diff --git "a/Case_Study_II_Salomé_Cohen).ipynb" "b/Case_Study_II_Salomé_Cohen).ipynb"
new file mode 100644--- /dev/null
+++ "b/Case_Study_II_Salomé_Cohen).ipynb"
@@ -0,0 +1,7584 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "ZuVfUTtPb_Ut"
+      },
+      "source": [
+        "# **Case Study II: What Could Some Future Jobs with AI Be?**"
+      ],
+      "id": "ZuVfUTtPb_Ut"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "jhLITaPkb_Uu"
+      },
+      "source": [
+        "This case study is inspired by the guest speaker session with **Derya Karli**, founder & CEO of [Sirius Quantum Solutions](https://siriusquantum.com/), a quantum AI startup building an open ecosystem for quantum data annotation with 100+ domain experts.\n",
+        "\n",
+        "> 🎥 *If you missed the live session, watch the recording when it is uploaded to Blackboard*\n",
+        "\n",
+        "**You will play three roles:**\n",
+        "1. **Data Annotation Worker** — label customer reviews by hand\n",
+        "2. **AI Analyst** — use MiniLM to analyze text similarity\n",
+        "3. **Custom GPT Builder** — each member builds a custom GPT in ChatGPT"
+      ],
+      "id": "jhLITaPkb_Uu"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "pcewF2Crb_Uu"
+      },
+      "source": [
+        "### *Group Information*\n",
+        "\n",
+        "*Double-click this cell and fill in your group details.*\n",
+        "\n",
+        "**Group Number:** [H]\n",
+        "\n",
+        "**Members (5 required):**\n",
+        "1. Alice Stenuit — GPT Role: A\n",
+        "\n",
+        "2. Alessandro TOGNETTI— GPT Role: B\n",
+        "\n",
+        "3. Salomé COHEN— GPT Role: C\n",
+        "\n",
+        "4. Fares Mohamed BOUARBI BARBARA — GPT Role: D\n",
+        "\n",
+        "5. Ange Tapiau — GPT Role: E\n"
+      ],
+      "id": "pcewF2Crb_Uu"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Owei459Ib_Uv"
+      },
+      "source": [
+        "## **1.** 📦 Install & import required packages"
+      ],
+      "id": "Owei459Ib_Uv"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "ruVb8eXEb_Uv"
+      },
+      "source": [
+        "### *a. Install and import*"
+      ],
+      "id": "ruVb8eXEb_Uv"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "uhj2ewnFb_Uv",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "b8b7b171-ffc2-46a0-da62-292137bf57d2"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "✅ All libraries installed and imported!\n"
+          ]
+        }
+      ],
+      "source": [
+        "!pip install -q pandas matplotlib seaborn scikit-learn sentence-transformers\n",
+        "\n",
+        "import pandas as pd\n",
+        "import numpy as np\n",
+        "import matplotlib.pyplot as plt\n",
+        "import seaborn as sns\n",
+        "from sentence_transformers import SentenceTransformer\n",
+        "from sklearn.metrics.pairwise import cosine_similarity\n",
+        "from collections import Counter\n",
+        "\n",
+        "sns.set_style(\"whitegrid\")\n",
+        "print(\"✅ All libraries installed and imported!\")"
+      ],
+      "id": "uhj2ewnFb_Uv"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "3ONZdK58b_Uw"
+      },
+      "source": [
+        "## **2.** ✅ Load & inspect datasets"
+      ],
+      "id": "3ONZdK58b_Uw"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "rEo24ioSb_Uw"
+      },
+      "source": [
+        "### *a. ✋🏻🛑⛔️ Upload all 5 CSV files and load them into dataframes*\n",
+        "\n",
+        "Upload: `reviews_dataset.csv`, `annotation_guidelines.csv`, `worker_performance_detailed.csv`, `competitor_pricing.csv`, `ai_replacement_timeline.csv`"
+      ],
+      "id": "rEo24ioSb_Uw"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "Q3TWxYAbb_Uw",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "1bc1b1a3-5fe3-48b3-b583-bfa81bfd1af9"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "✅ Fichiers déjà disponibles dans l'environnement Colab\n"
+          ]
+        }
+      ],
+      "source": [
+        "import io\n",
+        "print(\"✅ Fichiers déjà disponibles dans l'environnement Colab\")"
+      ],
+      "id": "Q3TWxYAbb_Uw"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "CRgEv6Bxb_Uw"
+      },
+      "source": [
+        "### *b. ✋🏻🛑⛔️ Create dataframes reviews_df, guidelines_df, workers_df, competitors_df, and timeline_df from each CSV file*"
+      ],
+      "id": "CRgEv6Bxb_Uw"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "FK1tEgWIb_Uw",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 367
+        },
+        "outputId": "9cfe7cda-bc4a-4805-d8b7-db5b57855c7c"
+      },
+      "outputs": [
+        {
+          "output_type": "error",
+          "ename": "FileNotFoundError",
+          "evalue": "[Errno 2] No such file or directory: 'reviews_dataset.csv'",
+          "traceback": [
+            "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+            "\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)",
+            "\u001b[0;32m/tmp/ipykernel_1383/2066144445.py\u001b[0m in \u001b[0;36m<cell line: 0>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mreviews_df\u001b[0m     \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'reviews_dataset.csv'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m      2\u001b[0m \u001b[0mguidelines_df\u001b[0m  \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'annotation_guidelines.csv'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      3\u001b[0m \u001b[0mworkers_df\u001b[0m     \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'worker_performance_detailed.csv'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      4\u001b[0m \u001b[0mcompetitors_df\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'competitor_pricing.csv'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mtimeline_df\u001b[0m    \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'ai_replacement_timeline.csv'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+            "\u001b[0;32m/usr/local/lib/python3.12/dist-packages/pandas/io/parsers/readers.py\u001b[0m in \u001b[0;36mread_csv\u001b[0;34m(filepath_or_buffer, sep, delimiter, header, names, index_col, usecols, dtype, engine, converters, true_values, false_values, skipinitialspace, skiprows, skipfooter, nrows, na_values, keep_default_na, na_filter, verbose, skip_blank_lines, parse_dates, infer_datetime_format, keep_date_col, date_parser, date_format, dayfirst, cache_dates, iterator, chunksize, compression, thousands, decimal, lineterminator, quotechar, quoting, doublequote, escapechar, comment, encoding, encoding_errors, dialect, on_bad_lines, delim_whitespace, low_memory, memory_map, float_precision, storage_options, dtype_backend)\u001b[0m\n\u001b[1;32m   1024\u001b[0m     \u001b[0mkwds\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mupdate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mkwds_defaults\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1025\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1026\u001b[0;31m     \u001b[0;32mreturn\u001b[0m \u001b[0m_read\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath_or_buffer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1027\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1028\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
+            "\u001b[0;32m/usr/local/lib/python3.12/dist-packages/pandas/io/parsers/readers.py\u001b[0m in \u001b[0;36m_read\u001b[0;34m(filepath_or_buffer, kwds)\u001b[0m\n\u001b[1;32m    618\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    619\u001b[0m     \u001b[0;31m# Create the parser.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 620\u001b[0;31m     \u001b[0mparser\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mTextFileReader\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath_or_buffer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    621\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    622\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mchunksize\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0miterator\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+            "\u001b[0;32m/usr/local/lib/python3.12/dist-packages/pandas/io/parsers/readers.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, f, engine, **kwds)\u001b[0m\n\u001b[1;32m   1618\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1619\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mhandles\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0mIOHandles\u001b[0m \u001b[0;34m|\u001b[0m \u001b[0;32mNone\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1620\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_engine\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_make_engine\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mengine\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1621\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1622\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mclose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m->\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+            "\u001b[0;32m/usr/local/lib/python3.12/dist-packages/pandas/io/parsers/readers.py\u001b[0m in \u001b[0;36m_make_engine\u001b[0;34m(self, f, engine)\u001b[0m\n\u001b[1;32m   1878\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0;34m\"b\"\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mmode\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1879\u001b[0m                     \u001b[0mmode\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;34m\"b\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1880\u001b[0;31m             self.handles = get_handle(\n\u001b[0m\u001b[1;32m   1881\u001b[0m                 \u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1882\u001b[0m                 \u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+            "\u001b[0;32m/usr/local/lib/python3.12/dist-packages/pandas/io/common.py\u001b[0m in \u001b[0;36mget_handle\u001b[0;34m(path_or_buf, mode, encoding, compression, memory_map, is_text, errors, storage_options)\u001b[0m\n\u001b[1;32m    871\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mioargs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mencoding\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0;34m\"b\"\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mioargs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    872\u001b[0m             \u001b[0;31m# Encoding\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 873\u001b[0;31m             handle = open(\n\u001b[0m\u001b[1;32m    874\u001b[0m                 \u001b[0mhandle\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    875\u001b[0m                 \u001b[0mioargs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
+            "\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: 'reviews_dataset.csv'"
+          ]
+        }
+      ],
+      "source": [
+        "reviews_df     = pd.read_csv('reviews_dataset.csv')\n",
+        "guidelines_df  = pd.read_csv('annotation_guidelines.csv')\n",
+        "workers_df     = pd.read_csv('worker_performance_detailed.csv')\n",
+        "competitors_df = pd.read_csv('competitor_pricing.csv')\n",
+        "timeline_df    = pd.read_csv('ai_replacement_timeline.csv')\n",
+        "\n",
+        "print(\"✅ 5 dataframes créés !\")\n",
+        "print(f\"   reviews_df : {reviews_df.shape[0]} lignes\")"
+      ],
+      "id": "FK1tEgWIb_Uw"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "jsXX80w9b_Uw"
+      },
+      "source": [
+        "### *c. ✋🏻🛑⛔️ Preview the first few rows of reviews_df, guidelines_df, and workers_df*"
+      ],
+      "id": "jsXX80w9b_Uw"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "id": "1bf47e6a",
+      "metadata": {
+        "id": "1bf47e6a",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 768
+        },
+        "outputId": "512158b6-ea32-4dc9-9879-ff0a8fc84894"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "📋 reviews_df :\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "  review_id                                        review_text  label  tag\n",
+              "0      R043  The laptop stand wobbles so badly my monitor l...    NaN  NaN\n",
+              "1      R042  bought this bc of the tiktok hype. dont believ...    NaN  NaN\n",
+              "2      R092  Functional. Not beautiful, not ugly. Not fast,...    NaN  NaN\n",
+              "3      R010  I've bought a LOT of phone cases over the year...    NaN  NaN\n",
+              "4      R066  Started out amazing but the quality has degrad...    NaN  NaN"
+            ],
+            "text/html": [
+              "\n",
+              "  <div id=\"df-30cea1b9-1342-45cf-9f51-8086c0341177\" 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>review_id</th>\n",
+              "      <th>review_text</th>\n",
+              "      <th>label</th>\n",
+              "      <th>tag</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>R043</td>\n",
+              "      <td>The laptop stand wobbles so badly my monitor l...</td>\n",
+              "      <td>NaN</td>\n",
+              "      <td>NaN</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>R042</td>\n",
+              "      <td>bought this bc of the tiktok hype. dont believ...</td>\n",
+              "      <td>NaN</td>\n",
+              "      <td>NaN</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>R092</td>\n",
+              "      <td>Functional. Not beautiful, not ugly. Not fast,...</td>\n",
+              "      <td>NaN</td>\n",
+              "      <td>NaN</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>R010</td>\n",
+              "      <td>I've bought a LOT of phone cases over the year...</td>\n",
+              "      <td>NaN</td>\n",
+              "      <td>NaN</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>R066</td>\n",
+              "      <td>Started out amazing but the quality has degrad...</td>\n",
+              "      <td>NaN</td>\n",
+              "      <td>NaN</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-30cea1b9-1342-45cf-9f51-8086c0341177')\"\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-30cea1b9-1342-45cf-9f51-8086c0341177 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-30cea1b9-1342-45cf-9f51-8086c0341177');\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>\n",
+              "  </div>\n"
+            ],
+            "application/vnd.google.colaboratory.intrinsic+json": {
+              "type": "dataframe",
+              "summary": "{\n  \"name\": \"display(workers_df\",\n  \"rows\": 5,\n  \"fields\": [\n    {\n      \"column\": \"review_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"R042\",\n          \"R066\",\n          \"R092\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"review_text\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"bought this bc of the tiktok hype. dont believe the influencers, this is trash\",\n          \"Started out amazing but the quality has degraded noticeably after 6 months of daily use. Rubber grips peeling, color fading. Disappointed trajectory.\",\n          \"Functional. Not beautiful, not ugly. Not fast, not slow. Just a portable charger.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"label\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": null,\n        \"min\": null,\n        \"max\": null,\n        \"num_unique_values\": 0,\n        \"samples\": [],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"tag\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": null,\n        \"min\": null,\n        \"max\": null,\n        \"num_unique_values\": 0,\n        \"samples\": [],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "📋 guidelines_df :\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "  rule_id   category                                          guideline  \\\n",
+              "0     G01  Sentiment  If a review mentions BOTH a positive product A...   \n",
+              "1     G02  Sentiment  Sarcasm and rhetorical exaggeration should be ...   \n",
+              "2     G03  Sentiment  Reviews that only state facts without opinion ...   \n",
+              "3     G04  Sentiment  If a review compares favorably to competitors ...   \n",
+              "4     G05  Sentiment  Reviews with conditional satisfaction ('good B...   \n",
+              "\n",
+              "                                             example correct_label  \n",
+              "0  Decent product but won't last very long if you...         mixed  \n",
+              "1  The description list these as large and they a...      negative  \n",
+              "2       These are thick and I haven't broken any yet       neutral  \n",
+              "3           Great value. High quality. Fit perfectly      positive  \n",
+              "4  It looks a little cheap for the price and colo...         mixed  "
+            ],
+            "text/html": [
+              "\n",
+              "  <div id=\"df-7d45197d-3bec-431b-9551-50fffc7c7306\" 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>rule_id</th>\n",
+              "      <th>category</th>\n",
+              "      <th>guideline</th>\n",
+              "      <th>example</th>\n",
+              "      <th>correct_label</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>G01</td>\n",
+              "      <td>Sentiment</td>\n",
+              "      <td>If a review mentions BOTH a positive product A...</td>\n",
+              "      <td>Decent product but won't last very long if you...</td>\n",
+              "      <td>mixed</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>G02</td>\n",
+              "      <td>Sentiment</td>\n",
+              "      <td>Sarcasm and rhetorical exaggeration should be ...</td>\n",
+              "      <td>The description list these as large and they a...</td>\n",
+              "      <td>negative</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>G03</td>\n",
+              "      <td>Sentiment</td>\n",
+              "      <td>Reviews that only state facts without opinion ...</td>\n",
+              "      <td>These are thick and I haven't broken any yet</td>\n",
+              "      <td>neutral</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>G04</td>\n",
+              "      <td>Sentiment</td>\n",
+              "      <td>If a review compares favorably to competitors ...</td>\n",
+              "      <td>Great value. High quality. Fit perfectly</td>\n",
+              "      <td>positive</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>G05</td>\n",
+              "      <td>Sentiment</td>\n",
+              "      <td>Reviews with conditional satisfaction ('good B...</td>\n",
+              "      <td>It looks a little cheap for the price and colo...</td>\n",
+              "      <td>mixed</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-7d45197d-3bec-431b-9551-50fffc7c7306')\"\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-7d45197d-3bec-431b-9551-50fffc7c7306 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-7d45197d-3bec-431b-9551-50fffc7c7306');\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>\n",
+              "  </div>\n"
+            ],
+            "application/vnd.google.colaboratory.intrinsic+json": {
+              "type": "dataframe",
+              "summary": "{\n  \"name\": \"display(workers_df\",\n  \"rows\": 5,\n  \"fields\": [\n    {\n      \"column\": \"rule_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"G02\",\n          \"G05\",\n          \"G03\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"category\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 1,\n        \"samples\": [\n          \"Sentiment\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"guideline\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"Sarcasm and rhetorical exaggeration should be labeled by the INTENDED meaning, not the literal words.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"example\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"The description list these as large and they are small and cheap\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"correct_label\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 4,\n        \"samples\": [\n          \"negative\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "📋 workers_df :\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "  worker_id       name    background            domain  months_active  \\\n",
+              "0       W01     Ana L.   PhD Physics  Quantum Circuits             18   \n",
+              "1       W02   Kenji M.        MSc CS          NLP Text             12   \n",
+              "2       W03   Priya R.   BSc Biology    Medical Images              8   \n",
+              "3       W04  Carlos D.   Self-taught         Sentiment             14   \n",
+              "4       W05   Fatma K.  MSc Data Sci       Multi-modal             10   \n",
+              "\n",
+              "   total_tasks  accuracy_pct  avg_time_min  cost_per_task  \\\n",
+              "0         2556          94.2           4.2            1.3   \n",
+              "1         3468          87.6           2.1            0.8   \n",
+              "2          784          91.0           6.8            2.1   \n",
+              "3         5740          78.3           1.3            0.5   \n",
+              "4         2150          89.7           3.5            1.1   \n",
+              "\n",
+              "  error_type_most_common quality_trend  \n",
+              "0         Boundary cases        Stable  \n",
+              "1        Context misread     Improving  \n",
+              "2        Rare conditions        Stable  \n",
+              "3          Sarcasm/irony     Declining  \n",
+              "4   Cross-modal conflict     Improving  "
+            ],
+            "text/html": [
+              "\n",
+              "  <div id=\"df-4c8c2bf7-1ce2-469f-941b-0b55b59d1dbf\" 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>worker_id</th>\n",
+              "      <th>name</th>\n",
+              "      <th>background</th>\n",
+              "      <th>domain</th>\n",
+              "      <th>months_active</th>\n",
+              "      <th>total_tasks</th>\n",
+              "      <th>accuracy_pct</th>\n",
+              "      <th>avg_time_min</th>\n",
+              "      <th>cost_per_task</th>\n",
+              "      <th>error_type_most_common</th>\n",
+              "      <th>quality_trend</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>W01</td>\n",
+              "      <td>Ana L.</td>\n",
+              "      <td>PhD Physics</td>\n",
+              "      <td>Quantum Circuits</td>\n",
+              "      <td>18</td>\n",
+              "      <td>2556</td>\n",
+              "      <td>94.2</td>\n",
+              "      <td>4.2</td>\n",
+              "      <td>1.3</td>\n",
+              "      <td>Boundary cases</td>\n",
+              "      <td>Stable</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>W02</td>\n",
+              "      <td>Kenji M.</td>\n",
+              "      <td>MSc CS</td>\n",
+              "      <td>NLP Text</td>\n",
+              "      <td>12</td>\n",
+              "      <td>3468</td>\n",
+              "      <td>87.6</td>\n",
+              "      <td>2.1</td>\n",
+              "      <td>0.8</td>\n",
+              "      <td>Context misread</td>\n",
+              "      <td>Improving</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>W03</td>\n",
+              "      <td>Priya R.</td>\n",
+              "      <td>BSc Biology</td>\n",
+              "      <td>Medical Images</td>\n",
+              "      <td>8</td>\n",
+              "      <td>784</td>\n",
+              "      <td>91.0</td>\n",
+              "      <td>6.8</td>\n",
+              "      <td>2.1</td>\n",
+              "      <td>Rare conditions</td>\n",
+              "      <td>Stable</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>W04</td>\n",
+              "      <td>Carlos D.</td>\n",
+              "      <td>Self-taught</td>\n",
+              "      <td>Sentiment</td>\n",
+              "      <td>14</td>\n",
+              "      <td>5740</td>\n",
+              "      <td>78.3</td>\n",
+              "      <td>1.3</td>\n",
+              "      <td>0.5</td>\n",
+              "      <td>Sarcasm/irony</td>\n",
+              "      <td>Declining</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>W05</td>\n",
+              "      <td>Fatma K.</td>\n",
+              "      <td>MSc Data Sci</td>\n",
+              "      <td>Multi-modal</td>\n",
+              "      <td>10</td>\n",
+              "      <td>2150</td>\n",
+              "      <td>89.7</td>\n",
+              "      <td>3.5</td>\n",
+              "      <td>1.1</td>\n",
+              "      <td>Cross-modal conflict</td>\n",
+              "      <td>Improving</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-4c8c2bf7-1ce2-469f-941b-0b55b59d1dbf')\"\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-4c8c2bf7-1ce2-469f-941b-0b55b59d1dbf 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-4c8c2bf7-1ce2-469f-941b-0b55b59d1dbf');\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>\n",
+              "  </div>\n"
+            ],
+            "application/vnd.google.colaboratory.intrinsic+json": {
+              "type": "dataframe",
+              "summary": "{\n  \"name\": \"display(workers_df\",\n  \"rows\": 5,\n  \"fields\": [\n    {\n      \"column\": \"worker_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"W02\",\n          \"W05\",\n          \"W03\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"name\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"Kenji M.\",\n          \"Fatma K.\",\n          \"Priya R.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"background\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"MSc CS\",\n          \"MSc Data Sci\",\n          \"BSc Biology\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"domain\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"NLP Text\",\n          \"Multi-modal\",\n          \"Medical Images\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"months_active\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 3,\n        \"min\": 8,\n        \"max\": 18,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          12,\n          10,\n          8\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"total_tasks\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 1839,\n        \"min\": 784,\n        \"max\": 5740,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          3468,\n          2150,\n          784\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"accuracy_pct\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 6.009409288773734,\n        \"min\": 78.3,\n        \"max\": 94.2,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          87.6,\n          89.7,\n          91.0\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"avg_time_min\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 2.1300234740490533,\n        \"min\": 1.3,\n        \"max\": 6.8,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          2.1,\n          3.5,\n          6.8\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"cost_per_task\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.6066300355241241,\n        \"min\": 0.5,\n        \"max\": 2.1,\n        \"num_unique_values\": 5,\n        \"samples\": [\n          0.8,\n          1.1,\n          2.1\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"error_type_most_common\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 5,\n        \"samples\": [\n          \"Context misread\",\n          \"Cross-modal conflict\",\n          \"Rare conditions\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"quality_trend\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Stable\",\n          \"Improving\",\n          \"Declining\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
+            }
+          },
+          "metadata": {}
+        }
+      ],
+      "source": [
+        "print(\"📋 reviews_df :\")\n",
+        "display(reviews_df.head())\n",
+        "\n",
+        "print(\"\\n📋 guidelines_df :\")\n",
+        "display(guidelines_df.head())\n",
+        "\n",
+        "print(\"\\n📋 workers_df :\")\n",
+        "display(workers_df.head())"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "_Cw5rLJyb_Ux"
+      },
+      "source": [
+        "## **3.** 🏷️ Be the Annotator (Exercise 1) — 3 points\n",
+        "\n",
+        "**Every member** labels 10 reviews independently. This is what real annotation workers do at Scale AI, Sama, and Appen. When annotators disagree, it reveals ambiguity, exactly where AI models struggle most.\n",
+        "\n",
+        "Labels: `\"positive\"`, `\"negative\"`, `\"neutral\"`, or `\"mixed\"`"
+      ],
+      "id": "_Cw5rLJyb_Ux"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "E5uev4zgb_Ux"
+      },
+      "source": [
+        "### *a. Display your group's 10 reviews*"
+      ],
+      "id": "E5uev4zgb_Ux"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "FQs2bBBUb_Ux",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "e691679b-faa5-4ef9-ad3c-7c7203d68ad6"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "--- Review 1 | ID: R093 ---\n",
+            "    second time buying this brand. first one lasted 2 years which is about right for the price. no strong feelings\n",
+            "\n",
+            "--- Review 2 | ID: R097 ---\n",
+            "    bought it, used it, forgot to review it for 6 months. that tells you how memorable this product is\n",
+            "\n",
+            "--- Review 3 | ID: R042 ---\n",
+            "    bought this bc of the tiktok hype. dont believe the influencers, this is trash\n",
+            "\n",
+            "--- Review 4 | ID: R019 ---\n",
+            "    The yoga mat is thick enough that my knees don't hurt anymore during planks. Finally.\n",
+            "\n",
+            "--- Review 5 | ID: R039 ---\n",
+            "    shipping took 3 weeks, arrived in a crushed box, product was scratched. 0/10 experience from start to finish\n",
+            "\n",
+            "--- Review 6 | ID: R082 ---\n",
+            "    its a water bottle. it holds water. not much else to say\n",
+            "\n",
+            "--- Review 7 | ID: R055 ---\n",
+            "    Half the buttons on the keyboard dont even register properly. Tested on 2 computers same problem. Defective\n",
+            "\n",
+            "--- Review 8 | ID: R034 ---\n",
+            "    worst purchase of 2024 easily. the bluetooth speaker stopped connecting after a week and customer service just keeps sending me the same troubleshooting email over and over\n",
+            "\n",
+            "--- Review 9 | ID: R007 ---\n",
+            "    perfect gift for my husband. he hasnt stopped using this air fryer since christmas\n",
+            "\n",
+            "--- Review 10 | ID: R064 ---\n",
+            "    4 stars for the product, 1 star for the company. Net: somewhere in the middle. Would I buy again? Maybe if it was on sale\n"
+          ]
+        }
+      ],
+      "source": [
+        "GROUP_NUMBER = 8\n",
+        "sample = reviews_df.sample(n=10, random_state=8).reset_index(drop=True)\n",
+        "for i, row in sample.iterrows():\n",
+        "    print(f\"\\n--- Review {i+1} | ID: {row['review_id']} ---\")\n",
+        "    print(f\"    {row['review_text']}\")"
+      ],
+      "id": "FQs2bBBUb_Ux"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "NN_hqikAb_Ux"
+      },
+      "source": [
+        "### *b. ✋🏻🛑⛔️ Set your group size and enter each member's name and labels*\n",
+        "\n",
+        "Every member annotates independently — do not look at each other's answers!"
+      ],
+      "id": "NN_hqikAb_Ux"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "umGHL_K8b_Ux"
+      },
+      "outputs": [],
+      "source": [
+        "GROUP_SIZE = 5  # ← CHANGE THIS to 5 or 6\n",
+        "\n",
+        "# ═══ MEMBER 1 ═══\n",
+        "member_1_name = \"Alice\"\n",
+        "member_1_labels = [\"positive\",\"mixed\", \"negative\", \"positive\", \"positive\", \"neutral\", \"positive\", \"positive\", \"neutral\", \"negative\",]\n",
+        "\n",
+        "# ═══ MEMBER 2 ═══\n",
+        "member_2_name = \"Alessandro\"\n",
+        "member_2_labels = [\"positive\", \"mixed\",\"positive\",\"negative\",\"mixed\",\"neutral\", \"positive\",\"positive\",\"negative\", \"negative\"]\n",
+        "\n",
+        "# ═══ MEMBER 3 ═══\n",
+        "member_3_name = \"Salome\"\n",
+        "member_3_labels = [\"neutral\",\"neutral\", \"negative\", \"positive\",\"negative\",\"neutral\",\"negative\",\"negative\",\"positive\",\"mixed\"]\n",
+        "\n",
+        "# ═══ MEMBER 4 ═══\n",
+        "member_4_name = \"Fares\"\n",
+        "member_4_labels = [\"positive\",\"neutral\",\"negative\",\"positive\",\"positive\",\"mixed\",\"positive\",\"negative\",\"positive\",\"negative\"]\n",
+        "\n",
+        "# ═══ MEMBER 5 ═══\n",
+        "member_5_name = \"Ange\"\n",
+        "member_5_labels = [\"positive\", \"negative\" \"positive\", \"positive\",\"negative\",\"mixed\", \"mixed\", \"positive\", \"negative\", \"mixed\"]\n"
+      ],
+      "id": "umGHL_K8b_Ux"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "JYknv76cb_Ux"
+      },
+      "source": [
+        "### *c. Compare all annotations*"
+      ],
+      "id": "JYknv76cb_Ux"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "7Kx8u7Hab_Ux",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "198e4ac2-28a0-4e7a-9f87-5f7a4f22bf55"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Review 1\n",
+            "Labels: ['positive', 'positive', 'neutral', 'positive', 'positive']\n",
+            "Counts: {'positive': 4, 'neutral': 1}\n",
+            "Majority label: positive\n",
+            "\n",
+            "Review 2\n",
+            "Labels: ['negative', 'mixed', 'neutral', 'neutral', 'mixed']\n",
+            "Counts: {'negative': 1, 'mixed': 2, 'neutral': 2}\n",
+            "Majority label: mixed\n",
+            "\n",
+            "Review 3\n",
+            "Labels: ['positive', 'positive', 'negative', 'negative', 'negative']\n",
+            "Counts: {'positive': 2, 'negative': 3}\n",
+            "Majority label: negative\n",
+            "\n",
+            "Review 4\n",
+            "Labels: ['positive', 'negative', 'positive', 'positive', 'positive']\n",
+            "Counts: {'positive': 4, 'negative': 1}\n",
+            "Majority label: positive\n",
+            "\n",
+            "Review 5\n",
+            "Labels: ['negative', 'mixed', 'negative', 'positive', 'positive']\n",
+            "Counts: {'negative': 2, 'mixed': 1, 'positive': 2}\n",
+            "Majority label: negative\n",
+            "\n",
+            "Review 6\n",
+            "Labels: ['mixed', 'neutral', 'neutral', 'mixed', 'neutral']\n",
+            "Counts: {'mixed': 2, 'neutral': 3}\n",
+            "Majority label: neutral\n",
+            "\n",
+            "Review 7\n",
+            "Labels: ['mixed', 'positive', 'negative', 'positive', 'positive']\n",
+            "Counts: {'mixed': 1, 'positive': 3, 'negative': 1}\n",
+            "Majority label: positive\n",
+            "\n",
+            "Review 8\n",
+            "Labels: ['positive', 'positive', 'negative', 'negative', 'positive']\n",
+            "Counts: {'positive': 3, 'negative': 2}\n",
+            "Majority label: positive\n",
+            "\n",
+            "Review 9\n",
+            "Labels: ['negative', 'negative', 'positive', 'positive', 'neutral']\n",
+            "Counts: {'negative': 2, 'positive': 2, 'neutral': 1}\n",
+            "Majority label: negative\n",
+            "\n",
+            "Review 10\n",
+            "Labels: ['mixed', 'negative', 'mixed', 'negative', 'negative']\n",
+            "Counts: {'mixed': 2, 'negative': 3}\n",
+            "Majority label: negative\n",
+            "\n"
+          ]
+        }
+      ],
+      "source": [
+        "from collections import Counter\n",
+        "\n",
+        "# All annotations\n",
+        "annotations = [\n",
+        "    member_1_labels,\n",
+        "    member_2_labels,\n",
+        "    member_3_labels,\n",
+        "    member_4_labels,\n",
+        "    member_5_labels\n",
+        "]\n",
+        "\n",
+        "num_reviews = len(member_1_labels)\n",
+        "\n",
+        "for i in range(num_reviews):\n",
+        "    labels = [member[i] for member in annotations]\n",
+        "    counts = Counter(labels)\n",
+        "    majority = counts.most_common(1)[0][0]\n",
+        "\n",
+        "    print(f\"Review {i+1}\")\n",
+        "    print(\"Labels:\", labels)\n",
+        "    print(\"Counts:\", dict(counts))\n",
+        "    print(\"Majority label:\", majority)\n",
+        "    print()"
+      ],
+      "id": "7Kx8u7Hab_Ux"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "yFWECyJib_Ux"
+      },
+      "source": [
+        "### *d. Display the industry annotation guidelines*"
+      ],
+      "id": "yFWECyJib_Ux"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "-SgJTwVfb_Ux",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "39926ecd-54f4-4ee7-f2cb-7e8bc4b26eff"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "📋 ANNOTATION GUIDELINES\n",
+            "================================================================================\n",
+            "\n",
+            "R1 [Positive]\n",
+            "  Rule: Use 'positive' when the review expresses clear praise or satisfaction.\n",
+            "  Example: \"Great product, I love it!\" → positive\n",
+            "\n",
+            "R2 [Negative]\n",
+            "  Rule: Use 'negative' when the review expresses complaints or dissatisfaction.\n",
+            "  Example: \"This product is terrible.\" → negative\n",
+            "\n",
+            "R3 [Neutral]\n",
+            "  Rule: Use 'neutral' when the review states facts without emotional opinion.\n",
+            "  Example: \"The package arrived yesterday.\" → neutral\n",
+            "\n",
+            "R4 [Mixed]\n",
+            "  Rule: Use 'mixed' when the review contains both positive and negative opinions.\n",
+            "  Example: \"Good quality but too expensive.\" → mixed\n"
+          ]
+        }
+      ],
+      "source": [
+        "import pandas as pd\n",
+        "\n",
+        "# Create the guidelines dataset\n",
+        "guidelines_data = [\n",
+        "    {\n",
+        "        \"rule_id\": \"R1\",\n",
+        "        \"category\": \"Positive\",\n",
+        "        \"guideline\": \"Use 'positive' when the review expresses clear praise or satisfaction.\",\n",
+        "        \"example\": \"Great product, I love it!\",\n",
+        "        \"correct_label\": \"positive\"\n",
+        "    },\n",
+        "    {\n",
+        "        \"rule_id\": \"R2\",\n",
+        "        \"category\": \"Negative\",\n",
+        "        \"guideline\": \"Use 'negative' when the review expresses complaints or dissatisfaction.\",\n",
+        "        \"example\": \"This product is terrible.\",\n",
+        "        \"correct_label\": \"negative\"\n",
+        "    },\n",
+        "    {\n",
+        "        \"rule_id\": \"R3\",\n",
+        "        \"category\": \"Neutral\",\n",
+        "        \"guideline\": \"Use 'neutral' when the review states facts without emotional opinion.\",\n",
+        "        \"example\": \"The package arrived yesterday.\",\n",
+        "        \"correct_label\": \"neutral\"\n",
+        "    },\n",
+        "    {\n",
+        "        \"rule_id\": \"R4\",\n",
+        "        \"category\": \"Mixed\",\n",
+        "        \"guideline\": \"Use 'mixed' when the review contains both positive and negative opinions.\",\n",
+        "        \"example\": \"Good quality but too expensive.\",\n",
+        "        \"correct_label\": \"mixed\"\n",
+        "    }\n",
+        "]\n",
+        "\n",
+        "# Convert to DataFrame\n",
+        "guidelines_df = pd.DataFrame(guidelines_data)\n",
+        "\n",
+        "# Display the guidelines\n",
+        "print(\"📋 ANNOTATION GUIDELINES\")\n",
+        "print(\"=\" * 80)\n",
+        "\n",
+        "for _, row in guidelines_df.iterrows():\n",
+        "    print(f\"\\n{row['rule_id']} [{row['category']}]\")\n",
+        "    print(f\"  Rule: {row['guideline']}\")\n",
+        "    print(f\"  Example: \\\"{row['example']}\\\" → {row['correct_label']}\")"
+      ],
+      "id": "-SgJTwVfb_Ux"
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "SECTION COMMENT :\n",
+        "This section was about creating our own GPT, for each members.\n",
+        "Each one of us had a different GPT with a different mission : ANNOTATION QUALITY REVIEWER, AI Companion Designer, Career navigator IA\n",
+        "Platform Strategist GPT,Workforce Automation Analyst GPT.\n",
+        "We typed in specific instructions for it to answer a certain way, we uploaded the knowledge which consisted of the different CSV files.\n",
+        "Therefore we had 5 different GPTs which could answer with different perspectives, and different formats."
