diff --git "a/notebooks/06b_Open-Source-Models_analysis.ipynb" "b/notebooks/06b_Open-Source-Models_analysis.ipynb"
--- "a/notebooks/06b_Open-Source-Models_analysis.ipynb"
+++ "b/notebooks/06b_Open-Source-Models_analysis.ipynb"
@@ -1 +1,3 @@
-{"cells":[{"cell_type":"code","execution_count":1,"metadata":{"application/vnd.databricks.v1+cell":{"cellMetadata":{},"inputWidgets":{},"nuid":"0ea8b46b-839b-445b-8043-ccdf4e920ace","showTitle":false,"title":""},"id":"YLH80COBzi_F"},"outputs":[],"source":["%load_ext autoreload\n","%autoreload 2"]},{"cell_type":"code","execution_count":2,"metadata":{"id":"63B5exAuzq4M"},"outputs":[],"source":["from pathlib import Path\n","\n","if \"workding_dir\" not in locals():\n","    try:\n","        from google.colab import drive\n","        drive.mount('/content/drive')\n","        workding_dir = \"/content/drive/MyDrive/logical-reasoning/\"\n","    except ModuleNotFoundError:\n","        workding_dir = str(Path.cwd().parent)"]},{"cell_type":"code","execution_count":3,"metadata":{"executionInfo":{"elapsed":368,"status":"ok","timestamp":1719461634865,"user":{"displayName":"Donghao Huang","userId":"00463591218503521679"},"user_tz":-480},"id":"zFulf0bg0H-9","outputId":"debdd535-c828-40b9-efc0-8a180e5830dd"},"outputs":[{"name":"stdout","output_type":"stream","text":["workding dir: /Users/inflaton/code/engd/projects/logical-reasoning\n"]}],"source":["import os\n","import sys\n","\n","os.chdir(workding_dir)\n","sys.path.append(workding_dir)\n","print(\"workding dir:\", workding_dir)"]},{"cell_type":"code","execution_count":4,"metadata":{"application/vnd.databricks.v1+cell":{"cellMetadata":{},"inputWidgets":{},"nuid":"9f67ec60-2f24-411c-84eb-0dd664b44775","showTitle":false,"title":""},"colab":{"base_uri":"https://localhost:8080/"},"executionInfo":{"elapsed":589,"status":"ok","timestamp":1719462011879,"user":{"displayName":"Donghao Huang","userId":"00463591218503521679"},"user_tz":-480},"id":"DIUiweYYzi_I","outputId":"e16e9247-9077-4b0c-f8ea-17059f05a1c4"},"outputs":[{"name":"stdout","output_type":"stream","text":["loading env vars from: /Users/inflaton/code/engd/projects/logical-reasoning/.env\n"]},{"data":{"text/plain":["True"]},"execution_count":4,"metadata":{},"output_type":"execute_result"}],"source":["from dotenv import find_dotenv, load_dotenv\n","\n","found_dotenv = find_dotenv(\".env\")\n","\n","if len(found_dotenv) == 0:\n","    found_dotenv = find_dotenv(\".env.example\")\n","print(f\"loading env vars from: {found_dotenv}\")\n","load_dotenv(found_dotenv, override=True)"]},{"cell_type":"code","execution_count":5,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat datasets/mgtv data/open_source_models_few_shots_results.csv 2048\n"]}],"source":["import os\n","\n","model_name = os.getenv(\"MODEL_NAME\")\n","data_path = os.getenv(\"LOGICAL_REASONING_DATA_PATH\")\n","results_path = os.getenv(\"LOGICAL_REASONING_RESULTS_PATH\")\n","max_new_tokens = int(os.getenv(\"MAX_NEW_TOKENS\", 2048))\n","\n","print(model_name, data_path, results_path, max_new_tokens)"]},{"cell_type":"code","execution_count":6,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["loading /Users/inflaton/code/engd/projects/logical-reasoning/llm_toolkit/logical_reasoning_utils.py\n"]}],"source":["from llm_toolkit.logical_reasoning_utils import *"]},{"cell_type":"code","execution_count":7,"metadata":{},"outputs":[],"source":["model_names = {\n","    \"gpt-4o-mini\": \"gpt-4o-mini\",\n","    \"gpt-4o\": \"gpt-4o\",\n","    \"o1-mini\": \"o1-mini\",\n","    \"o1-preview\": \"o1-preview\",\n","    \"Llama3.1-8B-Chinese-Chat\": \"Llama3.1-8B\",\n","    \"Llama3.1-70B-Chinese-Chat\": \"Llama3.1-70B\",\n","    \"Mistral-7B-v0.3-Chinese-Chat\": \"Mistral-7B\",\n","    \"internlm2_5-7b-chat\": \"InternLM2.5-7B\",\n","    \"internlm2_5-7b-chat-1m\": \"InternLM2.5-7B-1M\",\n","    \"internlm2_5-20b-chat\": \"InternLM2.5-20B\",\n","    \"Qwen2.5-0.5B-Instruct\": \"Qwen2.5-0.5B\",\n","    \"Qwen2.5-1.5B-Instruct\": \"Qwen2.5-1.5B\",\n","    \"Qwen2.5-3B-Instruct\": \"Qwen2.5-3B\",\n","    \"Qwen2.5-7B-Instruct\": \"Qwen2.5-7B\",\n","    \"Qwen2.5-72B-Instruct\": \"Qwen2.5-72B\",\n","}"]},{"cell_type":"code","execution_count":9,"metadata":{},"outputs":[{"data":{"text/plain":["{'Llama3.1-8B-Chinese-Chat': 4,\n"," 'Llama3.1-70B-Chinese-Chat': 5,\n"," 'Mistral-7B-v0.3-Chinese-Chat': 6,\n"," 'internlm2_5-7b-chat': 7,\n"," 'internlm2_5-7b-chat-1m': 8,\n"," 'internlm2_5-20b-chat': 9,\n"," 'Qwen2.5-0.5B-Instruct': 10,\n"," 'Qwen2.5-1.5B-Instruct': 11,\n"," 'Qwen2.5-3B-Instruct': 12,\n"," 'Qwen2.5-7B-Instruct': 13,\n"," 'Qwen2.5-72B-Instruct': 14}"]},"execution_count":9,"metadata":{},"output_type":"execute_result"}],"source":["model_orders = {k: i for i, k in enumerate(model_names.keys()) if i > 3}\n","model_orders"]},{"cell_type":"code","execution_count":10,"metadata":{},"outputs":[],"source":["markers = [\n","    \"o\",\n","    \"x\",\n","    \"^\",\n","    \"s\",\n","    \"d\",\n","    \"P\",\n","    \"X\",\n","    \"*\",\n","    \"v\",\n","    \">\",\n","    \"<\",\n","    \"p\",\n","    \"h\",\n","    \"H\",\n","    \"+\",\n","    \"|\",\n","    \"_\",\n","]\n","model_markers = {k: markers[i] for i, k in enumerate(model_orders.keys())}"]},{"cell_type":"code","execution_count":11,"metadata":{},"outputs":[{"data":{"text/html":["<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>shots</th>\n","      <th>model</th>\n","      <th>run</th>\n","      <th>accuracy</th>\n","      <th>precision</th>\n","      <th>recall</th>\n","      <th>f1</th>\n","      <th>ratio_valid_classifications</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>0</td>\n","      <td>Llama3.1-8B-Chinese-Chat</td>\n","      <td>shenzhi-wang/Llama3.1-8B-Chinese-Chat/shots-00</td>\n","      <td>0.734333</td>\n","      <td>0.737575</td>\n","      <td>0.734333</td>\n","      <td>0.727028</td>\n","      <td>0.803333</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>5</td>\n","      <td>Llama3.1-8B-Chinese-Chat</td>\n","      <td>shenzhi-wang/Llama3.1-8B-Chinese-Chat/shots-05</td>\n","      <td>0.705667</td>\n","      <td>0.750852</td>\n","      <td>0.705667</td>\n","      <td>0.723057</td>\n","      <td>0.988667</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>10</td>\n","      <td>Llama3.1-8B-Chinese-Chat</td>\n","      <td>shenzhi-wang/Llama3.1-8B-Chinese-Chat/shots-10</td>\n","      <td>0.673667</td>\n","      <td>0.777600</td>\n","      <td>0.673667</td>\n","      <td>0.709410</td>\n","      <td>0.962333</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>20</td>\n","      <td>Llama3.1-8B-Chinese-Chat</td>\n","      <td>shenzhi-wang/Llama3.1-8B-Chinese-Chat/shots-20</td>\n","      <td>0.767000</td>\n","      <td>0.764983</td>\n","      <td>0.767000</td>\n","      <td>0.763847</td>\n","      <td>0.979000</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>30</td>\n","      <td>Llama3.1-8B-Chinese-Chat</td>\n","      <td>shenzhi-wang/Llama3.1-8B-Chinese-Chat/shots-30</td>\n","      <td>0.771333</td>\n","      <td>0.772569</td>\n","      <td>0.771333</td>\n","      <td>0.769269</td>\n","      <td>0.732667</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>30</td>\n","      <td>Qwen2.5-7B-Instruct</td>\n","      <td>Qwen/Qwen2.5-7B-Instruct/shots-30</td>\n","      <td>0.764667</td>\n","      <td>0.778792</td>\n","      <td>0.764667</td>\n","      <td>0.752765</td>\n","      <td>0.805000</td>\n","    </tr>\n","    <tr>\n","      <th>5</th>\n","      <td>40</td>\n","      <td>Qwen2.5-7B-Instruct</td>\n","      <td>Qwen/Qwen2.5-7B-Instruct/shots-40</td>\n","      <td>0.759000</td>\n","      <td>0.773685</td>\n","      <td>0.759000</td>\n","      <td>0.747225</td>\n","      <td>0.854667</td>\n","    </tr>\n","    <tr>\n","      <th>6</th>\n","      <td>50</td>\n","      <td>Qwen2.5-7B-Instruct</td>\n","      <td>Qwen/Qwen2.5-7B-Instruct/shots-50</td>\n","      <td>0.758667</td>\n","      <td>0.764043</td>\n","      <td>0.758667</td>\n","      <td>0.741433</td>\n","      <td>0.756333</td>\n","    </tr>\n","    <tr>\n","      <th>0</th>\n","      <td>0</td>\n","      <td>Qwen2.5-72B-Instruct</td>\n","      <td>Qwen/Qwen2.5-72B-Instruct/shots-00</td>\n","      <td>0.795667</td>\n","      <td>0.809807</td>\n","      <td>0.795667</td>\n","      <td>0.777132</td>\n","      <td>0.994000</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>5</td>\n","      <td>Qwen2.5-72B-Instruct</td>\n","      <td>Qwen/Qwen2.5-72B-Instruct/shots-05</td>\n","      <td>0.819000</td>\n","      <td>0.818232</td>\n","      <td>0.819000</td>\n","      <td>0.809537</td>\n","      <td>0.941667</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>62 rows × 8 columns</p>\n","</div>"],"text/plain":["    shots                     model  \\\n","0       0  Llama3.1-8B-Chinese-Chat   \n","1       5  Llama3.1-8B-Chinese-Chat   \n","2      10  Llama3.1-8B-Chinese-Chat   \n","3      20  Llama3.1-8B-Chinese-Chat   \n","4      30  Llama3.1-8B-Chinese-Chat   \n","..    ...                       ...   \n","4      30       Qwen2.5-7B-Instruct   \n","5      40       Qwen2.5-7B-Instruct   \n","6      50       Qwen2.5-7B-Instruct   \n","0       0      Qwen2.5-72B-Instruct   \n","1       5      Qwen2.5-72B-Instruct   \n","\n","                                               run  accuracy  precision  \\\n","0   shenzhi-wang/Llama3.1-8B-Chinese-Chat/shots-00  0.734333   0.737575   \n","1   shenzhi-wang/Llama3.1-8B-Chinese-Chat/shots-05  0.705667   0.750852   \n","2   shenzhi-wang/Llama3.1-8B-Chinese-Chat/shots-10  0.673667   0.777600   \n","3   shenzhi-wang/Llama3.1-8B-Chinese-Chat/shots-20  0.767000   0.764983   \n","4   shenzhi-wang/Llama3.1-8B-Chinese-Chat/shots-30  0.771333   0.772569   \n","..                                             ...       ...        ...   \n","4                Qwen/Qwen2.5-7B-Instruct/shots-30  0.764667   0.778792   \n","5                Qwen/Qwen2.5-7B-Instruct/shots-40  0.759000   0.773685   \n","6                Qwen/Qwen2.5-7B-Instruct/shots-50  0.758667   0.764043   \n","0               Qwen/Qwen2.5-72B-Instruct/shots-00  0.795667   0.809807   \n","1               Qwen/Qwen2.5-72B-Instruct/shots-05  0.819000   0.818232   \n","\n","      recall        f1  ratio_valid_classifications  \n","0   0.734333  0.727028                     0.803333  \n","1   0.705667  0.723057                     0.988667  \n","2   0.673667  0.709410                     0.962333  \n","3   0.767000  0.763847                     0.979000  \n","4   0.771333  0.769269                     0.732667  \n","..       ...       ...                          ...  \n","4   0.764667  0.752765                     0.805000  \n","5   0.759000  0.747225                     0.854667  \n","6   0.758667  0.741433                     0.756333  \n","0   0.795667  0.777132                     0.994000  \n","1   0.819000  0.809537                     0.941667  \n","\n","[62 rows x 8 columns]"]},"execution_count":11,"metadata":{},"output_type":"execute_result"}],"source":["import pandas as pd\n","\n","perf_df = None\n","model_perf_dfs = {}\n","for model_name in model_orders.keys():\n","    metrics_csv = f\"data/{model_name}_shots_metrics.csv\"\n","    if not Path(metrics_csv).exists():\n","        print(f\"metrics file not found: {metrics_csv}\")\n","        continue\n","    df = pd.read_csv(metrics_csv)\n","    model_perf_dfs[model_name] = df\n","    perf_df = df if perf_df is None else pd.concat([perf_df, df])\n","\n","perf_df"]},{"cell_type":"code","execution_count":12,"metadata":{},"outputs":[],"source":["import matplotlib.pyplot as plt\n","from matplotlib.ticker import MultipleLocator\n","\n","\n","def plot_perf(\n","    model_perf_dfs,\n","    model_markers,\n","    x_major_locator=5,\n","    y_offset=0.005,\n","    variant=\"shots\"\n","):\n","    fig, ax = plt.subplots(1, 1, figsize=(12, 6))\n","\n","    for model_name, perf_df in model_perf_dfs.items():\n","        # Ensure the lengths of perf_df[\"epoch\"], perf_df[\"accuracy\"], and perf_df[\"f1\"] are the same\n","        min_length = min(\n","            len(perf_df[variant]), len(perf_df[\"accuracy\"]), len(perf_df[\"f1\"])\n","        )\n","        perf_df = perf_df.iloc[:min_length]\n","\n","        (line,) = ax.plot(\n","            perf_df[variant],\n","            perf_df[\"f1\"],\n","            marker=model_markers[model_name],\n","            label=model_name,\n","        )\n","\n","        line_color = line.get_color()\n","\n","        best_f1 = perf_df[\"f1\"].idxmax()\n","        print(\n","            f\"Best F1 for {model_name} @ {perf_df[variant].iloc[best_f1]:.2f} {variant}: {perf_df['f1'].iloc[best_f1]}\"\n","        )\n","        ax.annotate(\n","            f\"{perf_df['f1'].iloc[best_f1]*100:.2f}%\",\n","            (perf_df[variant].iloc[best_f1], perf_df[\"f1\"].iloc[best_f1]),\n","            ha=\"center\",\n","            va=\"bottom\",\n","            xytext=(0, 0),\n","            textcoords=\"offset points\",\n","            fontsize=10,\n","            color=line_color,\n","        )\n","\n","    # Set y-axis limit\n","    y_scales = ax.get_ylim()\n","    ax.set_ylim(y_scales[0], y_scales[1] + y_offset)\n","\n","    # Add title and labels\n","    ax.set_xlabel(\"Number of Shots\")\n","    ax.set_ylabel(\"F1 Score\")\n","\n","    # Set x-axis grid spacing to 0.2\n","    ax.xaxis.set_major_locator(MultipleLocator(x_major_locator))\n","    ax.set_title(\"Performance Analysis Across Shots for Models\")\n","\n","    # Rotate x labels\n","    plt.xticks(rotation=0)\n","    plt.grid(True)\n","    # plt.tight_layout()\n","\n","    # Set legend at the right to avoid overlapping with lines\n","    plt.legend(loc=\"center left\", bbox_to_anchor=(1.0, 0.5))\n","\n","    plt.show()"]},{"cell_type":"code","execution_count":13,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["Best F1 for Llama3.1-8B-Chinese-Chat @ 30.00 shots: 0.7692692690410152\n","Best F1 for Llama3.1-70B-Chinese-Chat @ 30.00 shots: 0.7570501796584528\n","Best F1 for Mistral-7B-v0.3-Chinese-Chat @ 30.00 shots: 0.6872462947319797\n","Best F1 for internlm2_5-7b-chat @ 5.00 shots: 0.7232456014841266\n","Best F1 for internlm2_5-7b-chat-1m @ 5.00 shots: 0.7665405919258307\n","Best F1 for internlm2_5-20b-chat @ 0.00 shots: 0.6416875854199033\n","Best F1 for Qwen2.5-0.5B-Instruct @ 50.00 shots: 0.5069942984615308\n","Best F1 for Qwen2.5-1.5B-Instruct @ 10.00 shots: 0.459589777544246\n","Best F1 for Qwen2.5-3B-Instruct @ 50.00 shots: 0.6451959368825358\n","Best F1 for Qwen2.5-7B-Instruct @ 30.00 shots: 0.7527649874769439\n","Best F1 for Qwen2.5-72B-Instruct @ 5.00 shots: 0.8095367865845521\n"]},{"data":{"image/png":"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","text/plain":["<Figure size 1200x600 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":["plot_perf(model_perf_dfs, model_markers)"]},{"cell_type":"code","execution_count":14,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-8B-Chinese-Chat, Shots: 0\n","count    3000.000000\n","mean      571.091000\n","std         9.115687\n","min       512.000000\n","25%       570.000000\n","50%       571.000000\n","75%       574.000000\n","max       652.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-70B-Chinese-Chat, Shots: 0\n","count    3000.000000\n","mean      571.091000\n","std         9.115687\n","min       512.000000\n","25%       570.000000\n","50%       571.000000\n","75%       574.000000\n","max       652.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Mistral-7B-v0.3-Chinese-Chat, Shots: 0\n","count    3000.000000\n","mean      799.354000\n","std        15.567385\n","min       694.000000\n","25%       798.000000\n","50%       802.000000\n","75%       806.000000\n","max       928.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat, Shots: 0\n","count    3000.000000\n","mean      461.917667\n","std         7.767732\n","min       426.000000\n","25%       459.000000\n","50%       461.000000\n","75%       463.000000\n","max       511.