
      "eval_samples_per_second": 3.187,
      "eval_steps_per_second": 3.187,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.634579062461853,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.5268,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.636789858341217,
      "learning_rate": 7e-05,
      "loss": 0.7165,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 1.1011930704116821,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.55,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.7425286769866943,
      "learning_rate": 8e-05,
      "loss": 0.5697,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5897098183631897,
      "learning_rate": 8.5e-05,
      "loss": 0.3776,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.551551878452301,
      "eval_runtime": 11.2818,
      "eval_samples_per_second": 3.191,
      "eval_steps_per_second": 3.191,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.4286981225013733,
      "learning_rate": 9e-05,
      "loss": 0.5132,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.5576430559158325,
      "learning_rate": 9.5e-05,
      "loss": 0.5236,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.43634840846061707,
      "learning_rate": 0.0001,
      "loss": 0.4267,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.36028972268104553,
      "learning_rate": 9.999238475781957e-05,
      "loss": 0.3705,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 0.48932600021362305,
      "eval_runtime": 11.2995,
      "eval_samples_per_second": 3.186,
      "eval_steps_per_second": 3.186,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.47366511821746826,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.444,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.6494544148445129,
      "learning_rate": 9.99314767377287e-05,
      "loss": 0.2437,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.5118721723556519,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.323,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.5225521922111511,
      "learning_rate": 9.980973490458728e-05,
      "loss": 0.4428,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.458663672208786,
      "eval_runtime": 11.302,
      "eval_samples_per_second": 3.185,
      "eval_steps_per_second": 3.185,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.5786665678024292,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.3419,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.3510679602622986,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.3884,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.37410250306129456,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.2686,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.47454220056533813,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.264,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 0.4528728723526001,
      "eval_runtime": 11.2942,
      "eval_samples_per_second": 3.187,
      "eval_steps_per_second": 3.187,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.6065816283226013,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.4503,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.38948529958724976,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.3219,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 1.886899471282959,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.2721,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.4307541847229004,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.1738,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.7600765228271484,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.2488,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.46797963976860046,
      "eval_runtime": 11.2843,
      "eval_samples_per_second": 3.19,
      "eval_steps_per_second": 3.19,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.5920466780662537,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.2145,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.6227742433547974,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.2745,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.39027008414268494,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.1845,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 0.6960971355438232,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.211,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 0.4984787702560425,
      "eval_runtime": 11.2957,
      "eval_samples_per_second": 3.187,
      "eval_steps_per_second": 3.187,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.5394251942634583,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.1749,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.8638529181480408,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.1724,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.9163314700126648,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.1293,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.5777033567428589,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.1326,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.5513709187507629,
      "eval_runtime": 11.2958,
      "eval_samples_per_second": 3.187,
      "eval_steps_per_second": 3.187,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.5982895493507385,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.1137,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 2.035459280014038,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0532,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 1.205432415008545,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.1232,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.8699578642845154,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.1028,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 0.6646276116371155,
      "eval_runtime": 11.2936,
      "eval_samples_per_second": 3.188,
      "eval_steps_per_second": 3.188,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.547171950340271,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.0922,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 0.8112836480140686,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0587,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 1.217887282371521,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.0558,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 1.0182747840881348,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0371,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.9864963889122009,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.0306,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.811215341091156,
      "eval_runtime": 11.293,
      "eval_samples_per_second": 3.188,
      "eval_steps_per_second": 3.188,
      "step": 102
    },
    {
      "epoch": 12.235294117647058,
      "grad_norm": 0.9888170957565308,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0409,
      "step": 104
    },
    {
      "epoch": 12.470588235294118,
      "grad_norm": 0.963406503200531,
      "learning_rate": 9.193352839727121e-05,
      "loss": 0.0283,
      "step": 106
    },
    {
      "epoch": 12.705882352941176,
      "grad_norm": 0.670022189617157,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.0178,
      "step": 108
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 0.7791545391082764,
      "learning_rate": 9.09576022144496e-05,
      "loss": 0.0188,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "eval_loss": 0.9621884226799011,
      "eval_runtime": 11.2889,
      "eval_samples_per_second": 3.189,
      "eval_steps_per_second": 3.189,
      "step": 110
    },
    {
      "epoch": 13.176470588235293,
      "grad_norm": 0.39974257349967957,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0133,
      "step": 112
    },
    {
      "epoch": 13.411764705882353,
      "grad_norm": 0.20974881947040558,
      "learning_rate": 8.993177550236464e-05,
      "loss": 0.0075,
      "step": 114
    },
    {
      "epoch": 13.647058823529411,
      "grad_norm": 0.8507333397865295,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.0176,
      "step": 116
    },
    {
      "epoch": 13.882352941176471,
      "grad_norm": 1.422010064125061,
      "learning_rate": 8.885729807284856e-05,
      "loss": 0.0227,
      "step": 118
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.9967208504676819,
      "eval_runtime": 11.301,
      "eval_samples_per_second": 3.186,
      "eval_steps_per_second": 3.186,
      "step": 119
    },
    {
      "epoch": 14.0,
      "step": 119,
      "total_flos": 5.134565931116134e+16,
      "train_loss": 0.63686583302289,
      "train_runtime": 2156.2694,
      "train_samples_per_second": 3.154,
      "train_steps_per_second": 0.186
    }
  ],
  "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": 5.134565931116134e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}