reft_with_lora.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "c2fcc080-fd76-455e-82d8-2255f3043d88",
      "metadata": {
        "id": "c2fcc080-fd76-455e-82d8-2255f3043d88"
      },
      "source": [
        "### ReFT is complimentary to existing LoRAs (or other PEFTs)\n",
        "\n",
        "You can wrap a PEFT model as a ReFT model and take advantages of both worlds, as light-weight LoRA has no inference overhead while might provide performance gains. Note that LoRA is coupled with model weights, yet it can also been seens as an intervention which edits the original representation.\n",
        "\n",
        "It's very easy to combine PEFTs with ReFT by using existing library such as `peft`:\n",
        "\n",
        "```py\n",
        "import pyreft\n",
        "from peft import LoraConfig, get_peft_model\n",
        "\n",
        "peft_config = LoraConfig(...)\n",
        "model = get_peft_model(model, peft_config)\n",
        "\n",
        "reft_config = pyreft.ReftConfig(...)\n",
        "reft_model = pyreft.get_reft_model(model, reft_config)\n",
        "```"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "id": "4221ca53-0c2d-4288-b2fd-cace955d48a4",
      "metadata": {
        "id": "4221ca53-0c2d-4288-b2fd-cace955d48a4"
      },
      "outputs": [],
      "source": [
        "try:\n",
        "    import peft\n",
        "\n",
        "except ModuleNotFoundError:\n",
        "    !pip install peft"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "try:\n",
        "    import pyreft\n",
        "\n",
        "except ModuleNotFoundError:\n",
        "    !pip install pyreft"
      ],
      "metadata": {
        "id": "kTjQGj_SVW3E"
      },
      "id": "kTjQGj_SVW3E",
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "id": "a31814f4-b9ce-46c4-a77d-b1d0b0c11ae3",
      "metadata": {
        "id": "a31814f4-b9ce-46c4-a77d-b1d0b0c11ae3"
      },
      "source": [
        "### Loading our LM"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "id": "c29576a9-bd64-4d79-a033-4255e4b21553",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 34,
          "referenced_widgets": [
            "56fe2e0af9044fe7a1dd1208e8823e84",
            "14e32aaa8bde45c2bfd2a221d1c54861",
            "4f8c06fd6a25461baa6b1b8a632fc4df",
            "f95dcb4f140244a39c8c7bad30e0cb1a",
            "168523a1c7e243a788a27b36148528b6",
            "732c7288b30149da9ce8b42856f8b9e4",
            "a6fa2bd3ecaa4f3a88be64ffae37fb3e",
            "33d71ae97084435dad2861ad896a57a7",
            "82680331a8234062978d7cc62760599f",
            "8663bd74a3b840369eb1082650997601",
            "301caf0f897047408ff67b8b1a6a6991"
          ]
        },
        "id": "c29576a9-bd64-4d79-a033-4255e4b21553",
        "outputId": "c7efae54-ecaf-4b85-b178-b46f1a8fc05a"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Loading checkpoint shards:   0%|          | 0/4 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "56fe2e0af9044fe7a1dd1208e8823e84"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n"
          ]
        }
      ],
      "source": [
        "import torch, transformers\n",
        "\n",
        "from peft import LoraConfig, get_peft_model\n",
        "\n",
        "model_name_or_path = \"meta-llama/Meta-Llama-3-8B-Instruct\"\n",
        "model = transformers.AutoModelForCausalLM.from_pretrained(model_name_or_path, torch_dtype=torch.bfloat16, device_map=\"cuda\")\n",
        "\n",
        "# get tokenizer\n",
        "tokenizer = transformers.AutoTokenizer.from_pretrained(\n",
        "    model_name_or_path,\n",
        "    model_max_length=512,\n",
        "    use_fast=False)\n",
        "\n",
        "terminators = [\n",
        "    tokenizer.eos_token_id,\n",
        "    tokenizer.convert_tokens_to_ids(\"<|eot_id|>\")\n",
        "]"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "0f870b16",
      "metadata": {
        "id": "0f870b16"
      },
      "source": [
        "## Prepare ReCOGS dataset"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "id": "63e4acd4",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "63e4acd4",
        "outputId": "3178b2f2-e0df-46b8-dedc-a889cf26966d"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: wget in /usr/local/lib/python3.10/dist-packages (3.2)\n",
            "Requirement already satisfied: pandas in /usr/local/lib/python3.10/dist-packages (2.0.3)\n",
            "Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.10/dist-packages (from pandas) (2.8.2)\n",
            "Requirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.10/dist-packages (from pandas) (2023.4)\n",
            "Requirement already satisfied: tzdata>=2022.1 in /usr/local/lib/python3.10/dist-packages (from pandas) (2024.1)\n",
            "Requirement already satisfied: numpy>=1.21.0 in /usr/local/lib/python3.10/dist-packages (from pandas) (1.26.4)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.2->pandas) (1.16.0)\n"
          ]
        }
      ],
      "source": [
        "!pip install wget pandas"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!wget https://raw.githubusercontent.com/cgpotts/cs224u/main/compgen.py"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aPxP0u2E9izT",
        "outputId": "f5f48f85-d9f0-4b4a-e49a-195d19047029"
      },
      "id": "aPxP0u2E9izT",
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--2024-05-31 14:25:46--  https://raw.githubusercontent.com/cgpotts/cs224u/main/compgen.py\n",
            "Resolving raw.githubusercontent.com (raw.githubusercontent.com)... 185.199.108.133, 185.199.109.133, 185.199.110.133, ...\n",
            "Connecting to raw.githubusercontent.com (raw.githubusercontent.com)|185.199.108.133|:443... connected.\n",
            "HTTP request sent, awaiting response... 200 OK\n",
            "Length: 3669 (3.6K) [text/plain]\n",
            "Saving to: ‘compgen.py.2’\n",
            "\n",
            "compgen.py.2        100%[===================>]   3.58K  --.-KB/s    in 0s      \n",
            "\n",
            "2024-05-31 14:25:47 (69.3 MB/s) - ‘compgen.py.2’ saved [3669/3669]\n",
            "\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "id": "ee21a440",
      "metadata": {
        "id": "ee21a440"
      },
      "outputs": [],
      "source": [
        "import os\n",
        "import wget\n",
        "\n",
        "SRC_DIRNAME = os.path.join(\"data\", \"recogs\")\n",
        "if not os.path.exists(SRC_DIRNAME):\n",
        "    os.makedirs('data', exist_ok=True)\n",
        "    wget.download('https://web.stanford.edu/class/cs224u/data/recogs.tgz', out='data/')\n",
        "    !tar xvf data/recogs.tgz -C data/"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "id": "a8283e3a",
      "metadata": {
        "id": "a8283e3a"
      },
      "outputs": [],
      "source": [
        "import pandas as pd\n",
        "def load_split(filename):\n",
        "    return pd.read_csv(\n",
        "        filename,\n",
        "        delimiter=\"\\t\",\n",
        "        names=['input', 'output', 'category'])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 7,
      "id": "8486d47a",
      "metadata": {
        "id": "8486d47a"
      },
      "outputs": [],
      "source": [
        "dataset = {}\n",
        "\n",
        "for splitname in (\"train\", \"dev\", \"gen\"):\n",
        "    dataset[splitname] = load_split(f\"{SRC_DIRNAME}/{splitname}.tsv\")\n",
        "\n",
        "# Data prep\n",
        "from datasets import Dataset\n",
        "trainset = dataset['train']\n",
        "trainset = Dataset.from_pandas(trainset) # convert from pandas to pyarrow format\n",
        "trainset = trainset.remove_columns('category') # remove the category column"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 9,
      "id": "9dfca072",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 112
        },
        "id": "9dfca072",
        "outputId": "3cb0fa52-9054-4764-b158-2084e5bb32f7"
      },
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                          input  \\\n",
              "0  A rose was helped by a dog .   \n",
              "1     The sailor dusted a boy .   \n",
              "\n",
              "                                              output         category  \n",
              "0  rose ( 53 ) ; dog ( 38 ) ; help ( 7 ) AND them...  in_distribution  \n",
              "1  * sailor ( 48 ) ; boy ( 53 ) ; dust ( 10 ) AND...  in_distribution  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-0aca1ed3-40e5-4627-abd8-b7701865136c\" class=\"colab-df-container\">\n",
              "    <div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>input</th>\n",
              "      <th>output</th>\n",
              "      <th>category</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>A rose was helped by a dog .</td>\n",
              "      <td>rose ( 53 ) ; dog ( 38 ) ; help ( 7 ) AND them...</td>\n",
              "      <td>in_distribution</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>The sailor dusted a boy .</td>\n",
              "      <td>* sailor ( 48 ) ; boy ( 53 ) ; dust ( 10 ) AND...</td>\n",
              "      <td>in_distribution</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "    <div class=\"colab-df-buttons\">\n",
              "\n",
              "  <div class=\"colab-df-container\">\n",
              "    <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-0aca1ed3-40e5-4627-abd8-b7701865136c')\"\n",
              "            title=\"Convert this dataframe to an interactive table.\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\" viewBox=\"0 -960 960 960\">\n",
              "    <path d=\"M120-120v-720h720v720H120Zm60-500h600v-160H180v160Zm220 220h160v-160H400v160Zm0 220h160v-160H400v160ZM180-400h160v-160H180v160Zm440 0h160v-160H620v160ZM180-180h160v-160H180v160Zm440 0h160v-160H620v160Z\"/>\n",
              "  </svg>\n",
              "    </button>\n",
              "\n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    .colab-df-buttons div {\n",
              "      margin-bottom: 4px;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "    <script>\n",
              "      const buttonEl =\n",
              "        document.querySelector('#df-0aca1ed3-40e5-4627-abd8-b7701865136c button.colab-df-convert');\n",
              "      buttonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "      async function convertToInteractive(key) {\n",
              "        const element = document.querySelector('#df-0aca1ed3-40e5-4627-abd8-b7701865136c');\n",
              "        const dataTable =\n",
              "          await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                    [key], {});\n",
              "        if (!dataTable) return;\n",
              "\n",
              "        const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "          '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "          + ' to learn more about interactive tables.';\n",