+      ],
+      "metadata": {
+        "id": "UUQEf3ucNltt"
+      },
+      "id": "UUQEf3ucNltt",
+      "execution_count": null,
+      "outputs": []
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "YmCSf7vsb_Ux"
+      },
+      "source": [
+        "### ✅ Checkpoint 1 — 3 points\n",
+        "\n",
+        "*Double-click and answer:*\n",
+        "\n",
+        "**1a.** On which reviews did your group disagree most? What words or phrases made them ambiguous? *(1 pt)*\n",
+        "\n",
+        "My group disagreed the most on Review 2 (R097), Review 5 (R039), and Review 9 (R007).\n",
+        "For Review 2, we thought the confusing part was “forgot to review it for 6 months. that tells you how memorable this product is”. This sounds sarcastic, so it was hard to know if it should be negative, neutral, or mixed.\n",
+        "For Review 5, the review says “shipping took 3 weeks, arrived in a crushed box, product was scratched. 0/10 experience from start to finish.” Some of us saw it as fully negative, but others may have hesitated because it talks about both the delivery and the product, it's not always the fault of the company if the product comes out in a bad shape.\n",
+        "For Review 9, the sentence “perfect gift for my husband. he hasnt stopped using this air fryer since christmas” made people disagree because it sounds clearly positive, but it is also kind of just describing what happened, so some people saw it as more neutral. Moreover, the review is about a husband, so we couldn't all succeed to identify.\n",
+        "\n",
+        "---\n",
+        "\n",
+        "**1b.** Look at the pairwise agreement matrix. Did certain members consistently agree with each other? Why? *(1 pt)*\n",
+        "\n",
+        "Yes, some members agreed more than others. I agreed the most with Alessandro and Fares. I think that’s because they looked more at the overall vibe of the review instead of overthinking every word. People usually agreed more on reviews that were super obvious, like really positive or really negative ones. The biggest disagreements happened when the review was sarcastic or a little unclear.\n",
+        "\n",
+        "---\n",
+        "\n",
+        "**1c.** Which guideline from `annotation_guidelines.csv` (above) would have resolved your biggest disagreement? Write 1 additional guideline of your own. *(1 pt)*\n",
+        "\n",
+        "The guideline that would have helped the most is G02, because it says that sarcasm and rhetorical exaggeration should be labeled by the real meaning, not the literal words. That would have helped a lot with Review 2. We think that if a review sounds nice at first but is obviously making fun of the product or being sarcastic, the label should be negative or mixed."
+      ],
+      "id": "YmCSf7vsb_Ux"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Sn9Etpx6b_Uy"
+      },
+      "source": [
+        "## **4.** 🧠 MiniLM & Cosine Similarity (Exercise 2) — 4 points\n",
+        "\n",
+        "MiniLM converts any sentence into a **vector** (a list of 384 numbers). If two sentences mean similar things, their vectors are close together. **Cosine similarity** measures this closeness: 1.0 = identical, 0.0 = unrelated."
+      ],
+      "id": "Sn9Etpx6b_Uy"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "OLttbeswb_Uy"
+      },
+      "source": [
+        "### *a. Load the MiniLM model*"
+      ],
+      "id": "OLttbeswb_Uy"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "kv52_ifWb_Uy",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 693,
+          "referenced_widgets": [
+            "1cc1db50c75649f7b0ced93d97f3e37c",
+            "d6df2f50329c4c96888eef5759b52530",
+            "9aad2aff693f42deab0ae506fb428ff9",
+            "25575c06b16847289915a68ef3262bea",
+            "f3397a560afd4a78a9ee8f6f1a3ab74f",
+            "1bdd47f33ddc456bacca88664e5dbbb6",
+            "d5d5913d3f484354addebc8e8d687d0f",
+            "1a8ccdc20f734f92988048c58e90d209",
+            "6a244b679b43419e9565b66afb8ed629",
+            "faf79610c1ef43008cef0be8da4cea65",
+            "419e3b69fecc4a95b88e9a7a548f8fb1",
+            "84d8e5e12761434e92e912c88b96fabb",
+            "d354075fde764868a321503a15fa94c5",
+            "cf58e65a1b08467ca8d71aa42b563918",
+            "07c547bd892d4edaa98f147d3fce7686",
+            "72721fd4bb7e4beca10d41b526d18b4f",
+            "5a4e69bec4c0483da1db8709ec028470",
+            "9b24b0d864a04181bf06655435106b8c",
+            "852bd86df04b4bee97072e2e9c63baf8",
+            "9eb874c27933468db94e2356e9ba0662",
+            "2fd549d2c315488ead302c743f8da933",
+            "e82edca607d542f3b9c5643875b45b56",
+            "16e2128bff8145628ac81cb0671079c2",
+            "f820281003904478b64ac92853fa6817",
+            "f4f2933f99db4e10957206251ed0dcdf",
+            "66f25761226746cbb9dafa2c970b0726",
+            "2ee256326b114269bb5a37e1a09dc5c9",
+            "c5ad0470b5314cbe8b3bd8f3f9b0abe6",
+            "a00a9affea804da994315360e5345293",
+            "0d6df43ab65a454b91c8e7598c95ff3e",
+            "022b55b95a0f410db7a84aa64f8fa9e6",
+            "9e215fe87976472c9de2eee9c5e837b7",
+            "e55453865ed2494b8a43dbbc7f082bd6",
+            "67314936052b4f45af63c3d969b85cbd",
+            "b465bb52c3004574bb2738d7f4f8c060",
+            "1d1d131fdc114f8ab978f6aff02f544b",
+            "a1650def53304242a12aa20d01e32d73",
+            "892841e82f804e72b96a70746b256580",
+            "35afad1035e5485e818bcc3e0b2246bd",
+            "4b5c55174a374649acc7c9677d644514",
+            "e7e73104d6db448faae9ba7667f03a27",
+            "a1908d51f7654364b87cb77a62a85bb0",
+            "63cfe38a8ac54d88a118bb3dc44094c8",
+            "b491658f6814422282396129e859c2f1",
+            "9f7dbf11128344f2be123b7dca512a5f",
+            "54d05885283c4887a7715206b2a5c763",
+            "68e7791b2af74ac0bde83768d27c4f38",
+            "f420bc6ad4244b008e03fc4486fdb7d2",
+            "776a2d7a839447f79e180b4a6291f177",
+            "0af009446e0f45488cd8ba2d5eb103b0",
+            "ae5cbff3686344c6906099ed2b6f98a9",
+            "f568dd11f7a347598299083a74eff064",
+            "0e18cbf8b03840fdae402b88e283b0e9",
+            "0b5135b7ca1b416db6679df7a7207cce",
+            "06a7376c429d46769420e546149c428e",
+            "68fc9bb13ed54338bec33b20cd73df59",
+            "8524a46a0467482f89e4f340d7fbebed",
+            "c467010cde554bc697864588c0439294",
+            "1e0cdc611546414a936b52948d0ff4f5",
+            "d02f493a6def45b99745387b8b32a77b",
+            "c86151d5a69e4e2b8d8d10d26fe5ee8b",
+            "b3c355a20d4a4997a18ac0b9978cc54e",
+            "7bb9719ab8054b77b0c1ad8fe0bd74e3",
+            "eaa987bc8b04432a84195504a90e7852",
+            "bc3b4482d673427f9e242bb4c14b94de",
+            "04b28ec852844cb8888bf403e964b605",
+            "b6c877d4fcda44d4b45d3add81a633f4",
+            "9743fefcd4ac43c6bdafc8a8bb50ec86",
+            "4e93bc2d06124bf9b896ae879777ef1e",
+            "83da641e6a9f4cb895ddb1795525a9e5",
+            "8df7510d2a0c418aab05581b123a851f",
+            "1972bf6bae5747b88bf3d66ec87aa966",
+            "2989bc6a2bcb4071a37a552fd28c4af4",
+            "610ebd6ae3c042a992baf596fa5d80a4",
+            "acccc4e7d1944150b505d4862b93fb77",
+            "ca0a5555d6234314a0ac7fbff7cd0873",
+            "ce66757f685f48c7a9db8a11f55a60cc",
+            "cf4ba8afab434f869a4a533913be8189",
+            "145e8e9fbffb43b8b0281604b5e62064",
+            "1a060e0c295c4d29b24f55de298dae01",
+            "27e99785224b4b448c6b0ff30be815fd",
+            "bd97516347c141ecbd6eea7307b8ef6d",
+            "5786780980fe463cbd06bb64dcc0d0bc",
+            "babe357164f8449bac3cdfaee3c98541",
+            "b20942603f744f59bf86ce0d5f8a1cc5",
+            "0404c8e259974005b56c5995f7cc1bfb",
+            "a02f9cda29dc49b19eae8b00f85e3c4d",
+            "2734046898b04a9f8dd45b584b5a6c11",
+            "a3e8c5f7a15e492aa59b1459963da66d",
+            "a6869838ad6b46cb9800021b4a43f616",
+            "9791194c93f6436eb6b33cc48647e7ec",
+            "c41fdf9ed74b41d08c4ef6c56c1d6a5a",
+            "e1e497e798524c9cb66c83c1813baed4",
+            "438b6b76332d4b70a43666373493b4ab",
+            "d1ff9f9c30ca4df2b6837b952c6238ac",
+            "acd318bacf0b4f578cb2d060846ffe46",
+            "f16a5c763cdc4170b51056efe68ab9c2",
+            "e61da32034e140dea96472fe9009112e",
+            "b5ca5bb022004a29bf7acb2f8514540d",
+            "3712dba5d01247298810249a327b27b4",
+            "ad8351537ec44a9c954f296839ad43ff",
+            "317978aa6f3b47cbbb2125a73b2bd11e",
+            "59e8ccedfd9c4aef8bc1beb361b3b67b",
+            "2ababd05c70f45cd8ba35f0714764b4f",
+            "fdde2ca71ad44daea2f3b963f95b1882",
+            "9a4af32cdbad41d5a6f3506323646ad5",
+            "bc617d06504d44718094911a1b02c606",
+            "85f11a7b7a2c48438f60ae62fbec77b2",
+            "4829fd0c03e44ab192789debcc04ca23",
+            "4451ba9d6a78497697ee28308e2af75f",
+            "2991c94441984658b3f24ca4be3bafc3",
+            "c8d18487ba984ffeb1b0f3b88dcafd83",
+            "2a10b85054b7487c8b612f1575b81439",
+            "1663e299d8b048e593e77a747e24d2aa",
+            "952eb895b6e842d7a5f4cafd0f99a113",
+            "394b581443144907aab8252375a5215a",
+            "bc9d41acbf7d4964bf9cee55c734b6ab",
+            "7eed38abc58a402493930b101d0ba967",
+            "c7988a7b1b2a4de99acabb10ba4b327b",
+            "ebd71a7b681d48a8b5e8979081c24ea8",
+            "a533b4fc5a2e496e8963314cf3322290",
+            "18774a626601487a8ff0bfc87cffa014",
+            "adaaf806513f4e6aa1c52c675e0885d1",
+            "1442a0f8db40493199fd3fffb1e79826",
+            "d7384e36485d4d0cae1b52ef0503730f",
+            "f8c4273fd71f4142a5921f2c72572f52",
+            "74bc5a2d01874f19836d89a26fe12518",
+            "02935cb32d67492bb4651b051dec5d6b",
+            "7babe961f29243af841cba0d3f55fab1",
+            "3ede47b700b541579c273da30ba18b88",
+            "ad3d3465c58e46e6959af1f4a97881c8",
+            "e9dcf02de017417cb8128dc8bff98d1d"
+          ]
+        },
+        "outputId": "933898d0-c2ea-48db-94da-3a20c2d97ca5"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "/usr/local/lib/python3.12/dist-packages/huggingface_hub/utils/_auth.py:94: UserWarning: \n",
+            "The secret `HF_TOKEN` does not exist in your Colab secrets.\n",
+            "To authenticate with the Hugging Face Hub, create a token in your settings tab (https://huggingface.co/settings/tokens), set it as secret in your Google Colab and restart your session.\n",
+            "You will be able to reuse this secret in all of your notebooks.\n",
+            "Please note that authentication is recommended but still optional to access public models or datasets.\n",
+            "  warnings.warn(\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "modules.json:   0%|          | 0.00/349 [00:00<?, ?B/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "1cc1db50c75649f7b0ced93d97f3e37c"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.\n",
+            "WARNING:huggingface_hub.utils._http:Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "config_sentence_transformers.json:   0%|          | 0.00/116 [00:00<?, ?B/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "84d8e5e12761434e92e912c88b96fabb"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "README.md: 0.00B [00:00, ?B/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "16e2128bff8145628ac81cb0671079c2"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "sentence_bert_config.json:   0%|          | 0.00/53.0 [00:00<?, ?B/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "67314936052b4f45af63c3d969b85cbd"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "config.json:   0%|          | 0.00/612 [00:00<?, ?B/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "9f7dbf11128344f2be123b7dca512a5f"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "model.safetensors:   0%|          | 0.00/90.9M [00:00<?, ?B/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "68fc9bb13ed54338bec33b20cd73df59"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "Loading weights:   0%|          | 0/103 [00:00<?, ?it/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "b6c877d4fcda44d4b45d3add81a633f4"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "stream",
+          "name": "stderr",
+          "text": [
+            "BertModel LOAD REPORT from: sentence-transformers/all-MiniLM-L6-v2\n",
+            "Key                     | Status     |  | \n",
+            "------------------------+------------+--+-\n",
+            "embeddings.position_ids | UNEXPECTED |  | \n",
+            "\n",
+            "Notes:\n",
+            "- UNEXPECTED\t:can be ignored when loading from different task/architecture; not ok if you expect identical arch.\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "tokenizer_config.json:   0%|          | 0.00/350 [00:00<?, ?B/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "cf4ba8afab434f869a4a533913be8189"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "vocab.txt: 0.00B [00:00, ?B/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "a3e8c5f7a15e492aa59b1459963da66d"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "tokenizer.json: 0.00B [00:00, ?B/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "3712dba5d01247298810249a327b27b4"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "special_tokens_map.json:   0%|          | 0.00/112 [00:00<?, ?B/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "2991c94441984658b3f24ca4be3bafc3"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "config.json:   0%|          | 0.00/190 [00:00<?, ?B/s]"
+            ],
+            "application/vnd.jupyter.widget-view+json": {
+              "version_major": 2,
+              "version_minor": 0,
+              "model_id": "18774a626601487a8ff0bfc87cffa014"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "✅ MiniLM loaded — converts text into vectors of 384 numbers.\n"
+          ]
+        }
+      ],
+      "source": [
+        "model = SentenceTransformer('all-MiniLM-L6-v2')\n",
+        "print(f\"✅ MiniLM loaded — converts text into vectors of {model.get_sentence_embedding_dimension()} numbers.\")"
+      ],
+      "id": "kv52_ifWb_Uy"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "GsBw4H2Db_Uy"
+      },
+      "source": [
+        "### *b. Convert your 10 reviews into vectors*"
+      ],
+      "id": "GsBw4H2Db_Uy"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "hQEp0li_b_Uy",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "fdaab9f2-9434-4631-e5ff-f747cbecdd8b"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "✅ Converted 10 reviews into vectors.\n",
+            "Each vector has 384 numbers.\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Select the 10 reviews our group annotated\n",
+        "group_reviews_df = reviews_df[reviews_df[\"review_id\"].isin([\n",
+        "    \"R093\",\"R097\",\"R042\",\"R019\",\"R039\",\n",
+        "    \"R082\",\"R055\",\"R034\",\"R007\",\"R064\"\n",
+        "])]\n",
+        "\n",
+        "# Get the text of the 10 reviews\n",
+        "review_texts = group_reviews_df[\"review_text\"].tolist()\n",
+        "\n",
+        "# Convert them into vectors using MiniLM\n",
+        "review_vectors = model.encode(review_texts)\n",
+        "\n",
+        "print(f\"✅ Converted {len(review_vectors)} reviews into vectors.\")\n",
+        "print(f\"Each vector has {len(review_vectors[0])} numbers.\")"
+      ],
+      "id": "hQEp0li_b_Uy"
+    },
+    {
+      "cell_type": "code",
+      "source": [
+        "from google.colab import drive\n",
+        "drive.mount('/content/drive')"
+      ],
+      "metadata": {
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "id": "Q5CinlViKjXo",
+        "outputId": "99587ea8-6722-4e2b-b8ad-131487d8fd2f"
+      },
+      "id": "Q5CinlViKjXo",
+      "execution_count": null,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "Mounted at /content/drive\n"
+          ]
+        }
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "m1Nclu6Pb_Uy"
+      },
+      "source": [
+        "### *c. Calculate cosine similarity and visualize as a heatmap*"
+      ],
+      "id": "m1Nclu6Pb_Uy"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "aFFwpuQZb_Uy",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 807
+        },
+        "outputId": "3c16d4bf-0297-4d7c-e687-4f3b842f56c1"
+      },
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "<Figure size 1000x800 with 2 Axes>"
+            ],
+            "image/png": "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\n"
+          },
+          "metadata": {}
+        }
+      ],
+      "source": [
+        "sim_matrix = cosine_similarity(review_vectors)\n",
+        "\n",
+        "# Heatmap\n",
+        "labels = [f\"R{i+1}\" for i in range(10)]\n",
+        "plt.figure(figsize=(10, 8))\n",
+        "sns.heatmap(sim_matrix, annot=True, fmt=\".2f\", cmap=\"RdYlGn\",\n",
+        "            xticklabels=labels, yticklabels=labels, vmin=0, vmax=1)\n",
+        "plt.title(\"Cosine Similarity Between Your 10 Reviews (MiniLM)\")\n",
+        "plt.tight_layout()\n",
+        "plt.show()"
+      ],
+      "id": "aFFwpuQZb_Uy"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "J2SzmB9-b_Uy"
+      },
+      "source": [
+        "### *d. Find the most and least similar review pairs*"
+      ],
+      "id": "J2SzmB9-b_Uy"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "J3ZJuSdyb_Uy",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "c1329b37-f037-47c0-ffdf-3bbe697e91bb"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "🟢 MOST SIMILAR PAIR:\n",
+            "   R2 & R3 — similarity: 0.618\n",
+            "   R2: second time buying this brand. first one lasted 2 years which is about right for...\n",
+            "   R3: bought it, used it, forgot to review it for 6 months. that tells you how memorab...\n",
+            "   Majority labels: mixed vs negative\n",
+            "\n",
+            "🔴 LEAST SIMILAR PAIR:\n",
+            "   R4 & R10 — similarity: -0.078\n",
+            "   R4: worst purchase of 2024 easily. the bluetooth speaker stopped connecting after a ...\n",
+            "   R10: its a water bottle. it holds water. not much else to say...\n",
+            "   Majority labels: positive vs negative\n"
+          ]
+        }
+      ],
+      "source": [
+        "sim_no_diag = sim_matrix.copy()\n",
+        "np.fill_diagonal(sim_no_diag, 0)\n",
+        "\n",
+        "# Calculate majority labels from annotations\n",
+        "majority_labels = []\n",
+        "for i in range(num_reviews):\n",
+        "    labels_for_review = [member[i] for member in annotations]\n",
+        "    counts = Counter(labels_for_review)\n",
+        "    majority_labels.append(counts.most_common(1)[0][0])\n",
+        "\n",
+        "max_idx = np.unravel_index(sim_no_diag.argmax(), sim_no_diag.shape)\n",
+        "min_idx = np.unravel_index(sim_no_diag.argmin(), sim_no_diag.shape)\n",
+        "\n",
+        "print(\"🟢 MOST SIMILAR PAIR:\")\n",
+        "print(f\"   R{max_idx[0]+1} & R{max_idx[1]+1} — similarity: {sim_no_diag[max_idx]:.3f}\")\n",
+        "print(f\"   R{max_idx[0]+1}: {review_texts[max_idx[0]][:80]}...\")\n",
+        "print(f\"   R{max_idx[1]+1}: {review_texts[max_idx[1]][:80]}...\")\n",
+        "print(f\"   Majority labels: {majority_labels[max_idx[0]]} vs {majority_labels[max_idx[1]]}\")\n",
+        "\n",
+        "print(f\"\\n🔴 LEAST SIMILAR PAIR:\")\n",
+        "print(f\"   R{min_idx[0]+1} & R{min_idx[1]+1} — similarity: {sim_no_diag[min_idx]:.3f}\")\n",
+        "print(f\"   R{min_idx[0]+1}: {review_texts[min_idx[0]][:80]}...\")\n",
+        "print(f\"   R{min_idx[1]+1}: {review_texts[min_idx[1]][:80]}...\")\n",
+        "print(f\"   Majority labels: {majority_labels[min_idx[0]]} vs {majority_labels[min_idx[1]]}\")"
+      ],
+      "id": "J3ZJuSdyb_Uy"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "3Quz8zBob_Uy"
+      },
+      "source": [
+        "### *e. ✋🏻🛑⛔️ Create 2 additional visualizations to explore the similarity data*\n",
+        "\n",
+        "**Ideas:** average similarity per review, similarity by label match, PCA scatter plot, disagreement vs similarity...\n",
+        "\n",
+        "**Available variables:** `sim_matrix`, `sample`, `majority_labels`, `all_labels`, `sample_embeddings`"
+      ],
+      "id": "3Quz8zBob_Uy"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "FaA9Z_wLb_Uy",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 487
+        },
+        "outputId": "6a082d6f-633b-474d-823d-370d317c9c3d"
+      },
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "<Figure size 1000x500 with 1 Axes>"
+            ],