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat-1m, Shots: 0\n","count    3000.000000\n","mean      461.917667\n","std         7.767732\n","min       426.000000\n","25%       459.000000\n","50%       461.000000\n","75%       463.000000\n","max       511.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-20b-chat, Shots: 0\n","count    3000.000000\n","mean      461.917667\n","std         7.767732\n","min       426.000000\n","25%       459.000000\n","50%       461.000000\n","75%       463.000000\n","max       511.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-0.5B-Instruct, Shots: 0\n","count    3000.000000\n","mean      465.338667\n","std         8.617118\n","min       426.000000\n","25%       462.000000\n","50%       464.000000\n","75%       467.000000\n","max       517.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-1.5B-Instruct, Shots: 0\n","count    3000.000000\n","mean      465.338667\n","std         8.617118\n","min       426.000000\n","25%       462.000000\n","50%       464.000000\n","75%       467.000000\n","max       517.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-3B-Instruct, Shots: 0\n","count    3000.000000\n","mean      465.338667\n","std         8.617118\n","min       426.000000\n","25%       462.000000\n","50%       464.000000\n","75%       467.000000\n","max       517.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-7B-Instruct, Shots: 0\n","count    3000.000000\n","mean      465.338667\n","std         8.617118\n","min       426.000000\n","25%       462.000000\n","50%       464.000000\n","75%       467.000000\n","max       517.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-72B-Instruct, Shots: 0\n","count    3000.000000\n","mean      465.338667\n","std         8.617118\n","min       426.000000\n","25%       462.000000\n","50%       464.000000\n","75%       467.000000\n","max       517.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-8B-Chinese-Chat, Shots: 5\n","count    3000.000000\n","mean     1737.091000\n","std         9.115687\n","min      1678.000000\n","25%      1736.000000\n","50%      1737.000000\n","75%      1740.000000\n","max      1818.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-70B-Chinese-Chat, Shots: 5\n","count    3000.000000\n","mean     1737.091000\n","std         9.115687\n","min      1678.000000\n","25%      1736.000000\n","50%      1737.000000\n","75%      1740.000000\n","max      1818.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Mistral-7B-v0.3-Chinese-Chat, Shots: 5\n","count    3000.000000\n","mean     2444.354000\n","std        15.567385\n","min      2339.000000\n","25%      2443.000000\n","50%      2447.000000\n","75%      2451.000000\n","max      2573.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat, Shots: 5\n","count    3000.000000\n","mean     1301.917667\n","std         7.767732\n","min      1266.000000\n","25%      1299.000000\n","50%      1301.000000\n","75%      1303.000000\n","max      1351.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat-1m, Shots: 5\n","count    3000.000000\n","mean     1301.917667\n","std         7.767732\n","min      1266.000000\n","25%      1299.000000\n","50%      1301.000000\n","75%      1303.000000\n","max      1351.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-20b-chat, Shots: 5\n","count    3000.000000\n","mean     1301.917667\n","std         7.767732\n","min      1266.000000\n","25%      1299.000000\n","50%      1301.000000\n","75%      1303.000000\n","max      1351.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-0.5B-Instruct, Shots: 5\n","count    3000.000000\n","mean     1329.338667\n","std         8.617118\n","min      1290.000000\n","25%      1326.000000\n","50%      1328.000000\n","75%      1331.000000\n","max      1381.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-1.5B-Instruct, Shots: 5\n","count    3000.000000\n","mean     1329.338667\n","std         8.617118\n","min      1290.000000\n","25%      1326.000000\n","50%      1328.000000\n","75%      1331.000000\n","max      1381.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-3B-Instruct, Shots: 5\n","count    3000.000000\n","mean     1329.338667\n","std         8.617118\n","min      1290.000000\n","25%      1326.000000\n","50%      1328.000000\n","75%      1331.000000\n","max      1381.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-7B-Instruct, Shots: 5\n","count    3000.000000\n","mean     1329.338667\n","std         8.617118\n","min      1290.000000\n","25%      1326.000000\n","50%      1328.000000\n","75%      1331.000000\n","max      1381.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-72B-Instruct, Shots: 5\n","count    3000.000000\n","mean     1329.338667\n","std         8.617118\n","min      1290.000000\n","25%      1326.000000\n","50%      1328.000000\n","75%      1331.000000\n","max      1381.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-8B-Chinese-Chat, Shots: 10\n","count    3000.000000\n","mean     2833.091000\n","std         9.115687\n","min      2774.000000\n","25%      2832.000000\n","50%      2833.000000\n","75%      2836.000000\n","max      2914.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-70B-Chinese-Chat, Shots: 10\n","count    3000.000000\n","mean     2833.091000\n","std         9.115687\n","min      2774.000000\n","25%      2832.000000\n","50%      2833.000000\n","75%      2836.000000\n","max      2914.