              "        element.innerHTML = '';\n",
              "        dataTable['output_type'] = 'display_data';\n",
              "        await google.colab.output.renderOutput(dataTable, element);\n",
              "        const docLink = document.createElement('div');\n",
              "        docLink.innerHTML = docLinkHtml;\n",
              "        element.appendChild(docLink);\n",
              "      }\n",
              "    </script>\n",
              "  </div>\n",
              "\n",
              "\n",
              "<div id=\"df-88c52948-62a6-45e0-97bc-35cc370fb66e\">\n",
              "  <button class=\"colab-df-quickchart\" onclick=\"quickchart('df-88c52948-62a6-45e0-97bc-35cc370fb66e')\"\n",
              "            title=\"Suggest charts\"\n",
              "            style=\"display:none;\">\n",
              "\n",
              "<svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "     width=\"24px\">\n",
              "    <g>\n",
              "        <path d=\"M19 3H5c-1.1 0-2 .9-2 2v14c0 1.1.9 2 2 2h14c1.1 0 2-.9 2-2V5c0-1.1-.9-2-2-2zM9 17H7v-7h2v7zm4 0h-2V7h2v10zm4 0h-2v-4h2v4z\"/>\n",
              "    </g>\n",
              "</svg>\n",
              "  </button>\n",
              "\n",
              "<style>\n",
              "  .colab-df-quickchart {\n",
              "      --bg-color: #E8F0FE;\n",
              "      --fill-color: #1967D2;\n",
              "      --hover-bg-color: #E2EBFA;\n",
              "      --hover-fill-color: #174EA6;\n",
              "      --disabled-fill-color: #AAA;\n",
              "      --disabled-bg-color: #DDD;\n",
              "  }\n",
              "\n",
              "  [theme=dark] .colab-df-quickchart {\n",
              "      --bg-color: #3B4455;\n",
              "      --fill-color: #D2E3FC;\n",
              "      --hover-bg-color: #434B5C;\n",
              "      --hover-fill-color: #FFFFFF;\n",
              "      --disabled-bg-color: #3B4455;\n",
              "      --disabled-fill-color: #666;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart {\n",
              "    background-color: var(--bg-color);\n",
              "    border: none;\n",
              "    border-radius: 50%;\n",
              "    cursor: pointer;\n",
              "    display: none;\n",
              "    fill: var(--fill-color);\n",
              "    height: 32px;\n",
              "    padding: 0;\n",
              "    width: 32px;\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart:hover {\n",
              "    background-color: var(--hover-bg-color);\n",
              "    box-shadow: 0 1px 2px rgba(60, 64, 67, 0.3), 0 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "    fill: var(--button-hover-fill-color);\n",
              "  }\n",
              "\n",
              "  .colab-df-quickchart-complete:disabled,\n",
              "  .colab-df-quickchart-complete:disabled:hover {\n",
              "    background-color: var(--disabled-bg-color);\n",
              "    fill: var(--disabled-fill-color);\n",
              "    box-shadow: none;\n",
              "  }\n",
              "\n",
              "  .colab-df-spinner {\n",
              "    border: 2px solid var(--fill-color);\n",
              "    border-color: transparent;\n",
              "    border-bottom-color: var(--fill-color);\n",
              "    animation:\n",
              "      spin 1s steps(1) infinite;\n",
              "  }\n",
              "\n",
              "  @keyframes spin {\n",
              "    0% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "      border-left-color: var(--fill-color);\n",
              "    }\n",
              "    20% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    30% {\n",
              "      border-color: transparent;\n",
              "      border-left-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    40% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-top-color: var(--fill-color);\n",
              "    }\n",
              "    60% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "    }\n",
              "    80% {\n",
              "      border-color: transparent;\n",
              "      border-right-color: var(--fill-color);\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "    90% {\n",
              "      border-color: transparent;\n",
              "      border-bottom-color: var(--fill-color);\n",
              "    }\n",
              "  }\n",
              "</style>\n",
              "\n",
              "  <script>\n",
              "    async function quickchart(key) {\n",
              "      const quickchartButtonEl =\n",
              "        document.querySelector('#' + key + ' button');\n",
              "      quickchartButtonEl.disabled = true;  // To prevent multiple clicks.\n",
              "      quickchartButtonEl.classList.add('colab-df-spinner');\n",
              "      try {\n",
              "        const charts = await google.colab.kernel.invokeFunction(\n",
              "            'suggestCharts', [key], {});\n",
              "      } catch (error) {\n",
              "        console.error('Error during call to suggestCharts:', error);\n",
              "      }\n",
              "      quickchartButtonEl.classList.remove('colab-df-spinner');\n",
              "      quickchartButtonEl.classList.add('colab-df-quickchart-complete');\n",
              "    }\n",
              "    (() => {\n",
              "      let quickchartButtonEl =\n",
              "        document.querySelector('#df-88c52948-62a6-45e0-97bc-35cc370fb66e button');\n",
              "      quickchartButtonEl.style.display =\n",
              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "    })();\n",
              "  </script>\n",
              "</div>\n",
              "    </div>\n",
              "  </div>\n"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "dataframe",
              "summary": "{\n  \"name\": \"dataset['train']\",\n  \"rows\": 2,\n  \"fields\": [\n    {\n      \"column\": \"input\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"The sailor dusted a boy .\",\n          \"A rose was helped by a dog .\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"output\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 2,\n        \"samples\": [\n          \"* sailor ( 48 ) ; boy ( 53 ) ; dust ( 10 ) AND agent ( 10 , 48 ) AND theme ( 10 , 53 )\",\n          \"rose ( 53 ) ; dog ( 38 ) ; help ( 7 ) AND theme ( 7 , 53 ) AND agent ( 7 , 38 )\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"category\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"in_distribution\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
            }
          },
          "metadata": {},
          "execution_count": 9
        }
      ],
      "source": [
        "dataset['train'].head(2)"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "a3b8a490-032c-4888-b2dd-649ec5a7e3b5",
      "metadata": {
        "id": "a3b8a490-032c-4888-b2dd-649ec5a7e3b5"
      },
      "source": [
        "### That's it! Wrap the LM as a `peft` LM, then as a `pyreft` LM"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 8,
      "id": "cfec23c5-18e7-4583-9df6-7e987b41705d",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cfec23c5-18e7-4583-9df6-7e987b41705d",
        "outputId": "7d0d1838-44a5-4d0f-ede9-d50dc9c32038"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "trainable intervention params: 32,772 || trainable model params: 32,768\n",
            "model params: 8,030,294,016 || trainable%: 0.0008161594067342304\n"
          ]
        }
      ],
      "source": [
        "include_peft = True\n",
        "layers_to_transform = [15]\n",
        "\n",
        "if include_peft:\n",
        "    peft_config = LoraConfig(\n",
        "        r=4,\n",
        "        lora_alpha=32,\n",
        "        target_modules=[\"o_proj\"],\n",
        "        layers_to_transform=layers_to_transform,\n",
        "        use_rslora=True,\n",
        "        lora_dropout=0.05,\n",
        "        bias=\"none\",\n",
        "        task_type=\"CAUSAL_LM\"\n",
        "    )\n",
        "    model = get_peft_model(model, peft_config)\n",
        "\n",
        "reft_config = pyreft.ReftConfig(representations=[{\n",
        "    \"layer\": l, \"component\": f\"base_model.model.model.layers[{l}].output\" if include_peft else \"block_output\",\n",
        "    \"low_rank_dimension\": 4,\n",
        "    \"intervention\": pyreft.LoreftIntervention(embed_dim=model.config.hidden_size,\n",
        "    low_rank_dimension=4)} for l in layers_to_transform])\n",
        "\n",
        "reft_model = pyreft.get_reft_model(model, reft_config)\n",
        "# you need to call this to re-enable lora grads!\n",
        "reft_model.model.enable_adapter_layers()\n",
        "reft_model.print_trainable_parameters()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "dbbd4238-97ff-44f8-b2ad-351f6128842e",
      "metadata": {
        "id": "dbbd4238-97ff-44f8-b2ad-351f6128842e"
      },
      "source": [
        "# Prompts"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "prompt_no_input_template = \"\"\"<s>[INST] <<SYS>>\n",
        "You are a helpful assistant. Translate english sentences to ReCOGS logical form.\n",
        "<</SYS>>\n",
        "\n",
        "%s [/INST]\n",
        "\"\"\""
      ],
      "metadata": {
        "id": "9KQwY_Uha5og"
      },
      "id": "9KQwY_Uha5og",
      "execution_count": 10,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": 11,
      "id": "46f47c05-77af-4309-8f6c-e2ab8b94c46c",
      "metadata": {
        "id": "46f47c05-77af-4309-8f6c-e2ab8b94c46c"
      },
      "outputs": [],
      "source": [
        "data_module = pyreft.make_last_position_supervised_data_module(\n",
        "    tokenizer,\n",
        "    model,\n",
        "    [prompt_no_input_template % i for i in dataset['train']['input'][:20000]],\n",
        "    [o for o in dataset['train']['output'][:20000]])"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "tokenizer.pad_token = tokenizer.eos_token"
      ],
      "metadata": {
        "id": "TVnegByRs1cA"
      },
      "id": "TVnegByRs1cA",
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "execution_count": 13,
      "id": "13d26c1a-2ff2-4b85-a00c-35591271f553",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "13d26c1a-2ff2-4b85-a00c-35591271f553",
        "outputId": "b0c5fd2d-62d0-4b50-c867-78ea2abf7d51"
      },
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "    <div>\n",
              "      \n",
              "      <progress value='20000' max='20000' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      [20000/20000 1:07:24, Epoch 10/10]\n",
              "    </div>\n",
              "    <table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              " <tr style=\"text-align: left;\">\n",
              "      <th>Step</th>\n",
              "      <th>Training Loss</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>50</td>\n",