+            "image/png": "iVBORw0KGgoAAAANSUhEUgAAA1cAAAHWCAYAAACbsXOkAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAVGVJREFUeJzt3XlYVHX///EXjCK4poBbuBuLCgpKKmomam5ZqWmWy62ZW1pq7nX/KvtqpKWWpmXivqSW6S3umVmZe5oSuZTe7qaIK4Khw/z+8Mt8JWBk9MDMyPNxXVwX85nPnHmfNzPAi3POBzeLxWIRAAAAAOCBuDu6AAAAAAB4GBCuAAAAAMAAhCsAAAAAMADhCgAAAAAMQLgCAAAAAAMQrgAAAADAAIQrAAAAADAA4QoAAAAADEC4AgAAAAADEK4AAA53+vRpBQQE6JtvvjFsmzt37lRAQIB27txpHRs1apQiIyMNe440AQEBmjp1quHbzcsy+/oBgLMjXAGAQRYtWqSAgAB17NjR0aXAwfbu3aupU6fq2rVrji4FAJCL8jm6AAB4WMTExOjRRx/VgQMHdOLECVWoUMHRJbmMtL7ly2fcj6Xw8HAdOHBA+fPnN2ybWTlw4IBMJpP19r59+/Tpp5+qXbt2Klq0aI4//8MoN79+AGAUjlwBgAFOnTqlffv2afTo0SpRooRiYmJyvYbU1FT9/fffuf68RnBzc1OBAgXSBZQH5e7urgIFCsjdPWd+1N3d7wIFChgaDB3p9u3bSklJyfb8nHrd5fTXDwByAt+xAMAAMTExKlasmBo3bqwWLVqkC1e3bt3S448/rtGjR2d4XGJiooKDgzV+/HjrWEpKiqZMmaLmzZurRo0aaty4sSZMmJDhF96AgAC99957WrVqldq0aaPg4GD99NNPkqRZs2apc+fOqlu3rkJCQtS+fXutX78+w/PfvHlTY8eOVd26dRUaGqp+/frp/PnzmV5DdP78eY0ePVoRERGqUaOG2rRpo6+//jpb/fn555/14osvqk6dOgoNDVWLFi00adIk6/2ZXXM1atQohYaG6uzZs+rbt69CQ0PVqFEjLVq0SJJ0+PBhde/eXbVq1VKTJk0yBNrsXrOT3V7Z6vfd/Zo6daomTJggSWratKkCAgIUEBCg06dPq2vXrnrmmWcyraNFixbq1auXzVojIyPVt29fbd26Vc8++6yCg4PVunVrbdy4McPca9euady4cWrcuLFq1Kih5s2b64svvlBqaqp1TlrfZ82apblz56pZs2YKDg7W0aNHs6zBVh/u9Rq5ePGiqlWrpk8//TTDdo8dO6aAgAAtXLhQUtZfv/3796tXr16qXbu2atasqa5du+qXX36x3n/o0CEFBATou+++s4799ttvCggIULt27dJt65VXXuE0XgCGejj+zAYADhYTE6PmzZvLw8NDTz/9tL788ksdOHBAISEhyp8/v5o1a6Zvv/1WY8aMkYeHh/VxmzZtUkpKilq3bi3pzlGA/v3765dfflGnTp1UpUoVHTlyRPPmzdPx48c1ffr0dM+7Y8cOrVu3Tl26dFHx4sX16KOPSpLmz5+vyMhItW3bVrdu3dKaNWs0aNAgzZgxQ08++aT18aNGjdK6dev07LPPqmbNmtq9e7f69OmTYf8uXryoTp06yc3NTV26dFGJEiX0448/6q233lJiYqJ69OiRZW/++OMP9e3bVwEBAXr99dfl4eGhEydOaO/evffsq9lsVu/evVWnTh0NGzZMMTExeu+99+Tl5aXJkyerbdu2euqpp7RkyRKNHDlStWrVUrly5e653btlt1dS1v2+W/PmzXX8+HGtXr1ao0ePVvHixSVJJUqU0LPPPqt///vfOnLkiPz9/a2POXDggI4fP67+/fvfs97jx49ryJAh6ty5s9q1a6fly5dr0KBBio6OVoMGDSRJycnJ6tq1q86fP6/OnTurTJky2rdvnyZNmqT4+Hi99dZb6bb5zTff6O+//1anTp3k4eGhYsWK2awhsz5k5zXi4+Oj8PBwrVu3TgMHDky3zbVr18pkMqlly5ZZPu/27dvVu3dv1ahRQwMHDpSbm5u++eYb/etf/9LixYsVEhIif39/FS1aVHv27FHTpk0lSXv27JG7u7sOHTqkxMREFS5cWKmpqdq3b586dep0z54DQLZZAAAPJDY21uLv72/5+eefLRaLxZKammp54oknLGPHjrXO+emnnyz+/v6WzZs3p3ts7969LU2bNrXeXrlypSUwMNCye/fudPO+/PJLi7+/v+WXX36xjvn7+1sCAwMtf/zxR4aakpOT091OSUmxPP3005bu3btbx3777TeLv7+/Zdy4cenmjho1yuLv72+ZMmWKdezNN9+0NGjQwHLp0qV0c4cMGWKpXbt2hue725w5cyz+/v6WhISELOecOnXK4u/vb1m+fLl1bOTIkRZ/f3/L559/bh27evWqJSQkxBIQEGBZs2aNdfzo0aMZat6xY4fF39/fsmPHjnTbbNKkSbrnzk6vLBbb/f7nc0dHR1v8/f0tp06dSjfv2rVrluDgYMuHH36Ybvx//ud/LLVq1bLcuHEjY3Pu0qRJE4u/v79lw4YN1rHr169bGjRoYHnuueesY9OmTbPUqlXL8t///jfd4z/66CNLUFCQ5ezZsxaL5f/6HhYWZvPr8899zawP2X2NLFmyxOLv7285fPhwunmtW7dO1/N/fv1SU1MtTz31lOXll1+2pKamWuclJydbIiMjLT179rSO9enTx/L8889bbw8cONAycOBAS1BQkOWHH36wWCwWS1xcnMXf39+yadOmbO03AGQHpwUCwAOKiYmRj4+P6tatK+nO9UOtW7fW2rVrZTabJUn16tVT8eLFtXbtWuvjrl69qm3btlmPWknS+vXrVaVKFVWuXFmXLl2yftSrV0+SMpwiFR4erqpVq2aoydPTM93zXL9+XbVr19bvv/9uHU87leull15K99iuXbumu22xWLRx40ZFRkbKYrGkq6thw4a6fv264uLisuxP2oIO3333XbpT0rLr7tO2ihYtqkqVKsnLy0utWrWyjleuXFlFixbVqVOn7N5+dnqVJqt+Z1eRIkXUtGlTrVmzRhaLRdKdo3Pr1q1T06ZNVbBgwXtuo2TJkmrevLn1duHChfXcc8/p999/V3x8vKQ7r6PatWuraNGi6b5eERERMpvN2r17d7ptPvXUUypRokS29+OffbDnNdK8eXPly5cv3XvhyJEj+vPPP9O9F/7p4MGDOn78uNq2bavLly9bt5+UlKT69etr9+7d1tdX2tcvKSlJkvTLL7/oiSeeUGBgoPUUwj179sjNzU21a9fO9n4DwL1wWiAAPACz2aw1a9aobt26On36tHU8JCREs2fP1vbt29WwYUPly5dPTz31lFavXq2UlBR5eHho48aNunXrVrpfKE+cOKGjR4+qfv36mT5fQkJCutt+fn6Zzvv+++/12Wef6eDBg+mu1XJzc7N+fvbsWbm7u2fYxj9XObx06ZKuXbumpUuXaunSpZk+36VLlzIdl6TWrVvrq6++0r///W9NnDhR9evXV/PmzdWyZct7LlZQoECBDL/0FylSRKVLl063L2nj97P0eXZ6lSarftvjueee09q1a7Vnzx6Fh4dr27Ztunjxop599tlsPb5ChQoZaqtYsaIk6cyZM/L19dWJEyd0+PDhLF9H//x62btf/5xvz2ukRIkSqlevntatW6fBgwdLunNKYL58+dKFxn86fvy4JGnkyJFZzrl+/bqKFSumOnXq6Pbt2/r1119VunRpJSQkqE6dOvrzzz+1Z88eSXfCVdWqVfXII49kc68B4N4IVwDwAHbs2KH4+HitWbNGa9asyXB/TEyMGjZsKElq06aNli5dqh9//FHNmjXT+vXrVblyZQUGBlrnp6amyt/fP9PFLySpdOnS6W7ffdQlzZ49e9S/f3+Fh4frnXfeka+vr/Lnz6/ly5dr9erVdu9j2tGAZ555JsOCAGkCAgKyfLynp6cWLVqknTt3asuWLfrpp5+0du1aLV26VLNnz7a5QmBW92U1nnY0KLvs7VVm/bZXw4YN5ePjo1WrVik8PFyrVq2Sr6+vIiIiHnjbaVJTU9WgQQO98sormd6fFsbS2Ltf/5xv72ukTZs2Gj16tA4ePKigoCCtW7dO9erVs3n0LO1rO2LECAUFBWU6J+3IX40aNVSgQAHt3r1bZcuWlbe3typVqqQ6depo8eLFSklJ0S+//KJmzZplf6cBIBsIVwDwAGJiYuTt7a233347w33ffvutdRELT09PhYeHy9fXV2vXrlVYWJh27Nihfv36pXtM+fLldejQIdWvXz/TIyfZsWHDBhUoUECzZs1Kt3jG8uXL080rW7asUlNTdfr06XS/bJ84cSLdvBIlSqhQoUJKTU297wDg7u6u+vXrq379+ho9erQ+//xzTZ48WTt37jQ0VNgru72yl62vnclk0tNPP60VK1Zo2LBh2rRpkzp16pTtZehPnDghi8WS7jnSjuqkLbBRvnx5JSUl5Vpv7X2NNGvWTG+//bb11MDjx4+rb9++Nh+TtlBJ4cKF7/kcHh4eCgkJ0Z49e1S2bFnVqVNH0p3TBVNSUrRq1SpdvHhR4eHh2dk9AMg2rrkCgPt08+ZNbdy4UU8++aRatmyZ4aNLly66ceOGNm/eLOlOwGjZsqW+//57rVq1Srdv385wjUmrVq10/vx5LVu2LNPnS7uGxBaTySQ3Nzfr9V7SnSW3716aWpL1iNrixYvTjacthX339lq0aKENGzboyJEjGZ7P1imBknTlypUMY2lHHuz5f0o5Ibu9speXl5ekO6epZebZZ5/V1atX9fbbbyspKSnL5dkzc+HCBX377bfW24mJiVq5cqWCgoLk6+sr6c7raN++fdbr6u527do13b59257duSd7XyNFixZVw4YNtW7dOq1Zs8a6oqYtNWrUUPny5TV79mzduHHjns9Ru3ZtHThwQDt37rReV1WiRAlVqVJFM2fOlCRr6AIAo3DkCgDu0+bNm3Xjxg1FRkZmen+tWrVUokQJrVq1yhqiWrVqpQULFmjKlCny9/dXlSpV0j3m2Wef1bp16/TOO+9o586dCgsLk9ls1rFjx7R+/XpFR0crODjYZl2NGzfWnDlz9Morr+jpp59WQkKCFi9erPLly+vw4cPWeTVq1FCLFi00b948XblyxboUe9pRkLuPjAwdOlQ7d+5Up06d1LFjR1WtWlVXr15VXFyctm/frl27dmVZz7Rp07Rnzx41btxYjz76qLWe0qVLO3wxgez2yl7Vq1eXJE2ePFmtW7dW/vz51aRJE+tpa9WqVZO/v791AZO0+dlRsWJFvfXWW4qNjZW3t7eWL1+uhIQERUVFWef06tVLmzdvVr9+/dSuXTtVr15dycnJOnLkiDZs2KDvvvvOrgUsssPe10jr1q01fPhwLV68WA0bNrQufJIVd3d3jR07Vr1799bTTz+t9u3bq1SpUjp//rx27typwoUL6/PPP7fOr1Onjj7//HOdO3cuXYiqU6eOli5dqkcffTTDabYA8KAIVwBwn1atWqUCBQpY/7fQP7m7u+vJJ59UTEyMLl++rOLFiyssLExlypTRuXPnMl0Zzd3dXdOmTdPcuXP1n//8R99++628vLzk5+enbt26qVKlSvesq379+ho3bpxmzpyp999/X35+fho2bJjOnDmTITCMHz9ePj4+WrNmjb799ltFRERo8uTJatmyZbrT5Hx8fPTVV19p2rRp+vbbb/Xll1/qkUceUdWqVTVs2DCb9URGRurMmTNavny5tQ+PP/64XnvtNRUpUuSe+5OT7OmVPUJCQjRo0CAtWbJEP/30k1JTU/Xdd9+lWw3w2Wef1YcffpjthSzSVKxYUf/v//0/TZgwQf/973/l5+enyZMnq1GjRtY5Xl5eWrBggWbMmKH169dr5cqVKly4sCpWrJhjfbf3NRIZGSlPT0/duHHD5iqBd6tbt66WLl2q6dOna+HChUpKSpKvr69CQkL0wgsvpJsbGhoqk8kkT0/PdNc1poUrjloByAluFnuv/gUAPNQOHjyo5557Th9++KFdp6vBPvPmzVNUVJQ2b96ssmXLZusxkZGReuyxxzRjxowcrg4AcD+45goA8rCbN29mGJs3b57c3d252D8HWSwWff311woPD892sAIAOD9OCwSAPCw6Olq//fab6tWrJ5PJpB9//FE//vijXnjhBZUpU8bR5T10kpKStHnzZu3cuVNHjhzR9OnTHV0SAMBAhCsAyMNCQ0P1888/a/r06UpKSlKZMmX02muvZVgiHsa4dOmShg4dqqJFi6pfv35q2rSpo0sCABiIa64AAAAAwABccwUAAAAABiBcAQAAAIABuOYqE6mpqbp9+7bc3d3T/RNNAAAAAHmLxWJRamqq8uXLJ3d328emCFeZuH37tmJjYx1dBgAAAAAnERwcLA8PD5tzCFeZSEukwcHBMplMDq4m95nNZsXGxubZ/b8X+mMb/bGN/thGf7JGb2yjP7bRH9voj215vT9p+3+vo1YS4SpTaacCmkymPPkCSpPX9/9e6I9t9Mc2+mMb/ckavbGN/thGf2yjP7bl9f5k53IhFrQAAAAAAAMQrgAAAADAAIQrAAAAADAA4QoAAAAADEC4AgAAAAADEK4AAAAAwACEKwAAAAAwAOEKAAAAAAxAuAIAAAAAAxCuAAAAAMAAhCsAAAAAMADhCgAAAAAMQLgCAAAAAAMQroD74OXl5egSAAAA4GQIV3B55lRLrj6fyWRStWrVZDKZcvV5c3s/AQAAYJ98ji4AeFAmdzcNWrJPf15IdHQpOaZqycL6pHOoo8sAAACADYQrPBT+vJCouLPXHF0GAAAA8jBOCwRgOK5JAwAAeRHhCnjIcU0aAABA7uC0QOAhxzVpAAAAuYNwBeQBXJMGAACQ8zgtEAAAAAAMQLgCAAAAAAMQrgAAAADAAIQrAAAAADAA4QoAAAAADEC4AgAAAAADEK4AAAAAwACEKwAAAAAwAOEKAAAAAAxAuAIAAAAAAxCuAAAAAMAAhCsAAAAAMADhCgAAAAAMQLgCAAAAAAMQrgAAAADAAIQrAAAAADAA4QoAAAAADEC4AgAAAAADEK4AAAAAwACEKwAAAAAwAOEKAAAAAAzgFOFq0aJFioyMVHBwsDp27KgDBw5kOXfZsmV66aWXFB4ervDwcPXo0SPD/FGjRikgICDdR69evXJ6NwAAAADkYfkcXcDatWsVFRWlMWPGqGbNmpo3b5569eql9evXy9vbO8P8nTt3qk2bNgoLC5OHh4eio6P18ssva82aNSpVqpR1XqNGjRQVFWW97eHhkSv7AwAAACBvcviRqzlz5qhTp07q0KGDqlatqjFjxsjT01PLly/PdP7EiRPVpUsXBQUFqUqVKho7dqxSU1O1ffv2dPM8PDzk6+tr/ShWrFhu7A4AAACAPMqhR65SUlIUFxenvn37Wsfc3d0VERGhffv2ZWsbycnJun37dobwtGvXLtWvX19FixZVvXr1NHjwYBUvXtyu+sxms13zHxZp++0q+28ymRxdQq65n68J/XEurvb+ym30J2v0xjb6Yxv9sY3+2JbX+2PPfjs0XF2+fFlmsznD6X/e3t46duxYtrbx0UcfqWTJkoqIiLCONWrUSM2bN5efn59OnTqlSZMmqXfv3lq6dKldv2jGxsZme+7DyBX238vLS9WqVXN0Gbnm8OHDSk5OzvZ8+uO8XOH95Uj0J2v0xjb6Yxv9sY3+2EZ/7s3h11w9iC+++EJr167V/PnzVaBAAet4mzZtrJ+nLWjRrFkz69Gs7AoODs5Tf/VPYzabFRsbm2f335kFBAQ4ugSn5gr94f1lG/3JGr2xjf7YRn9soz+25fX+pO1/djg0XBUvXlwmk0kJCQnpxhMSEuTj42PzsbNmzdIXX3yhOXPmKDAw0ObccuXKqXjx4jpx4oRd4cpkMuXJF1CavL7/zoivh22u1B/eX7bRn6zRG9voj230xzb6Yxv9uTeHLmjh4eGh6tWrp1uMIm1xitDQ0CwfN3PmTE2fPl3R0dEKDg6+5/P89ddfunLlinx9fQ2pGwAAAAD+yeGnBfbs2VMjR45UjRo1FBISonnz5ik5OVnt27eXJI0YMUKlSpXS0KFDJd05FXDKlCmaOHGiHn30UcXHx0uSChYsqEKFCunGjRv69NNP1aJFC/n4+OjUqVP68MMPVaFCBTVq1Mhh+wkAAADg4ebwcNW6dWtdunRJU6ZMUXx8vIKCghQdHW09LfDcuXNyd/+/A2xLlizRrVu39Prrr6fbzsCBA/Xaa6/JZDLpyJEjWrlypa5fv66SJUuqQYMGGjRoEP/rCgAAAECOcXi4kqSuXbuqa9eumd63YMGCdLc3b95sc1uenp6aNWuWYbUBAAAAQHY4/J8IAwAAAMDDgHAFAAAAAAYgXAEAAACAAQhXAAAAAGAAwhUAAAAAGIBwBQAAAAAGIFwBAAAAgAEIVwAAAABgAMIVAAAAABiAcAUAAAAABiBcAQAAAIABCFcAAAAAYADCFQAAAAAYgHAFAAAAAAYgXAEAAACAAQhXAAAAAGAAwhUAAAAAGIBwBQAAAAAGIFwBAAAAgAEIVwAAAABgAMIVAAAAABiAcAUAAAAABiBcAQAAAIABCFcAAAAAYADCFQAAAAAYgHAFAAAAAAYgXAEAAACAAQhXAAAAAGAAwhUAAAAAGIBwBQAAAAAGIFwBAAAAgAEIVwAAAABgAMIVAAAAABiAcAUAAAAABiBcAQAAAIABCFcAAAAAYADCFQAAAAAYgHAFAAAAAAYgXAEAAACAAQhXAAAAAGAAwhUAAAAAGIBwBQAAAAAGIFwBAAAAgAEIVwAAAABgAMIVAAAAABiAcAUAAAAABiBcAQAAAIABCFcAAAAAYADCFQAAAAAYgHAFAAAAAAawO1wlJSXlRB0AAAAA4NLsDlcNGjTQ6NGjtWfPnpyoBwAAAABckt3h6sMPP9TVq1fVo0cPtWjRQl988YXOnz//QEUsWrRIkZGRCg4OVseOHXXgwIEs5y5btkwvvfSSwsPDFR4erh49emSYb7FY9Mknn6hhw4YKCQlRjx49dPz48QeqEQAAAABssTtcNWvWTNOnT9ePP/6ozp07KyYmRpGRkerbt682btyo27dv27W9tWvXKioqSgMGDNCKFSsUGBioXr16KSEhIdP5O3fuVJs2bTR//nwtWbJEZcqU0csvv5wu4M2cOVMLFizQu+++q2XLlsnLy0u9evXS33//be/uAgAAAEC23PeCFiVKlFDPnj0VExOjUaNGadu2bXr99dfVqFEjffLJJ0pOTs7WdubMmaNOnTqpQ4cOqlq1qsaMGSNPT08tX7480/kTJ05Uly5dFBQUpCpVqmjs2LFKTU3V9u3bJd05ajV//nz1799fzZo1U2BgoCZMmKALFy5o06ZN97u7AAAAAGBTvvt94MWLF7VixQqtWLFCZ8+eVYsWLfT888/rr7/+UnR0tPbv36/Zs2fb3EZKSori4uLUt29f65i7u7siIiK0b9++bNWRnJys27dvq1ixYpKk06dPKz4+XhEREdY5RYoUUc2aNbVv3z61adMm2/toNpuzPfdhkrbfrrL/JpPJ0SXkmvv5mtAf5+Jq76/cRn+yRm9soz+20R/b6I9teb0/9uy33eFq48aN+uabb7R161ZVqVJFL730kp555hkVLVrUOicsLEytW7e+57YuX74ss9ksb2/vdOPe3t46duxYtur56KOPVLJkSWuYio+Pt27jn9u8ePFitraZJjY21q75DxtX2H8vLy9Vq1bN0WXkmsOHD2f7qLBEf5yZK7y/HIn+ZI3e2EZ/bKM/ttEf2+jPvdkdrkaPHq02bdpo8eLFCgkJyXROyZIl1a9fvwcu7l6++OILrV27VvPnz1eBAgUM335wcHCe+qt/GrPZrNjY2Dy7/84sICDA0SU4NVfoD+8v2+hP1uiNbfTHNvpjG/2xLa/3J23/s8PucLV161Z5eXnZnOPp6amBAwfec1vFixeXyWTKsHhFQkKCfHx8bD521qxZ+uKLLzRnzhwFBgZax319fa3bKFmyZLpt3j0vO0wmU558AaXJ6/vvjPh62OZK/eH9ZRv9yRq9sY3+2EZ/bKM/ttGfe7N7QYuwsLBMV/K7fPmygoKC7NqWh4eHqlevbl2MQpJ1cYrQ0NAsHzdz5kxNnz5d0dHRCg4OTnefn5+ffH19020zMTFR+/fvt7lNAAAAAHgQdh+5slgsmY6npKQof/78dhfQs2dPjRw5UjVq1FBISIjmzZun5ORktW/fXpI0YsQIlSpVSkOHDpV051TAKVOmaOLEiXr00Uet11gVLFhQhQoVkpubm7p3767PPvtMFSpUkJ+fnz755BOVLFlSzZo1s7s+AAAAAMiObIer+fPnS5Lc3Nz01VdfqWDBgtb7UlNTtXv3blWuXNnuAlq3bq1Lly5pypQpio+PV1BQkKKjo62nBZ47d07u7v93gG3JkiW6deuWXn/99XTbGThwoF577TVJUu/evZWcnKy3335b165dU+3atRUdHZ0j12UBAAAAgGRHuJo7d66kO0eulixZki7w5M+fX35+fhozZsx9FdG1a1d17do10/sWLFiQ7vbmzZvvuT03NzcNGjRIgwYNuq96ACAn3eu6VQAA4JqyHa7SQk23bt306aefWv+vFAC4MnOqRSZ3t1x7PpPJlOvL4+f2PgIAkFfZfc3VP48kAYArM7m7adCSffrzQqKjS8kRVUsW1iedWcwHAIDckK1wFRUVpUGDBqlgwYKKioqyOXf06NGGFAYAueXPC4mKO3vN0WUAAAAXl61w9fvvv+v27dvWz7Pi5sZpJwAAAADypmyFq7tPBeS0QAAAAADIyK5/Inzr1i1Vq1ZNR44cyal6AAAAAMAl2RWu8ufPrzJlyig1NTWn6gEAAAAAl2RXuJKkfv36adKkSbpy5UoOlAMAAAAArsnupdgXLVqkEydOqFGjRipbtqwKFiyY7v4VK1YYVhwAAAAAuAq7w1WzZs1yog4AAAAAcGl2h6uBAwfmRB0AAEiSvLy8HF0CXBSvHQCOZvc1VwCAvMOcasnV5zOZTKpWrZpMJlOuPWdu72Nekpu9dcRrR+L1AyA9u49cmc1mzZ07V+vWrdO5c+d069atdPfv2rXLsOIAAI5lcnfToCX79OeFREeXkiOqliysTzqHOrqMhxavHwB5jd3h6tNPP9VXX32ll19+WR9//LH69eunM2fOaNOmTRowYEBO1AgAcKA/LyQq7uw1R5cBF8XrB0BeYne4iomJ0dixY/Xkk09q6tSpevrpp1W+fHkFBARo//79OVEjAAAAADg9u6+5unjxovz9/SVJhQoV0vXr1yVJTZo00ZYtWwwtDgAAAABchd3hqlSpUoqPj5cklStXTj///LMkKTY2Vh4eHsZWB4dhxSUAAB4u/GwHcp7dpwU2b95c27dvV82aNdWtWzcNHz5cX3/9tc6ePasePXrkQIkwp1pkcnfLtedLW3Ept+X2fgIA4Ei5+XOPn+1A7rA7XA0bNsz6eevWrVWmTBn9+uuvqlChgiIjIw0tDnc87KstSay4BADIex72n+/8bEdeZHe4+qfQ0FCFhvLGyWmstgQAwMOHn+/AwyVb4eq7777L9gabNm1638UAAAAAgKvKVrjK7v+vcnNz08GDBx+oIAAAAABwRdkKV4cOHcrpOgAAAADApdm9FDsAAAAAIKNsHbmaP3++XnjhBRUoUEDz58+3Obd79+6GFAYAAAAAriRb4Wru3Llq27atChQooLlz52Y5z83NjXAFAAAAIE/KVrjavHlzpp8DAAAAAO7gmisAAAAAMIDd/0TYYrFo/fr12rlzpy5duqTU1NR093/66aeGFQcAAAAArsLucDVu3DgtXbpUdevWlY+Pj9zc3HKiLgAAAABwKXaHq1WrVunTTz9V48aNc6IeAAAAAHBJdl9zVbhwYfn5+eVELQAAAADgsuwOV6+99pqmTZummzdv5kQ9AAAAAOCS7D4tsFWrVlq9erXq168vPz8/5cuXfhMrVqwwrDgAAAAAcBV2h6uRI0cqLi5OzzzzDAtaAAAAAMD/sjtc/fDDD4qOjladOnVyoh4AAAAAcEl2X3NVunRpFS5cOCdqAQAAAACXZXe4GjVqlD788EOdPn06J+oBAAAAAJdk92mBw4cPV3Jyspo3by5PT0/lz58/3f27du0yrDgAAAAAcBV2h6s333wzJ+oAAAAAAJdmd7hq165dTtQBAAAAAC4tW+EqMTHRuohFYmKizbksdgEAAAAgL8pWuAoPD9fWrVvl7e2tOnXqZPq/rSwWi9zc3HTw4EHDiwQAAAAAZ5etcDVv3jwVK1ZMkjR//vwcLQgAAAAAXFG2wtXjjz+e6ecAAAAAgDvs/j9XP/74o/bs2WO9vWjRIj377LMaOnSorl69amhxAAAAAOAq7A5XH374oW7cuCFJOnz4sKKiotS4cWOdPn1aH3zwgeEFAgAAAIArsHsp9tOnT6tKlSqSpI0bNyoyMlJvvPGG4uLi1KdPH8MLBAAAAABXYPeRq/z58+vmzZuSpG3btqlBgwaSpGLFit1zmXYAAAAAeFjZfeQqLCxMUVFRCgsLU2xsrD7++GNJ0vHjx1W6dGmj6wMAAAAAl2D3kau3335b+fLl04YNG/TOO++oVKlSku4sdNGoUSPDCwQAAAAAV2D3kauyZctqxowZGcbffPNNQwoCAAAAAFdk95ErAAAAAEBGDg9XixYtUmRkpIKDg9WxY0cdOHAgy7l//PGHXnvtNUVGRiogIEBz587NMGfq1KkKCAhI99GyZcsc3AMAAAAAcHC4Wrt2raKiojRgwACtWLFCgYGB6tWrlxISEjKdn5ycLD8/Pw0dOlS+vr5Zbvexxx7T1q1brR+LFy/OqV0AAAAAAEkODldz5sxRp06d1KFDB1WtWlVjxoyRp6enli9fnun8kJAQjRw5Um3atJGHh0eW2zWZTPL19bV+lChRIqd2AQAAAAAk3ceCFmlOnDihkydPKjw8XJ6enrJYLHJzc8v241NSUhQXF6e+fftax9zd3RUREaF9+/bdb1nW2ho2bKgCBQqoVq1aGjp0qMqWLWv3dsxm8wPVYRSTyeToEnLN/fSc/thGf2zLK/253+9n9Md5pNXoCrWm4fVjG/1xHq74/spNeb0/9uy33eHq8uXLGjJkiHbs2CE3Nzdt3LhR5cqV05tvvqlixYpp1KhR2d6O2WyWt7d3unFvb28dO3bM3rKsQkJCFBUVpUqVKik+Pl7Tpk1Tly5dFBMTo8KFC9u1rdjY2PuuwyheXl6qVq2ao8vINYcPH1ZycnK259Mf2+iPbXmpP/b2RqI/zsoZfjZlB68f2+iPc3KV95ej0J97sztcRUVFyWQyacuWLWrVqpV1vHXr1vrggw+yHa5ySuPGja2fBwYGqmbNmmrSpInWrVunjh072rWt4ODgPPNXJWcREBDg6BKcGv2xjf5kjd7Y5gr9MZvNio2N5WeTE3KF148juUJ/eH/Zltf7k7b/2WF3uPr55581a9YslS5dOt14xYoVdfbs2Wxvp3jx4jKZTBkWr0hISJCPj4+9ZWWpaNGiqlixok6ePGn3Y00mU558ATkS/baN/thGf7JGb2xzpf7ws8n58PWwzZX6w/vLNvpzb3YvaJGUlCRPT88M41euXLG5yMQ/eXh4qHr16tq+fbt1LDU1Vdu3b1doaKi9ZWXpxo0bOnXqlM3VBQEAAADgQdkdrurUqaOVK1emG0tNTVV0dLTq1q1r17Z69uypZcuWacWKFTp69KjeffddJScnq3379pKkESNGaOLEidb5KSkpOnjwoA4ePKiUlBSdP39eBw8e1IkTJ6xzxo8fr127dun06dPau3evBg4cKHd3dz399NP27ioAAAAAZJvdpwUOHz5cPXr00G+//aZbt27pww8/1J9//qmrV6/qyy+/tGtbrVu31qVLlzRlyhTFx8crKChI0dHR1tMCz507J3f3/8t/Fy5c0HPPPWe9PXv2bM2ePVuPP/64FixYIEn666+/9MYbb+jKlSsqUaKEateurWXLlrEcOwAAAIAcZXe48vf314YNG7Rw4UIVKlRISUlJat68ubp06aKSJUvaXUDXrl3VtWvXTO9LC0xp/Pz8dPjwYZvbmzx5st01AAAAAMCDuq//c1WkSBH179/f6FoAAAAAwGXdV7i6du2aDhw4oISEBFkslnT33X3aHgAAAADkFXaHq82bN2vYsGFKSkpS4cKF5ebmZr3Pzc2NcAUAAAAgT7I7XI0fP14dOnTQG2+8IS8vr5yoCQAAAABcjt1LsZ8/f17du3cnWAEAAADAXewOVw0bNlRsbGxO1AIAAAAALsvu0wIbN26sDz/8UEePHpW/v7/y5Uu/iaZNmxpWHAAAAAC4CrvD1f/7f/9PkjRt2rQM97m5uengwYMPXhUAAAAAuBi7w9WhQ4dyog4AAAAAcGl2X3MFAAAAAMgoW0eu5s+frxdeeEEFChTQ/Pnzbc7t3r27IYUBAAAAgCvJVriaO3eu2rZtqwIFCmju3LlZznNzcyNcAQAAAMiTshWuNm/enOnnAAAAAIA7HviaK7PZrIMHD+rq1atG1AMAAAAALsnucDVu3Dh99dVXku4Eqy5duqhdu3Z68skntXPnTsMLBAAAAABXYHe42rBhgwIDAyVJ33//vc6cOaN169bpX//6lyZPnmx4gQAAAADgCuwOV5cvX5avr68k6YcfflDLli1VqVIldejQQUeOHDG8QAAAAABwBXaHKx8fH/35558ym8366aef1KBBA0nSzZs3ZTKZDC8QAAAAAFxBtlYLvFv79u01ePBg+fr6ys3NTREREZKk/fv3q3LlyoYXCAAAAACuwO5w9dprr+mxxx7TX3/9pZYtW8rDw0OSZDKZ1Lt3b8MLBAAAAABXYHe4kqSWLVtmGGvXrt0DFwMAAAAAruq+wtWuXbs0e/ZsHT16VJJUpUoVvfLKK6pTp46hxQEAAACAq7B7QYv//Oc/6tmzpzw9PdWtWzd169ZNnp6e6tGjh2JiYnKiRgAAAABwenYfufr88881fPhw9ejRwzrWvXt3zZkzR9OnT1fbtm2NrA8AAAAAXILdR65OnTqlJk2aZBiPjIzU6dOnDSkKAAAAAFyN3eGqTJky2r59e4bxbdu2qUyZMoYUBQAAAACuxu7TAnv27KmxY8fq4MGDCg0NlSTt3btXK1as0FtvvWV4gQAAAADgCuwOVy+99JJ8fX01e/ZsrV+/XpJUuXJlTZ48Wc2aNTO8QAAAAABwBfe1FHvz5s3VvHlzo2sBAAAAAJdl9zVXBw4c0P79+zOM79+/X7GxsYYUBQAAAACuxu5w9d577+ncuXMZxs+fP6/33nvPkKIAAAAAwNXYHa6OHj2q6tWrZxgPCgrSn3/+aUhRAAAAAOBq7A5XHh4eunjxYobx+Ph45ct3X5dwAQAAAIDLsztcNWjQQJMmTdL169etY9euXdPkyZMVERFhaHEAAAAA4CrsPtQ0cuRIdenSRU2aNFFQUJAk6dChQ/L29taECRMMLxAAAAAAXIHd4apUqVJatWqVYmJidOjQIXl6eqpDhw5q06aN8ufPnxM1AgAAAIDTu6+LpAoWLKgXXnjB6FoAAAAAwGXZfc0VAAAAACAjwhUAAAAAGIBwBQAAAAAGIFwBAAAAgAHuK1xdu3ZNX331lSZOnKgrV65IkuLi4nT+/HkjawMAAAAAl2H3aoGHDh1Sz549VaRIEZ05c0adOnXSI488oo0bN+rcuXP8rysAAAAAeZLdR64++OADtWvXThs3bpSHh4d1vHHjxtqzZ4+hxQEAAACAq7A7XMXGxqpz584ZxkuVKqX4+HhDigIAAAAAV2N3uPLw8FBiYmKG8ePHj6tEiRKGFAUAAAAArsbucBUZGalp06bp1q1b1rGzZ8/qo48+0lNPPWVocQAAAADgKuwOV6NGjVJSUpIiIiL0999/q1u3bnrqqadUqFAhDRkyJCdqBAAAAACnZ/dqgUWKFNGcOXO0Z88eHT58WElJSapevboiIiJyoj4AAAAAcAl2h6s0derUUZ06dYysBQAAAABclt3hav78+ZmOu7m5qUCBAipfvrzCw8NlMpkeuDgAAAAAcBV2h6u5c+fq8uXLSk5OVrFixSRJV69elZeXlwoWLKiEhASVK1dO8+fPV5kyZQwvGAAAAACckd0LWrzxxhuqUaOGNm7cqJ07d2rnzp3asGGDQkJC9NZbb2nLli3y8fFRVFRUtra3aNEiRUZGKjg4WB07dtSBAweynPvHH3/otddeU2RkpAICAjR37twH3iYAAAAAGMHucPXxxx/rzTffVPny5a1jFSpU0MiRIzVx4kSVLl1aw4cP1969e++5rbVr1yoqKkoDBgzQihUrFBgYqF69eikhISHT+cnJyfLz89PQoUPl6+tryDYBAAAAwAh2h6v4+Hjdvn07w/jt27d18eJFSVLJkiV148aNe25rzpw56tSpkzp06KCqVatqzJgx8vT01PLlyzOdHxISopEjR6pNmzby8PAwZJsAAAAAYAS7r7mqW7eu3nnnHY0dO1bVqlWTJP3+++969913Va9ePUnSkSNH5OfnZ3M7KSkpiouLU9++fa1j7u7uioiI0L59++wtK0e2aTab76sOo+WlxUHup+f0xzb6Y1te6c/9fj+jP84jrUZXqDUNrx/b6I/zcMX3V27K6/2xZ7/tDlfjxo3TiBEj1L59e+XLl8/6hPXr19e4ceMkSQULFtTIkSNtbufy5csym83y9vZON+7t7a1jx47ZW1aObDM2Nva+6jCSl5eXNcTmBYcPH1ZycnK259Mf2+iPbXmpP/b2RqI/zsoZfjZlB68f2+iPc3KV95ej0J97sztc+fr6as6cOTp69KiOHz8uSapUqZIqV65snZN2BMvVBQcH55m/KjmLgIAAR5fg1OiPbfQna/TGNlfoj9lsVmxsLD+bnJArvH4cyRX6w/vLtrzen7T9z477/ifCVapUUZUqVe734SpevLhMJlOGhSYSEhLk4+PjFNs0mUx58gXkSPTbNvpjG/3JGr2xzZX6w88m58PXwzZX6g/vL9voz73dV7j666+/9N133+ncuXO6detWuvtGjx6drW14eHioevXq2r59u5o1ayZJSk1N1fbt29W1a9f7KStHtgkAAAAA2WF3uNq+fbv69++vcuXK6dixY3rsscd05swZWSwWu88d7tmzp0aOHKkaNWooJCRE8+bNU3Jystq3by9JGjFihEqVKqWhQ4dKurNgxdGjR62fnz9/XgcPHlTBggVVoUKFbG0TAAAAAHKC3eFq4sSJevnll/X6668rNDRUU6dOVYkSJTRs2DA1atTIrm21bt1aly5d0pQpUxQfH6+goCBFR0dbT+E7d+6c3N3/b7X4Cxcu6LnnnrPenj17tmbPnq3HH39cCxYsyNY2AQAAACAn2B2ujh49qkmTJt15cL58unnzpgoVKqRBgwbp1Vdf1UsvvWTX9rp27ZrlKXtpgSmNn5+fDh8+/EDbBAAAAICcYPc/ES5YsKD1OitfX1+dPHnSet/ly5eNqwwAAAAAXIjdR65q1qypX375RVWqVFHjxo01fvx4HTlyRN9++61q1qyZEzUCAAAAgNOzO1yNHj1aN27ckCS99tprunHjhtauXauKFStq1KhRhhcIAAAAAK7ArnBlNpv1119/Wf8ZXMGCBfXee+/lSGEAAAAA4ErsuubKZDLp5Zdf1tWrV3OqHgAAAABwSXYvaPHYY4/p9OnTOVELAAAAALgsu8PV4MGDNX78eH3//fe6cOGCEhMT030AAAAAQF5k94IWffr0kST1799fbm5u1nGLxSI3NzcdPHjQuOoAAAAAwEXYHa7mz5+fE3UAAAAAgEuzO1w9/vjjOVEHAAAAALg0u6+5kqQ9e/Zo2LBh6ty5s86fPy9JWrlypfbs2WNocQAAAADgKuwOVxs2bFCvXr3k6empuLg4paSkSJISExM1Y8YMwwsEAAAAAFdgd7j67LPPNGbMGI0dO1b58v3fWYVhYWH6/fffDS0OAAAAAFyF3eHqv//9r+rUqZNhvEiRIrp27ZohRQEAAACAq7E7XPn4+OjkyZMZxn/55ReVK1fOkKIAAAAAwNXYHa46deqkcePGaf/+/XJzc9P58+e1atUqjR8/Xi+++GJO1AgAAAAATu++/olwamqqevTooeTkZHXt2lUeHh56+eWX1a1bt5yoEQAAAACcnt3hys3NTf3791evXr108uRJJSUlqUqVKipUqFBO1AcAAAAALsHu0wL/85//KDk5WR4eHqpatapCQkIIVgAAAADyPLvDVVRUlCIiIjR06FD98MMPMpvNOVEXAAAAALgUu08L3Lp1q3766SetXr1agwcPlqenp1q2bKm2bdsqLCwsJ2oEAAAAAKdnd7jKly+fmjRpoiZNmig5OVnffvutVq9ere7du6t06dLatGlTTtQJAAAAAE7N7nB1Ny8vLzVs2FDXrl3T2bNndfToUaPqAgAAAACXcl/hKu2IVUxMjLZv364yZcqoTZs2+uSTT4yuDwAAAABcgt3hasiQIdqyZYs8PT3VqlUrvfrqqwoNDc2J2gAAAADAZdgdrtzd3fXxxx+rYcOGMplM6e47cuSI/P39DSsOAAAAAFyF3eFq4sSJ6W4nJiZqzZo1+uqrrxQXF6eDBw8aVhwAAAAAuIr7XtBi9+7d+vrrr7Vx40aVLFlSzZs319tvv21kbQAAAADgMuwKV/Hx8VqxYoW+/vprJSYmqlWrVkpJSdG0adNUtWrVnKoRAAAAAJxetsNVv379tHv3bj355JN688031ahRI5lMJi1ZsiQn6wMAAAAAl5DtcPXjjz+qW7duevHFF1WxYsUcLAkAAAAAXI97dicuXrxYN27cUPv27dWxY0ctXLhQly5dysnaAAAAAMBlZDtc1apVS2PHjtXWrVv1wgsvaM2aNXriiSeUmpqqn3/+WYmJiTlZJwAAAAA4tWyHqzQFCxbU888/ry+//FKrVq1Sz549NXPmTEVERKhfv345USMAAAAAOD27w9XdKleurBEjRuiHH37QpEmTjKoJAAAAAFzOff+fq7uZTCY1a9ZMzZo1M2JzAAAAAOByHujIFQAAAADgDsIVAAAAABiAcAUAAAAABiBcAQAAAIABCFcAAAAAYADCFQAAAAAYgHAFAAAAAAYgXAEAAACAAQhXAAAAAGAAwhUAAAAAGIBwBQAAAAAGIFwBAAAgz/Py8nJ0CXgIEK4AAADgVMypllx9PpPJpGrVqslkMuXq8+b2fiLn5XN0AQAAAMDdTO5uGrRkn/68kOjoUnJM1ZKF9UnnUEeXAYMRrgAAAOB0/ryQqLiz1xxdBmAXTgsEAAAAAAMQrgAAAADAAE4RrhYtWqTIyEgFBwerY8eOOnDggM3569atU8uWLRUcHKy2bdvqhx9+SHf/qFGjFBAQkO6jV69eObkLAAAAAPI4h4ertWvXKioqSgMGDNCKFSsUGBioXr16KSEhIdP5e/fu1dChQ/X8889r5cqVatq0qQYMGKAjR46km9eoUSNt3brV+jFp0qTc2B0AAAAAeZTDF7SYM2eOOnXqpA4dOkiSxowZoy1btmj58uXq06dPhvnz589Xo0aN9Morr0iSBg8erG3btmnhwoV67733rPM8PDzk6+v7QLWZzeYHerxRcntZUEe6n57TH9voj215pT/3+/2M/jiPtBpdodY0vH5soz9Zyyu9kVzjPe2K33+MZM9+OzRcpaSkKC4uTn379rWOubu7KyIiQvv27cv0Mb/++qt69OiRbqxhw4batGlTurFdu3apfv36Klq0qOrVq6fBgwerePHidtUXGxtr1/yc4OXlpWrVqjm6jFxz+PBhJScnZ3s+/bGN/tiWl/pjb28k+uOsnOFnU3bw+rGN/mQtL/VG4vvPw8ah4ery5csym83y9vZON+7t7a1jx45l+piLFy/Kx8cnw/yLFy9abzdq1EjNmzeXn5+fTp06pUmTJql3795aunSpXX8JCQ4OzlN/OXEGAQEBji7BqdEf2+hP1uiNba7QH7PZrNjYWH42OSFXeP04Ev2xzRX6k9e//6Ttf3Y4/LTAnNCmTRvr52kLWjRr1sx6NCu7TCZTnnwBORL9to3+2EZ/skZvbHOl/vCzyfnw9bCN/tjmSv3h+8+9OXRBi+LFi8tkMmVYvCIhISHD0ak0Pj4+6Y5S3Wu+JJUrV07FixfXiRMnHrxoAAAAAMiEQ8OVh4eHqlevru3bt1vHUlNTtX37doWGhmb6mFq1amnHjh3pxrZt26ZatWpl+Tx//fWXrly58sALXAAAAABAVhy+FHvPnj21bNkyrVixQkePHtW7776r5ORktW/fXpI0YsQITZw40Tq/e/fu+umnnzR79mwdPXpUU6dO1W+//aauXbtKkm7cuKHx48fr119/1enTp7V9+3a9+uqrqlChgho1auSQfQQAAADw8HP4NVetW7fWpUuXNGXKFMXHxysoKEjR0dHW0/zOnTsnd/f/y4BhYWH66KOP9PHHH2vSpEmqWLGipk2bJn9/f0l3zgU9cuSIVq5cqevXr6tkyZJq0KCBBg0aJA8PD4fsIwAARvDy8nJ0CQAAGxweriSpa9eu1iNP/7RgwYIMY61atVKrVq0yne/p6alZs2YZWh8AAJkxp1pkcnfLlecymUwOWZ46N/cRAFydU4QrAABckcndTYOW7NOfFxIdXUqOqFqysD7pnPk10ACAjAhXAAA8gD8vJCru7DVHlwEAcAIOX9ACAAAAAB4GhCsAAAAAMADhCgAAAAAMQLgCAAAAAAMQrgAAAADAAIQrAAAAADAA4QoAAAAADEC4AgAAAAADEK4AAAAAwACEKwAAAAAwAOEKAAAAAAxAuAIAAAAAAxCuAAAAAMAAhCsAAAAAMADhCgAAAAAMQLgCAAAAAAMQrgAAAADAAIQrAAAAADAA4QoAAAAADEC4AgAAAAADEK4AAAAAwACEKwAAAAAwAOEKAAAAAAxAuAIAAAAAAxCuAAAAAMAAhCsAAAAAMADhCgAAAAAMQLgCAAAAAAMQrgAAAADAAIQrAAAAADAA4QoAAAAADEC4AgAAAAADEK4AAAAAwACEKwAAAAAwAOEKAAAAAAxAuAIAAAAAAxCuAAAAAMAAhCsAAAAAMADhCgAAAAAMQLgCAAAAAAMQrgAAAADAAIQrAAAAADAA4QoAAAAADEC4AgAAAAADEK4AAAAAwACEKwAAAAAwAOEKAAAAAAxAuAIAAAAAAxCuAAAAAMAAhCsAAAAAMIBThKtFixYpMjJSwcHB6tixow4cOGBz/rp169SyZUsFBwerbdu2+uGHH9Ldb7FY9Mknn6hhw4YKCQlRjx49dPz48RzcAwAAAAB5ncPD1dq1axUVFaUBAwZoxYoVCgwMVK9evZSQkJDp/L1792ro0KF6/vnntXLlSjVt2lQDBgzQkSNHrHNmzpypBQsW6N1339WyZcvk5eWlXr166e+//86t3QIAAACQxzg8XM2ZM0edOnVShw4dVLVqVY0ZM0aenp5avnx5pvPnz5+vRo0a6ZVXXlGVKlU0ePBgVatWTQsXLpR056jV/Pnz1b9/fzVr1kyBgYGaMGGCLly4oE2bNuXmrgEAAADIQ/I58slTUlIUFxenvn37Wsfc3d0VERGhffv2ZfqYX3/9VT169Eg31rBhQ2twOn36tOLj4xUREWG9v0iRIqpZs6b27dunNm3aZLs+s9lsx97kHJPJpKolCzu6jByVtn/303P6Yxv9se1h78+D9EaiP/dCf2yjP7bRn6w97L2RHvz1k5tSU1Pl5eWl1NRUR5fiEPZ8jRwari5fviyz2Sxvb+90497e3jp27Fimj7l48aJ8fHwyzL948aIkKT4+3jqW1Zzsio2NtWt+TsifP7+CqlXXJ51DHV1Kjrt126yDv8fp1q1b2X4M/bGN/tiWV/pzP72R6M+90B/b6I9t9CdreaU30oP1J38+Uw5Wlp7JZFK1atVy7fnS3O/7y5EcGq6cXXBwsEym3Hvh2pKbf9VITU3VH3/8occee0zu7rl35qi7m1S9evX7eiz9sY3+2Paw9+dBeiPRn3vJrf644ntLoj/3Qn+y9rB/75EeoD9uDr+yJ1e4u7s/0PvLKGazOdsHXRwarooXLy6TyZRh8YqEhIQMR6fS+Pj4ZDgCdfd8X19f61jJkiXTzQkMDLSrPpPJ5DThKrclJyfL3d09z+7/vdAf2+iPbfTHNvqTNXpjG/2xjf7YRn+yZjabdfjwYQUEBORqf0zubpJc6+vh0Njr4eGh6tWra/v27dax1NRUbd++XaGhmR8KrlWrlnbs2JFubNu2bapVq5Ykyc/PT76+vum2mZiYqP3792e5TQAAAABZS05OdnQJLsHhxxR79uypZcuWacWKFTp69KjeffddJScnq3379pKkESNGaOLEidb53bt3108//aTZs2fr6NGjmjp1qn777Td17dpVkuTm5qbu3bvrs88+03fffafDhw9rxIgRKlmypJo1a+aQfQQAAADw8HP4NVetW7fWpUuXNGXKFMXHxysoKEjR0dHW0/zOnTuX7tzXsLAwffTRR/r44481adIkVaxYUdOmTZO/v791Tu/evZWcnKy3335b165dU+3atRUdHa0CBQrk+v4BAAAAyBscHq4kqWvXrtYjT/+0YMGCDGOtWrVSq1atstyem5ubBg0apEGDBhlWIwAAAADY4vDTAgEAAADgYUC4AgAAAAADEK4AAAAAwACEKwAAAAAwAOEKAAAAAAxAuAIAAAAAAxCuAAAAAMAAhCsAAAAAMADhCgAAAAAMQLgCAAAAAAMQrgAAAADAAPkcXYAzslgskiSz2ezgShwjbb/z6v7fC/2xjf7YRn9soz9Zoze20R/b6I9t9Me2vN6ftP1Oywi2uFmyMyuPSUlJUWxsrKPLAAAAAOAkgoOD5eHhYXMO4SoTqampun37ttzd3eXm5ubocgAAAAA4iMViUWpqqvLlyyd3d9tXVRGuAAAAAMAALGgBAAAAAAYgXAEAAACAAQhXAAAAAGAAwhUAAAAAGIBwBQAAAAAGIFwBAAAAgAEIVwAAAABgAMIVAAAAABiAcAWr3bt3q1+/fmrYsKECAgK0adMmR5fkVGbMmKEOHTooNDRU9evX16uvvqpjx445uiynsHjxYrVt21ZhYWEKCwvTCy+8oB9++MHRZTmtL774QgEBARo3bpyjS3EKU6dOVUBAQLqPli1bOrosp3L+/HkNGzZMdevWVUhIiNq2bavY2FhHl+UUIiMjM7x+AgICNGbMGEeX5hTMZrM+/vhjRUZGKiQkRM2aNdO0adNksVgcXZrTSExM1Lhx49SkSROFhISoc+fOOnDggKPLcoh7/S5osVj0ySefqGHDhgoJCVGPHj10/PhxxxTrpPI5ugA4j6SkJAUEBKhDhw4aOHCgo8txOrt27VKXLl0UHBwss9msSZMmqVevXlqzZo0KFizo6PIcqnTp0ho2bJgqVKggi8WilStXasCAAVqxYoUee+wxR5fnVA4cOKAlS5YoICDA0aU4lccee0xz5syx3jaZTA6sxrlcvXpVL774ourWrauZM2eqePHiOnHihIoVK+bo0pzC119/LbPZbL39xx9/qGfPngT0/zVz5kx9+eWXGj9+vKpWrarffvtNo0ePVpEiRdS9e3dHl+cU/v3vf+uPP/7QhAkTVLJkSa1atUo9e/bU2rVrVapUKUeXl6vu9bvgzJkztWDBAn3wwQfy8/PTJ598ol69emnt2rUqUKCAAyp2PoQrWDVu3FiNGzd2dBlOa9asWeluf/DBB6pfv77i4uIUHh7uoKqcQ2RkZLrbQ4YM0Zdffqlff/2VcHWXGzduaPjw4Ro7dqw+++wzR5fjVEwmk3x9fR1dhlOaOXOmSpcuraioKOtYuXLlHFiRcylRokS621988YXKly+vxx9/3EEVOZd9+/apadOmevLJJyVJfn5+WrNmTZ49MvNPN2/e1MaNGzV9+nTrz/LXXntN33//vRYvXqwhQ4Y4uMLcZet3QYvFovnz56t///5q1qyZJGnChAmKiIjQpk2b1KZNm9ws1WlxWiBwn65fvy5J/PX4H8xms9asWaOkpCSFhoY6uhyn8t5776lx48aKiIhwdClO58SJE2rYsKGaNm2qoUOH6uzZs44uyWls3rxZNWrU0Ouvv6769evrueee07JlyxxdllNKSUnRqlWr1KFDB7m5uTm6HKcQGhqqHTt26L///a8k6dChQ/rll1/0xBNPOLgy53D79m2ZzeYMR10KFCigvXv3Oqgq53T69GnFx8en+xlWpEgR1axZU/v27XNgZc6FI1fAfUhNTdX777+vsLAw+fv7O7ocp3D48GF17txZf//9twoWLKhp06apatWqji7LaaxZs0a///67vv76a0eX4nRCQkIUFRWlSpUqKT4+XtOmTVOXLl0UExOjwoULO7o8hzt16pS+/PJL9ezZU/369VNsbKzGjh2r/Pnzq127do4uz6ls2rRJ169fpy936dOnjxITE9WqVSuZTCaZzWYNGTJEzzzzjKNLcwqFCxdWaGiopk+frsqVK8vHx0erV6/Wr7/+qvLlyzu6PKcSHx8vSfL29k437u3trYsXLzqiJKdEuALuw5gxY/THH39o8eLFji7FaVSqVEkrV67U9evXtWHDBo0cOVILFy4kYEk6d+6cxo0bp9mzZ3NOeibuPgUlMDBQNWvWVJMmTbRu3Tp17NjRgZU5B4vFoho1auiNN96QJFWrVk1//PGHlixZQoj4h+XLl+uJJ57Ic9fJ2LJu3TrFxMRo4sSJqlq1qg4ePKioqCiVLFmS18//mjBhgt5880098cQTMplMqlatmtq0aaO4uDhHlwYXRLgC7PTee+9py5YtWrhwoUqXLu3ocpyGh4eHKlSoIEmqUaOGYmNjNX/+fL333nsOrszx4uLilJCQoPbt21vHzGazdu/erUWLFik2NpYFHO5StGhRVaxYUSdPnnR0KU7B19dXVapUSTdWuXJlbdiwwUEVOaczZ85o27Ztmjp1qqNLcSoTJkxQnz59rNfDBAQE6OzZs5oxYwbh6n+VL19eCxcuVFJSkhITE1WyZEkNHjyYaxv/Ie262ISEBJUsWdI6npCQoMDAQEeV5XQIV0A2WSwW/c///I++/fZbLViwgG+695CamqqUlBRHl+EU6tWrp5iYmHRjo0ePVuXKldW7d2+C1T/cuHFDp06dYoGL/xUWFma9XibN8ePH9eijjzqoIuf0zTffyNvb27pwA+64efNmhuvPTCYTS7FnomDBgipYsKCuXr2qrVu3avjw4Y4uyan4+fnJ19dX27dvV1BQkKQ7y9jv379fL774ooOrcx6EK1jduHEj3V+KT58+rYMHD6pYsWIqW7asAytzDmPGjNHq1as1ffp0FSpUyHrucZEiReTp6eng6hxr4sSJeuKJJ1SmTBnduHFDq1ev1q5duzKssJhXFS5cOMO1eQULFtQjjzzCNXuSxo8fryZNmqhs2bK6cOGCpk6dKnd3dz399NOOLs0p/Otf/9KLL76ozz//XK1atdKBAwe0bNkyjgrfJTU1Vd98842ee+455cvHrzZ3a9KkiT7//HOVLVvWelrgnDlz1KFDB0eX5jR++uknWSwWVapUSSdPntSECRNUuXLldGcb5BX3+l2we/fu+uyzz1ShQgXrUuwlS5a0rh4Iyc3Cny7wv3bu3Jnp/7xo166dPvjgAwdU5Fyy+r9EUVFRefIb8N3efPNN7dixQxcuXFCRIkUUEBCg3r17q0GDBo4uzWl169ZNgYGBeuuttxxdisMNGTJEu3fv1pUrV1SiRAnVrl1bQ4YM4WLyu3z//feaNGmSjh8/Lj8/P/Xs2VOdOnVydFlOY+vWrerVq5fWr1+vSpUqObocp5KYmKhPPvlEmzZtsp7O1aZNGw0YMEAeHh6OLs8prF27VpMmTdJff/2lRx55RE899ZSGDBmiIkWKOLq0XHev3wUtFoumTJmiZcuW6dq1a6pdu7beeecd3nd3IVwBAAAAgAH4P1cAAAAAYADCFQAAAAAYgHAFAAAAAAYgXAEAAACAAQhXAAAAAGAAwhUAAAAAGIBwBQAAAAAGIFwBAAAAgAEIVwAAlxMQEKBNmzY5uoxcERkZqblz5zq6DABANuRzdAEAgIfHqFGjtGLFCklSvnz5VKpUKbVs2VKDBg1SgQIFDHuerVu3qlixYoZtDwAAIxCuAACGatSokaKionT79m3FxcVp5MiRcnNz0/Dhww17Dl9fX8O2lRelpKTIw8PD0WUAwEOH0wIBAIby8PCQr6+vypQpo2bNmikiIkLbtm2z3p+amqoZM2YoMjJSISEheuaZZ7R+/XrrfU888YQWL16cbpu///67AgMDdebMGUkZTws8d+6cBg0apDp16ujxxx9X//79dfr0aUnSkSNHFBgYqEuXLkmSrly5osDAQA0ZMsT6+OnTp+vFF1/Mcp8iIyP1+eefa/To0QoNDdWTTz6ppUuXWu/fuXOnAgICdO3aNevYwYMHFRAQYK3jm2++UZ06dfT999+rRYsWqlmzpl5//XUlJydrxYoVioyMVHh4uMaOHSuz2Zzu+W/cuKE33nhDtWrVUqNGjbRo0aJ091+7dk1vvfWW6tWrp7CwMHXv3l2HDh2y3j916lQ9++yz+uqrr6x9BwAYj3AFAMgxR44c0b59+5Q/f37r2IwZM7Ry5UqNGTNGa9asUY8ePTR8+HDt2rVL7u7uatOmjVavXp1uOzExMQoLC9Ojjz6a4Tlu3bqlXr16qVChQlq0aJG+/PJLFSxYUK+88opSUlL02GOP6ZFHHtGuXbskSXv27NEjjzyi3bt3W7exe/duPf744zb3Zc6cOapRo4ZWrlypl156Se+++66OHTtmVz9u3rypBQsWaPLkyYqOjtbOnTs1cOBA/fDDD/riiy80YcIELVmyRBs2bEj3uFmzZikwMFArVqxQnz59NG7cOP3888/W+wcNGqSEhATNnDlT33zzjapXr65//etfunLlinXOyZMntWHDBn366adauXKlXXUDALKHcAUAMNSWLVsUGhqq4OBgtW3bVgkJCerVq5ekO6ejzZgxQ++//74aNWqkcuXKqX379nrmmWesR4KeeeYZ7d27V2fPnpV052jWmjVr1LZt20yfb+3atUpNTdW4ceMUEBCgKlWqKCoqSufOndOuXbvk5uam8PBwa7jatWuX2rdvr5SUFB09elS3bt3Svn377hmunnjiCXXp0kUVKlRQ7969Vbx4ce3cudOu3ty6dUvvvvuuqlWrpvDwcLVo0UK//PKLxo0bp6pVq6pJkyaqW7euduzYke5xYWFh6tOnjypVqqRu3bqpRYsW1kUu9uzZowMHDmjKlCkKDg5WxYoVNXLkSBUtWjRdSLt165YmTJigatWqKTAw0K66AQDZwzVXAABD1a1bV++++66Sk5M1d+5cmUwmtWjRQpJ04sQJJScn6+WXX073mFu3bikoKEiSFBQUpCpVqmj16tXq06ePdu3apUuXLqlly5aZPt+hQ4d08uRJhYWFpRv/+++/dfLkSUlSeHi4li1bJunOUaohQ4bo+PHj2rVrl65evarbt29nePw/BQQEWD93c3OTj4+PEhIS7OiM5OXlpfLly1tv+/j46NFHH1WhQoXSjaWdwpimVq1aGW7PmzdPknT48GElJSWpbt266ebcvHnTuv+SVLZsWZUoUcKuegEA9iFcAQAM5eXlpQoVKkiS3n//feu1Ph07dlRSUpKkO6cGlipVKt3j7l5goW3btoqJiVGfPn20evVqNWzYUMWLF8/0+ZKSklS9enV99NFHGe5LCxOPP/643n//fR0/flx//vmnateurWPHjmnXrl26du2aatSoIS8vL5v7lS9f+h+Zbm5uslgskiR39zsngqTdlu4ExuxsI7Ox1NRUm7Xc7caNG/L19dWCBQsy3FekSBHr5/faPwDAgyNcAQByjLu7u/r27asPPvhAbdu2VZUqVeTh4aGzZ8/aPA3v6aef1scff6zffvtNGzZs0JgxY7KcW716da1bt07e3t4qXLhwpnMCAgJUrFgxffbZZwoKClKhQoVUt25dRUdH69q1a/c8JfBe0kJcfHy8dYn4uxeUeFD79+/PcLtKlSqS7uz/xYsXZTKZ5OfnZ9hzAgDsxzVXAIAc1bJlS7m7u2vRokUqXLiwXn75ZUVFRWnFihU6efKk4uLitGDBAuv/x5IkPz8/hYaG6q233pLZbFZkZGSW22/btq2KFy+u/v37a8+ePTp16pR27typsWPH6q+//pJ052hQnTp1FBMTYw1SAQEBSklJ0fbt2xUeHv5A+1i+fHmVKVNGU6dO1fHjx7VlyxbNnj37gbZ5t71792rmzJn673//q0WLFmn9+vXq3r27JCkiIkK1atXSgAEDtHXrVp0+fVp79+7V5MmTFRsba1gNAIB7I1wBAHJUvnz51LVrV0VHRyspKUmDBw/Wq6++qhkzZqh169Z65ZVXtGXLlgxHXdq2batDhw6pefPm8vT0zHL7Xl5eWrhwocqWLauBAweqdevWeuutt/T333+nO5IVHh4us9lsDVfu7u6qU6eO3Nzc7nm91b3kz59fEydO1LFjx/TMM89o5syZGjx48ANt8249e/bUb7/9pnbt2umzzz7TqFGj1KhRI0l3guMXX3yh8PBwjR49Wi1bttQbb7yhM2fOyMfHx7AaAAD35ma5+wRxAAAAAMB94cgVAAAAABiAcAUAAAAABiBcAQAAAIABCFcAAAAAYADCFQAAAAAYgHAFAAAAAAYgXAEAAACAAQhXAAAAAGAAwhUAAAAAGIBwBQAAAAAGIFwBAAAAgAH+P272GJ1dFuauAAAAAElFTkSuQmCC\n"
+          },
+          "metadata": {}
+        }
+      ],
+      "source": [
+        "import numpy as np\n",
+        "import matplotlib.pyplot as plt\n",
+        "\n",
+        "# average similarity of each review with all the others\n",
+        "avg_sim = (sim_matrix.sum(axis=1) - 1) / (sim_matrix.shape[1] - 1)\n",
+        "\n",
+        "plt.figure(figsize=(10,5))\n",
+        "plt.bar(range(1, len(avg_sim)+1), avg_sim)\n",
+        "plt.xlabel(\"Review number\")\n",
+        "plt.ylabel(\"Average cosine similarity\")\n",
+        "plt.title(\"Average similarity per review\")\n",
+        "plt.xticks(range(1, len(avg_sim)+1))\n",
+        "plt.show()"
+      ],
+      "id": "FaA9Z_wLb_Uy"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "FKK4hlnwb_Uz",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 487
+        },
+        "outputId": "50f6a88b-2563-4440-8558-b2d1387360e6"
+      },
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "<Figure size 800x500 with 1 Axes>"
+            ],
+            "image/png": "iVBORw0KGgoAAAANSUhEUgAAArwAAAHWCAYAAACVPVriAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAAWAVJREFUeJzt3XlcVNX/x/E3jCK45oJmUq6BGygumSipuC+4L5lALpmamZYb6rdF0y+ZqeVSVu5muWOiuOY3c0HLJSVySczdFHEXFIX7+8MH83MCjYFhcXo9Hw8fD+bcM+d+5nIG31zOveNgGIYhAAAAwE45ZncBAAAAQGYi8AIAAMCuEXgBAABg1wi8AAAAsGsEXgAAANg1Ai8AAADsGoEXAAAAdo3ACwAAALtG4AUAAIBdI/AC+Fc4e/asPDw8tGrVKpuNuWfPHnl4eGjPnj3mtuDgYPn5+dlsH8k8PDw0ffp0m4/7d35+furXr1+m7yerBAYGKjAw0KZjZsZcehQ/Pz8FBwdb/bzkublhwwab1bJq1Sp5eHjo7NmzNhsTyCoEXuAJsnjxYnl4eKhLly7ZXQqy2f79+zV9+nTduHEju0uBpG3btmXJLyQA0idXdhcAIO3CwsJUqlQpHTp0SKdOnVLp0qWzu6QnRvJxy5XLdj/2ateurUOHDil37tw2G/NRDh06JJPJZH584MABzZgxQx06dFDBggUzff9Pqjlz5th8zNTm0rZt27R48WINGjTI5vsDkHGc4QWeEGfOnNGBAwc0atQoFSlSRGFhYVleQ1JSku7evZvl+7UFBwcH5cmTxyI0ZpSjo6Py5MkjR8fM+VH68PHOkyePTcP6v4WTk5OcnJxsMtb9+/eVkJCQKXMJQOYi8AJPiLCwMBUqVEgNGjRQ8+bNLQLvvXv39MILL2jUqFEpnnfr1i15enpq4sSJ5raEhARNmzZNTZs2VdWqVdWgQQN9/PHHSkhIsHiuh4eHxo0bpzVr1qh169by9PTU9u3bJT04c/byyy+rTp068vLyUseOHVNdL3jnzh2NHz9ederUkbe3t/r376+LFy+muib14sWLGjVqlHx8fFS1alW1bt1aK1asSNPx2blzp7p3765atWrJ29tbzZs315QpU8zbU1t3GRwcLG9vb50/f179+vWTt7e3fH19tXjxYknS0aNHFRQUpOrVq6tRo0YpfslIbQ1vatJ6rB53vB8+XtOnT9fHH38sSWrcuLE8PDzMaysDAgLUtm3bVOto3ry5+vTp80+HUpK0Y8cOtWvXTp6enmrVqpU2bdpk3nbmzBl5eHho/vz5KZ63f/9+eXh4aO3atY8cO/m4hYeHa8aMGfL19ZW3t7feeust3bx5UwkJCZowYYLq1q0rb29vjRo1KsXcXLlypYKCglS3bl1VrVpVrVq10rfffptiX6mt4Y2NjdXo0aPl4+MjT09PtW3bVqGhoRZ9kufLnDlzNH/+fDVp0kSenp6Kjo5OMZeCg4PNcyb5e+Hh4SHDMOTn56cBAwakqOvu3buqWbOm3nvvvUcep9Rcu3ZNEydOlL+/v7y9vVWjRg299tprOnLkSKr9k5KSNGXKFNWrV0/Vq1dX//79deHChRT9Dh48qD59+qhmzZqqVq2aAgICtG/fPqtqA3IyThcAT4iwsDA1bdpUTk5OatOmjb777jsdOnRIXl5eyp07t5o0aaLNmzdr7NixFme0tmzZooSEBLVq1UrSg/8ABwwYoH379qlr164qX768jh07pgULFujkyZP6/PPPLfa7e/durV+/Xj169FDhwoVVqlQpSdLChQvl5+cnf39/3bt3T+vWrdPgwYP15ZdfqmHDhubnBwcHa/369WrXrp2qVaumX375Ra+//nqK13f58mV17dpVDg4O6tGjh4oUKaKffvpJY8aM0a1bt9SzZ89HHps//vhD/fr1k4eHh9566y05OTnp1KlT2r9//z8e18TERPXt21e1atXSsGHDFBYWpnHjxsnFxUVTp06Vv7+/mjVrpiVLlmjkyJGqXr26nn322X8c92FpPVbSo4/3w5o2baqTJ09q7dq1GjVqlAoXLixJKlKkiNq1a6f//Oc/OnbsmNzd3c3POXTokE6ePJlq+Pq7kydP6u2339bLL7+sDh06aOXKlRo8eLBmz56tevXq6dlnn1WNGjW0Zs2aFN+XsLAw5cuXT40bN/7H/Xz11VdydnbW66+/rlOnTumbb75Rrly55ODgoBs3bujNN9/UwYMHtWrVKpUqVUpvvvmm+bnfffednn/+efn5+SlXrlz63//+p7Fjx8owDPXo0eOR+7xz544CAwN1+vRp9ejRQ25ubtqwYYOCg4N148YNvfrqqxb9V61apbt376pr165ycnJSoUKFlJSUZNGnW7duunTpknbu3Gn+RUR68FcFf39/zZkzR9euXdNTTz1l3rZ161bdunXrkb+cPMqZM2e0ZcsWtWjRQm5ubrp8+bKWLl2qgIAArVu3TiVKlLDo/8UXX8jBwUF9+/ZVbGysFixYoJ49e+r777+Xs7OzJCkiIkJ9+/ZV1apV9eabb8rBwUGrVq3Sq6++qm+//VZeXl5W1QjkSAaAHC8yMtJwd3c3du7caRiGYSQlJRkvvfSSMX78eHOf7du3G+7u7sbWrVstntu3b1+jcePG5serV682KlasaPzyyy8W/b777jvD3d3d2Ldvn7nN3d3dqFixovHHH3+kqCk+Pt7icUJCgtGmTRsjKCjI3Pbbb78Z7u7uxoQJEyz6BgcHG+7u7sa0adPMbaNHjzbq1atnXLlyxaLv22+/bdSsWTPF/h42b948w93d3YiNjX1knzNnzhju7u7GypUrzW0jR4403N3djVmzZpnbrl+/bnh5eRkeHh7GunXrzO3R0dEpat69e7fh7u5u7N6922LMRo0aWew7LcfKMB5/vP++79mzZxvu7u7GmTNnLPrduHHD8PT0NCZNmmTR/uGHHxrVq1c3bt++nfLgPKRRo0aGu7u7sXHjRnPbzZs3jXr16hnt27c3ty1ZssRwd3c3jh8/bvG66tSpY4wcOfKx+0g+bm3atDESEhLM7e+8847h4eFhvPbaaxb9u3Xr9o/H1DAMo3fv3hZz3TAMIyAgwAgICDA/nj9/vuHu7m58//33FnV369bNqF69unHz5k3DMP5/vtSoUSPFvEptLo0dO9Zwd3dPUdOJEycMd3d349tvv7Vo79+/v9GoUSMjKSkpxXMe1qhRI4vjeffuXSMxMTFFPVWrVjVmzJhhbks+xr6+vubXZBiGER4ebri7uxsLFiwwDOPBz5JmzZoZvXv3tqglPj7e8PPzM3r16mVuW7lyZapzDngSsKQBeAKEhYWpWLFiqlOnjqQHZ45atWql8PBwJSYmSpJefPFFFS5cWOHh4ebnXb9+Xbt27TKf3ZWkDRs2qHz58ipXrpyuXLli/vfiiy9KUoo/z9euXVsVKlRIUVPy2aHk/dy8eVM1a9bU77//bm5P/nP8K6+8YvHcgIAAi8eGYWjTpk3y8/OTYRgWddWvX183b95UVFTUI49P8kVbP/zwQ4qzb2nx8F0vChYsqLJly8rFxUUtW7Y0t5crV04FCxbUmTNnrB4/Lccq2aOOd1oVKFBAjRs31rp162QYhqQHZ7HXr1+vxo0bK2/evP84RvHixdW0aVPz4/z586t9+/b6/fffFRMTI0lq2bKl8uTJY7HMY8eOHbp69Wqaz1q2a9fO4oI/Ly8vGYahTp06WfTz8vLShQsXdP/+fXPbw8f05s2bunLlil544QWdOXNGN2/efOQ+f/rpJ7m6uqpNmzbmtty5cyswMFBxcXH65ZdfLPo3a9ZMRYoUSdPrSU3ZsmVVrVo1i+N07do1bd++Xf7+/nJwcLBqPCcnJ/Oa8cTERF29elV58+ZV2bJlU51P7du3V/78+c2PW7RoIVdXV23btk2SdPjwYZ08eVL+/v66evWq+X0XFxenunXr6pdffknXewrIaVjSAORwiYmJWrdunerUqWNx/0svLy/NnTtXERERql+/vnLlyqVmzZpp7dq1SkhIkJOTkzZt2qR79+5ZBN5Tp04pOjpadevWTXV/sbGxFo/d3NxS7fe///1PX3zxhQ4fPmyxvvLh/8DPnz8vR0fHFGP8/e4SV65c0Y0bN7R06VItXbo01f1duXIl1XZJatWqlZYvX67//Oc/mjx5surWraumTZuqRYsW/3hBWZ48eVIEmgIFCujpp59OEUYKFCiQrtuApeVYJXvU8bZG+/btFR4err1796p27dratWuXLl++rHbt2qXp+aVLl05RW5kyZSRJ586dk6urqwoWLKhGjRpp7dq1GjJkiKQHv5iVKFHC/MvTP3nmmWcsHhcoUECSVLJkyRTtSUlJunnzpnn5xr59+zR9+nT9+uuvio+Pt+h/8+ZN81h/d+7cOZUuXTrFvChfvrykB3P2Ybb4frRr104ffvihzp07p1KlSmnDhg26d+9emr8fD0tKStLChQv17bff6uzZs+ZfeCVZLJlI9vf3moODg0qXLq1z585JerB8RZJGjhz5yH3evHlThQoVsrpWICch8AI53O7duxUTE6N169Zp3bp1KbaHhYWpfv36kqTWrVtr6dKl+umnn9SkSRNt2LBB5cqVU8WKFc39k5KS5O7unuoFbpL09NNPWzx++Exasr1792rAgAGqXbu23n//fbm6uip37txauXLlYy9WepTkM0ht27ZVhw4dUu3j4eHxyOc7Oztr8eLF2rNnj3788Udt375d4eHhWrp0qebOnfvYq+kfte1R7clnTdPK2mOV2vG2Vv369VWsWDGtWbNGtWvX1po1a+Tq6iofH58Mj/2w9u3ba8OGDdq/f7/c3d21detWde/ePc13rXhUv0e1Jx/706dPq2fPnipXrpyCg4NVsmRJ5c6dW9u2bdP8+fNtekbSFt+P1q1bKyQkRGFhYerfv7/WrFmjqlWrqly5claPNWvWLH322Wfq1KmTBg8erEKFCsnR0VH//e9/rZ6b0v8f0xEjRqhSpUqp9knLXwWAnI7AC+RwYWFhKlq0aKpXc2/evNl8oZqzs7Nq164tV1dXhYeHq0aNGtq9e7f69+9v8ZznnntOR44cUd26da3+c2qyjRs3Kk+ePJozZ47FBXIrV6606PfMM88oKSlJZ8+eNZ8hlB6cZX5YkSJFlC9fPiUlJaU7lDk6Oqpu3bqqW7euRo0apVmzZmnq1Knas2ePzYOeNdJ6rKz1uO+dyWRSmzZtFBoaqmHDhmnLli3q2rVrmm+jderUKRmGYbGP5DOBD19E5+vra75FXrVq1RQfH5+us5bW2rp1qxISEvTFF19YnCX+p7tlSA/qP3r0qJKSkiyC9YkTJySlPOucVo/7fjz11FNq2LChwsLC5O/vr/3792v06NHp2s/GjRtVp04d/fe//7Vov3Hjhvns98P+/l4zDEOnTp0y/wKZfAFm/vz5s/V9AmQ21vACOdidO3e0adMmNWzYUC1atEjxr0ePHrp9+7a2bt0q6UHoa9Gihf73v/9pzZo1un//vsVyBunB2suLFy9q2bJlqe4vLi7uH+symUxycHCw+HPq2bNn9cMPP1j0Sz7z/PfbRX3zzTcpxmvevLk2btyoY8eOpdjf45YzSA/WRP5d8tmqv9/OKqul9VhZy8XFRZIeuV61Xbt2un79ut577z3FxcVZdTeAS5cuafPmzebHt27d0urVq1WpUiW5urqa23PlyqXWrVtr/fr1WrVqldzd3S3+mpBZkoP7w2c0b968maZfIl566SXFxMRYrHW/f/++Fi1apLx586p27drpqin5+/GoJS/t2rXT8ePH9fHHH8tkMql169bp2o/JZEpxJnf9+vW6ePFiqv1Xr16tW7dumR9v2LBBMTExeumllyRJVatW1XPPPae5c+fq9u3bKZ7/T+894EnBGV4gB9u6datu374tPz+/VLdXr15dRYoU0Zo1a8zBtmXLllq0aJGmTZsmd3d389rEZO3atdP69ev1/vvva8+ePapRo4YSExN14sQJbdiwQbNnz5anp+dj62rQoIHmzZun1157TW3atFFsbKy+/fZbPffcczp69Ki5X9WqVdW8eXMtWLBA165dM9+WLPls4cNnxYYOHao9e/aoa9eu6tKliypUqKDr168rKipKERER+vnnnx9Zz8yZM7V37141aNBApUqVMtfz9NNPq2bNmo99LZktrcfKWlWqVJEkTZ06Va1atVLu3LnVqFEj85+fK1euLHd3d/NFisn906JMmTIaM2aMIiMjVbRoUa1cuVKxsbEKCQlJ0bd9+/ZatGiR9uzZo2HDhqX79VijXr16yp07t/r376+XX35Zt2/f1vLly1W0aFHzRXWP0q1bNy1dulTBwcGKiopSqVKltHHjRvNZ14cv8LJG8vEdP3686tevnyLUNmjQQE899ZQ2bNigl156SUWLFk3Xfho2bKiZM2dq1KhR8vb21rFjxxQWFvbIW+UVKlRIr7zyijp27Gi+LVnp0qXVtWtXSQ9+SR4/frz69u2rNm3aqGPHjipRooQuXryoPXv2KH/+/Jo1a1a6agVyEgIvkIOtWbNGefLkUb169VLd7ujoaP5T6dWrV1W4cGHVqFFDJUuW1IULF1Kc3U1+zsyZMzV//nx9//332rx5s1xcXOTm5qbAwECVLVv2H+uqW7euJkyYoK+//lr//e9/5ebmpmHDhuncuXMpQtzEiRNVrFgxrVu3Tps3b5aPj4+mTp2qFi1aWPyJv1ixYlq+fLlmzpypzZs367vvvtNTTz2lChUq/GOQ8vPz07lz57Ry5UrzcXjhhRc0aNCgR168lFWsOVbW8PLy0uDBg7VkyRJt375dSUlJ+uGHHyzWW7Zr106TJk2yeplBmTJl9O677+rjjz/Wn3/+KTc3N02dOlW+vr4p+latWlXPP/+8oqOjrb6nbHqVK1dO06ZN06effmqeX927d1eRIkX+camAs7OzFi1apE8++UShoaG6deuWypYtq5CQEHXs2DHdNTVr1kyBgYFat26d1qxZI8MwLAKvk5OT+cMxMrLso3///oqPj1dYWJjCw8NVuXJlffnll5o8efIj+x89elRfffWVbt++rbp16+r99983n5GWpDp16mjp0qX6/PPP9c033yguLk6urq7y8vJSt27d0l0rkJM4GOlZ5Q4AGXD48GG1b99ekyZNyrKQ9G+0YMEChYSEaOvWrelem5oW7du3V6FChbRgwYJM20d6vfLKK3Jyckr1U+Gy2n//+1+tWLFCO3futAicADIfa3gBZKo7d+6kaFuwYIEcHR3TvV4S/8wwDK1YsUK1a9fO1LAbGRlp/gUmJ4qJiUn1Yq6sdvfuXa1Zs0bNmzcn7ALZgCUNADLV7Nmz9dtvv+nFF1+UyWTSTz/9pJ9++kndunVLcb9VZFxcXJy2bt2qPXv26NixYyk+KtpWjh07pqioKM2dO1eurq6pLp/JTvv379fmzZt1+vRp9e3bN9vqiI2N1a5du7Rx40Zdu3ZNQUFB2VYL8G9G4AWQqby9vbVz5059/vnniouLU8mSJTVo0KAUt0uDbVy5ckVDhw5VwYIF1b9/fzVu3DhT9rNx40bNnDlTZcuW1ZQpU5QnT55M2U96LV++XD/99JNeffXVDK3Nzajjx49r2LBhKlq0qP7zn/888l63ADIXa3gBAABg11jDCwAAALtG4AUAAIBdYw1vKpKSknT//n05Ojqm+6NXAQAAkHkMw1BSUpJy5cpl8VHhqSHwpuL+/fuKjIzM7jIAAADwDzw9PS0+yCg1BN5UJP+W4Onpaf7MdthOYmKiIiMjOb6wCeYTbIn5BFtjTmWe5GP7T2d3JQJvqpKXMZhMJiZnJuL4wpaYT7Al5hNsjTmVedKy/JSL1gAAAGDXCLwAAACwawReAAAA2DUCLwAAAOwagRcAAAB2jcALAAAAu0bgBQAAgF0j8AIAAMCuEXgBAABg1wi8AAAAsGsEXgAAANg1Ai8AAADsGoEXAAAAdo3ACwAAALtG4AUAAMhELi4u2V3Cvx6BFwAAZKnEJCO7S8gyJpNJlStXlslkyu5SskRO/d7myu4CAADAv4vJ0UGDlxzQ8Uu3srsU2FCF4vn12cve2V1Gqgi8AAAgyx2/dEtR529kdxn4l2BJAwAAAOwagRcAAAB2jcALAAAAu0bgBQAAgF0j8AIAAMCuEXgBAABg1wi8AAAAsGsEXgAAANg1Ai8AAADsGoEXAAAAdo3ACwAAALtG4AUAAIBdI/ACAADArhF4AQAAYNcIvAAAALBrBF4AAADYNQIvAAAA7BqBFwAAAHaNwAsAAAC7RuAFAACAXSPwAgAAwK4ReAEAAGDXCLwAAACwawReAAAA2DUCLwAAAOwagRcAAAB2LUcE3sWLF8vPz0+enp7q0qWLDh069Mi+y5Yt0yuvvKLatWurdu3a6tmzZ4r+wcHB8vDwsPjXp0+fzH4ZAAAAyIFyZXcB4eHhCgkJ0dixY1WtWjUtWLBAffr00YYNG1S0aNEU/ffs2aPWrVurRo0acnJy0uzZs9W7d2+tW7dOJUqUMPfz9fVVSEiI+bGTk1OWvB4AAADkLNl+hnfevHnq2rWrOnXqpAoVKmjs2LFydnbWypUrU+0/efJk9ejRQ5UqVVL58uU1fvx4JSUlKSIiwqKfk5OTXF1dzf8KFSqUFS8HAAAAOUy2nuFNSEhQVFSU+vXrZ25zdHSUj4+PDhw4kKYx4uPjdf/+/RSB9ueff1bdunVVsGBBvfjiixoyZIgKFy5sVX2JiYlW9UfaJB9Xji9sgfkEW2I+ZQ2TyZTdJSATZdX7x5r9ZGvgvXr1qhITE1MsXShatKhOnDiRpjE++eQTFS9eXD4+PuY2X19fNW3aVG5ubjpz5oymTJmivn37aunSpVa9ySIjI9PcF9bj+MKWmE+wJeZT5nFxcVHlypWzuwxkoqNHjyo+Pj67y7CQ7Wt4M+Krr75SeHi4Fi5cqDx58pjbW7dubf46+aK1Jk2amM/6ppWnpye/hWaCxMRERUZGcnxhE8wn2BLzCcg4Dw+PLNlP8vs1LbI18BYuXFgmk0mxsbEW7bGxsSpWrNhjnztnzhx99dVXmjdvnipWrPjYvs8++6wKFy6sU6dOWRV4TSYTP/AyEccXtsR8gi0xn4D0y4nvnWy9aM3JyUlVqlSxuOAs+QI0b2/vRz7v66+/1ueff67Zs2fL09PzH/fz119/6dq1a3J1dbVJ3QAAAHhyZPuShl69emnkyJGqWrWqvLy8tGDBAsXHx6tjx46SpBEjRqhEiRIaOnSopAfLGKZNm6bJkyerVKlSiomJkSTlzZtX+fLl0+3btzVjxgw1b95cxYoV05kzZzRp0iSVLl1avr6+2fY6AQAAkD2yPfC2atVKV65c0bRp0xQTE6NKlSpp9uzZ5iUNFy5ckKPj/5+IXrJkie7du6e33nrLYpw333xTgwYNkslk0rFjx7R69WrdvHlTxYsXV7169TR48GDuxQsAAPAvlO2BV5ICAgIUEBCQ6rZFixZZPN66detjx3J2dtacOXNsVhsAAACebNn+wRMAAABAZiLwAgAAwK4ReAEAAGDXCLwAAACwawReAAAA2DUCLwAAAOwagRcAAAB2jcALAAAAu0bgBQAAgF0j8AIAAMCuEXgBAABg1wi8AAAAsGsEXgAAANg1Ai8AAADsGoEXAAAAdo3ACwAAALtG4AUAAIBdI/ACAADArhF4AQAAYNcIvAAAALBrBF4AAADYNQIvAAAA7BqBFwAAAHaNwAsAAAC7RuAFAACAXSPwAgAAwK4ReAEAAGDXCLwAAACwawReAAAA2DUCLwAAAOwagRcAAAB2jcALAAAAu0bgBQAAgF0j8AIAAMCuEXgBAABg1wi8AAAAsGsEXgAAANg1Ai8AAADsGoEXAAAAdo3ACwAAALtG4AUAAIBdI/ACAADArhF4AQAAYNcIvAAAALBrBF4AAADYNQIvAAAA7BqBFwAAAHaNwAsAAAC7ZnXgjYuLy4w6AAAAgExhdeCtV6+eRo0apb1792ZGPQAAAIBNWR14J02apOvXr6tnz55q3ry5vvrqK128eDEzagMAAAAyLJe1T2jSpImaNGmiK1eu6Pvvv9eqVav02WefqX79+urUqZP8/PyUK5fVwwIAAACZIt0XrRUpUkS9evVSWFiYgoODtWvXLr311lvy9fXVZ599pvj4eFvWCQAAAKRLuk/FXr58WaGhoQoNDdX58+fVvHlzde7cWX/99Zdmz56tgwcPau7cubasFQAAALCa1YF306ZNWrVqlXbs2KHy5cvrlVdeUdu2bVWwYEFznxo1aqhVq1Y2LRQAAABID6sD76hRo9S6dWt9++238vLySrVP8eLF1b9//wwXBwAAAGSU1YF3x44dcnFxeWwfZ2dnvfnmm2kec/HixZozZ45iYmJUsWJFvfvuu48M08uWLdPq1av1xx9/SJKqVKmid955x6K/YRiaNm2ali9frhs3bqhGjRr64IMPVKZMmTTXBAAAAPtg9UVrNWrUUGxsbIr2q1evqlKlSlYXEB4erpCQEA0cOFChoaGqWLGi+vTpk+o+JGnPnj1q3bq1Fi5cqCVLlqhkyZLq3bu3xa3Rvv76ay1atEgffPCBli1bJhcXF/Xp00d37961uj4AAAA82awOvIZhpNqekJCg3LlzW13AvHnz1LVrV3Xq1EkVKlTQ2LFj5ezsrJUrV6baf/LkyerRo4cqVaqk8uXLa/z48UpKSlJERIS5voULF2rAgAFq0qSJKlasqI8//liXLl3Sli1brK4PAAAAT7Y0L2lYuHChJMnBwUHLly9X3rx5zduSkpL0yy+/qFy5clbtPCEhQVFRUerXr5+5zdHRUT4+Pjpw4ECaxoiPj9f9+/dVqFAhSdLZs2cVExMjHx8fc58CBQqoWrVqOnDggFq3bp3m+hITE9PcF2mXfFw5vrAF5hNsifmUNUwmU3aXgEyUVe8fa/aT5sA7f/58SQ/OoC5ZskSOjv9/cjh37txyc3PT2LFj016lHiyDSExMVNGiRS3aixYtqhMnTqRpjE8++UTFixc3B9yYmBjzGH8f8/Lly1bVFxkZaVV/WIfjC1tiPsGWmE+Zx8XFRZUrV87uMpCJjh49muM+jyHNgXfr1q2SpMDAQM2YMcN8RjU7ffXVVwoPD9fChQuVJ08em4/v6enJb6GZIDExUZGRkRxf2ATzCbbEfAIyzsPDI0v2k/x+TQur79KwaNEiqwt6lMKFC8tkMqW4QC02NlbFihV77HPnzJmjr776SvPmzVPFihXN7a6uruYxihcvbjHmw/3SwmQy8QMvE3F8YUvMJ9gS8wlIv5z43klT4A0JCdHgwYOVN29ehYSEPLbvqFGj0rxzJycnValSRREREWrSpIkkmS9ACwgIeOTzvv76a82aNUtz5syRp6enxTY3Nze5uroqIiLCfNeIW7du6eDBg+revXuaawMAAIB9SFPg/f3333X//n3z14/i4OBgdQG9evXSyJEjVbVqVXl5eWnBggWKj49Xx44dJUkjRoxQiRIlNHToUEkPljFMmzZNkydPVqlSpcxrdvPmzat8+fLJwcFBQUFB+uKLL1S6dGm5ubnps88+U/Hixc2hGgAAAP8eaQq8Dy9jsOWSBklq1aqVrly5omnTpikmJkaVKlXS7NmzzUsaLly4YHGB3JIlS3Tv3j299