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Mistral-7B-v0.3-Chinese-Chat, Shots: 10\n","count    3000.000000\n","mean     3990.354000\n","std        15.567385\n","min      3885.000000\n","25%      3989.000000\n","50%      3993.000000\n","75%      3997.000000\n","max      4119.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat, Shots: 10\n","count    3000.000000\n","mean     2195.917667\n","std         7.767732\n","min      2160.000000\n","25%      2193.000000\n","50%      2195.000000\n","75%      2197.000000\n","max      2245.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat-1m, Shots: 10\n","count    3000.000000\n","mean     2195.917667\n","std         7.767732\n","min      2160.000000\n","25%      2193.000000\n","50%      2195.000000\n","75%      2197.000000\n","max      2245.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-20b-chat, Shots: 10\n","count    3000.000000\n","mean     2195.917667\n","std         7.767732\n","min      2160.000000\n","25%      2193.000000\n","50%      2195.000000\n","75%      2197.000000\n","max      2245.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"4a5d93c643ba401ba00c3d9099e4be3c","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"9bf533ad0e064d768e215dfc6dac4c8c","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-0.5B-Instruct, Shots: 10\n","count    3000.000000\n","mean     2237.338667\n","std         8.617118\n","min      2198.000000\n","25%      2234.000000\n","50%      2236.000000\n","75%      2239.000000\n","max      2289.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"c966d4370e7e404e87abc210685cb80e","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"53bfd02c002d4737a16905d7c0b4aa90","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-1.5B-Instruct, Shots: 10\n","count    3000.000000\n","mean     2237.338667\n","std         8.617118\n","min      2198.000000\n","25%      2234.000000\n","50%      2236.000000\n","75%      2239.000000\n","max      2289.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"36c65b6b931b4c8ab915d3b8b95d8d77","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"8ac092b4189b4e8fa19ea4ede0b6d559","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-3B-Instruct, Shots: 10\n","count    3000.000000\n","mean     2237.338667\n","std         8.617118\n","min      2198.000000\n","25%      2234.000000\n","50%      2236.000000\n","75%      2239.000000\n","max      2289.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-7B-Instruct, Shots: 10\n","count    3000.000000\n","mean     2237.338667\n","std         8.617118\n","min      2198.000000\n","25%      2234.000000\n","50%      2236.000000\n","75%      2239.000000\n","max      2289.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-72B-Instruct, Shots: 10\n","count    3000.000000\n","mean     2237.338667\n","std         8.617118\n","min      2198.000000\n","25%      2234.000000\n","50%      2236.000000\n","75%      2239.000000\n","max      2289.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-8B-Chinese-Chat, Shots: 20\n","count    3000.000000\n","mean     5202.091000\n","std         9.115687\n","min      5143.000000\n","25%      5201.000000\n","50%      5202.000000\n","75%      5205.000000\n","max      5283.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-70B-Chinese-Chat, Shots: 20\n","count    3000.000000\n","mean     5202.091000\n","std         9.115687\n","min      5143.000000\n","25%      5201.000000\n","50%      5202.000000\n","75%      5205.000000\n","max      5283.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Mistral-7B-v0.3-Chinese-Chat, Shots: 20\n","count    3000.000000\n","mean     7263.354000\n","std        15.567385\n","min      7158.000000\n","25%      7262.000000\n","50%      7266.000000\n","75%      7270.000000\n","max      7392.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat, Shots: 20\n","count    3000.000000\n","mean     4015.917667\n","std         7.767732\n","min      3980.000000\n","25%      4013.000000\n","50%      4015.000000\n","75%      4017.000000\n","max      4065.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat-1m, Shots: 20\n","count    3000.000000\n","mean     4015.917667\n","std         7.767732\n","min      3980.000000\n","25%      4013.000000\n","50%      4015.000000\n","75%      4017.000000\n","max      4065.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-20b-chat, Shots: 20\n","count    3000.000000\n","mean     4015.917667\n","std         7.767732\n","min      3980.000000\n","25%      4013.000000\n","50%      4015.000000\n","75%      4017.000000\n","max      4065.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"f41a7aacc870496eaae4e67d3796826a","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"1f405d1f86bf494599ce97758bee3abe","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-0.5B-Instruct, Shots: 20\n","count    3000.000000\n","mean     4124.338667\n","std         8.617118\n","min      4085.000000\n","25%      4121.000000\n","50%      4123.000000\n","75%      4126.000000\n","max      4176.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"d97b558254eb4596b3754ab2b1b8eadb","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"2d1a46fcef47448d91a263645e46e056","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-1.5B-Instruct, Shots: 20\n","count    3000.000000\n","mean     4124.338667\n","std         8.617118\n","min      4085.000000\n","25%      4121.000000\n","50%      4123.000000\n","75%      4126.000000\n","max      4176.