              "      <td>0.986500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>100</td>\n",
              "      <td>0.564200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>150</td>\n",
              "      <td>0.552300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>200</td>\n",
              "      <td>0.514800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>250</td>\n",
              "      <td>0.497700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>300</td>\n",
              "      <td>0.497200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>350</td>\n",
              "      <td>0.498900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>400</td>\n",
              "      <td>0.489400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>450</td>\n",
              "      <td>0.492300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>500</td>\n",
              "      <td>0.473100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>550</td>\n",
              "      <td>0.480700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>600</td>\n",
              "      <td>0.464800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>650</td>\n",
              "      <td>0.472800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>700</td>\n",
              "      <td>0.485700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>750</td>\n",
              "      <td>0.454400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>800</td>\n",
              "      <td>0.452300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>850</td>\n",
              "      <td>0.470900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>900</td>\n",
              "      <td>0.466200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>950</td>\n",
              "      <td>0.465300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1000</td>\n",
              "      <td>0.469300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1050</td>\n",
              "      <td>0.483200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1100</td>\n",
              "      <td>0.456600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1150</td>\n",
              "      <td>0.480800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1200</td>\n",
              "      <td>0.462800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1250</td>\n",
              "      <td>0.458100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1300</td>\n",
              "      <td>0.445600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1350</td>\n",
              "      <td>0.463100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1400</td>\n",
              "      <td>0.456000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1450</td>\n",
              "      <td>0.446100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1500</td>\n",
              "      <td>0.453500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1550</td>\n",
              "      <td>0.464400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1600</td>\n",
              "      <td>0.444800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1650</td>\n",
              "      <td>0.459200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1700</td>\n",
              "      <td>0.449700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1750</td>\n",
              "      <td>0.450000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1800</td>\n",
              "      <td>0.442900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1850</td>\n",
              "      <td>0.437500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1900</td>\n",
              "      <td>0.457100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1950</td>\n",
              "      <td>0.445800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2000</td>\n",
              "      <td>0.423300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2050</td>\n",
              "      <td>0.433600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2100</td>\n",
              "      <td>0.458900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2150</td>\n",
              "      <td>0.426100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2200</td>\n",
              "      <td>0.439700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2250</td>\n",
              "      <td>0.427400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2300</td>\n",
              "      <td>0.428700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2350</td>\n",
              "      <td>0.425300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2400</td>\n",
              "      <td>0.451700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2450</td>\n",
              "      <td>0.430200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2500</td>\n",
              "      <td>0.431400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2550</td>\n",
              "      <td>0.441300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2600</td>\n",
              "      <td>0.441500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2650</td>\n",
              "      <td>0.431500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2700</td>\n",
              "      <td>0.432500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2750</td>\n",
              "      <td>0.443200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2800</td>\n",
              "      <td>0.459000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2850</td>\n",
              "      <td>0.442600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2900</td>\n",
              "      <td>0.429700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>2950</td>\n",
              "      <td>0.430800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3000</td>\n",
              "      <td>0.431700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3050</td>\n",
              "      <td>0.438600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3100</td>\n",
              "      <td>0.410500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3150</td>\n",
              "      <td>0.435800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3200</td>\n",
              "      <td>0.425900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3250</td>\n",
              "      <td>0.422100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3300</td>\n",
              "      <td>0.437300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3350</td>\n",
              "      <td>0.429600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3400</td>\n",
              "      <td>0.415100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3450</td>\n",
              "      <td>0.421200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3500</td>\n",
              "      <td>0.420200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3550</td>\n",
              "      <td>0.413900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3600</td>\n",
              "      <td>0.436400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3650</td>\n",
              "      <td>0.434000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3700</td>\n",
              "      <td>0.411600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3750</td>\n",
              "      <td>0.412800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3800</td>\n",
              "      <td>0.424600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3850</td>\n",
              "      <td>0.421900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3900</td>\n",
              "      <td>0.412700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>3950</td>\n",
              "      <td>0.424200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4000</td>\n",
              "      <td>0.433500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4050</td>\n",
              "      <td>0.416900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4100</td>\n",
              "      <td>0.410600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4150</td>\n",
              "      <td>0.437400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4200</td>\n",
              "      <td>0.417300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4250</td>\n",
              "      <td>0.418900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4300</td>\n",
              "      <td>0.411300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4350</td>\n",
              "      <td>0.435100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4400</td>\n",
              "      <td>0.405800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4450</td>\n",
              "      <td>0.401900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4500</td>\n",
              "      <td>0.396600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4550</td>\n",
              "      <td>0.407000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4600</td>\n",
              "      <td>0.395900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4650</td>\n",
              "      <td>0.405500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4700</td>\n",
              "      <td>0.424600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4750</td>\n",
              "      <td>0.424900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4800</td>\n",
              "      <td>0.431800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4850</td>\n",
              "      <td>0.402300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4900</td>\n",
              "      <td>0.414400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>4950</td>\n",
              "      <td>0.404600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5000</td>\n",
              "      <td>0.426800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5050</td>\n",
              "      <td>0.411100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5100</td>\n",
              "      <td>0.402300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5150</td>\n",
              "      <td>0.408400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5200</td>\n",
              "      <td>0.385500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5250</td>\n",
              "      <td>0.403700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5300</td>\n",
              "      <td>0.408200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5350</td>\n",
              "      <td>0.396300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5400</td>\n",
              "      <td>0.389600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5450</td>\n",
              "      <td>0.405200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5500</td>\n",
              "      <td>0.398100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5550</td>\n",