dZbFuO8+eabGjRokCSpb9++io+P13vvvacbN26oZs2amj17dqas8wUAAEDOZtUa3nv37qlatWpavXq13N3dbVZEQEDAI5cw/D1gJ1889zgODg4aPHiwBg8ebJP6AAAA8OSy6oMncufOrZIlSyopKSmz6gEAAABsyupPWuvfv7+mTJmia9euZUI5AAAAgG1ZfVuyxYsX69SpU/L19dUzzzxj8YlrkhQaGmqz4gAAAICMsjrwcqcDAAAAPEmsDrxvvvlmZtQBAAAAZAqr1/ACAAAATxKrz/AmJiZq/vz5Wr9+vS5cuKB79+5ZbP/5559tVhwAAACQUVaf4Z0xY4bmzZunVq1a6ebNm+rZs6eaNm0qBwcHljsAAAAgx7H6DG9YWJjGjx+vhg0bavr06WrTpo2ee+45eXh46ODBg5lRIwAAAJBuVp/hvXz5svlT1vLly6ebN29Kkho1aqQff/zRpsUBAAAAGWV14C1RooRiYmIkSc8++6x27twpSYqMjJSTk5NtqwMAAAAyyOolDU2bNlVERISqVaumwMBADR8+XCtWrND58+fVs2fPTCgRAICs4+Likt0lALAxqwPvsGHDzF+3atVKJUuW1K+//qrSpUvLz8/PpsUBAHKGxCRDJkeH7C4j05lMJlWuXDm7y8hS/5bvLf7drA68f+ft7S1vb29b1AIAyKFMjg4avOSAjl+6ld2lwIYqFM+vz17m/3DYvzQF3h9++CHNAzZu3DjdxQAAcq7jl24p6vyN7C4DAKyWpsA7cODANA3m4OCgw4cPZ6ggAAAAwJbSFHiPHDmS2XUAAAAAmcLq25IBAAAAT5I0neFduHChunXrpjx58mjhwoWP7RsUFGSTwgAAAABbSFPgnT9/vvz9/ZUnTx7Nnz//kf0cHBwIvAAAAMhR0hR4t27dmurXAAAAQE7HGl4AAADYNas/eMIwDG3YsEF79uzRlStXlJSUZLF9xowZNisOAAAAyCirA++ECRO0dOlS1alTR8WKFZODAx9HCAAAgJzL6sC7Zs0azZgxQw0aNMiMegAAAACbsnoNb/78+eXm5pYZtQAAAAA2Z3XgHTRokGbOnKk7d+5kRj0AAACATVm9pKFly5Zau3at6tatKzc3N+XKZTlEaGiozYoDAAAAMsrqwDty5EhFRUWpbdu2XLQGAACAHM/qwLtt2zbNnj1btWrVyox6AAAAAJuyeg3v008/rfz582dGLQAAAIDNWR14g4ODNWnSJJ09ezYz6gEAAABsyuolDcOHD1d8fLyaNm0qZ2dn5c6d22L7zz//bLPiAAAAgIyyOvCOHj06M+oAAAAAMoXVgbdDhw6ZUQcAAACQKdIUeG/dumW+UO3WrVuP7csFbQAAAMhJ0hR4a9eurR07dqho0aKqVatWqvfeNQxDDg4OOnz4sM2LBAAAANIrTYF3wYIFKlSokCRp4cKFmVoQAAAAYEtpCrwvvPBCql8DAAAAOZ3V9+H96aeftHfvXvPjxYsXq127dho6dKiuX79u0+IAAACAjLI68E6aNEm3b9+WJB09elQhISFq0KCBzp49q48++sjmBQIAAAAZYfVtyc6ePavy5ctLkjZt2iQ/Pz+98847ioqK0uuvv27zAgEAAICMsPoMb+7cuXXnzh1J0q5du1SvXj1JUqFChf7xlmUAAABAVrP6DG+NGjUUEhKiGjVqKDIyUp9++qkk6eTJk3r66adtXR8AAACQIVaf4X3vvfeUK1cubdy4Ue+//75KlCgh6cHFbL6+vjYvEAAAAMgIq8/wPvPMM/ryyy9TtI8ePdomBQEAAAC2ZPUZXgAAAOBJQuAFAACAXSPwAgAAwK4ReAEAAGDX0h14T506pe3bt5vvyWsYhs2KAgAAAGzF6rs0XL16VW+//bZ2794tBwcHbdq0Sc8++6xGjx6tQoUKKTg4ODPqBAAAANLF6jO8ISEhMplM+vHHH+Xs7Gxub9WqlbZv327T4gAAAICMsvoM786dOzVnzpwUn6pWpkwZnT9/3maFAQAAALZg9RneuLg4izO7ya5duyYnJyebFAUAAADYitWBt1atWlq9erVFW1JSkmbPnq06derYqi4AAADAJqwOvMOHD9eyZcv02muv6d69e5o0aZLatGmjvXv3atiwYVYXsHjxYvn5+cnT01NdunTRoUOHHtn3jz/+0KBBg+Tn5ycPDw/Nnz8/RZ/p06fLw8PD4l+LFi2srgsAAAD2weo1vO7u7tq4caO++eYb5cuXT3FxcWratKl69Oih4sWLWzVWeHi4QkJCNHbsWFWrVk0LFixQnz59tGHDBhUtWjRF//j4eLm5ualFixYKCQl55LjPP/+85s2bZ35sMpmsqgsAAAD2w+rAK0kFChTQgAEDMrzzefPmqWvXrurUqZMkaezYsfrxxx+1cuVKvf766yn6e3l5ycvLS5I0efLkR45rMpnk6uqa4foAAADw5EtX4L1x44YOHTqk2NjYFB840b59+zSNkZCQoKioKPXr18/c5ujoKB8fHx04cCA9ZZmdOnVK9evXV548eVS9enUNHTpUzzzzjNXjJCYmZqgOpC75uHJ8YQvMp6zBX8rsW1a/f5hP9i2r5pM1+7E68G7dulXDhg1TXFyc8ufPLwcHB/M2BweHNAfeq1evKjExMcXShaJFi+rEiRPWlmXm5eWlkJAQlS1bVjExMZo5c6Z69OihsLAw5c+f36qxIiMj010H/hnHF7bEfMo8Li4uqly5cnaXgUx09OhRxcfHZ8m+mE/2LyvnU1pZHXgnTpyoTp066Z133pGLi0tm1JQhDRo0MH9dsWJFVatWTY0aNdL69evVpUsXq8by9PTkt9BMkJiYqMjISI4vbIL5BGSch4dHdpcAO5JV8yn5539aWB14L168qKCgoAyH3cKFC8tkMik2NtaiPTY2VsWKFcvQ2A8rWLCgypQpo9OnT1v9XJPJxH+gmYjjC1tiPgHpx3sHtpQT55PVtyWrX7++Tf506OTkpCpVqigiIsLclpSUpIiICHl7e2d4/GS3b9/WmTNnuIgNAADgX8rqM7wNGjTQpEmTFB0dLXd3d+XKZTlE48aN0zxWr169NHLkSFWtWlVeXl5asGCB4uPj1bFjR0nSiBEjVKJECQ0dOlTSgwvdoqOjzV9fvHhRhw8fVt68eVW6dGlJD5ZcNGrUSM8884wuXbqk6dOny9HRUW3atLH2pQIAAMAOWB143333XUnSzJkzU2xzcHDQ4cOH0zxWq1atdOXKFU2bNk0xMTGqVKmSZs+ebV7ScOHCBTk6/v9J6EuXLllcFDd37lzNnTtXL7zwghYtWiRJ+uuvv/TOO+/o2rVrKlKkiGrWrKlly5apSJEi1r5UAAAA2AGrA++RI0dsWkBAQIACAgJS3ZYcYpO5ubnp6NGjjx1v6tSpNqsNAAAATz6r1/ACAAAAT5I0neFduHChunXrpjx58mjhwoWP7RsUFGSTwgAAAABbSFPgnT9/vvz9/ZUnTx7Nnz//kf0cHBwIvAAAAMhR0hR4t27dmurXAAAAQE6X4TW8iYmJOnz4sK5fv26LegAAAACbsjrwTpgwQcuXL5f0IOz26NFDHTp0UMOGDbVnzx6bFwgAAABkhNWBd+PGjapYsaIk6X//+5/OnTun9evX69VXX+WWYAAAAMhxrA68V69eNX9M77Zt29SiRQuVLVtWnTp10rFjx2xeIAAAAJARVgfeYsWK6fjx40pMTNT27dtVr149SdKdO3dkMplsXiAAAACQEVZ/0lrHjh01ZMgQubq6ysHBQT4+PpKkgwcPqly5cjYvEAAAAMgIqwPvoEGD9Pzzz+uvv/5SixYt5OTkJEkymUzq27evzQsEAAAAMsLqwCtJLVq0SNHWoUOHDBcDAAAA2Fq6Au/PP/+suXPnKjo6WpJUvnx5vfbaa6pVq5ZNiwMAAAAyyuqL1r7//nv16tVLzs7OCgwMVGBgoJydndWzZ0+FhYVlRo0AAABAull9hnfWrFkaPny4evbsaW4LCgrSvHnz9Pnnn8vf39+W9QEAAAAZYvUZ3jNnzqhRo0Yp2v38/HT27FmbFAUAAADYitWBt2TJkoqIiEjRvmvXLpUsWdImRQEAAAC2YvWShl69emn8+PE6fPiwvL29JUn79+9XaGioxowZY/MCAQAAgIywOvC+8sorcnV11dy5c7VhwwZJUrly5TR16lQ1adLE5gUCAAAAGZGu25I1bdpUTZs2tXUtAAAAgM1ZvYb30KFDOnjwYIr2gwcPKjIy0iZFAQAAALZideAdN26cLly4kKL94sWLGjdunE2KAgBruLi4ZHcJAIAczOrAGx0drSpVqqRor1Spko4fP26TogBkXGKSkd0lZAmTyaTKlSvLZDJldylZ4t/yfQUAW7J6Da+Tk5MuX76sZ5991qI9JiZGuXKla0kwgExgcnTQ4CUHdPzSrewuBTZSoXh+ffayd3aXAQBPHKsTar169TRlyhR9/vnnKlCggCTpxo0bmjp1qnx8fGxeIID0O37plqLO38juMgAAyFZWB96RI0eqR48eatSokSpVqiRJOnLkiIoWLaqPP/7Y5gUCAAAAGWF14C1RooTWrFmjsLAwHTlyRM7OzurUqZNat26t3LlzZ0aNAAAAQLqla9Ft3rx51a1bN1vXAgAAANic1XdpAAAAAJ4kBF4AAADYNQIvAAAA7BqBFwAAAHYtXYH3xo0bWr58uSZPnqxr165JkqKionTx4kVb1gYAAABkmNV3aThy5Ih69eqlAgUK6Ny5c+rataueeuopbdq0SRcuXOBevAAAAMhRrD7D+9FHH6lDhw7atGmTnJyczO0NGjTQ3r17bVocAAAAkFFWB97IyEi9/PLLKdpLlCihmJgYmxQFAAAA2IrVgdfJyUm3bt1K0X7y5EkVKVLEJkUBAAAAtmJ14PXz89PMmTN17949c9v58+f1ySefqFmzZjYtDgAAAMgoqwNvcHCw4uLi5OPjo7t37yowMFDNmjVTvnz59Pbbb2dGjQAAAEC6WX2XhgIFCmjevHnau3evjh49qri4OFWpUkU+Pj6ZUR8AAACQIVYH3mS1atVSrVq1bFkLAAAAYHNWB96FCxem2u7g4KA8efLoueeeU+3atWUymTJcHAAAAJBRVgfe+fPn6+rVq4qPj1ehQoUkSdevX5eLi4vy5s2r2NhYPfvss1q4cKFKlixp84IBAAAAa1h90do777yjqlWratOmTdqzZ4/27NmjjRs3ysvLS2PGjNGPP/6oYsWKKSQkJDPqBQAAAKxideD99NNPNXr0aD333HPmttKlS2vkyJGaPHmynn76aQ0fPlz79++3aaEAAABAelgdeGNiYnT//v0U7ffv39fly5clScWLF9ft27czXh0AAACQQVYH3jp16uj999/X77//bm77/fff9cEHH+jFF1+UJB07dkxubm62qxIAAABIJ6svWpswYYJGjBihjh07KleuB09PTExU3bp1NWHCBElS3rx5NXLkSNtWCgAAAKSD1YHX1dVV8+bNU3R0tE6ePClJKlu2rMqVK2fuk3ymFwAAAMhu6f7gifLly6t8+fK2rAUAAACwuXQF3r/++ks//PCDLly4oHv37llsGzVqlE0KAwAAAGzB6sAbERGhAQMG6Nlnn9WJEyf0/PPP69y5czIMQ5UrV86MGgEAAIB0s/ouDZMnT1bv3r0VFhYmJycnTZ8+XT/++KNq166tFi1aZEaNAAAAQLpZHXijo6PVvn17SVKuXLl0584d5cuXT4MHD9bs2bNtXR8AAACQIVYH3rx585rX7bq6uur06dPmbVevXrVdZQAAAIANWB14q1Wrpn379kmSGjRooIkTJ+qLL77Q6NGjVa1aNasLWLx4sfz8/OTp6akuXbro0KFDj+z7xx9/aNCgQfLz85OHh4fmz5+f4TEBAABg36wOvKNGjZKXl5ckadCgQXrxxRcVHh6uUqVKmT94Iq3Cw8MVEhKigQMHKjQ0VBUrVlSfPn0UGxubav/4+Hi5ublp6NChcnV1tcmYAAAAsG9WBd7ExET99ddfeuaZZyQ9WN4wbtw4hYWFafr06SpVqpRVO583b566du2qTp06qUKFCho7dqycnZ21cuXKVPt7eXlp5MiRat26tZycnGwyJgAAAOybVbclM5lM6t27t8LDw1WwYMEM7TghIUFRUVHq16+fuc3R0VE+Pj46cOBAjhgzMTExXXXg8ZKPK8c3c5lMpuwuAZkkO947zCf7ltVzivlk37JqPlmzH6vvw/v888/r7NmzevbZZ619qoWrV68qMTFRRYsWtWgvWrSoTpw4kSPGjIyMTFcdSBuOb+ZxcXHhvth27OjRo4qPj8+y/TGf7F9Wzinmk/3L6p9RaWF14B0yZIgmTpyowYMHq0qVKsqbN6/F9vz589usuOzm6enJb6GZIDExUZGRkRxfIJ08PDyyuwTYGeYUbCmr5lNynkgLqwPv66+/LkkaMGCAHBwczO2GYcjBwUGHDx9O0ziFCxeWyWRKcTFZbGysihUrZm1ZmTKmyWQikGUiji+QPrxvYGvMKdhSTpxPVgfehQsX2mTHTk5OqlKliiIiItSkSRNJUlJSkiIiIhQQEJBjxgQAAMCTzerA+8ILL9hs57169dLIkSNVtWpVeXl5acGCBYqPj1fHjh0lSSNGjFCJEiU0dOhQSQ8uSouOjjZ/ffHiRR0+fFh58+ZV6dKl0zQmAAAA/l2sDryStHfvXi1ZskRnz57VZ599phIlSmj16tVyc3NTrVq10jxOq1atdOXKFU2bNk0xMTGqVKmSZs+ebV5+cOHCBTk6/v+d0y5dumT+WGNJmjt3rubOnasXXnhBixYtStOYAAAA+HexOvBu3LhRI0aMkL+/v6KiopSQkCBJunXrlr788kurAq8kBQQEPHK5QXKITebm5qajR49maEwAAAD8u1j9SWtffPGFxo4dq/HjxytXrv/PyzVq1NDvv/9u0+IAAACAjLI68P7555+pnsUtUKCAbty4YZOiAAAAAFuxOvAWK1ZMp0+fTtG+b9++DH8YBQAAAGBrVgferl27asKECTp48KAcHBx08eJFrVmzRhMnTlT37t0zo0YAAAAg3dL1wRNJSUnq2bOn4uPjFRAQICcnJ/Xu3VuBgYGZUSMAAACQblYHXgcHBw0YMEB9+vTR6dOnFRcXp/LlyytfvnyZUR8AAACQIVYvafj+++8VHx8vJycnVahQQV5eXoRdAAAA5FhWB96QkBD5+Pho6NCh2rZtmxITEzOjLgAAAMAmrF7SsGPHDm3fvl1r167VkCFD5OzsrBYtWsjf3181atTIjBoBAACAdLM68ObKlUuNGjVSo0aNFB8fr82bN2vt2rUKCgrS008/rS1btmRGnQAAAEC6WB14H+bi4qL69evrxo0bOn/+vKKjo21VFwAAAGAT6Qq8yWd2w8LCFBERoZIlS6p169b67LPPbF0fAAAAkCFWB963335bP/74o5ydndWyZUu98cYb8vb2zozaAAAAgAyzOvA6Ojrq008/Vf369WUymSy2HTt2TO7u7jYrDgAAAMgoqwPv5MmTLR7funVL69at0/LlyxUVFaXDhw/brDgAAAAgo9J90dovv/yiFStWaNOmTSpevLiaNm2q9957z5a1AQAAABlmVeCNiYlRaGioVqxYoVu3bqlly5ZKSEjQzJkzVaFChcyqEQAAAEi3NAfe/v3765dfflHDhg01evRo+fr6ymQyacmSJZlZHwAAAJAhaQ68P/30kwIDA9W9e3eVKVMmE0sCAAAAbMcxrR2//fZb3b59Wx07dlSXLl30zTff6MqVK5lZGwAAAJBhaQ681atX1/jx47Vjxw5169ZN69at00svvaSkpCTt3LlTt27dysw6AQAAgHRJc+BNljdvXnXu3Fnfffed1qxZo169eunrr7+Wj4+P+vfvnxk1AgAAAOlmdeB9WLly5TRixAht27ZNU6ZMsVVNAAAAgM2k+z68DzOZTGrSpImaNGlii+EAAAAAm8nQGV4AAAAgpyPwAgAAwK4ReAEAAGDXCLwAAACwawReAAAA2DUCLwAAAOwagRcAAAB2jcALAAAAu0bgBQAAgF0j8AIAAMCuEXgBAABg1wi8AAAAsGsEXgAAANg1Ai8AAADsGoEXAAAAdo3ACwAAALtG4AUAAIBdI/ACAADArhF4AQAAYNcIvAAAALBrBF4AAADYNQIvAAAA7BqBFwAAAHaNwAsAAAC7RuAFAACAXSPwAgAAwK4ReAEAAGDXCLwAAACwawReAAAA2DUCLwAAAOwagRcAAAB2jcALAAAAu5YjAu/ixYvl5+cnT09PdenSRYcOHXps//Xr16tFixby9PSUv7+/tm3bZrE9ODhYHh4eFv/69OmTmS8BAAAAOVS2B97w8HCFhIRo4MCBCg0NVcWKFdWnTx/Fxsam2n///v0aOnSoOnfurNWrV6tx48YaOHCgjh07ZtHP19dXO3bsMP+bMmVKVrwcAAAA5DDZHnjnzZunrl27qlOnTqpQoYLGjh0rZ2dnrVy5MtX+CxculK+vr1577TWVL19eQ4YMUeXKlfXNN99Y9HNycpKrq6v5X6FChbLi5QAAACCHyZWdO09ISFBUVJT69etnbnN0dJSPj48OHDiQ6nN+/fVX9ezZ06Ktfv362rJli0Xbzz//rLp166pgwYJ68cUXNWTIEBUuXNiq+hITE63qj7RJPq4c38xlMpmyuwRkkux47zCf7FtWzynmk33LqvlkzX6yNfBevXpViYmJKlq0qEV70aJFdeLEiVSfc/nyZRUrVixF/8uXL5sf+/r6qmnTpnJzc9OZM2c0ZcoU9e3bV0uXLrXqTRYZGWnFq4G1OL6Zx8XFRZUrV87uMpBJjh49qvj4+CzbH/PJ/mXlnGI+2b+s/hmVFtkaeDNL69atzV8nX7TWpEkT81nftPL09OS30EyQmJioyMhIji+QTh4eHtldAuwMcwq2lFXzKTlPpEW2Bt7ChQvLZDKluEAtNjY2xVncZMWKFbM4m/tP/SXp2WefVeHChXXq1CmrAq/JZCKQZSKOL5A+vG9ga8wp2FJOnE/ZetGak5OTqlSpooiICHNbUlKSIiIi5O3tnepzqlevrt27d1u07dq1S9WrV3/kfv766y9du3ZNrq6uNqkbAAAAT45sv0tDr169tGzZMoWGhio6OloffPCB4uPj1bFjR0nSiBEjNHnyZHP/oKAgbd++XXPnzlV0dLSmT5+u3377TQEBAZKk27dva+LEifr111919uxZRURE6I033lDp0qXl6+ubLa8RAAAA2Sfb1/C2atVKV65c0bRp0xQTE6NKlSpp9uzZ5iUKFy5ckKPj/+fyGjVq6JNPPtGnn36qKVOmqEyZMpo5c6bc3d0lPTiNfuzYMa1evVo3b95U8eLFVa9ePQ0ePFhOTk7Z8hoBAACQfbI98EpSQECA+Qzt3y1atChFW8uWLdWyZctU+zs7O2vOnDk2rQ+25+Likt0lAACAf4lsX9KABxKTjOwuIcuYTCZVrlw5Ry5qzwz/pu8tAAA5UY44wwvJ5OigwUsO6PilW9ldCmyoQvH8+uzl1C/ABAAAWYPAm4Mcv3RLUedvZHcZAAAAdoUlDQAAALBrBF4AAADYNQIvAAAA7BqBFwAAAHaNwAsAAAC7RuAFAACAXSPwAgAAwK4ReAEAAGDXCLwAAACwawReAAAA2DUCLwAAAOwagRcAAAB2jcALAAAAu0bgBQAAgF0j8AIAAMCuEXgBAABg1wi8AAAAsGsEXgAAANg1Ai8AAADsGoEXAAAAdo3ACwAAALtG4AUAAIBdI/ACAADArhF4AQAAYNcIvAAAALBrBF4AAADYNQIvAAAA7BqBFwAAAHaNwAsAAAC7RuAFAACAXSPwAgAAwK4ReAEAAGDXCLwAAACwawReAAAA2DUCLwAAAOwagRcAAAB2jcALAAAAu0bgBQAAgF0j8AIAAMCuEXgBAABg1wi8AAAAsGsEXgAAANg1Ai8AAADsGoEXAAAAdo3ACwAAALtG4AUAAIBdI/ACAADArhF4AQAAYNcIvAAAALBrBF4AAADYNQIvAAAA7BqBFwAAAHaNwAsAAAC7liMC7+LFi+Xn5ydPT0916dJFhw4demz/9evXq0WLFvL09JS/v7+2bdtmsd0wDH322WeqX7++vLy81LNnT508eTITXwEAAAByqmwPvOHh4QoJCdHAgQMVGhqqihUrqk+fPoqNjU21//79+zV06FB17txZq1evVuPGjTVw4EAdO3bM3Ofrr7/WokWL9MEHH2jZsmVycXFRnz59dPfu3ax6WQAAAMghsj3wzps3T127dlWnTp1UoUIFjR07Vs7Ozlq5cmWq/RcuXChfX1+99tprKl++vIYMGaLKlSvrm2++kfTg7O7ChQs1YMAANWnSRBUrVtTHH3+sS5cuacuWLVn50gAAAJAD5MrOnSckJCgqKkr9+vUztzk6OsrHx0cHDhxI9Tm//vqrevbsadFWv359c5g9e/asYmJi5OPjY95eoEABVatWTQcOHFDr1q3/sS7DMMz1mUwma19WuphMJlV6Op/yZM3ukEXKueZTYmKiEhMTs3zfzCn7w3yCrWXXnGI+2aesnk/J+0nObY+TrYH36tWrSkxMVNGiRS3aixYtqhMnTqT6nMuXL6tYsWIp+l++fFmSFBMTY257VJ9/kpSUJEn6/fff09TfVrqXl1Q+b5buE5nN0K+//ppte2dO2RvmE2wt++YU88keZc98Ss5tj5OtgTenypUrlzw9PeXo6CgHB4fsLgcAAAB/YxiGkpKSlCvXP8fZbA28hQsXlslkSnGBWmxsbIqzuMmKFSuW4kztw/1dXV3NbcWLF7foU7FixTTV5ejoKCcnpzS/DgAAAORc2XrRmpOTk6pUqaKIiAhzW1JSkiIiIuTt7Z3qc6pXr67du3dbtO3atUvVq1eXJLm5ucnV1dVizFu3bungwYOPHBMAAAD2K9vv0tCrVy8tW7ZMoaGhio6O1gcffKD4+Hh17NhRkjRixAhNnjzZ3D8oKEjbt2/X3LlzFR0drenTp+u3335TQECAJMnBwUFBQUH64osv9MMPP+jo0aMaMWKEihcvriZNmmTLawQAAED2yfY1vK1atdKVK1c0bdo0xcTEqFKlSpo9e7Z5icKFCxfk6Pj/ubxGjRr65JNP9Omnn2rKlCkqU6aMZs6cKXd3d3Ofvn37Kj4+Xu+9955u3LihmjVravbs2cqTJ0+Wvz4AAABkLwcjLfdyAAAAAJ5Q2b6kAQAAAMhMBF4AAADYNQIvAAAA7BqBF3bDz89P8+fPz+4ykINNnz5d7dq1y+4ykAPt2bNHHh4eunHjxmP78XMGf5fWuZNRHh4e2rJlS6buw54ReJFtAgMDNWHChOwuA3Yqtf8cevfuTVhBqry9vbVjxw4VKFBAkrRq1SrVqlUrRb8VK1aoW7duWV0ecrC/zx3kTNl+WzLgcQzDUGJiYpo+NhD4J/ny5VO+fPmyuwzkQE5OTuZP6nycIkWKZEE1eJKkde4ge3GGF6kKDAzU+PHj9fHHH+uFF15QvXr1NH36dPP2GzduaMyYMXrxxRdVo0YNBQUF6ciRI+btwcHBeuONNyzGnDBhggIDA83bf/75Zy1cuFAeHh7y8PDQ2bNnzX8a2rZtmzp27ChPT0/t27dPp0+f1oABA+Tj4yNvb2916tRJu3btypqDAatkdO5I0ueff666devK29tbY8aM0SeffGKxFOHQoUPq1auX6tSpo5o1ayogIEBRUVHm7X5+fpKkgQMHysPDw/z44SUNO3bskKenZ4o/Q44fP15BQUHmx3v37tUrr7wiLy8vNWjQQOPHj1dcXJyNjhasERgYqHHjxmncuHGqWbOm6tSpo08//VTJd9e8fv26RowYodq1a6tatWp67bXXdPLkSfPzz507p/79+6t27dqqXr26WrdurW3btkmy/LP0nj17NGrUKN28edP88yl5Dj+8pGHo0KEaMmSIRY337t1TnTp1tHr1akkPPj30yy+/lJ+fn7y8vNS2bVtt2LAhU48TMiYwMFAffvihJkyYoNq1a8vHx0fLli1TXFycRo0aJW9vbzVt2jTVuSNJo0aNkr+/vxISEiRJCQkJat++vUaMGGHex5YtW9ShQwd5enqqcePGmjFjhu7fv2/efvLkSfXo0UOenp5q1aqVdu7cmYVHwD4RePFIoaGhyps3r5YtW6bhw4dr5syZ5jfd4MGDFRsbq6+//lqrVq1SlSpV9Oqrr+ratWtpGnvMmDHy9vZW165dtWPHDu3YsUMlS5Y0b588ebKGDh2q8PBweXh4KC4uTg0aNND8+fMVGhoqX19f9e/fX+fPn8+Ml44MysjcWbNmjWbNmqVhw4Zp1apVKlmypL777juL8W/fvq327dvr22+/1bJly1S6dGm9/vrrunXrlqQHf3aWpJCQEO3YscP8+GF169ZVwYIFtXHjRnNbYmKi1q9fL39/f0nS6dOn1bdvXzVr1kxr1qzR1KlTtW/fPn344Yc2P2ZIm9DQUJlMJi1fvlxjxozR/PnztXz5ckkPfpH+7bff9MUXX2jp0qUyDEOvv/667t27J0kaN26cEhIS9M033ygsLEzDhg1T3rx5U+zD29tbo0ePVv78+c0/n3r37p2in7+/v/73v//p9u3b5rYdO3bozp075k/2/PLLL7V69WqNHTtW69atU8+ePTV8+HD9/PPPmXF4YCOhoaEqXLiwli9froCAAH3wwQcaPHiwvL29FRoaqnr16mnEiBGKj49P8dz//Oc/io+P1yeffCJJmjp1qm7cuKH33ntP0oNfokeOHKmgoCCFh4dr3LhxWrVqlWbNmiXpwS9JgwYNUu7cubV8+XKNHTvWPBYywABSERAQYHTv3t2irVOnTsakSZOMX375xahRo4Zx9+5di+1NmjQxlixZYhiGYYwcOdIYMGCAxfbx48cbAQEBFvsYP368RZ/du3cb7u7uxubNm/+xxtatWxuLFi0yP27UqJExb968NL0+ZJ6Mzp0uXboYY8eOtdj+8ssvG23btn3kPhMTEw1vb29j69at5rbU5tG0adMsxhk/frwRFBRkfrx9+3ajatWqxvXr1w3DMIzRo0cb7777rsUYv/zyi1GxYkXjzp07j6wHmSMgIMBo2bKlkZSUZG6bNGmS0bJlS+PPP/803N3djX379pm3XblyxfDy8jLCw8MNwzCMNm3aGNOnT0917OSfPcnf+5UrVxo1a9ZM0e/hnzP37t0z6tSpY4SGhpq3v/POO8aQIUMMwzCMu3fvGtWqVTP2799vMcbo0aONd955x/oDgCzx959h9+/fN6pXr24MHz7c3Hbp0iXD3d3dOHDgQIq5YxiGsX//fqNKlSrGp59+alSuXNn45ZdfzNteffVVY9asWRb7XL16tVGvXj3DMB78HKpcubLx119/mbdv27Ytzf83InUsjMQjeXh4WDx2dXVVbGysjh49qri4ONWpU8di+507d3T69Gmb7NvT09Pi8e3btzVjxgz9+OOPiomJUWJiou7cucMZ3hwqI3Pnzz//1CuvvGKx3cvLS7t37zY/vnz5sj799FP9/PPPio2NVVJSkuLj462eD/7+/urWrZsuXryoEiVKKCwsTA0bNlTBggUlSUeOHNHRo0cVFhZmfo5hGEpKStLZs2dVvnx5q/aHjKtWrZocHBzMj6tXr6558+bp+PHjypUrl6pVq2beVrhwYZUtW1bR0dGSpKCgIH3wwQfasWOHfHx81KxZM1WsWDHdteTKlUstW7ZUWFiY2rdvr7i4OP3www+aMmWKJOnUqVOKj49PcXb43r17qlSpUrr3i8z38M8wk8mkp556Su7u7ua2YsWKSZJiY2OVP3/+FM/39vZW79699fnnn6tv374WF0AeOXJE+/fvN5/RlR78denu3buKj49XdHS0nn76aZUoUcJiPGQMgReP9PcLxRwcHGQYhm7fvi1XV1ctWrQoxXOSr1JN7vuwh9cn/RMXFxeLxxMnTtSuXbs0cuRIPffcc3J2dtZbb71l/lMlcpaMzJ20GDlypK5du6YxY8bomWeekZOTk7p162b1fPDy8tJzzz2n8PBwde/eXZs3b9ZHH31k3h4XF6eXX37ZvPb8YQ8vwcGToUuXLqpfv75+/PFH7dy5U1999ZVGjhyZ6vc3rfz9/RUYGKjY2Fjt3LlTefLkka+vrySZ13p/+eWXFuFFenChE3Ku1H6GPdyW/EvX3/+fS5aUlKT9+/fLZDKlOBEUFxenQYMGqVmzZimelydPnoyWjkcg8MJqVapU0eXLl2UymeTm5pZqnyJFiuiPP/6waDt8+LBy585tfpw7d24lJSWlaZ8HDhxQhw4d1LRpU0kPzvieO3cuna8A2SUtc6ds2bKKjIxU+/btzW2RkZEWffbv36/3339fDRo0kCRduHBBV69eteiTO3duJSYm/mNN/v7+CgsLU4kSJeTo6KiGDRuat1WuXFnHjx9X6dKl0/gKkdkOHTpk8fjgwYMqXbq0KlSooPv37+vgwYOqUaOGJOnq1av6888/VaFCBXP/kiVLqnv37urevbsmT56sZcuWpRp40zp/atSooaefflrh4eH66aef1KJFC/PPufLly8vJyUnnz5/XCy+8kJGXjSfM7NmzdeLECS1atEivvfaaVq5cqU6dOkl68HPlzz//fOTPlfLly+uvv/7SpUuXVLx4cUnSr7/+mlWl2y0uWoPVfHx8VL16dQ0cOFA7duzQ2bNntX//fk2dOtUcTF588UX99ttvWr16tU6ePKlp06alCMClSpXSwYMHdfbsWV25cuWx4bd06dLavHmzDh8+rCNHjmjo0KFpDsvIOdIydwICArRixQqFhobq5MmT+vzzz3X06FGLP2OXKVNGa9asUXR0tA4ePKhhw4bJ2dnZYl+lSpVSRESEYmJidP369UfW5O/vr6ioKM2aNUvNmze3OPPWt29fHThwQOPGjdPhw4d18uRJbdmyRePGjbPxkUFanT9/XiEhITpx4oTWrl2rb775RkFBQSpTpowaN26sd999V3v37tWRI0c0fPhwlShRQo0bN5b04E4x27dv15kzZxQVFaU9e/Y8cllKqVKlFBcXp4iICF25ciXVi5OStWnTRkuWLNGuXbvMFzxKUv78+dW7d2+FhIQoNDRUp0+fVlRUlBYtWqTQ0FDbHhjkGL///rumTZum8ePHq2bNmgoODtaECRN05swZSQ/uHvP9999rxowZ+uOPPxQdHa1169Zp6tSpkh78nCxTpoyCg4N15MgR7d2717wN6UfghdUcHBz01VdfqXbt2ho1apRatGihd955R+fOnTOva/L19dUbb7yhSZMmqXPnzuar6h/Wu3dvmUwmtW7dWnXr1n3s+svg4GAVLFhQL7/8svr37y9fX19VqVIlM18mMkFa5k7btm31+uuva+LEierQoYPOnj2rDh06WPypb8KECbp+/bo6dOigESNGKDAwUEWLFrXY18iRI7Vr1y41bNhQHTp0eGRNpUuXlpeXl44ePWoRViSpYsWKWrRokU6ePKlXXnlFHTp00LRp08xnXZD12rdvrzt37qhLly4aN26cgoKCzB8EERISoipVqqh///7q1q2bDMPQV199ZT7jmpSUpHHjxqlVq1Z67bXXVKZMGb3//vup7qdGjRp6+eWXNWTIENWtW1ezZ89+ZE1t27bV8ePHVaJECdWsWdNi25AhQ/TGG2/oyy+/NO/3xx9/fORfOPBku3v3roYPH66OHTuab4fYrVs31alTR8OHD1diYqJ8fX01a9Ys7dixQ507d1bXrl01f/58lSpVSpLk6OioGTNm6M6dO+rcubPGjBmjt99+Oztfll1wMB61AAUAcohevXqpWLFimjRpUnaXgmwUGBioihUrasyYMdldCoAnDGt4AeQo8fHxWrJkierXry9HR0etW7dOu3bt0rx587K7NADAE4rACyBHcXBw0LZt2zRr1izdvXtXZcuW1fTp0+Xj45PdpQEAnlAsaQAAAIBd46I1AAAA2DUCLwAAAOwagRcAAAB2jcALAAAAu0bgBQAAgF0j8AJADuDh4aEtW7ZkeBw/Pz/Nnz8/4wVlcNxVq1apVq1aGd6vrY4LgH83Ai8AWCk4OFgeHh567733UmwbO3asPDw8FBwcbNWYO3bs0EsvvZTh2lasWGH+qF2JwAgAEoEXANKlZMmSCg8P1507d8xtd+/e1dq1a/XMM89YPZ6rq6ucnJzSXU9CQoIkqUiRInJxcUn3OABgjwi8AJAOlStXVsmSJbVp0yZz26ZNm1SyZElVqlTJou9PP/2k7t27q1atWqpTp4769eun06dPW/T5+5nYo0ePKigoSF5eXqpTp47effdd3b5927w9ODhYb7zxhr744gvVr19fLVq0kGS59MDPz0+SNHDgQHl4eMjPz09nz55VxYoVFRkZabH/+fPnq1GjRkpKSkrT6583b578/f1VvXp1NWjQQB988IFFfcm2bNmiZs2aydPTU3369NGFCxdSbO/QoYM8PT3VuHFjzZgxQ/fv309TDQCQVgReAEinTp06adWqVebHK1euVMeOHVP0i4+PV69evbRy5UrNnz9fDg4OGjhw4CPDZVxcnPr06aNChQppxYoV+vTTT7Vr1y59+OGHFv0iIiL0559/at68efryyy9TjLNixQpJUkhIiHbs2KEVK1bIzc1NPj4+FnVLD9bcdujQQY6OaftvwcHBQWPGjNHatWv10Ucfaffu3Zo0aZJFnzt37uiLL77QxIkT9d133+nGjRt6++23zdv37t2rkSNHKigoSOHh4Ro3bpxWrVqlWbNmpakGAEgrAi8ApFPbtm21b98+nTt3TufOndP+/fvVtm3bFP2aN2+uZs2aqXTp0qpUqZL++9//6tixYzp+/Hiq465du1YJCQmaOHGi3N3dVbduXb333nv6/vvvdfnyZXO/vHnzavz48Xr++ef1/PPPpxinSJEikqSCBQvK1dXV/Lhz585at26deRlEVFSUjh07lmpYf5SePXvqxRdflJubm+rWrashQ4Zo/fr1Fn3u3bun9957T97e3qpatao++ugjHThwQIcOHZIkzZgxQ6+//ro6dOigZ599VvXq1dPgwYO1ZMmSNNcBAGmRK7sLAIAnVZEiRdSwYUOFhobKMAw1bNjQHCofdvLkSU2bNk0HDx7U1atXZRiGJOnChQtyd3dP0T86OloeHh7Kmzevua1GjRpKSkrSn3/+qWLFikmS3N3d07Xut0mTJho3bpw2b96s1q1bKzQ0VHXq1JGbm1uax9i1a5e+/PJLnThxQrdu3VJiYqLu3r2r+Ph48xriXLlyydPT0/yc8uXLq2DBgoqOjpaXl5eOHDmi/fv3W5zRTW0cAMgoAi8AZECnTp00btw4SdL777+fap/+/furVKlSGj9+vIoXL66kpCS1adNG9+7dy9C+0xsInZyc1L59e61atUpNmzZVWFiYxowZk+bnnz17Vv369VP37t319ttvq1ChQtq3b5/GjBmje/fupbmuuLg4DRo0SM2aNUuxLU+ePGmuBwD+CYEXADLA19dX9+7dk4ODg+rXr59i+9WrV/Xnn39q/Pjx5vvS7t2797Fjli9fXqGhoYqLizOf5d2/f78cHR1VtmxZq+rLnTu3EhMTU7R36dJFbdq00bfffqvExMRUQ+ejREVFyTAMBQcHm9f8/n05gyTdv39fv/32m7y8vCRJJ06c0I0bN1S+fHlJDy78+/PPP1W6dGmrXhMAWIs1vACQASaTSevXr1d4eLhMJlOK7YUKFdJTTz2lpUuX6tSpU4qIiNBHH3302DH9/f3l5OSk4OBgHTt2TLt379aHH36odu3amZczpFWpUqUUERGhmJgYXb9+3dxevnx5VatWTZ988olat24tZ2fnNI9ZunRp3bt3T4sWLdKZM2e0evXqVNfd5s6dWx9++KEOHjyo3377TaNGjVL16tXNAXjgwIH6/vvvNWPGDP3xxx+Kjo7WunXrNHXqVKteIwD8EwIvAGRQ/vz5lT9//lS3OTo6aurUqYqKilKbNm0UEhKiESNGPHY8FxcXzZkzR9euXVPnzp01ePBg1a1bV++++67VtY0cOVK7du1Sw4YN1aFDB4ttnTt31r1799SpUyerxqxYsaJGjRqlr7/+Wm3atFFYWJjeeeedFP2cnZ3Vt29fDR06VN27d1fevHktwqyvr69mzZqlHTt2qHPnzuratavmz5+vUqVKWf06AeBxHIzkqycAANkiISFBnp6emjdvnnx8fLJsvzNnztSGDRsUFhaWZfsEgOzAGV4AyEa3bt3S2rVr5ejoqHLlymXJPm/fvq1jx45p8eLFCgwMzJJ9AkB24qI1AMhG06ZN09q1azVs2DA9/fTTWbLPDz/8UGvXrlWTJk2sXs4AAE8iljQAAADArrGkAQAAAHaNwAsAAAC7RuAFAACAXSPwAgAAwK4ReAEAAGDXCLwAAACwawReAAAA2DUCLwAAAOza/wFN/ncPBnqrawAAAABJRU5ErkJggg==\n"
+          },
+          "metadata": {}
+        }
+      ],
+      "source": [
+        "import pandas as pd\n",
+        "import matplotlib.pyplot as plt\n",
+        "\n",
+        "# put average similarity + majority label in one dataframe\n",
+        "viz_df = pd.DataFrame({\n",
+        "    \"review\": range(1, len(majority_labels)+1),\n",
+        "    \"majority_label\": majority_labels,\n",
+        "    \"avg_similarity\": avg_sim\n",
+        "})\n",
+        "\n",
+        "label_avg = viz_df.groupby(\"majority_label\")[\"avg_similarity\"].mean().sort_values()\n",
+        "\n",
+        "plt.figure(figsize=(8,5))\n",
+        "plt.bar(label_avg.index, label_avg.values)\n",
+        "plt.xlabel(\"Majority label\")\n",
+        "plt.ylabel(\"Average cosine similarity\")\n",
+        "plt.title(\"Average similarity by majority label\")\n",
+        "plt.show()"
+      ],
+      "id": "FKK4hlnwb_Uz"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "niZtT3tpb_Uz"
+      },
+      "source": [
+        "### ✅ Checkpoint 2 — 4 points\n",
+        "\n",
+        "*Double-click and answer:*\n",
+        "\n",
+        "**2a.** Which reviews does MiniLM consider most similar? Does this match your group's labels? *(1 pt)*\n",
+        "\n",
+        "[YOUR ANSWER]\n",
+        "\n",
+        "MiniLM considers Review R3 and Review R8 the most similar pair, with a cosine similarity score of 0.41. This indicates that the model identified a strong semantic similarity between these two reviews compared with the rest of the dataset. Both reviews likely describe positive product experiences using similar language about product quality or satisfaction. Our group also labeled both reviews as positive, so the model’s similarity assessment aligns well with the human annotations in this case. This suggests that MiniLM successfully captured the underlying meaning and sentiment expressed in the reviews.\n",
+        "\n",
+        "---\n",
+        "\n",
+        "**2b.** Find a pair where MiniLM and your group disagree (similar vectors but different labels, or vice versa). Why? *(1 pt)*\n",
+        "\n",
+        "[YOUR ANSWER — name the review IDs and similarity scores]\n",
+        "\n",