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"06cd22b5216944dcb45b8ad44a7451c4","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"3ba05b7053444f0a9189d753f91baed3","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-3B-Instruct, Shots: 20\n","count    3000.000000\n","mean     4124.338667\n","std         8.617118\n","min      4085.000000\n","25%      4121.000000\n","50%      4123.000000\n","75%      4126.000000\n","max      4176.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-7B-Instruct, Shots: 20\n","count    3000.000000\n","mean     4124.338667\n","std         8.617118\n","min      4085.000000\n","25%      4121.000000\n","50%      4123.000000\n","75%      4126.000000\n","max      4176.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-72B-Instruct, Shots: 20\n","count    3000.000000\n","mean     4124.338667\n","std         8.617118\n","min      4085.000000\n","25%      4121.000000\n","50%      4123.000000\n","75%      4126.000000\n","max      4176.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-8B-Chinese-Chat, Shots: 30\n","count    3000.000000\n","mean     7687.091000\n","std         9.115687\n","min      7628.000000\n","25%      7686.000000\n","50%      7687.000000\n","75%      7690.000000\n","max      7768.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-70B-Chinese-Chat, Shots: 30\n","count    3000.000000\n","mean     7687.091000\n","std         9.115687\n","min      7628.000000\n","25%      7686.000000\n","50%      7687.000000\n","75%      7690.000000\n","max      7768.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Mistral-7B-v0.3-Chinese-Chat, Shots: 30\n","count     3000.000000\n","mean     10675.354000\n","std         15.567385\n","min      10570.000000\n","25%      10674.000000\n","50%      10678.000000\n","75%      10682.000000\n","max      10804.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat, Shots: 30\n","count    3000.000000\n","mean     5853.917667\n","std         7.767732\n","min      5818.000000\n","25%      5851.000000\n","50%      5853.000000\n","75%      5855.000000\n","max      5903.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat-1m, Shots: 30\n","count    3000.000000\n","mean     5853.917667\n","std         7.767732\n","min      5818.000000\n","25%      5851.000000\n","50%      5853.000000\n","75%      5855.000000\n","max      5903.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-20b-chat, Shots: 30\n","count    3000.000000\n","mean     5853.917667\n","std         7.767732\n","min      5818.000000\n","25%      5851.000000\n","50%      5853.000000\n","75%      5855.000000\n","max      5903.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"3653e64e34334d5fa27ef82076f9e172","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"63f88697fbc3406784f6b588275d1ad4","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-0.5B-Instruct, Shots: 30\n","count    3000.000000\n","mean     6055.338667\n","std         8.617118\n","min      6016.000000\n","25%      6052.000000\n","50%      6054.000000\n","75%      6057.000000\n","max      6107.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"d137cdc92b7a4b88a9b47da5f70d0a38","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"af1a32214fd840bfb27a89e5aa4022ef","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-1.5B-Instruct, Shots: 30\n","count    3000.000000\n","mean     6055.338667\n","std         8.617118\n","min      6016.000000\n","25%      6052.000000\n","50%      6054.000000\n","75%      6057.000000\n","max      6107.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"dfab7500c59e414d99fe9a7756bdcead","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"3e17d62403664c0a8100b7b68d51dd83","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-3B-Instruct, Shots: 30\n","count    3000.000000\n","mean     6055.338667\n","std         8.617118\n","min      6016.000000\n","25%      6052.000000\n","50%      6054.000000\n","75%      6057.000000\n","max      6107.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-7B-Instruct, Shots: 30\n","count    3000.000000\n","mean     6055.338667\n","std         8.617118\n","min      6016.000000\n","25%      6052.000000\n","50%      6054.000000\n","75%      6057.000000\n","max      6107.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-72B-Instruct, Shots: 30\n","count    3000.000000\n","mean     6055.338667\n","std         8.617118\n","min      6016.000000\n","25%      6052.000000\n","50%      6054.000000\n","75%      6057.000000\n","max      6107.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-8B-Chinese-Chat, Shots: 40\n","count     3000.000000\n","mean     10136.091000\n","std          9.115687\n","min      10077.000000\n","25%      10135.000000\n","50%      10136.000000\n","75%      10139.000000\n","max      10217.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-70B-Chinese-Chat, Shots: 40\n","count     3000.000000\n","mean     10136.091000\n","std          9.115687\n","min      10077.000000\n","25%      10135.000000\n","50%      10136.000000\n","75%      10139.000000\n","max      10217.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Mistral-7B-v0.3-Chinese-Chat, Shots: 40\n","count     3000.000000\n","mean     14023.354000\n","std         15.567385\n","min      13918.000000\n","25%      14022.000000\n","50%      14026.000000\n","75%      14030.000000\n","max      14152.