              "      <td>0.402600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5600</td>\n",
              "      <td>0.406600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5650</td>\n",
              "      <td>0.387600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5700</td>\n",
              "      <td>0.386800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5750</td>\n",
              "      <td>0.398200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5800</td>\n",
              "      <td>0.414900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5850</td>\n",
              "      <td>0.400800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5900</td>\n",
              "      <td>0.399700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>5950</td>\n",
              "      <td>0.395500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6000</td>\n",
              "      <td>0.403100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6050</td>\n",
              "      <td>0.394900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6100</td>\n",
              "      <td>0.383800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6150</td>\n",
              "      <td>0.397100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6200</td>\n",
              "      <td>0.384500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6250</td>\n",
              "      <td>0.404400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6300</td>\n",
              "      <td>0.382300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6350</td>\n",
              "      <td>0.390200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6400</td>\n",
              "      <td>0.387500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6450</td>\n",
              "      <td>0.378000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6500</td>\n",
              "      <td>0.389400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6550</td>\n",
              "      <td>0.381400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6600</td>\n",
              "      <td>0.375600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6650</td>\n",
              "      <td>0.388100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6700</td>\n",
              "      <td>0.395000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6750</td>\n",
              "      <td>0.399000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6800</td>\n",
              "      <td>0.399500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6850</td>\n",
              "      <td>0.410200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6900</td>\n",
              "      <td>0.404600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>6950</td>\n",
              "      <td>0.392200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7000</td>\n",
              "      <td>0.382400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7050</td>\n",
              "      <td>0.391700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7100</td>\n",
              "      <td>0.381100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7150</td>\n",
              "      <td>0.381600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7200</td>\n",
              "      <td>0.385100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7250</td>\n",
              "      <td>0.387500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7300</td>\n",
              "      <td>0.393500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7350</td>\n",
              "      <td>0.382000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7400</td>\n",
              "      <td>0.375000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7450</td>\n",
              "      <td>0.374500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7500</td>\n",
              "      <td>0.381300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7550</td>\n",
              "      <td>0.368500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7600</td>\n",
              "      <td>0.396100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7650</td>\n",
              "      <td>0.374200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7700</td>\n",
              "      <td>0.387900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7750</td>\n",
              "      <td>0.365700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7800</td>\n",
              "      <td>0.382900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7850</td>\n",
              "      <td>0.378100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7900</td>\n",
              "      <td>0.388600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>7950</td>\n",
              "      <td>0.381900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8000</td>\n",
              "      <td>0.373600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8050</td>\n",
              "      <td>0.380300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8100</td>\n",
              "      <td>0.383400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8150</td>\n",
              "      <td>0.379600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8200</td>\n",
              "      <td>0.385200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8250</td>\n",
              "      <td>0.363900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8300</td>\n",
              "      <td>0.378100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8350</td>\n",
              "      <td>0.380700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8400</td>\n",
              "      <td>0.378100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8450</td>\n",
              "      <td>0.365400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8500</td>\n",
              "      <td>0.377600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8550</td>\n",
              "      <td>0.374000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8600</td>\n",
              "      <td>0.373000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8650</td>\n",
              "      <td>0.374300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8700</td>\n",
              "      <td>0.381900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8750</td>\n",
              "      <td>0.363800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8800</td>\n",
              "      <td>0.374200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8850</td>\n",
              "      <td>0.365800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8900</td>\n",
              "      <td>0.367000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>8950</td>\n",
              "      <td>0.362900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9000</td>\n",
              "      <td>0.364900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9050</td>\n",
              "      <td>0.362000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9100</td>\n",
              "      <td>0.370300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9150</td>\n",
              "      <td>0.370200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9200</td>\n",
              "      <td>0.365400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9250</td>\n",
              "      <td>0.364300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9300</td>\n",
              "      <td>0.375800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9350</td>\n",
              "      <td>0.361000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9400</td>\n",
              "      <td>0.360900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9450</td>\n",
              "      <td>0.368100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9500</td>\n",
              "      <td>0.379300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9550</td>\n",
              "      <td>0.357500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9600</td>\n",
              "      <td>0.374500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9650</td>\n",
              "      <td>0.357900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9700</td>\n",
              "      <td>0.363700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9750</td>\n",
              "      <td>0.374500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9800</td>\n",
              "      <td>0.367100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9850</td>\n",
              "      <td>0.374800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9900</td>\n",
              "      <td>0.358300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>9950</td>\n",
              "      <td>0.357900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10000</td>\n",
              "      <td>0.365100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10050</td>\n",
              "      <td>0.371300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10100</td>\n",
              "      <td>0.362200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10150</td>\n",
              "      <td>0.364900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10200</td>\n",
              "      <td>0.359100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10250</td>\n",
              "      <td>0.364500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10300</td>\n",
              "      <td>0.364300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10350</td>\n",
              "      <td>0.367100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10400</td>\n",
              "      <td>0.360900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10450</td>\n",
              "      <td>0.361500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10500</td>\n",
              "      <td>0.363900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10550</td>\n",
              "      <td>0.355300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10600</td>\n",
              "      <td>0.363100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10650</td>\n",
              "      <td>0.361600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10700</td>\n",
              "      <td>0.357800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10750</td>\n",
              "      <td>0.360400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10800</td>\n",
              "      <td>0.355900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10850</td>\n",
              "      <td>0.358000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10900</td>\n",
              "      <td>0.374400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>10950</td>\n",
              "      <td>0.349000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11000</td>\n",
              "      <td>0.360600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11050</td>\n",