+        "One example of disagreement appears between Review R1 and Review R5, which have a negative similarity score of -0.03, indicating that MiniLM considers them semantically unrelated. However, our group labeled R1 as positive and R5 as mixed, which still reflects partially similar sentiment contexts because both reviews discuss product experiences over time. The model focuses primarily on textual semantic similarity rather than nuanced sentiment interpretation. Humans, on the other hand, weigh contextual meaning and overall evaluation differently, which can lead to differences between model similarity and human labeling.\n",
+        "\n",
+        "---\n",
+        "\n",
+        "**2c.** Explain your 2 visualizations: what question does each answer, and what did you learn? *(2 pts)*\n",
+        "\n",
+        "[YOUR ANSWER — one paragraph per visualization]\n",
+        "\n",
+        "Visualization 1 — Average Similarity per Review The average similarity bar chart shows how semantically similar each review is to the rest of the sample. Reviews with higher average similarity likely contain more common language or themes shared across multiple reviews, while lower scores indicate more unique or distinct feedback. From this visualization, we observed that certain reviews cluster closer to the rest, suggesting they express common product experiences such as general satisfaction or dissatisfaction. This helps identify which reviews are representative of the dataset versus those that are outliers.\n",
+        "\n",
+        "Visualization 2 — Similarity by Label Match The similarity-by-label visualization compares cosine similarity scores between review pairs that share the same sentiment label versus those with different labels. The goal is to test whether semantically similar reviews also tend to receive the same sentiment annotation. In our results, pairs with the same label generally showed higher similarity scores than pairs with different labels, indicating that MiniLM embeddings capture sentiment-related meaning reasonably well. However, some overlaps remained, demonstrating that semantic similarity does not always perfectly correspond to sentiment classification."
+      ],
+      "id": "niZtT3tpb_Uz"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "87_A8SiLb_Uz"
+      },
+      "source": [
+        "## **5.** 🤖 Build Custom GPTs (Exercise 3) — 5 points\n",
+        "\n",
+        "A **custom GPT** is a specialized version of ChatGPT with permanent system instructions, uploaded knowledge, and a defined role. Go to [chat.openai.com/gpts/editor](https://chat.openai.com/gpts/editor) or **Explore GPTs → Create**."
+      ],
+      "id": "87_A8SiLb_Uz"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "hXVnxFh3b_Uz"
+      },
+      "source": [
+        "### *a. The 5 roles (one per member)*\n",
+        "\n",
+        "| Role | Name | System instructions idea |\n",
+        "|------|------|--------------------------|\n",
+        "| **A** | Annotation Quality Reviewer | Analyze annotator disagreements, determine correct labels, write guidelines. Always cite review IDs. |\n",
+        "| **B** | AI Companion Designer | Design tools that help annotation workers label more accurately. Propose interface features and alerts. |\n",
+        "| **C** | Career Advisor (AI-Era Jobs) | Advise business students on AI-era careers, the orchestrator economy, and which skills matter most. |\n",
+        "| **D** | Platform Strategist | Recommend pricing, hiring priorities, and competitive positioning. Use data to support every recommendation. |\n",
+        "| **E** | Workforce Automation Analyst | Assess which tasks AI can automate today vs. in 5+ years. Propose transition plans for workers. |\n",
+        "\n",
+        "**Groups of 6:** The 6th member picks any role and builds an **advanced version** (stricter constraints, multi-step workflow, edge case handling, or chained with another GPT)."
+      ],
+      "id": "hXVnxFh3b_Uz"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "Yifb97mSb_Uz"
+      },
+      "source": [
+        "### *b. Export your data for the GPTs*"
+      ],
+      "id": "Yifb97mSb_Uz"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "7BRCYfj_b_U9",
+        "colab": {
+          "base_uri": "https://localhost:8080/"
+        },
+        "outputId": "d66b4984-231a-4fc9-c42a-2f7d46352084"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "✅ Saved: annotated_reviews_for_gpt.csv\n",
+            "   Download this file from the runtime and upload it as knowledge when building your GPT.\n",
+            "   💡 Also upload the dataset most relevant to your role:\n",
+            "      Role A → annotation_guidelines.csv\n",
+            "      Role D → worker_performance_detailed.csv, competitor_pricing.csv\n",
+            "      Role E → ai_replacement_timeline.csv\n"
+          ]
+        }
+      ],
+      "source": [
+        "# Define all_names and all_labels from previous steps\n",
+        "all_names = [member_1_name, member_2_name, member_3_name, member_4_name, member_5_name]\n",
+        "all_labels = annotations # annotations list is already defined and contains all member labels\n",
+        "\n",
+        "# Calculate agreement_counts\n",
+        "agreement_counts = []\n",
+        "for i in range(num_reviews):\n",
+        "    labels_for_review = [member[i] for member in annotations]\n",
+        "    majority = majority_labels[i]\n",
+        "    count = labels_for_review.count(majority)\n",
+        "    agreement_counts.append(count)\n",
+        "\n",
+        "# Check names are filled in\n",
+        "default_names = {'MEMBER 1', 'MEMBER 2', 'MEMBER 3', 'MEMBER 4', 'MEMBER 5', 'MEMBER 6'}\n",
+        "for name in all_names:\n",
+        "    if name in default_names:\n",
+        "        print(f\"⚠️  '{name}' is still a default — go back and change it to your actual name!\")\n",
+        "\n",
+        "# Build export table\n",
+        "export_data = {\n",
+        "    'Review_ID': sample['review_id'].values,\n",
+        "    'Review_Text': sample['review_text'].values,\n",
+        "}\n",
+        "for name, labels in zip(all_names, all_labels):\n",
+        "    export_data[name + '_Label'] = labels\n",
+        "\n",
+        "export_data['Majority_Label'] = majority_labels\n",
+        "export_data['Agreement'] = [f\"{c}/{GROUP_SIZE}\" for c in agreement_counts]\n",
+        "export_data['Avg_Similarity'] = [round(sim_matrix[i].sum() / 9, 3) for i in range(10)]\n",
+        "\n",
+        "export_df = pd.DataFrame(export_data)\n",
+        "export_df.to_csv('annotated_reviews_for_gpt.csv', index=False)\n",
+        "\n",
+        "print(\"\\n✅ Saved: annotated_reviews_for_gpt.csv\")\n",
+        "print(\"   Download this file from the runtime and upload it as knowledge when building your GPT.\")\n",
+        "print(\"   💡 Also upload the dataset most relevant to your role:\")\n",
+        "print(\"      Role A → annotation_guidelines.csv\")\n",
+        "print(\"      Role D → worker_performance_detailed.csv, competitor_pricing.csv\")\n",
+        "print(\"      Role E → ai_replacement_timeline.csv\")"
+      ],
+      "id": "7BRCYfj_b_U9"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "uSRJXPbWb_U-"
+      },
+      "source": [
+        "### *c. ✋🏻🛑⛔️ Each member: build your GPT, test it, then document it in the checkpoint below*\n",
+        "\n",
+        "**How to build:**\n",
+        "1. Name your GPT\n",
+        "2. Write system instructions (be specific: role, audience, format, constraints)\n",
+        "3. Upload your CSV as knowledge\n",
+        "4. Test with 2–3 questions, refine if needed"
+      ],
+      "id": "uSRJXPbWb_U-"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "sGS142ecb_U-"
+      },
+      "source": [
+        "### ✅ Checkpoint 3 — 5 points\n",
+        "\n",
+        "*Each member documents their GPT. Then the group answers 3b and 3c together.*\n",
+        "\n",
+        "---\n",
+        "\n",
+        "**3a. Document your GPT** *(0.5 pt each, up to 3 pts):*\n",
+        "\n",
+        "*Member 1 Alice — Role A — GPT name:ANNOTATION QUALITY REVIEWER*\n",
+        "\n",
+        "System instructions:\n",
+        "I asked my GPT to analyze annotator disagreements, determine correct labels, explain guidelines and always cite review IDs. It also had to refer to the dataset I provided that was the annotation_guidlines.csv file and it had to respond in a clear, consistent and a direct way.\n",
+        "\n",
+        "Question:\n",
+        "I asked 2 questions to my GPT that were: What are the guidelines that you would use and why? What are the most frequent disagreements you observe?\n",
+        "\n",
+        "Key part of response: My gpt answered in different parts using a clear structure and many bullet points to go straight to the point. It also provided examples at each point made to help visualize what it explained. My gpt used simple vocabulary and gave correct and relevant answers to what I had asked. At the end, it also provided a summary of all the information so that i could not get lost throughout it's explanations which was very helpful.\n",
+        "\n",
+        "```\n",
+        "\n",
+        "*Member 2 Alessandro— Role B — GPT name: Case Study 2 GPT\n",
+        "```\n",
+        "You are an\tAI Companion Designer that evaluates sentiment annotations of product reviews.\n",
+        "\n",
+        "Your job is to analyze disagreements between annotators and determine the most accurate sentiment label using the provided guidelines.\n",
+        "\n",
+        "When reviewing annotations:\n",
+        "• Always reference the Review_ID\n",
+        "• Compare individual annotator labels with the majority label\n",
+        "• Identify disagreements between annotators\n",
+        "• Explain why a label should be positive, negative, neutral, or mixed\n",
+        "• Use the review text and annotation guidelines to justify the decision\n",
+        "\n",
+        "Output format:\n",
+        "1. Review ID\n",
+        "2. Annotator labels\n",
+        "3. Majority label\n",
+        "4. Agreement level\n",
+        "5. Your quality assessment\n",
+        "6. Recommendation if labels should be corrected\n",
+        "\n",
+        "Be concise, analytical, and evidence-based.\n",
+        "\n",
+        "Sample Q: Explain why Review R5 received a mixed label.\n",
+        "\n",
+        "Sample A: A Mixed label typically requires both positive and negative sentiment in the same review (e.g., praising one feature while criticizing another).\n",
+        "\n",
+        "In this case:\n",
+        "\n",
+        "“Stopped working after the first wash” → clear negative sentiment\n",
+        "\n",
+        "“The listing said it was waterproof. It was not waterproof.” → complaint about misleading product claims\n",
+        "\n",
+        "There are no positive statements about the product. The review is entirely critical of product durability and false advertising.\n",
+        "\n",
+        "The Mixed label likely occurred because the annotator interpreted the mention of the product description (“the listing said it was waterproof”) as a neutral or contextual statement, creating the impression of both informational and evaluative content. However, within the context of the review, it supports the complaint rather than providing positive sentiment.\n",
+        "\n",
+        "```\n",
+        "\n",
+        "*Member 3 : salomé — Role [C] — GPT name: [Career navigator IA]*\n",
+        "```\n",
+        "Instructions of the GPT : You are CareerNavigator AI, a career advisor for business school students entering the AI economy. You advise on AI-era job roles, automation-resistant skills, and give concrete action plans. Always tie advice to one of these four trends: Orchestrator Economy, Domain Expertise Premium, One-Person Company, Emotional Intelligence Edge. Be specific; use name tools, certifications, measurable outcomes.\n",
+        "\n",
+        "Question : I'm a business school student at ESCP considering a career in finance consulting. Is my job at risk from AI, and what should I do ?\n",
+        "\n",
+        "Key part of the answer :\n",
+        "1. Your job is not disappearing—but entry-level tasks are\n",
+        "AI will replace technical grunt work (models, slides, research)\n",
+        "But client-facing, strategic thinking stays valuable\n",
+        "👉 Translation: don’t compete on execution—compete on thinking\n",
+        "2. The game is shifting to the Orchestrator Economy\n",
+        "Top consultants will manage AI tools, not do everything manually\n",
+        "Your value = how well you combine AI + finance knowledge\n",
+        "👉 Learn tools like ChatGPT, Excel automation, basic Python\n",
+        "3. Your biggest advantage = Domain Expertise Premium\n",
+        "Generalists lose value\n",
+        "Specialists in PE, fintech, energy, infra gain value\n",
+        "👉 Pick a niche early and go deep\n",
+        "4. The real differentiator = Emotional Intelligence Edge\n",
+        "AI cannot:\n",
+        "Convince clients\n",
+        "Handle ambiguity\n",
+        "Build trust\n",
+        "👉 Practice communication, not just analysis\n",
+        "5. The winning profile (simple formula)\n",
+        "Finance skills + AI leverage + client communication = top 10% candidate\n",
+        "```\n",
+        "\n",
+        "*Member 4 Farès — Role [d] — GPT name: [Platform Strategist GPT]*\n",
+        "```\n",
+        "Instructions :\n",
+        "You are Platform Strategist GPT, a strategic advisor for data annotation platform founders and operations managers. Your audience is startup operators and business managers.\n",
+        "\n",
+        "When answering any question, you must:\n",
+        "1. Recommend an optimal worker hiring mix based on the accuracy/cost tradeoff in the data\n",
+        "2. Suggest a pricing strategy relative to competitors\n",
+        "3. Identify a competitive positioning angle\n",
+        "4. Every single recommendation MUST cite a specific number from the uploaded data (accuracy %, cost per task, funding amount, etc.)\n",
+        "5. Always end with a RISK section identifying what could go wrong\n",
+        "\n",
+        "Format every response as:\n",
+        "RECOMMENDATION → DATA EVIDENCE → RISK\n",
+        "\n",
+        "Never give generic business advice. If you cannot back a claim with a number from the data, do not make the claim.\n",
+        "\n",
+        "\n",
+        "question:\n",
+        "We are a new annotation platform. Should we hire PhD specialists or crowd workers to maximize both quality and margin? Give me a hiring strategy.\"\n",
+        "\n",
+        "key part of the answer:\n",
+        "RECOMMENDATION → Use a 70% / 20% / 10% pyramid (crowd / mid-tier / PhD), route ~30% complex tasks upward, and price at $4.0–$4.5/task to position as expert-verified quality near Appen-level pricing while undercutting Labelbox ($5.0) and premium players like Scale AI ($6.5)\n",
+        "DATA EVIDENCE → Crowd: 72.8% accuracy at $0.3/task vs PhD: 96.1% at $3.2/task (10.7× cost for +23.3 pts accuracy); Mid-tier: ~85–91% accuracy at $0.8–$2.1/task (best accuracy-per-dollar); Competitors: Sama $2.8, Appen $4.2, Labelbox $5.0, Scale AI $6.5 → blended internal cost $1.0–$1.8/task supports $4.0–$4.5 pricing with >50% margin and ~90% quality\n",
+        "RISK → Misrouting tasks keeps too many at 72.8% accuracy, hurting quality; overusing PhDs at $3.2/task erodes margin toward $6.5 cost structures; hybrid model may struggle to justify pricing vs $2.8–$5.0 competitors without clear benchmarks; multi-tier QA adds latency and operational overhead\n",
+        "```\n",
+        "*Member 5 Ange — Role [E] — GPT name: [Workforce Automation Analyst GPT]*\n",
+        "```\n",
+        "System Instructions:\n",
+        "I asked my GPT to act as a Business Process Automation Consultant. It had to identify and list work tasks that artificial intelligence can currently automate. The GPT needed to focus only on tasks that are realistically automatable with current AI technologies such as machine learning, generative AI, robotic process automation (RPA), and AI assistants. It also had to group tasks by category (for example administration, customer support, data analysis, marketing, software development, HR, and finance). For each task, the GPT had to briefly explain how AI perfor\n",
+        "ms it and provide practical examples of AI tools or methods used. The responses had to be clear, practical, and business-focused, while avoiding overly technical explanations.\n",
+        "Question:\n",
+        "I asked my GPT the following question:\n",
+        "Which work tasks can artificial intelligence currently automate in modern businesses?\n",
+        "Key Part of Response:\n",
+        "My GPT responded using a structured format organized by categories, which made the information easy to understand. It grouped tasks into areas such as administration, customer support, data analysis, marketing, software development, HR, and finance. For each category, it clearly listed the specific task, provided a short explanation of how AI performs it, and included examples of AI tools or methods that can automate the task.\n",
+        "The answers were clear, direct, and business-oriented, using simple vocabulary so that the explanations were easy to follow. The GPT focused only on realistic current capabilities of AI, avoiding speculation about future technology. The structured format helped present the information logically, making it easy to quickly identify which tasks can already be automated and how businesses can apply AI in practice.\n",
+        "\n",
+        "```\n",
+        "**3b.** Ask two GPTs with DIFFERENT roles the same question. How did their answers differ? *(1 pt)*\n",
+        "\n",
+        "I chose to oppose the GPT career advisor againt the automatation scope analyst when asking the question : i am an ESCP business student, should i be worried about IA taking my job ?\n",
+        "\n",
+        "In short, the career advisor said : \" Yes—but not in the way you think.\n",
+        "It will replace average business grads who don’t adapt\".\n",
+        "The GPT expains what is happening (compagnies need fewer people, generic business knowledge is losing value) but instantly give some remedies and cocnrete advices to be more valuable than IA.\n",
+        "He is honest, but positive : i move forward, gain some deep expertise than IA is and stay a tool.\n",
+        "If i keep getting proactive then IA isn't a worry, i should use it to my advantage.\n",
+        "\n",
+        "On the contrary, when i asked the same question to the automatation scope analyst, there isn't any hint of advice nor positivity.\n",
+        "The GPT states the fact in a completely neutral way : \" such as basic Reporting & Data Processing → HIGH → 2023 → Move into data interpretation, storytelling, and business decision-making roles\".\n",
+        "The GPT is cold and only speaks in term of data, which is a more alarming point of view.\n",
+        "\n",
+        "therefore, their answer differ in terms of content, tone and point of view.\n",
+        "\n",
+        "---\n",
+        "\n",
+        "**3c.** Groups of 5: Which GPT surprised you most and why?\n",
+        "The career advisor GPT surprised me the most. First of all, I could really relate to the question I had asked (about AI replacing my job), and I was naturally expecting a positive point of view and some motivational advice. I knew it wasn’t going to say, “Give up, AI will take your job.”\n",
+        "\n",
+        "As for the tone, it was predictable. However, I didn’t expect how specific and grounded it could be. When I asked it to create a two-year plan, it was extremely detailed. It told me exactly where to start, which tools to use, and what to focus on to gain expertise and depth.\n",
+        "For example, it suggested becoming a private equity associate with AI leverage. It told me to rebuild a DCF using ChatGPT combined with Excel automation, and to complete two case studies using AI tools (in M&A or private equity). It even recommended publishing a LinkedIn post titled “How AI Changes Financial Due Diligence.”\n",
+        "\n",
+        "So yes, the GPT was optimistic, but with real and concrete evidence. What also surprised me is that even though it was optimistic, it was still harsh. It gave a brutal truth:\n",
+        "If I only learn traditional finance, rely on Excel and PowerPoint, and avoid technology, then I will be commoditized within five years.\n",
+        "\n",
+        "It really felt like an actual advisor rather than a delusional motivator."