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat, Shots: 40\n","count    3000.000000\n","mean     7659.917667\n","std         7.767732\n","min      7624.000000\n","25%      7657.000000\n","50%      7659.000000\n","75%      7661.000000\n","max      7709.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat-1m, Shots: 40\n","count    3000.000000\n","mean     7659.917667\n","std         7.767732\n","min      7624.000000\n","25%      7657.000000\n","50%      7659.000000\n","75%      7661.000000\n","max      7709.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-20b-chat, Shots: 40\n","count    3000.000000\n","mean     7659.917667\n","std         7.767732\n","min      7624.000000\n","25%      7657.000000\n","50%      7659.000000\n","75%      7661.000000\n","max      7709.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"74cad7e27d1747a9b922b21ca4455205","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"8458ed58792e4c8ca6d2aedd731ad21a","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-0.5B-Instruct, Shots: 40\n","count    3000.000000\n","mean     7958.338667\n","std         8.617118\n","min      7919.000000\n","25%      7955.000000\n","50%      7957.000000\n","75%      7960.000000\n","max      8010.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"a84ce4d14b8942daad8a0146700250cf","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"80a7a5b379434e0f9677905d7967c40e","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-1.5B-Instruct, Shots: 40\n","count    3000.000000\n","mean     7958.338667\n","std         8.617118\n","min      7919.000000\n","25%      7955.000000\n","50%      7957.000000\n","75%      7960.000000\n","max      8010.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"8fdd0391cb034eb28ae6ea6c77a75492","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"d5c0356e6c9e4d28982b23f08699af6f","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-3B-Instruct, Shots: 40\n","count    3000.000000\n","mean     7958.338667\n","std         8.617118\n","min      7919.000000\n","25%      7955.000000\n","50%      7957.000000\n","75%      7960.000000\n","max      8010.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-7B-Instruct, Shots: 40\n","count    3000.000000\n","mean     7958.338667\n","std         8.617118\n","min      7919.000000\n","25%      7955.000000\n","50%      7957.000000\n","75%      7960.000000\n","max      8010.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-72B-Instruct, Shots: 40\n","count    3000.000000\n","mean     7958.338667\n","std         8.617118\n","min      7919.000000\n","25%      7955.000000\n","50%      7957.000000\n","75%      7960.000000\n","max      8010.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-8B-Chinese-Chat, Shots: 50\n","count     3000.000000\n","mean     12638.091000\n","std          9.115687\n","min      12579.000000\n","25%      12637.000000\n","50%      12638.000000\n","75%      12641.000000\n","max      12719.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Llama3.1-70B-Chinese-Chat, Shots: 50\n","count     3000.000000\n","mean     12638.091000\n","std          9.115687\n","min      12579.000000\n","25%      12637.000000\n","50%      12638.000000\n","75%      12641.000000\n","max      12719.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Mistral-7B-v0.3-Chinese-Chat, Shots: 50\n","count     3000.000000\n","mean     17459.354000\n","std         15.567385\n","min      17354.000000\n","25%      17458.000000\n","50%      17462.000000\n","75%      17466.000000\n","max      17588.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat, Shots: 50\n","count    3000.000000\n","mean     9511.917667\n","std         7.767732\n","min      9476.000000\n","25%      9509.000000\n","50%      9511.000000\n","75%      9513.000000\n","max      9561.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-7b-chat-1m, Shots: 50\n","count    3000.000000\n","mean     9511.917667\n","std         7.767732\n","min      9476.000000\n","25%      9509.000000\n","50%      9511.000000\n","75%      9513.000000\n","max      9561.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: internlm2_5-20b-chat, Shots: 50\n","count    3000.000000\n","mean     9511.917667\n","std         7.767732\n","min      9476.000000\n","25%      9509.000000\n","50%      9511.000000\n","75%      9513.000000\n","max      9561.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"1ed79f4e6862426ab1739eb15e1505f5","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"e63f59a07b7a42d6a143c53e04028316","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-0.5B-Instruct, Shots: 50\n","count    3000.000000\n","mean     9909.338667\n","std         8.617118\n","min      9870.000000\n","25%      9906.000000\n","50%      9908.000000\n","75%      9911.000000\n","max      9961.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"3b44e26442bc4f9dbac30f96d0d81a78","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"e1211bd4a3e44a77860eb4617a51b530","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-1.5B-Instruct, Shots: 50\n","count    3000.000000\n","mean     9909.338667\n","std         8.617118\n","min      9870.000000\n","25%      9906.000000\n","50%      9908.000000\n","75%      9911.000000\n","max      9961.