              "      <td>0.359700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11100</td>\n",
              "      <td>0.360700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11150</td>\n",
              "      <td>0.352600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11200</td>\n",
              "      <td>0.356300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11250</td>\n",
              "      <td>0.360500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11300</td>\n",
              "      <td>0.361400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11350</td>\n",
              "      <td>0.370400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11400</td>\n",
              "      <td>0.355200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11450</td>\n",
              "      <td>0.361000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11500</td>\n",
              "      <td>0.354700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11550</td>\n",
              "      <td>0.359800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11600</td>\n",
              "      <td>0.351100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11650</td>\n",
              "      <td>0.351600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11700</td>\n",
              "      <td>0.351800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11750</td>\n",
              "      <td>0.348400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11800</td>\n",
              "      <td>0.351600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11850</td>\n",
              "      <td>0.350100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11900</td>\n",
              "      <td>0.346800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>11950</td>\n",
              "      <td>0.345000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12000</td>\n",
              "      <td>0.360300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12050</td>\n",
              "      <td>0.348900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12100</td>\n",
              "      <td>0.344300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12150</td>\n",
              "      <td>0.343800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12200</td>\n",
              "      <td>0.351000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12250</td>\n",
              "      <td>0.344200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12300</td>\n",
              "      <td>0.342300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12350</td>\n",
              "      <td>0.347200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12400</td>\n",
              "      <td>0.347300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12450</td>\n",
              "      <td>0.344400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12500</td>\n",
              "      <td>0.358000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12550</td>\n",
              "      <td>0.343700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12600</td>\n",
              "      <td>0.355100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12650</td>\n",
              "      <td>0.342500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12700</td>\n",
              "      <td>0.344500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12750</td>\n",
              "      <td>0.350000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12800</td>\n",
              "      <td>0.363100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12850</td>\n",
              "      <td>0.352900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12900</td>\n",
              "      <td>0.339700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>12950</td>\n",
              "      <td>0.343800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13000</td>\n",
              "      <td>0.347800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13050</td>\n",
              "      <td>0.358300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13100</td>\n",
              "      <td>0.351600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13150</td>\n",
              "      <td>0.341600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13200</td>\n",
              "      <td>0.351500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13250</td>\n",
              "      <td>0.359200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13300</td>\n",
              "      <td>0.362400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13350</td>\n",
              "      <td>0.341200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13400</td>\n",
              "      <td>0.342500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13450</td>\n",
              "      <td>0.335000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13500</td>\n",
              "      <td>0.342000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13550</td>\n",
              "      <td>0.339200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13600</td>\n",
              "      <td>0.344200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13650</td>\n",
              "      <td>0.340700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13700</td>\n",
              "      <td>0.339900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13750</td>\n",
              "      <td>0.339800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13800</td>\n",
              "      <td>0.337400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13850</td>\n",
              "      <td>0.340800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13900</td>\n",
              "      <td>0.340600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>13950</td>\n",
              "      <td>0.337900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14000</td>\n",
              "      <td>0.337800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14050</td>\n",
              "      <td>0.342300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14100</td>\n",
              "      <td>0.338900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14150</td>\n",
              "      <td>0.340300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14200</td>\n",
              "      <td>0.347500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14250</td>\n",
              "      <td>0.335900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14300</td>\n",
              "      <td>0.343900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14350</td>\n",
              "      <td>0.335200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14400</td>\n",
              "      <td>0.338900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14450</td>\n",
              "      <td>0.339900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14500</td>\n",
              "      <td>0.333100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14550</td>\n",
              "      <td>0.341000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14600</td>\n",
              "      <td>0.338600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14650</td>\n",
              "      <td>0.339600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14700</td>\n",
              "      <td>0.335000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14750</td>\n",
              "      <td>0.330500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14800</td>\n",
              "      <td>0.326500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14850</td>\n",
              "      <td>0.350500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14900</td>\n",
              "      <td>0.333900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>14950</td>\n",
              "      <td>0.335200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15000</td>\n",
              "      <td>0.327800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15050</td>\n",
              "      <td>0.339500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15100</td>\n",
              "      <td>0.340000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15150</td>\n",
              "      <td>0.342900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15200</td>\n",
              "      <td>0.328900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15250</td>\n",
              "      <td>0.337500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15300</td>\n",
              "      <td>0.339300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15350</td>\n",
              "      <td>0.334200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15400</td>\n",
              "      <td>0.332900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15450</td>\n",
              "      <td>0.330000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15500</td>\n",
              "      <td>0.321900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15550</td>\n",
              "      <td>0.332600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15600</td>\n",
              "      <td>0.349200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15650</td>\n",
              "      <td>0.336300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15700</td>\n",
              "      <td>0.339300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15750</td>\n",
              "      <td>0.330200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15800</td>\n",
              "      <td>0.325400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15850</td>\n",
              "      <td>0.331100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15900</td>\n",
              "      <td>0.326100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>15950</td>\n",
              "      <td>0.333500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16000</td>\n",
              "      <td>0.330600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16050</td>\n",
              "      <td>0.328700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16100</td>\n",
              "      <td>0.327800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16150</td>\n",
              "      <td>0.324500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16200</td>\n",
              "      <td>0.327700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16250</td>\n",
              "      <td>0.324400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16300</td>\n",
              "      <td>0.332200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16350</td>\n",
              "      <td>0.335700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16400</td>\n",
              "      <td>0.327100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16450</td>\n",
              "      <td>0.333700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16500</td>\n",