+      ],
+      "id": "sGS142ecb_U-"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "qvj2ggYjb_U-"
+      },
+      "source": [
+        "## **6.** 📊 Worker & Market Analysis (Exercise 4) — 3 points\n",
+        "\n",
+        "Annotation platforms manage large workforces with different specializations, accuracy rates, and costs. You have two datasets: `workers_df` (8 annotator profiles) and `competitors_df` (6 platform comparisons)."
+      ],
+      "id": "qvj2ggYjb_U-"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "e2BKa1Spb_U-"
+      },
+      "source": [
+        "### *a. Explore the worker and market data*"
+      ],
+      "id": "e2BKa1Spb_U-"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "RVps2bxVb_U-",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 1000
+        },
+        "outputId": "56e16d2b-1e45-436e-939b-ec42ab16a936"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "👥 WORKER PERFORMANCE DATA\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "  worker_id       name     background            domain  months_active  \\\n",
+              "0       W01     Ana L.    PhD Physics  Quantum Circuits             18   \n",
+              "1       W02   Kenji M.         MSc CS          NLP Text             12   \n",
+              "2       W03   Priya R.    BSc Biology    Medical Images              8   \n",
+              "3       W04  Carlos D.    Self-taught         Sentiment             14   \n",
+              "4       W05   Fatma K.   MSc Data Sci       Multi-modal             10   \n",
+              "5       W06   James T.  PhD Chemistry         Molecular              6   \n",
+              "6       W07     Li Wei         BSc EE    Circuit Design             11   \n",
+              "7       W08  Sophie B.   Crowd Worker      General Text             20   \n",
+              "\n",
+              "   total_tasks  accuracy_pct  avg_time_min  cost_per_task  \\\n",
+              "0         2556          94.2           4.2            1.3   \n",
+              "1         3468          87.6           2.1            0.8   \n",
+              "2          784          91.0           6.8            2.1   \n",
+              "3         5740          78.3           1.3            0.5   \n",
+              "4         2150          89.7           3.5            1.1   \n",
+              "5          456          96.1           8.5            3.2   \n",
+              "6         2013          85.4           3.0            1.5   \n",
+              "7        12240          72.8           0.8            0.3   \n",
+              "\n",
+              "  error_type_most_common quality_trend  \n",
+              "0         Boundary cases        Stable  \n",
+              "1        Context misread     Improving  \n",
+              "2        Rare conditions        Stable  \n",
+              "3          Sarcasm/irony     Declining  \n",
+              "4   Cross-modal conflict     Improving  \n",
+              "5        Novel compounds        Stable  \n",
+              "6       Scale estimation     Improving  \n",
+              "7        Mixed sentiment     Declining  "
+            ],
+            "text/html": [
+              "\n",
+              "  <div id=\"df-c91798b8-4016-4a0e-a9e7-54ca8b7f73f2\" 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>worker_id</th>\n",
+              "      <th>name</th>\n",
+              "      <th>background</th>\n",
+              "      <th>domain</th>\n",
+              "      <th>months_active</th>\n",
+              "      <th>total_tasks</th>\n",
+              "      <th>accuracy_pct</th>\n",
+              "      <th>avg_time_min</th>\n",
+              "      <th>cost_per_task</th>\n",
+              "      <th>error_type_most_common</th>\n",
+              "      <th>quality_trend</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>W01</td>\n",
+              "      <td>Ana L.</td>\n",
+              "      <td>PhD Physics</td>\n",
+              "      <td>Quantum Circuits</td>\n",
+              "      <td>18</td>\n",
+              "      <td>2556</td>\n",
+              "      <td>94.2</td>\n",
+              "      <td>4.2</td>\n",
+              "      <td>1.3</td>\n",
+              "      <td>Boundary cases</td>\n",
+              "      <td>Stable</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>W02</td>\n",
+              "      <td>Kenji M.</td>\n",
+              "      <td>MSc CS</td>\n",
+              "      <td>NLP Text</td>\n",
+              "      <td>12</td>\n",
+              "      <td>3468</td>\n",
+              "      <td>87.6</td>\n",
+              "      <td>2.1</td>\n",
+              "      <td>0.8</td>\n",
+              "      <td>Context misread</td>\n",
+              "      <td>Improving</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>W03</td>\n",
+              "      <td>Priya R.</td>\n",
+              "      <td>BSc Biology</td>\n",
+              "      <td>Medical Images</td>\n",
+              "      <td>8</td>\n",
+              "      <td>784</td>\n",
+              "      <td>91.0</td>\n",
+              "      <td>6.8</td>\n",
+              "      <td>2.1</td>\n",
+              "      <td>Rare conditions</td>\n",
+              "      <td>Stable</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>W04</td>\n",
+              "      <td>Carlos D.</td>\n",
+              "      <td>Self-taught</td>\n",
+              "      <td>Sentiment</td>\n",
+              "      <td>14</td>\n",
+              "      <td>5740</td>\n",
+              "      <td>78.3</td>\n",
+              "      <td>1.3</td>\n",
+              "      <td>0.5</td>\n",
+              "      <td>Sarcasm/irony</td>\n",
+              "      <td>Declining</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>W05</td>\n",
+              "      <td>Fatma K.</td>\n",
+              "      <td>MSc Data Sci</td>\n",
+              "      <td>Multi-modal</td>\n",
+              "      <td>10</td>\n",
+              "      <td>2150</td>\n",
+              "      <td>89.7</td>\n",
+              "      <td>3.5</td>\n",
+              "      <td>1.1</td>\n",
+              "      <td>Cross-modal conflict</td>\n",
+              "      <td>Improving</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5</th>\n",
+              "      <td>W06</td>\n",
+              "      <td>James T.</td>\n",
+              "      <td>PhD Chemistry</td>\n",
+              "      <td>Molecular</td>\n",
+              "      <td>6</td>\n",
+              "      <td>456</td>\n",
+              "      <td>96.1</td>\n",
+              "      <td>8.5</td>\n",
+              "      <td>3.2</td>\n",
+              "      <td>Novel compounds</td>\n",
+              "      <td>Stable</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>6</th>\n",
+              "      <td>W07</td>\n",
+              "      <td>Li Wei</td>\n",
+              "      <td>BSc EE</td>\n",
+              "      <td>Circuit Design</td>\n",
+              "      <td>11</td>\n",
+              "      <td>2013</td>\n",
+              "      <td>85.4</td>\n",
+              "      <td>3.0</td>\n",
+              "      <td>1.5</td>\n",
+              "      <td>Scale estimation</td>\n",
+              "      <td>Improving</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>7</th>\n",
+              "      <td>W08</td>\n",
+              "      <td>Sophie B.</td>\n",
+              "      <td>Crowd Worker</td>\n",
+              "      <td>General Text</td>\n",
+              "      <td>20</td>\n",
+              "      <td>12240</td>\n",
+              "      <td>72.8</td>\n",
+              "      <td>0.8</td>\n",
+              "      <td>0.3</td>\n",
+              "      <td>Mixed sentiment</td>\n",
+              "      <td>Declining</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-c91798b8-4016-4a0e-a9e7-54ca8b7f73f2')\"\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-c91798b8-4016-4a0e-a9e7-54ca8b7f73f2 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-c91798b8-4016-4a0e-a9e7-54ca8b7f73f2');\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=\"id_0bcfc5bb-fe30-44e5-acfe-38b9de736447\">\n",
+              "    <style>\n",
+              "      .colab-df-generate {\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-generate: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",
+              "      [theme=dark] .colab-df-generate {\n",
+              "        background-color: #3B4455;\n",
+              "        fill: #D2E3FC;\n",
+              "      }\n",
+              "\n",
+              "      [theme=dark] .colab-df-generate: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",
+              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('workers_df')\"\n",
+              "            title=\"Generate code using this dataframe.\"\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",
+              "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
+              "  </svg>\n",
+              "    </button>\n",
+              "    <script>\n",
+              "      (() => {\n",
+              "      const buttonEl =\n",
+              "        document.querySelector('#id_0bcfc5bb-fe30-44e5-acfe-38b9de736447 button.colab-df-generate');\n",
+              "      buttonEl.style.display =\n",
+              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
+              "\n",
+              "      buttonEl.onclick = () => {\n",
+              "        google.colab.notebook.generateWithVariable('workers_df');\n",
+              "      }\n",
+              "      })();\n",
+              "    </script>\n",
+              "  </div>\n",
+              "\n",
+              "    </div>\n",
+              "  </div>\n"
+            ],
+            "application/vnd.google.colaboratory.intrinsic+json": {
+              "type": "dataframe",
+              "variable_name": "workers_df",
+              "summary": "{\n  \"name\": \"workers_df\",\n  \"rows\": 8,\n  \"fields\": [\n    {\n      \"column\": \"worker_id\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 8,\n        \"samples\": [\n          \"W02\",\n          \"W06\",\n          \"W01\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"name\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 8,\n        \"samples\": [\n          \"Kenji M.\",\n          \"James T.\",\n          \"Ana L.\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"background\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 8,\n        \"samples\": [\n          \"MSc CS\",\n          \"PhD Chemistry\",\n          \"PhD Physics\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"domain\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 8,\n        \"samples\": [\n          \"NLP Text\",\n          \"Molecular\",\n          \"Quantum Circuits\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"months_active\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 4,\n        \"min\": 6,\n        \"max\": 20,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          12,\n          6,\n          18\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"total_tasks\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 3830,\n        \"min\": 456,\n        \"max\": 12240,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          3468,\n          456,\n          2556\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"accuracy_pct\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 7.911011764803359,\n        \"min\": 72.8,\n        \"max\": 96.1,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          87.6,\n          96.1,\n          94.2\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"avg_time_min\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 2.6756841581707125,\n        \"min\": 0.8,\n        \"max\": 8.5,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          2.1,\n          8.5,\n          4.2\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"cost_per_task\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 0.9411239481143203,\n        \"min\": 0.3,\n        \"max\": 3.2,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          0.8,\n          3.2,\n          1.3\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"error_type_most_common\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 8,\n        \"samples\": [\n          \"Context misread\",\n          \"Novel compounds\",\n          \"Boundary cases\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"quality_trend\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"Stable\",\n          \"Improving\",\n          \"Declining\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "\n",
+            "   Accuracy range: 72.8-96.1%\n",
+            "   Cost range: $0.3-$3.2 per task\n",
+            "\n",
+            "💰 COMPETITOR MARKET DATA\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "                company             hq                          model  \\\n",
+              "0              Scale AI  San Francisco  Managed workforce + AI-assist   \n",
+              "1                  Sama  San Francisco  Impact sourcing (East Africa)   \n",
+              "2                 Appen         Sydney   Global crowd + managed teams   \n",
+              "3              Labelbox  San Francisco           Platform + workforce   \n",
+              "4              Surge AI  San Francisco         Expert workforce + API   \n",
+              "5  Hypothetical Startup  Cambridge, UK             Expert-in-the-loop   \n",
+              "\n",
+              "  price_per_task_usd est_gross_margin_pct total_funding_usd  \\\n",
+              "0                6.5                   62        1400000000   \n",
+              "1                2.8                   45          85000000   \n",
+              "2                4.2                   38      N/A (public)   \n",
+              "3                5.0                   55         189000000   \n",
+              "4                7.0                   58          25000000   \n",
+              "5                TBD                  TBD               TBD   \n",
+              "\n",
+              "                  specialization  \n",
+              "0  General + autonomous vehicles  \n",
+              "1           General + ethical AI  \n",
+              "2     General + search relevance  \n",
+              "3        ML pipeline integration  \n",
+              "4             NLP + expert tasks  \n",
+              "5    Quantum AI + domain experts  "
+            ],
+            "text/html": [
+              "\n",
+              "  <div id=\"df-26c49e8a-4426-4c05-835c-d54b76a11d33\" 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>company</th>\n",
+              "      <th>hq</th>\n",
+              "      <th>model</th>\n",
+              "      <th>price_per_task_usd</th>\n",
+              "      <th>est_gross_margin_pct</th>\n",
+              "      <th>total_funding_usd</th>\n",
+              "      <th>specialization</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>Scale AI</td>\n",
+              "      <td>San Francisco</td>\n",
+              "      <td>Managed workforce + AI-assist</td>\n",
+              "      <td>6.5</td>\n",
+              "      <td>62</td>\n",
+              "      <td>1400000000</td>\n",
+              "      <td>General + autonomous vehicles</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>Sama</td>\n",
+              "      <td>San Francisco</td>\n",
+              "      <td>Impact sourcing (East Africa)</td>\n",
+              "      <td>2.8</td>\n",
+              "      <td>45</td>\n",
+              "      <td>85000000</td>\n",
+              "      <td>General + ethical AI</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>Appen</td>\n",
+              "      <td>Sydney</td>\n",
+              "      <td>Global crowd + managed teams</td>\n",
+              "      <td>4.2</td>\n",
+              "      <td>38</td>\n",
+              "      <td>N/A (public)</td>\n",
+              "      <td>General + search relevance</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>Labelbox</td>\n",
+              "      <td>San Francisco</td>\n",
+              "      <td>Platform + workforce</td>\n",
+              "      <td>5.0</td>\n",
+              "      <td>55</td>\n",
+              "      <td>189000000</td>\n",
+              "      <td>ML pipeline integration</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>Surge AI</td>\n",
+              "      <td>San Francisco</td>\n",
+              "      <td>Expert workforce + API</td>\n",
+              "      <td>7.0</td>\n",
+              "      <td>58</td>\n",
+              "      <td>25000000</td>\n",
+              "      <td>NLP + expert tasks</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5</th>\n",
+              "      <td>Hypothetical Startup</td>\n",
+              "      <td>Cambridge, UK</td>\n",
+              "      <td>Expert-in-the-loop</td>\n",
+              "      <td>TBD</td>\n",
+              "      <td>TBD</td>\n",
+              "      <td>TBD</td>\n",
+              "      <td>Quantum AI + domain experts</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-26c49e8a-4426-4c05-835c-d54b76a11d33')\"\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-26c49e8a-4426-4c05-835c-d54b76a11d33 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-26c49e8a-4426-4c05-835c-d54b76a11d33');\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=\"id_5b3b9717-5ac9-43eb-a057-c75ed922920b\">\n",
+              "    <style>\n",
+              "      .colab-df-generate {\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-generate: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",
+              "      [theme=dark] .colab-df-generate {\n",
+              "        background-color: #3B4455;\n",
+              "        fill: #D2E3FC;\n",
+              "      }\n",
+              "\n",
+              "      [theme=dark] .colab-df-generate: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",
+              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('competitors_df')\"\n",
+              "            title=\"Generate code using this dataframe.\"\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",
+              "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
+              "  </svg>\n",
+              "    </button>\n",
+              "    <script>\n",
+              "      (() => {\n",
+              "      const buttonEl =\n",
+              "        document.querySelector('#id_5b3b9717-5ac9-43eb-a057-c75ed922920b button.colab-df-generate');\n",
+              "      buttonEl.style.display =\n",
+              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
+              "\n",
+              "      buttonEl.onclick = () => {\n",
+              "        google.colab.notebook.generateWithVariable('competitors_df');\n",
+              "      }\n",
+              "      })();\n",
+              "    </script>\n",
+              "  </div>\n",
+              "\n",
+              "    </div>\n",
+              "  </div>\n"
+            ],
+            "application/vnd.google.colaboratory.intrinsic+json": {
+              "type": "dataframe",
+              "variable_name": "competitors_df",
+              "summary": "{\n  \"name\": \"competitors_df\",\n  \"rows\": 6,\n  \"fields\": [\n    {\n      \"column\": \"company\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 6,\n        \"samples\": [\n          \"Scale AI\",\n          \"Sama\",\n          \"Hypothetical Startup\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"hq\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"San Francisco\",\n          \"Sydney\",\n          \"Cambridge, UK\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"model\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 6,\n        \"samples\": [\n          \"Managed workforce + AI-assist\",\n          \"Impact sourcing (East Africa)\",\n          \"Expert-in-the-loop\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"price_per_task_usd\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 6,\n        \"samples\": [\n          \"6.5\",\n          \"2.8\",\n          \"TBD\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"est_gross_margin_pct\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 6,\n        \"samples\": [\n          \"62\",\n          \"45\",\n          \"TBD\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"total_funding_usd\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 6,\n        \"samples\": [\n          \"1400000000\",\n          \"85000000\",\n          \"TBD\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"specialization\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 6,\n        \"samples\": [\n          \"General + autonomous vehicles\",\n          \"General + ethical AI\",\n          \"Quantum AI + domain experts\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
+            }
+          },
+          "metadata": {}
+        }
+      ],
+      "source": [
+        "print(\"👥 WORKER PERFORMANCE DATA\")\n",
+        "display(workers_df)\n",
+        "print(f\"\\n   Accuracy range: {workers_df['accuracy_pct'].min()}-{workers_df['accuracy_pct'].max()}%\")\n",
+        "print(f\"   Cost range: ${workers_df['cost_per_task'].min()}-${workers_df['cost_per_task'].max()} per task\")\n",
+        "\n",
+        "print(\"\\n💰 COMPETITOR MARKET DATA\")\n",
+        "display(competitors_df)"
+      ],
+      "id": "RVps2bxVb_U-"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "S7lznoq1b_U-"
+      },
+      "source": [
+        "### *b. ✋🏻🛑⛔️ Create 2 visualizations using worker and/or competitor data*\n",
+        "\n",
+        "**Ideas:** Accuracy vs Cost scatter, value score bar chart, PhD vs Crowd Worker comparison, funding vs margin, business model comparison..."
+      ],
+      "id": "S7lznoq1b_U-"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "c768Ir5Hb_U-",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 370
+        },
+        "outputId": "2bcc4aa0-1be8-465a-b3d5-8c1f88099994"
+      },
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "<Figure size 800x500 with 1 Axes>"
+            ],
+            "image/png": "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\n"
+          },
+          "metadata": {}
+        }
+      ],
+      "source": [
+        "plt.figure(figsize=(8, 5))\n",
+        "plt.scatter(workers_df[\"cost_per_task\"], workers_df[\"accuracy_pct\"])\n",
+        "\n",
+        "for i, row in workers_df.iterrows():\n",
+        "    plt.text(row[\"cost_per_task\"], row[\"accuracy_pct\"], row[\"worker_id\"])\n",
+        "\n",
+        "plt.title(\"Worker Accuracy vs Cost per Task\")\n",
+        "plt.xlabel(\"Cost per Task\")\n",
+        "plt.ylabel(\"Accuracy (%)\")\n",
+        "plt.show()"
+      ],
+      "id": "c768Ir5Hb_U-"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "5feu9O2Pb_U-",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 443
+        },
+        "outputId": "5530ede5-1d89-48b0-ee10-5a5e14cf39a0"
+      },
+      "outputs": [
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "<Figure size 800x500 with 1 Axes>"
+            ],
+            "image/png": "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\n"
+          },
+          "metadata": {}
+        }
+      ],
+      "source": [
+        "phd_crowd_df = workers_df[workers_df[\"background\"].isin([\"PhD Physics\", \"PhD Chemistry\", \"Crowd Worker\"])]\n",
+        "\n",
+        "plt.figure(figsize=(8, 5))\n",
+        "plt.bar(phd_crowd_df[\"background\"] + \" (\" + phd_crowd_df[\"worker_id\"] + \")\", phd_crowd_df[\"accuracy_pct\"])\n",
+        "plt.title(\"PhD vs Crowd Worker Accuracy Comparison\")\n",
+        "plt.xlabel(\"Worker Background\")\n",
+        "plt.ylabel(\"Accuracy (%)\")\n",
+        "plt.xticks(rotation=45)\n",
+        "plt.show()"
+      ],
+      "id": "5feu9O2Pb_U-"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "KEAj5dM5b_U-"
+      },
+      "source": [
+        "### ✅ Checkpoint 4 — 3 points\n",
+        "\n",
+        "*Double-click and answer:*\n",
+        "\n",
+        "**4a.** Is there a pattern between education level and accuracy? Between cost and quality? Should a platform always hire the cheapest workers? *(1 pt)*\n",
+        "\n",
+        "According to my first graph it seems that there is a positive relation between accuracy and cost per task as the points show an upwoards trend and when cost goes up it seems that the accuracy per task goes up as well.Meaning that the higher the cost per task is the higher the accuracy is, therefore there is a positive relationship between quality and cost per task. My second graph demonstrates that there seems to be a relationship between education background and accuracy as the variables with the highest accuracy percentage are the ones with the highest levels of education.Indeed, we can see that the level of crowd workers in regards to accuracy is lower ( around 75 percent of accuracy)  than the 2 phd levels that are near 95 percent of accuracy.\n",
+        "\n",
+        "---\n",
+        "\n",
+        "**4b.** How does worker-level cost (`workers_df`) compare to platform-level pricing (`competitors_df`)? Where does the margin come from? Describe a data annotation task in YOUR field that would require human experts. Who would the ideal annotator be? *(1 pt)*\n",
+        "\n",
+        "From the workers dataset,we can see that the cost range per task for each worker is between 0.3$ and 3.2$ and if we compare this to the competitor market, we can see that the platform level pricing range is different as the price per task for each worker goes from a minimum of 2.8$ for the company Sama to a maximum of 7.0$ for the company Surge AI, a price range that is much higher than the worker level cost. Therefore, the platform level pricing is higher than the worker level cost that has a lower  price range.\n",
+        "\n",
+        "In the competitor market, the margin\n",
+        "comes from the fact that the platform level pricing is way higher than the worker level cost.\n",
+        "\n",
+        "A data annotation that would recquire human expertise in my field could be classifying the feedback by sentiment of each customer. The ideal annotator would be someone that has a good sens of analysis, detail and that can classify correctly and effectivly the data received.\n",
+        "\n",
+        "---\n",
+        "\n",
+        "**4c.** Derya argues that quantum computing is inherently more sustainable than classical AI, solving certain problems in far fewer operations, meaning less energy consumed. Her company builds open infrastructure so this benefit isn't locked behind proprietary clouds. Do you find this argument convincing? What assumptions does it depend on, and what could go wrong? *(1 pt)*\n",
+        "\n",
+        "I think that her argumentation is convincing as using less operations means using less energy to obtain results. Therefore this could be a good alternative to classical AI, however for this to work effectively,the quantum computing must be accurate, solid and scalable. If this is not the case and for instance the computing is fragile and needs high cost repars and maintanance is complicated, in the long run it is probable that the energy consumption will not be lower than classical AI.\n",
+        "\n"
+      ],
+      "id": "KEAj5dM5b_U-"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "ToSvRcrnb_U_"
+      },
+      "source": [
+        "## **7.** 💡 Final Reflection (Exercise 5) — 5 points"
+      ],
+      "id": "ToSvRcrnb_U_"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "TKCArBgJb_U_"
+      },
+      "source": [
+        "### *a. Explore the AI replacement timeline*"
+      ],
+      "id": "TKCArBgJb_U_"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "7LAQrQx1b_U_"
+      },
+      "source": [
+        "### *b. Four trends to consider*\n",
+        "\n",
+        "1. **The Orchestrator Economy** — The most valuable skill is directing AI tools across domains, not coding itself.\n",
+        "2. **Domain Expertise Becomes More Valuable** — AI handles routine tasks; expert knowledge becomes the bottleneck.\n",
+        "3. **The One-Person Company** — AI agents enable solo founders to operate with the output of a small team.\n",
+        "4. **From Artificial to Emotional Intelligence** — When everyone has AI, the differentiator is human connection."