              "      <td>0.324800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16550</td>\n",
              "      <td>0.326200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16600</td>\n",
              "      <td>0.325200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16650</td>\n",
              "      <td>0.323500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16700</td>\n",
              "      <td>0.324600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16750</td>\n",
              "      <td>0.326300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16800</td>\n",
              "      <td>0.326600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16850</td>\n",
              "      <td>0.327600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16900</td>\n",
              "      <td>0.321200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>16950</td>\n",
              "      <td>0.322700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17000</td>\n",
              "      <td>0.320500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17050</td>\n",
              "      <td>0.319900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17100</td>\n",
              "      <td>0.324500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17150</td>\n",
              "      <td>0.320500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17200</td>\n",
              "      <td>0.320900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17250</td>\n",
              "      <td>0.321900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17300</td>\n",
              "      <td>0.326100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17350</td>\n",
              "      <td>0.321800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17400</td>\n",
              "      <td>0.321200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17450</td>\n",
              "      <td>0.323100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17500</td>\n",
              "      <td>0.318000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17550</td>\n",
              "      <td>0.316900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17600</td>\n",
              "      <td>0.330700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17650</td>\n",
              "      <td>0.321600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17700</td>\n",
              "      <td>0.317000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17750</td>\n",
              "      <td>0.318900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17800</td>\n",
              "      <td>0.318100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17850</td>\n",
              "      <td>0.315100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17900</td>\n",
              "      <td>0.319700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>17950</td>\n",
              "      <td>0.317800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18000</td>\n",
              "      <td>0.320000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18050</td>\n",
              "      <td>0.315200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18100</td>\n",
              "      <td>0.316600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18150</td>\n",
              "      <td>0.317400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18200</td>\n",
              "      <td>0.317100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18250</td>\n",
              "      <td>0.318300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18300</td>\n",
              "      <td>0.316800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18350</td>\n",
              "      <td>0.313100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18400</td>\n",
              "      <td>0.317800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18450</td>\n",
              "      <td>0.317200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18500</td>\n",
              "      <td>0.318100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18550</td>\n",
              "      <td>0.314700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18600</td>\n",
              "      <td>0.313600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18650</td>\n",
              "      <td>0.311500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18700</td>\n",
              "      <td>0.315700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18750</td>\n",
              "      <td>0.313200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18800</td>\n",
              "      <td>0.312500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18850</td>\n",
              "      <td>0.312600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18900</td>\n",
              "      <td>0.310100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>18950</td>\n",
              "      <td>0.312100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19000</td>\n",
              "      <td>0.312100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19050</td>\n",
              "      <td>0.311900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19100</td>\n",
              "      <td>0.312000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19150</td>\n",
              "      <td>0.310100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19200</td>\n",
              "      <td>0.312400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19250</td>\n",
              "      <td>0.313300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19300</td>\n",
              "      <td>0.310200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19350</td>\n",
              "      <td>0.310600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19400</td>\n",
              "      <td>0.310200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19450</td>\n",
              "      <td>0.313900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19500</td>\n",
              "      <td>0.310400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19550</td>\n",
              "      <td>0.308600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19600</td>\n",
              "      <td>0.309800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19650</td>\n",
              "      <td>0.308500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19700</td>\n",
              "      <td>0.307400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19750</td>\n",
              "      <td>0.309000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19800</td>\n",
              "      <td>0.310100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19850</td>\n",
              "      <td>0.309700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19900</td>\n",
              "      <td>0.307300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>19950</td>\n",
              "      <td>0.307100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>20000</td>\n",
              "      <td>0.307600</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table><p>"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Directory './tmp/checkpoint-500/intervenable_model' already exists.\n",
            "Directory './tmp/checkpoint-1000/intervenable_model' already exists.\n",
            "Directory './tmp/checkpoint-1500/intervenable_model' already exists.\n",
            "Directory './tmp/checkpoint-2000/intervenable_model' already exists.\n",
            "Directory './tmp/checkpoint-2500/intervenable_model' already exists.\n",
            "Directory './tmp/checkpoint-3000/intervenable_model' already exists.\n",
            "Directory './tmp/checkpoint-3500/intervenable_model' already exists.\n",
            "Directory './tmp/checkpoint-4000/intervenable_model' already exists.\n",
            "Directory './tmp/checkpoint-4500/intervenable_model' already exists.\n",
            "Directory './tmp/checkpoint-5000/intervenable_model' already exists.\n",
            "Directory './tmp/checkpoint-5500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-6000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-6500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-7000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-7500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-8000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-8500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-9000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-9500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-10000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-10500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-11000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-11500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-12000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-12500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-13000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-13500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-14000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-14500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-15000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-15500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-16000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-16500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-17000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-17500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-18000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-18500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-19000/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-19500/intervenable_model' created successfully.\n",
            "Directory './tmp/checkpoint-20000/intervenable_model' created successfully.\n"
          ]
        }
      ],
      "source": [
        "# train\n",
        "training_args = transformers.TrainingArguments(\n",
        "    num_train_epochs=10.0,\n",
        "    output_dir=\"./tmp\",\n",
        "    per_device_train_batch_size=10,\n",
        "    learning_rate=4e-3,\n",
        "    logging_steps=50,\n",
        "    report_to=[])\n",
        "trainer = pyreft.ReftTrainerForCausalLM(\n",
        "    model=reft_model,\n",
        "    tokenizer=tokenizer,\n",
        "    args=training_args,\n",
        "    **data_module)\n",
        "_ = trainer.train()\n",
        "\n",
        "# ensure everything is in eval mode\n",
        "reft_model.model.eval()\n",
        "for k,v in reft_model.interventions.items():\n",
        "    _ = v[0].eval()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "8f6a7d11-963a-4a36-ad8d-3417ec15012c",
      "metadata": {
        "id": "8f6a7d11-963a-4a36-ad8d-3417ec15012c"
      },
      "source": [
        "**Note**: `loss` looks a bit different if you compare these with the ones in the original ReFT-only training."
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from transformers import logging\n",
        "logging.set_verbosity_error()\n",
        "\n",
        "def predict(instruction):\n",
        "    # Define the target sequence\n",
        "    target_sequence = \"[/INST]\\n\"\n",
        "\n",
        "    # tokenize and prepare the input\n",
        "    prompt = prompt_no_input_template % instruction\n",
        "    prompt = tokenizer(prompt, return_tensors=\"pt\").to(\"cuda\")\n",
        "\n",
        "    base_unit_location = prompt[\"input_ids\"].shape[-1] - 1  # last position\n",
        "    _, reft_response = reft_model.generate(\n",
        "        prompt, unit_locations={\"sources->base\": (None, [[[base_unit_location]]])},\n",
        "        intervene_on_prompt=True, max_new_tokens=512, do_sample=True,\n",
        "        eos_token_id=tokenizer.eos_token_id, early_stopping=True, num_beams=5)\n",
        "\n",
        "    input_string = tokenizer.decode(reft_response[0], skip_special_tokens=True)\n",
        "\n",
        "    # Find the index of the target sequence in the input string\n",
        "    start_index = input_string.find(target_sequence)\n",
        "\n",
        "    # If the target sequence is not found, return an empty string\n",
        "    if start_index == -1:\n",
        "        return \"\"\n",
        "\n",
        "    # Calculate the start index of the substring after the target sequence\n",
        "    start_index += len(target_sequence)\n",
        "\n",
        "    # Return the substring from the calculated start index to the end of the string\n",
        "    return input_string[start_index:]"
      ],
      "metadata": {
        "id": "sB72i4gO3AzB"
      },
      "id": "sB72i4gO3AzB",
      "execution_count": 14,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from compgen import recogs_exact_match"
      ],
      "metadata": {
        "id": "qB5kCCTh5EDC"
      },
      "id": "qB5kCCTh5EDC",
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "dataset['dev']['output'][0]"
      ],
      "metadata": {
        "outputId": "ec2ee3f0-440b-40b7-bb82-991dcf7a18a6",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 35
        },
        "id": "TP099XdR1Dwy"
      },
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'Liam ( 30 ) ; box ( 25 ) ; girl ( 21 ) ; hope ( 33 ) AND agent ( 33 , 30 ) AND ccomp ( 33 , 24 ) AND burn ( 24 ) AND theme ( 24 , 25 ) AND agent ( 24 , 21 )'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 22
        }
      ],
      "id": "TP099XdR1Dwy"
    },
    {
      "cell_type": "code",
      "source": [
        "query = dataset['dev'].sample(1).iloc[0]\n",
        "expected = query['output']\n",
        "input = query['input']\n",
        "\n",
        "predicted = predict(input)\n",
        "print(f\"Query: {input}\")\n",
        "print(f\"Expected: {expected}\")\n",
        "print(f\"Predicted: {predicted}\")\n",
        "print(f\"Correct: {recogs_exact_match(predicted, expected)}\")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UgBEr5N-ytwD",
        "outputId": "0775c2f9-d788-4f9a-b4ac-320afb19d847"
      },
      "id": "UgBEr5N-ytwD",
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Query: The girl said that a lollipop froze .\n",
            "Expected: * girl ( 22 ) ; lollipop ( 48 ) ; say ( 6 ) AND agent ( 6 , 22 ) AND ccomp ( 6 , 52 ) AND freeze ( 52 ) AND theme ( 52 , 48 )\n",
            "Predicted: * girl ( 45 ) ; lollipop ( 22 ) ; say ( 46 ) AND agent ( 46, 45 ) AND ccomp ( 46, 27 ) AND freeze ( 27 ) AND theme ( 27, 22 )\n",
            "Correct: True\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "ssamp = dataset['dev'].sample(200)"
      ],
      "metadata": {
        "id": "ID6NewnO6DL4"
      },
      "id": "ID6NewnO6DL4",
      "execution_count": 16,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "ssamp['prediction'] = ssamp.input.apply(lambda x: predict(x))"
      ],
      "metadata": {
        "id": "P1usKnlF6H8L"
      },
      "id": "P1usKnlF6H8L",
      "execution_count": 17,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "ssamp['correct'] = ssamp.apply(lambda row: recogs_exact_match(row['output'], row['prediction']), axis=1)"
      ],
      "metadata": {
        "id": "U_wjVXzh6iBu"
      },
      "id": "U_wjVXzh6iBu",
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "ssamp['correct'].sum() / ssamp.shape[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "807Rchoe6tg3",
        "outputId": "e83b4b90-90e0-4605-9e31-54ea8c97a0c5"
      },
      "id": "807Rchoe6tg3",
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0.985"
            ]
          },
          "metadata": {},
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "model.push_to_hub(\"ricardo-larosa/recogs-Meta-Llama-3-8B-Instruct\", token = \"\") # save to HF"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 52,
          "referenced_widgets": [
            "cdaadb2d4ee442f08f516f09db9e179b",
            "2143cf4a5085498b90fc26d6ee3233b1",
            "73215571e490481aae8448c0908a161d",
            "0a0296ee55e04de98b2bc986755bd1bc",
            "337cf05e9c66417393a92f80862d0622",
            "c44c30be11be4c21ae4b23c23704f1d7",
            "d3af487a46904aac8e0f6e5b6e173795",
            "9d76c6f2ed63448a8954afd837b96dcb",
            "9327f5a00a0e4095b67f21582ad12766",
            "4fd8ca98f726465887bf5b460bf4457f",
            "0c32972b51f447e7b9fa30fc9c7314e8"
          ]
        },
        "id": "CJGMW73FWhZt",
        "outputId": "196b79d5-163d-4b9b-ed39-36c53ecc38c0"
      },
      "id": "CJGMW73FWhZt",
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "adapter_model.safetensors:   0%|          | 0.00/65.8k [00:00<?, ?B/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "cdaadb2d4ee442f08f516f09db9e179b"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "CommitInfo(commit_url='https://huggingface.co/ricardo-larosa/recogs-Meta-Llama-3-8B-Instruct/commit/af36e473f8d9fc9b9834be73aa5496081426404b', commit_message='Upload model', commit_description='', oid='af36e473f8d9fc9b9834be73aa5496081426404b', pr_url=None, pr_revision=None, pr_num=None)"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "qbwtN9XvXKft"
      },
      "id": "qbwtN9XvXKft",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.10.13"
    },
    "colab": {
      "provenance": [],
      "gpuType": "A100"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "56fe2e0af9044fe7a1dd1208e8823e84": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "2.0.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "2.0.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_14e32aaa8bde45c2bfd2a221d1c54861",
              "IPY_MODEL_4f8c06fd6a25461baa6b1b8a632fc4df",
              "IPY_MODEL_f95dcb4f140244a39c8c7bad30e0cb1a"
            ],
            "layout": "IPY_MODEL_168523a1c7e243a788a27b36148528b6",
            "tabbable": null,
            "tooltip": null
          }
        },
        "14e32aaa8bde45c2bfd2a221d1c54861": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "2.0.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "2.0.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_allow_html": false,
            "layout": "IPY_MODEL_732c7288b30149da9ce8b42856f8b9e4",
            "placeholder": "​",
            "style": "IPY_MODEL_a6fa2bd3ecaa4f3a88be64ffae37fb3e",
            "tabbable": null,
            "tooltip": null,
            "value": "Loading checkpoint shards: 100%"
          }
        },