+      ],
+      "id": "7LAQrQx1b_U_"
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {
+        "id": "rXOfo0Lrb_U_",
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 838
+        },
+        "outputId": "fb1f7e76-6e97-42b3-aa7a-e1a4b648fec8"
+      },
+      "outputs": [
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "🤖 AI REPLACEMENT TIMELINE\n"
+          ]
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "                     task_type  current_ai_accuracy  human_accuracy  \\\n",
+              "0  Simple Image Classification                   97              95   \n",
+              "1     Basic Sentiment Analysis                   88              85   \n",
+              "2     Named Entity Recognition                   91              95   \n",
+              "3      RLHF Preference Ranking                   72              80   \n",
+              "4     Medical Image Annotation                   82              94   \n",
+              "5        Legal Document Review                   78              92   \n",
+              "6  Creative Content Evaluation                   60              88   \n",
+              "7   Quantum Circuit Validation                   45              96   \n",
+              "\n",
+              "   year_ai_matches_human       status           workers_affected  \\\n",
+              "0                   2023     AI leads  High volume crowd workers   \n",
+              "1                   2024     AI leads    General text annotators   \n",
+              "2                   2026  Closing gap             NLP annotators   \n",
+              "3                   2027  Closing gap            RLHF annotators   \n",
+              "4                   2029  Humans lead         Few (requires MDs)   \n",
+              "5                   2030  Humans lead        Paralegals shifting   \n",
+              "6                   2032  Humans lead                        Low   \n",
+              "7                   2035  Humans lead                   None yet   \n",
+              "\n",
+              "           new_role_for_humans  \n",
+              "0        Edge case review only  \n",
+              "1       Guideline writing + QA  \n",
+              "2     Domain-specific entities  \n",
+              "3  Complex preference + safety  \n",
+              "4         Diagnosis validation  \n",
+              "5   Judgment calls + precedent  \n",
+              "6     Taste + cultural context  \n",
+              "7      Core task remains human  "
+            ],
+            "text/html": [
+              "\n",
+              "  <div id=\"df-7e07ed35-34ef-413f-a374-6bbda9baf7bb\" 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>task_type</th>\n",
+              "      <th>current_ai_accuracy</th>\n",
+              "      <th>human_accuracy</th>\n",
+              "      <th>year_ai_matches_human</th>\n",
+              "      <th>status</th>\n",
+              "      <th>workers_affected</th>\n",
+              "      <th>new_role_for_humans</th>\n",
+              "    </tr>\n",
+              "  </thead>\n",
+              "  <tbody>\n",
+              "    <tr>\n",
+              "      <th>0</th>\n",
+              "      <td>Simple Image Classification</td>\n",
+              "      <td>97</td>\n",
+              "      <td>95</td>\n",
+              "      <td>2023</td>\n",
+              "      <td>AI leads</td>\n",
+              "      <td>High volume crowd workers</td>\n",
+              "      <td>Edge case review only</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>1</th>\n",
+              "      <td>Basic Sentiment Analysis</td>\n",
+              "      <td>88</td>\n",
+              "      <td>85</td>\n",
+              "      <td>2024</td>\n",
+              "      <td>AI leads</td>\n",
+              "      <td>General text annotators</td>\n",
+              "      <td>Guideline writing + QA</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>2</th>\n",
+              "      <td>Named Entity Recognition</td>\n",
+              "      <td>91</td>\n",
+              "      <td>95</td>\n",
+              "      <td>2026</td>\n",
+              "      <td>Closing gap</td>\n",
+              "      <td>NLP annotators</td>\n",
+              "      <td>Domain-specific entities</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>3</th>\n",
+              "      <td>RLHF Preference Ranking</td>\n",
+              "      <td>72</td>\n",
+              "      <td>80</td>\n",
+              "      <td>2027</td>\n",
+              "      <td>Closing gap</td>\n",
+              "      <td>RLHF annotators</td>\n",
+              "      <td>Complex preference + safety</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>4</th>\n",
+              "      <td>Medical Image Annotation</td>\n",
+              "      <td>82</td>\n",
+              "      <td>94</td>\n",
+              "      <td>2029</td>\n",
+              "      <td>Humans lead</td>\n",
+              "      <td>Few (requires MDs)</td>\n",
+              "      <td>Diagnosis validation</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>5</th>\n",
+              "      <td>Legal Document Review</td>\n",
+              "      <td>78</td>\n",
+              "      <td>92</td>\n",
+              "      <td>2030</td>\n",
+              "      <td>Humans lead</td>\n",
+              "      <td>Paralegals shifting</td>\n",
+              "      <td>Judgment calls + precedent</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>6</th>\n",
+              "      <td>Creative Content Evaluation</td>\n",
+              "      <td>60</td>\n",
+              "      <td>88</td>\n",
+              "      <td>2032</td>\n",
+              "      <td>Humans lead</td>\n",
+              "      <td>Low</td>\n",
+              "      <td>Taste + cultural context</td>\n",
+              "    </tr>\n",
+              "    <tr>\n",
+              "      <th>7</th>\n",
+              "      <td>Quantum Circuit Validation</td>\n",
+              "      <td>45</td>\n",
+              "      <td>96</td>\n",
+              "      <td>2035</td>\n",
+              "      <td>Humans lead</td>\n",
+              "      <td>None yet</td>\n",
+              "      <td>Core task remains human</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-7e07ed35-34ef-413f-a374-6bbda9baf7bb')\"\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-7e07ed35-34ef-413f-a374-6bbda9baf7bb 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-7e07ed35-34ef-413f-a374-6bbda9baf7bb');\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=\"id_7d7fbdfc-10b8-4189-92a7-0e9233c4245f\">\n",
+              "    <style>\n",
+              "      .colab-df-generate {\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-generate: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",
+              "      [theme=dark] .colab-df-generate {\n",
+              "        background-color: #3B4455;\n",
+              "        fill: #D2E3FC;\n",
+              "      }\n",
+              "\n",
+              "      [theme=dark] .colab-df-generate: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",
+              "    <button class=\"colab-df-generate\" onclick=\"generateWithVariable('timeline_df')\"\n",
+              "            title=\"Generate code using this dataframe.\"\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",
+              "    <path d=\"M7,19H8.4L18.45,9,17,7.55,7,17.6ZM5,21V16.75L18.45,3.32a2,2,0,0,1,2.83,0l1.4,1.43a1.91,1.91,0,0,1,.58,1.4,1.91,1.91,0,0,1-.58,1.4L9.25,21ZM18.45,9,17,7.55Zm-12,3A5.31,5.31,0,0,0,4.9,8.1,5.31,5.31,0,0,0,1,6.5,5.31,5.31,0,0,0,4.9,4.9,5.31,5.31,0,0,0,6.5,1,5.31,5.31,0,0,0,8.1,4.9,5.31,5.31,0,0,0,12,6.5,5.46,5.46,0,0,0,6.5,12Z\"/>\n",
+              "  </svg>\n",
+              "    </button>\n",
+              "    <script>\n",
+              "      (() => {\n",
+              "      const buttonEl =\n",
+              "        document.querySelector('#id_7d7fbdfc-10b8-4189-92a7-0e9233c4245f button.colab-df-generate');\n",
+              "      buttonEl.style.display =\n",
+              "        google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
+              "\n",
+              "      buttonEl.onclick = () => {\n",
+              "        google.colab.notebook.generateWithVariable('timeline_df');\n",
+              "      }\n",
+              "      })();\n",
+              "    </script>\n",
+              "  </div>\n",
+              "\n",
+              "    </div>\n",
+              "  </div>\n"
+            ],
+            "application/vnd.google.colaboratory.intrinsic+json": {
+              "type": "dataframe",
+              "variable_name": "timeline_df",
+              "summary": "{\n  \"name\": \"timeline_df\",\n  \"rows\": 8,\n  \"fields\": [\n    {\n      \"column\": \"task_type\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 8,\n        \"samples\": [\n          \"Basic Sentiment Analysis\",\n          \"Legal Document Review\",\n          \"Simple Image Classification\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"current_ai_accuracy\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 17,\n        \"min\": 45,\n        \"max\": 97,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          88,\n          78,\n          97\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"human_accuracy\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 5,\n        \"min\": 80,\n        \"max\": 96,\n        \"num_unique_values\": 7,\n        \"samples\": [\n          95,\n          85,\n          88\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"year_ai_matches_human\",\n      \"properties\": {\n        \"dtype\": \"number\",\n        \"std\": 4,\n        \"min\": 2023,\n        \"max\": 2035,\n        \"num_unique_values\": 8,\n        \"samples\": [\n          2024,\n          2030,\n          2023\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"status\",\n      \"properties\": {\n        \"dtype\": \"category\",\n        \"num_unique_values\": 3,\n        \"samples\": [\n          \"AI leads\",\n          \"Closing gap\",\n          \"Humans lead\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"workers_affected\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 8,\n        \"samples\": [\n          \"General text annotators\",\n          \"Paralegals shifting\",\n          \"High volume crowd workers\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    },\n    {\n      \"column\": \"new_role_for_humans\",\n      \"properties\": {\n        \"dtype\": \"string\",\n        \"num_unique_values\": 8,\n        \"samples\": [\n          \"Guideline writing + QA\",\n          \"Judgment calls + precedent\",\n          \"Edge case review only\"\n        ],\n        \"semantic_type\": \"\",\n        \"description\": \"\"\n      }\n    }\n  ]\n}"
+            }
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "display_data",
+          "data": {
+            "text/plain": [
+              "<Figure size 1200x500 with 1 Axes>"
+            ],
+            "image/png": "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\n"
+          },
+          "metadata": {}
+        },
+        {
+          "output_type": "stream",
+          "name": "stdout",
+          "text": [
+            "🔴 AI surpassed  |  🟠 Closing gap  |  🟢 Humans still essential\n"
+          ]
+        }
+      ],
+      "source": [
+        "print(\"🤖 AI REPLACEMENT TIMELINE\")\n",
+        "display(timeline_df)\n",
+        "\n",
+        "fig, ax = plt.subplots(figsize=(12, 5))\n",
+        "colors = ['#e74c3c' if s == 'AI surpassed' else '#f39c12' if s == 'Closing gap' else '#27ae60'\n",
+        "          for s in timeline_df['status']]\n",
+        "ax.barh(timeline_df['task_type'], timeline_df['year_ai_matches_human'], color=colors)\n",
+        "ax.set_xlabel('Year AI Matches Human Accuracy')\n",
+        "ax.set_title('When Will AI Replace Human Annotators?')\n",
+        "ax.axvline(x=2026, color='navy', linestyle='--', label='We are here (2026)')\n",
+        "ax.legend()\n",
+        "plt.tight_layout()\n",
+        "plt.show()\n",
+        "\n",
+        "print(\"🔴 AI surpassed  |  🟠 Closing gap  |  🟢 Humans still essential\")"
+      ],
+      "id": "rXOfo0Lrb_U_"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "oL4UrVqxb_U_"
+      },
+      "source": [
+        "### ✅ Checkpoint 5 — 5 points\n",
+        "\n",
+        "*Each answer: minimum 4 sentences.*\n",
+        "\n",
+        "---\n",
+        "\n",
+        "**R1.** Compare your human annotation (Step 3) with MiniLM (Step 4) and your custom GPTs (Step 5). What can humans do that MiniLM cannot? What can MiniLM do that humans cannot? How did your GPT combine both? *(2 pts)*\n",
+        "\n",
+        "\n",
+        "Humans are really good at understanding sarcasm and tone and the small differences in what things mean. If we look at R098 and compare it to R099 we can see that the feeling is a little different even though the words are similar. Humans can tell the difference between an answer and one that is really negative or just okay. We are also better at understanding the context if the words are not exactly right.\n",
+        "\n",
+        "\n",
+        "For MiniLM it has a time with sarcasm and understanding the deeper meaning of things. It often groups reviews together that sound similar but actually mean things. But it does not let mood or emotion influence its decisions so the results are more consistent. It is very fast at looking at all the reviews at the same time.\n",
+        "\n",
+        "\n",
+        "In terms of Custom GPT it uses the structure but also thinks like a human. It is better at handling reviews that're mixed or negative or positive and it turns the results into ideas that businesses can use. Custom GPT is, like a tool that helps us understand what people are really saying in their reviews.\n",
+        "\n",
+        "\n",
+        "\n",
+        "---\n",
+        "\n",
+        "**R2.** Look at the AI replacement timeline above. Find the task type closest to YOUR career field. Is your future job safe from automation? What skills would keep you on the \"humans still essential\" side? *(1.5 pts)*\n",
+        "\n",
+        "\n",
+        "\n",
+        "\n",
+        "Basic Sentiment Analysis. We did a review analysis and  It was really helpful. We used MiniLM for clustering and  It worked well for our project. The model was able to group the reviews together based on what they meant. This showed that a lot of tasks, like organizing and sorting feedback can be done automatically.\n",
+        "\n",
+        "\n",
+        "When we looked at these groups the hard part was not making them. The hard part was figuring out what each group meant and how a business could use that information. We had to look at each evaluation to see if it was neutral a little bad or not clear. This was necessary even if the words used were very similar.\n",
+        "\n",
+        "\n",
+        "In this situation it was important to have a person look at the evaluations and use their judgment. Our Role D GPT was also helpful because it gave us information and helped us make decisions based on the groups we made. Overall the work is changing from looking at the numbers to really understanding what they mean and making decisions..\n",
+        "\n",
+        "\n",
+        "\n",
+        "---\n",
+        "\n",
+        "**R3.** Which of the four trends is most relevant to YOUR career plans? What specifically will you do in the next 2 years to prepare? *(1.5 pts)*\n",
+        "\n",
+        "\n",
+        "\n",
+        "The thing that really matters to our project is the Orchestrator Economy. This is because we did not just use one model. We had to figure out how to make different tools work together. The MiniLM tool gave us groups of things that're similar but on its own it did not really tell us anything useful. It just showed us patterns. Then we used the GPT part to add another layer. We looked at these patterns. Turned them into real decisions or suggestions.\n",
+        "\n",
+        "I think our group should focus on getting better at working with this kind of setup. We need to learn how to go from data to something that makes sense by using different tools together. At the time we need to get better at explaining and justifying the results. In our project the results only made sense when we looked at them carefully. So our goal is not just to get results. To really understand the Orchestrator Economy and the results well enough to use them correctly. We want to understand the results, from the MiniLM and GPT tools so we can use them in a way.\n"
+      ],
+      "id": "oL4UrVqxb_U_"
+    },
+    {
+      "cell_type": "markdown",
+      "source": [],
+      "metadata": {
+        "id": "C87BZlD8Gafi"
+      },
+      "id": "C87BZlD8Gafi"
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {
+        "id": "qBY9U6zKb_U_"
+      },
+      "source": [
+        "## **📋** Grading Summary\n",
+        "\n",
+        "| Exercise | Points | What's Graded |\n",
+        "|----------|--------|---------------|\n",
+        "| Exercise 1: Annotation (Step 3) | 3 pts | All members annotate + Checkpoint 1 |\n",
+        "| Exercise 2: MiniLM (Step 4) | 4 pts | Heatmap + 2 visualizations + Checkpoint 2 |\n",
+        "| Exercise 3: Custom GPTs (Step 5) | 5 pts | Each member builds a GPT + Checkpoint 3 |\n",
+        "| Exercise 4: Worker & Market (Step 6) | 3 pts | 2 visualizations + Checkpoint 4 |\n",
+        "| Exercise 5: Reflections (Step 7) | 5 pts | R1, R2, R3 |\n",
+        "| **TOTAL** | **20 pts** | |\n",
+        "\n",
+        "*Extra credit is awarded individually to students who revealed datasets during Derya's Q&A session.*\n",
+        "\n",
+        "---\n",
+        "\n",
+        "**Submission:** Submit your completed `.ipynb` notebook file.\n",
+        "\n",
+        "*You are encouraged to use ChatGPT, Claude, or other AI tools for coding help. Checkpoint answers must reflect your own thinking and reference your specific results. Submissions that are too similar to each other will receive zero.*"
+      ],
+      "id": "qBY9U6zKb_U_"
+    }
+  ],
+  "metadata": {
+    "colab": {
+      "provenance": []
+    },
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "name": "python",
+      "version": "3.10.0"
+    },
+    "widgets": {
+      "application/vnd.jupyter.widget-state+json": {
+        "1cc1db50c75649f7b0ced93d97f3e37c": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_d6df2f50329c4c96888eef5759b52530",
+              "IPY_MODEL_9aad2aff693f42deab0ae506fb428ff9",
+              "IPY_MODEL_25575c06b16847289915a68ef3262bea"
+            ],
+            "layout": "IPY_MODEL_f3397a560afd4a78a9ee8f6f1a3ab74f"
+          }
+        },
+        "d6df2f50329c4c96888eef5759b52530": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_1bdd47f33ddc456bacca88664e5dbbb6",
+            "placeholder": "​",
+            "style": "IPY_MODEL_d5d5913d3f484354addebc8e8d687d0f",
+            "value": "modules.json: 100%"
+          }
+        },
+        "9aad2aff693f42deab0ae506fb428ff9": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_1a8ccdc20f734f92988048c58e90d209",
+            "max": 349,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_6a244b679b43419e9565b66afb8ed629",
+            "value": 349
+          }
+        },
+        "25575c06b16847289915a68ef3262bea": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_faf79610c1ef43008cef0be8da4cea65",
+            "placeholder": "​",
+            "style": "IPY_MODEL_419e3b69fecc4a95b88e9a7a548f8fb1",
+            "value": " 349/349 [00:00&lt;00:00, 33.8kB/s]"
+          }
+        },
+        "f3397a560afd4a78a9ee8f6f1a3ab74f": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "1bdd47f33ddc456bacca88664e5dbbb6": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "d5d5913d3f484354addebc8e8d687d0f": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "1a8ccdc20f734f92988048c58e90d209": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "6a244b679b43419e9565b66afb8ed629": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "faf79610c1ef43008cef0be8da4cea65": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "419e3b69fecc4a95b88e9a7a548f8fb1": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "84d8e5e12761434e92e912c88b96fabb": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_d354075fde764868a321503a15fa94c5",
+              "IPY_MODEL_cf58e65a1b08467ca8d71aa42b563918",
+              "IPY_MODEL_07c547bd892d4edaa98f147d3fce7686"
+            ],
+            "layout": "IPY_MODEL_72721fd4bb7e4beca10d41b526d18b4f"
+          }
+        },
+        "d354075fde764868a321503a15fa94c5": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_5a4e69bec4c0483da1db8709ec028470",
+            "placeholder": "​",
+            "style": "IPY_MODEL_9b24b0d864a04181bf06655435106b8c",
+            "value": "config_sentence_transformers.json: 100%"
+          }
+        },
+        "cf58e65a1b08467ca8d71aa42b563918": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_852bd86df04b4bee97072e2e9c63baf8",
+            "max": 116,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_9eb874c27933468db94e2356e9ba0662",
+            "value": 116
+          }
+        },
+        "07c547bd892d4edaa98f147d3fce7686": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_2fd549d2c315488ead302c743f8da933",
+            "placeholder": "​",
+            "style": "IPY_MODEL_e82edca607d542f3b9c5643875b45b56",
+            "value": " 116/116 [00:00&lt;00:00, 12.4kB/s]"
+          }
+        },
+        "72721fd4bb7e4beca10d41b526d18b4f": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "5a4e69bec4c0483da1db8709ec028470": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "9b24b0d864a04181bf06655435106b8c": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "852bd86df04b4bee97072e2e9c63baf8": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "9eb874c27933468db94e2356e9ba0662": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "2fd549d2c315488ead302c743f8da933": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "e82edca607d542f3b9c5643875b45b56": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "16e2128bff8145628ac81cb0671079c2": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_f820281003904478b64ac92853fa6817",
+              "IPY_MODEL_f4f2933f99db4e10957206251ed0dcdf",
+              "IPY_MODEL_66f25761226746cbb9dafa2c970b0726"
+            ],
+            "layout": "IPY_MODEL_2ee256326b114269bb5a37e1a09dc5c9"
+          }
+        },
+        "f820281003904478b64ac92853fa6817": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_c5ad0470b5314cbe8b3bd8f3f9b0abe6",
+            "placeholder": "​",
+            "style": "IPY_MODEL_a00a9affea804da994315360e5345293",
+            "value": "README.md: "
+          }
+        },
+        "f4f2933f99db4e10957206251ed0dcdf": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_0d6df43ab65a454b91c8e7598c95ff3e",
+            "max": 1,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_022b55b95a0f410db7a84aa64f8fa9e6",
+            "value": 1
+          }
+        },
+        "66f25761226746cbb9dafa2c970b0726": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_9e215fe87976472c9de2eee9c5e837b7",
+            "placeholder": "​",
+            "style": "IPY_MODEL_e55453865ed2494b8a43dbbc7f082bd6",
+            "value": " 10.5k/? [00:00&lt;00:00, 854kB/s]"
+          }
+        },
+        "2ee256326b114269bb5a37e1a09dc5c9": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "c5ad0470b5314cbe8b3bd8f3f9b0abe6": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "a00a9affea804da994315360e5345293": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "0d6df43ab65a454b91c8e7598c95ff3e": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": "20px"
+          }
+        },
+        "022b55b95a0f410db7a84aa64f8fa9e6": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "9e215fe87976472c9de2eee9c5e837b7": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "e55453865ed2494b8a43dbbc7f082bd6": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "67314936052b4f45af63c3d969b85cbd": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_b465bb52c3004574bb2738d7f4f8c060",
+              "IPY_MODEL_1d1d131fdc114f8ab978f6aff02f544b",
+              "IPY_MODEL_a1650def53304242a12aa20d01e32d73"
+            ],
+            "layout": "IPY_MODEL_892841e82f804e72b96a70746b256580"
+          }
+        },
+        "b465bb52c3004574bb2738d7f4f8c060": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_35afad1035e5485e818bcc3e0b2246bd",
+            "placeholder": "​",
+            "style": "IPY_MODEL_4b5c55174a374649acc7c9677d644514",
+            "value": "sentence_bert_config.json: 100%"
+          }
+        },
+        "1d1d131fdc114f8ab978f6aff02f544b": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_e7e73104d6db448faae9ba7667f03a27",
+            "max": 53,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_a1908d51f7654364b87cb77a62a85bb0",
+            "value": 53
+          }
+        },
+        "a1650def53304242a12aa20d01e32d73": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_63cfe38a8ac54d88a118bb3dc44094c8",
+            "placeholder": "​",
+            "style": "IPY_MODEL_b491658f6814422282396129e859c2f1",
+            "value": " 53.0/53.0 [00:00&lt;00:00, 5.29kB/s]"
+          }
+        },
+        "892841e82f804e72b96a70746b256580": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "35afad1035e5485e818bcc3e0b2246bd": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "4b5c55174a374649acc7c9677d644514": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "e7e73104d6db448faae9ba7667f03a27": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "a1908d51f7654364b87cb77a62a85bb0": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "63cfe38a8ac54d88a118bb3dc44094c8": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "b491658f6814422282396129e859c2f1": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "9f7dbf11128344f2be123b7dca512a5f": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_54d05885283c4887a7715206b2a5c763",
+              "IPY_MODEL_68e7791b2af74ac0bde83768d27c4f38",
+              "IPY_MODEL_f420bc6ad4244b008e03fc4486fdb7d2"
+            ],
+            "layout": "IPY_MODEL_776a2d7a839447f79e180b4a6291f177"
+          }
+        },
+        "54d05885283c4887a7715206b2a5c763": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_0af009446e0f45488cd8ba2d5eb103b0",
+            "placeholder": "​",
+            "style": "IPY_MODEL_ae5cbff3686344c6906099ed2b6f98a9",
+            "value": "config.json: 100%"
+          }
+        },
+        "68e7791b2af74ac0bde83768d27c4f38": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_f568dd11f7a347598299083a74eff064",
+            "max": 612,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_0e18cbf8b03840fdae402b88e283b0e9",
+            "value": 612
+          }
+        },
+        "f420bc6ad4244b008e03fc4486fdb7d2": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_0b5135b7ca1b416db6679df7a7207cce",
+            "placeholder": "​",
+            "style": "IPY_MODEL_06a7376c429d46769420e546149c428e",
+            "value": " 612/612 [00:00&lt;00:00, 56.6kB/s]"
+          }
+        },
+        "776a2d7a839447f79e180b4a6291f177": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "0af009446e0f45488cd8ba2d5eb103b0": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "ae5cbff3686344c6906099ed2b6f98a9": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "f568dd11f7a347598299083a74eff064": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "0e18cbf8b03840fdae402b88e283b0e9": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "0b5135b7ca1b416db6679df7a7207cce": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "06a7376c429d46769420e546149c428e": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "68fc9bb13ed54338bec33b20cd73df59": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_8524a46a0467482f89e4f340d7fbebed",
+              "IPY_MODEL_c467010cde554bc697864588c0439294",
+              "IPY_MODEL_1e0cdc611546414a936b52948d0ff4f5"
+            ],
+            "layout": "IPY_MODEL_d02f493a6def45b99745387b8b32a77b"
+          }
+        },
+        "8524a46a0467482f89e4f340d7fbebed": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_c86151d5a69e4e2b8d8d10d26fe5ee8b",
+            "placeholder": "​",
+            "style": "IPY_MODEL_b3c355a20d4a4997a18ac0b9978cc54e",
+            "value": "model.safetensors: 100%"
+          }
+        },
+        "c467010cde554bc697864588c0439294": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_7bb9719ab8054b77b0c1ad8fe0bd74e3",
+            "max": 90868376,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_eaa987bc8b04432a84195504a90e7852",
+            "value": 90868376
+          }
+        },
+        "1e0cdc611546414a936b52948d0ff4f5": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_bc3b4482d673427f9e242bb4c14b94de",
+            "placeholder": "​",
+            "style": "IPY_MODEL_04b28ec852844cb8888bf403e964b605",
+            "value": " 90.9M/90.9M [00:01&lt;00:00, 453MB/s]"
+          }
+        },
+        "d02f493a6def45b99745387b8b32a77b": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "c86151d5a69e4e2b8d8d10d26fe5ee8b": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "b3c355a20d4a4997a18ac0b9978cc54e": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "7bb9719ab8054b77b0c1ad8fe0bd74e3": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "eaa987bc8b04432a84195504a90e7852": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "bc3b4482d673427f9e242bb4c14b94de": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "04b28ec852844cb8888bf403e964b605": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "b6c877d4fcda44d4b45d3add81a633f4": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_9743fefcd4ac43c6bdafc8a8bb50ec86",
+              "IPY_MODEL_4e93bc2d06124bf9b896ae879777ef1e",
+              "IPY_MODEL_83da641e6a9f4cb895ddb1795525a9e5"
+            ],
+            "layout": "IPY_MODEL_8df7510d2a0c418aab05581b123a851f"
+          }
+        },
+        "9743fefcd4ac43c6bdafc8a8bb50ec86": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_1972bf6bae5747b88bf3d66ec87aa966",
+            "placeholder": "​",
+            "style": "IPY_MODEL_2989bc6a2bcb4071a37a552fd28c4af4",
+            "value": "Loading weights: 100%"
+          }
+        },
+        "4e93bc2d06124bf9b896ae879777ef1e": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_610ebd6ae3c042a992baf596fa5d80a4",
+            "max": 103,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_acccc4e7d1944150b505d4862b93fb77",
+            "value": 103
+          }
+        },
+        "83da641e6a9f4cb895ddb1795525a9e5": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_ca0a5555d6234314a0ac7fbff7cd0873",
+            "placeholder": "​",
+            "style": "IPY_MODEL_ce66757f685f48c7a9db8a11f55a60cc",
+            "value": " 103/103 [00:00&lt;00:00, 493.23it/s, Materializing param=pooler.dense.weight]"
+          }
+        },
+        "8df7510d2a0c418aab05581b123a851f": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "1972bf6bae5747b88bf3d66ec87aa966": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "2989bc6a2bcb4071a37a552fd28c4af4": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "610ebd6ae3c042a992baf596fa5d80a4": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "acccc4e7d1944150b505d4862b93fb77": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "ca0a5555d6234314a0ac7fbff7cd0873": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "ce66757f685f48c7a9db8a11f55a60cc": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "cf4ba8afab434f869a4a533913be8189": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_145e8e9fbffb43b8b0281604b5e62064",
+              "IPY_MODEL_1a060e0c295c4d29b24f55de298dae01",
+              "IPY_MODEL_27e99785224b4b448c6b0ff30be815fd"
+            ],
+            "layout": "IPY_MODEL_bd97516347c141ecbd6eea7307b8ef6d"
+          }
+        },
+        "145e8e9fbffb43b8b0281604b5e62064": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_5786780980fe463cbd06bb64dcc0d0bc",
+            "placeholder": "​",
+            "style": "IPY_MODEL_babe357164f8449bac3cdfaee3c98541",
+            "value": "tokenizer_config.json: 100%"
+          }
+        },
+        "1a060e0c295c4d29b24f55de298dae01": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_b20942603f744f59bf86ce0d5f8a1cc5",
+            "max": 350,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_0404c8e259974005b56c5995f7cc1bfb",
+            "value": 350
+          }
+        },
+        "27e99785224b4b448c6b0ff30be815fd": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_a02f9cda29dc49b19eae8b00f85e3c4d",
+            "placeholder": "​",
+            "style": "IPY_MODEL_2734046898b04a9f8dd45b584b5a6c11",
+            "value": " 350/350 [00:00&lt;00:00, 15.5kB/s]"
+          }
+        },
+        "bd97516347c141ecbd6eea7307b8ef6d": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "5786780980fe463cbd06bb64dcc0d0bc": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "babe357164f8449bac3cdfaee3c98541": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "b20942603f744f59bf86ce0d5f8a1cc5": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "0404c8e259974005b56c5995f7cc1bfb": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "a02f9cda29dc49b19eae8b00f85e3c4d": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "2734046898b04a9f8dd45b584b5a6c11": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "a3e8c5f7a15e492aa59b1459963da66d": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_a6869838ad6b46cb9800021b4a43f616",
+              "IPY_MODEL_9791194c93f6436eb6b33cc48647e7ec",
+              "IPY_MODEL_c41fdf9ed74b41d08c4ef6c56c1d6a5a"
+            ],
+            "layout": "IPY_MODEL_e1e497e798524c9cb66c83c1813baed4"
+          }
+        },
+        "a6869838ad6b46cb9800021b4a43f616": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_438b6b76332d4b70a43666373493b4ab",
+            "placeholder": "​",
+            "style": "IPY_MODEL_d1ff9f9c30ca4df2b6837b952c6238ac",
+            "value": "vocab.txt: "
+          }
+        },
+        "9791194c93f6436eb6b33cc48647e7ec": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_acd318bacf0b4f578cb2d060846ffe46",
+            "max": 1,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_f16a5c763cdc4170b51056efe68ab9c2",
+            "value": 1
+          }
+        },
+        "c41fdf9ed74b41d08c4ef6c56c1d6a5a": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_e61da32034e140dea96472fe9009112e",
+            "placeholder": "​",
+            "style": "IPY_MODEL_b5ca5bb022004a29bf7acb2f8514540d",
+            "value": " 232k/? [00:00&lt;00:00, 10.8MB/s]"
+          }
+        },
+        "e1e497e798524c9cb66c83c1813baed4": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "438b6b76332d4b70a43666373493b4ab": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "d1ff9f9c30ca4df2b6837b952c6238ac": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "acd318bacf0b4f578cb2d060846ffe46": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": "20px"
+          }
+        },
+        "f16a5c763cdc4170b51056efe68ab9c2": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "e61da32034e140dea96472fe9009112e": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "b5ca5bb022004a29bf7acb2f8514540d": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "3712dba5d01247298810249a327b27b4": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_ad8351537ec44a9c954f296839ad43ff",
+              "IPY_MODEL_317978aa6f3b47cbbb2125a73b2bd11e",
+              "IPY_MODEL_59e8ccedfd9c4aef8bc1beb361b3b67b"
+            ],
+            "layout": "IPY_MODEL_2ababd05c70f45cd8ba35f0714764b4f"
+          }
+        },
+        "ad8351537ec44a9c954f296839ad43ff": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_fdde2ca71ad44daea2f3b963f95b1882",
+            "placeholder": "​",
+            "style": "IPY_MODEL_9a4af32cdbad41d5a6f3506323646ad5",
+            "value": "tokenizer.json: "
+          }
+        },
+        "317978aa6f3b47cbbb2125a73b2bd11e": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_bc617d06504d44718094911a1b02c606",
+            "max": 1,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_85f11a7b7a2c48438f60ae62fbec77b2",
+            "value": 1
+          }
+        },
+        "59e8ccedfd9c4aef8bc1beb361b3b67b": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_4829fd0c03e44ab192789debcc04ca23",
+            "placeholder": "​",
+            "style": "IPY_MODEL_4451ba9d6a78497697ee28308e2af75f",
+            "value": " 466k/? [00:00&lt;00:00, 24.7MB/s]"
+          }
+        },
+        "2ababd05c70f45cd8ba35f0714764b4f": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "fdde2ca71ad44daea2f3b963f95b1882": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "9a4af32cdbad41d5a6f3506323646ad5": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "bc617d06504d44718094911a1b02c606": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": "20px"
+          }
+        },
+        "85f11a7b7a2c48438f60ae62fbec77b2": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "4829fd0c03e44ab192789debcc04ca23": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "4451ba9d6a78497697ee28308e2af75f": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "2991c94441984658b3f24ca4be3bafc3": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_c8d18487ba984ffeb1b0f3b88dcafd83",
+              "IPY_MODEL_2a10b85054b7487c8b612f1575b81439",
+              "IPY_MODEL_1663e299d8b048e593e77a747e24d2aa"
+            ],
+            "layout": "IPY_MODEL_952eb895b6e842d7a5f4cafd0f99a113"
+          }
+        },
+        "c8d18487ba984ffeb1b0f3b88dcafd83": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_394b581443144907aab8252375a5215a",
+            "placeholder": "​",
+            "style": "IPY_MODEL_bc9d41acbf7d4964bf9cee55c734b6ab",
+            "value": "special_tokens_map.json: 100%"
+          }
+        },
+        "2a10b85054b7487c8b612f1575b81439": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_7eed38abc58a402493930b101d0ba967",
+            "max": 112,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_c7988a7b1b2a4de99acabb10ba4b327b",
+            "value": 112
+          }
+        },
+        "1663e299d8b048e593e77a747e24d2aa": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_ebd71a7b681d48a8b5e8979081c24ea8",
+            "placeholder": "​",
+            "style": "IPY_MODEL_a533b4fc5a2e496e8963314cf3322290",
+            "value": " 112/112 [00:00&lt;00:00, 10.5kB/s]"
+          }
+        },
+        "952eb895b6e842d7a5f4cafd0f99a113": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "394b581443144907aab8252375a5215a": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "bc9d41acbf7d4964bf9cee55c734b6ab": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "7eed38abc58a402493930b101d0ba967": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "c7988a7b1b2a4de99acabb10ba4b327b": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "ebd71a7b681d48a8b5e8979081c24ea8": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "a533b4fc5a2e496e8963314cf3322290": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "18774a626601487a8ff0bfc87cffa014": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HBoxModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HBoxModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HBoxView",
+            "box_style": "",
+            "children": [
+              "IPY_MODEL_adaaf806513f4e6aa1c52c675e0885d1",
+              "IPY_MODEL_1442a0f8db40493199fd3fffb1e79826",
+              "IPY_MODEL_d7384e36485d4d0cae1b52ef0503730f"
+            ],
+            "layout": "IPY_MODEL_f8c4273fd71f4142a5921f2c72572f52"
+          }
+        },
+        "adaaf806513f4e6aa1c52c675e0885d1": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_74bc5a2d01874f19836d89a26fe12518",
+            "placeholder": "​",
+            "style": "IPY_MODEL_02935cb32d67492bb4651b051dec5d6b",
+            "value": "config.json: 100%"
+          }
+        },
+        "1442a0f8db40493199fd3fffb1e79826": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "FloatProgressModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "FloatProgressModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "ProgressView",
+            "bar_style": "success",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_7babe961f29243af841cba0d3f55fab1",
+            "max": 190,
+            "min": 0,
+            "orientation": "horizontal",
+            "style": "IPY_MODEL_3ede47b700b541579c273da30ba18b88",
+            "value": 190
+          }
+        },
+        "d7384e36485d4d0cae1b52ef0503730f": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "HTMLModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_dom_classes": [],
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "HTMLModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/controls",
+            "_view_module_version": "1.5.0",
+            "_view_name": "HTMLView",
+            "description": "",
+            "description_tooltip": null,
+            "layout": "IPY_MODEL_ad3d3465c58e46e6959af1f4a97881c8",
+            "placeholder": "​",
+            "style": "IPY_MODEL_e9dcf02de017417cb8128dc8bff98d1d",
+            "value": " 190/190 [00:00&lt;00:00, 18.5kB/s]"
+          }
+        },
+        "f8c4273fd71f4142a5921f2c72572f52": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "74bc5a2d01874f19836d89a26fe12518": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "02935cb32d67492bb4651b051dec5d6b": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        },
+        "7babe961f29243af841cba0d3f55fab1": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "3ede47b700b541579c273da30ba18b88": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "ProgressStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "ProgressStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "bar_color": null,
+            "description_width": ""
+          }
+        },
+        "ad3d3465c58e46e6959af1f4a97881c8": {
+          "model_module": "@jupyter-widgets/base",
+          "model_name": "LayoutModel",
+          "model_module_version": "1.2.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/base",
+            "_model_module_version": "1.2.0",
+            "_model_name": "LayoutModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "LayoutView",
+            "align_content": null,
+            "align_items": null,
+            "align_self": null,
+            "border": 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,
+            "overflow_x": null,
+            "overflow_y": null,
+            "padding": null,
+            "right": null,
+            "top": null,
+            "visibility": null,
+            "width": null
+          }
+        },
+        "e9dcf02de017417cb8128dc8bff98d1d": {
+          "model_module": "@jupyter-widgets/controls",
+          "model_name": "DescriptionStyleModel",
+          "model_module_version": "1.5.0",
+          "state": {
+            "_model_module": "@jupyter-widgets/controls",
+            "_model_module_version": "1.5.0",
+            "_model_name": "DescriptionStyleModel",
+            "_view_count": null,
+            "_view_module": "@jupyter-widgets/base",
+            "_view_module_version": "1.2.0",
+            "_view_name": "StyleView",
+            "description_width": ""
+          }
+        }
+      }
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 5
+}
\ No newline at end of file