        "4f8c06fd6a25461baa6b1b8a632fc4df": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "2.0.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "2.0.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_allow_html": false,
            "layout": "IPY_MODEL_33d71ae97084435dad2861ad896a57a7",
            "max": 4,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_82680331a8234062978d7cc62760599f",
            "tabbable": null,
            "tooltip": null,
            "value": 4
          }
        },
        "f95dcb4f140244a39c8c7bad30e0cb1a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "2.0.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "2.0.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_allow_html": false,
            "layout": "IPY_MODEL_8663bd74a3b840369eb1082650997601",
            "placeholder": "​",
            "style": "IPY_MODEL_301caf0f897047408ff67b8b1a6a6991",
            "tabbable": null,
            "tooltip": null,
            "value": " 4/4 [00:06&lt;00:00,  1.40s/it]"
          }
        },
        "168523a1c7e243a788a27b36148528b6": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "2.0.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border_bottom": null,
            "border_left": null,
            "border_right": null,
            "border_top": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "732c7288b30149da9ce8b42856f8b9e4": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "2.0.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border_bottom": null,
            "border_left": null,
            "border_right": null,
            "border_top": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a6fa2bd3ecaa4f3a88be64ffae37fb3e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLStyleModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "HTMLStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "StyleView",
            "background": null,
            "description_width": "",
            "font_size": null,
            "text_color": null
          }
        },
        "33d71ae97084435dad2861ad896a57a7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "2.0.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border_bottom": null,
            "border_left": null,
            "border_right": null,
            "border_top": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "82680331a8234062978d7cc62760599f": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "8663bd74a3b840369eb1082650997601": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "2.0.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border_bottom": null,
            "border_left": null,
            "border_right": null,
            "border_top": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "301caf0f897047408ff67b8b1a6a6991": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLStyleModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "HTMLStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "StyleView",
            "background": null,
            "description_width": "",
            "font_size": null,
            "text_color": null
          }
        },
        "cdaadb2d4ee442f08f516f09db9e179b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "2.0.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "2.0.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_2143cf4a5085498b90fc26d6ee3233b1",
              "IPY_MODEL_73215571e490481aae8448c0908a161d",
              "IPY_MODEL_0a0296ee55e04de98b2bc986755bd1bc"
            ],
            "layout": "IPY_MODEL_337cf05e9c66417393a92f80862d0622",
            "tabbable": null,
            "tooltip": null
          }
        },
        "2143cf4a5085498b90fc26d6ee3233b1": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "2.0.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "2.0.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_allow_html": false,
            "layout": "IPY_MODEL_c44c30be11be4c21ae4b23c23704f1d7",
            "placeholder": "​",
            "style": "IPY_MODEL_d3af487a46904aac8e0f6e5b6e173795",
            "tabbable": null,
            "tooltip": null,
            "value": "adapter_model.safetensors: 100%"
          }
        },
        "73215571e490481aae8448c0908a161d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "2.0.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "2.0.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_allow_html": false,
            "layout": "IPY_MODEL_9d76c6f2ed63448a8954afd837b96dcb",
            "max": 65832,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_9327f5a00a0e4095b67f21582ad12766",
            "tabbable": null,
            "tooltip": null,
            "value": 65832
          }
        },
        "0a0296ee55e04de98b2bc986755bd1bc": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "2.0.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "2.0.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_allow_html": false,
            "layout": "IPY_MODEL_4fd8ca98f726465887bf5b460bf4457f",
            "placeholder": "​",
            "style": "IPY_MODEL_0c32972b51f447e7b9fa30fc9c7314e8",
            "tabbable": null,
            "tooltip": null,
            "value": " 65.8k/65.8k [00:00&lt;00:00, 98.4kB/s]"
          }
        },
        "337cf05e9c66417393a92f80862d0622": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "2.0.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border_bottom": null,
            "border_left": null,
            "border_right": null,
            "border_top": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "c44c30be11be4c21ae4b23c23704f1d7": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "2.0.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border_bottom": null,
            "border_left": null,
            "border_right": null,
            "border_top": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "d3af487a46904aac8e0f6e5b6e173795": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLStyleModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "HTMLStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "StyleView",
            "background": null,
            "description_width": "",
            "font_size": null,
            "text_color": null
          }
        },
        "9d76c6f2ed63448a8954afd837b96dcb": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "2.0.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border_bottom": null,
            "border_left": null,
            "border_right": null,
            "border_top": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9327f5a00a0e4095b67f21582ad12766": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "4fd8ca98f726465887bf5b460bf4457f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "2.0.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border_bottom": null,
            "border_left": null,
            "border_right": null,
            "border_top": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0c32972b51f447e7b9fa30fc9c7314e8": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLStyleModel",
          "model_module_version": "2.0.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "2.0.0",
            "_model_name": "HTMLStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "2.0.0",
            "_view_name": "StyleView",
            "background": null,
            "description_width": "",
            "font_size": null,
            "text_color": null
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}