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n"]},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"03720de4193c4f90af36ad30d078f16a","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/25000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"data":{"application/vnd.jupyter.widget-view+json":{"model_id":"2ed60e4d9e944775a93932aac80e1cb5","version_major":2,"version_minor":0},"text/plain":["Map:   0%|          | 0/3000 [00:00<?, ? examples/s]"]},"metadata":{},"output_type":"display_data"},{"name":"stdout","output_type":"stream","text":["DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-3B-Instruct, Shots: 50\n","count    3000.000000\n","mean     9909.338667\n","std         8.617118\n","min      9870.000000\n","25%      9906.000000\n","50%      9908.000000\n","75%      9911.000000\n","max      9961.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-7B-Instruct, Shots: 50\n","count    3000.000000\n","mean     9909.338667\n","std         8.617118\n","min      9870.000000\n","25%      9906.000000\n","50%      9908.000000\n","75%      9911.000000\n","max      9961.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth', 'train_text', 'prompt'],\n","        num_rows: 3000\n","    })\n","})\n","Model: Qwen2.5-72B-Instruct, Shots: 50\n","count    3000.000000\n","mean     9909.338667\n","std         8.617118\n","min      9870.000000\n","25%      9906.000000\n","50%      9908.000000\n","75%      9911.000000\n","max      9961.000000\n","Name: num_tokens, dtype: float64\n"]}],"source":["from transformers import (\n","    AutoTokenizer,\n",")\n","\n","from llm_toolkit.llm_utils import print_row_details\n","\n","model_test_dfs = {}\n","\n","for num_shots in [0, 5, 10, 20, 30, 40, 50]:\n","    for model_name in model_orders.keys():\n","        model_id = (\n","            model_perf_dfs[model_name][\"run\"].unique()[0].split(model_name)[0]\n","            + model_name\n","        )\n","        tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True)\n","\n","        datasets = load_logical_reasoning_dataset(\n","            data_path,\n","            tokenizer=tokenizer,\n","            chinese_prompt=True,\n","            using_p1=False,\n","            num_shots=num_shots,\n","        )\n","        print(f\"Model: {model_name}, Shots: {num_shots}\")\n","        test_df = datasets[\"test\"].to_pandas()\n","        test_df[\"num_tokens\"] = test_df[\"prompt\"].apply(\n","            lambda x: len(tokenizer(x)[\"input_ids\"])\n","        )\n","        \n","        print(test_df[\"num_tokens\"].describe())\n","\n","        model_test_dfs[(model_name, num_shots)] = test_df"]},{"cell_type":"code","execution_count":15,"metadata":{},"outputs":[],"source":["import tiktoken\n","\n","\n","def num_tokens_from_text(text, model=\"gpt-4o\"):\n","    \"\"\"Return the number of tokens used by a list of messages.\"\"\"\n","    try:\n","        encoding = tiktoken.encoding_for_model(model)\n","    except KeyError:\n","        # print(\"Warning: model not found. Using cl100k_base encoding.\")\n","        encoding = tiktoken.get_encoding(\"cl100k_base\")\n","\n","    return len(encoding.encode(text))\n","\n","\n","def num_tokens_from_row(row, num_shots, train_dataset, model=\"gpt-4o\"):\n","    prompt = get_few_shot_prompt_template(\n","        num_shots, train_dataset\n","    )\n","    text = prompt.format(row[\"puzzle\"], row[\"truth\"], row[\"text\"])\n","    return num_tokens_from_text(text, model=model)"]},{"cell_type":"code","execution_count":16,"metadata":{},"outputs":[{"name":"stdout","output_type":"stream","text":["loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o, Shots: 0\n","count    3000.000000\n","mean      524.806333\n","std        10.057595\n","min       464.000000\n","25%       522.000000\n","50%       525.000000\n","75%       528.250000\n","max       606.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o-mini, Shots: 0\n","count    3000.000000\n","mean      524.806333\n","std        10.057595\n","min       464.000000\n","25%       522.000000\n","50%       525.000000\n","75%       528.250000\n","max       606.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-preview, Shots: 0\n","count    3000.000000\n","mean      797.595333\n","std        16.417250\n","min       682.000000\n","25%       797.000000\n","50%       799.000000\n","75%       803.000000\n","max       925.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-mini, Shots: 0\n","count    3000.000000\n","mean      797.595333\n","std        16.417250\n","min       682.000000\n","25%       797.000000\n","50%       799.000000\n","75%       803.000000\n","max       925.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o, Shots: 5\n","count    3000.000000\n","mean     1629.806333\n","std        10.057595\n","min      1569.000000\n","25%      1627.000000\n","50%      1630.000000\n","75%      1633.250000\n","max      1711.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o-mini, Shots: 5\n","count    3000.000000\n","mean     1629.806333\n","std        10.057595\n","min      1569.000000\n","25%      1627.000000\n","50%      1630.000000\n","75%      1633.250000\n","max      1711.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-preview, Shots: 5\n","count    3000.000000\n","mean     2521.595333\n","std        16.417250\n","min      2406.000000\n","25%      2521.000000\n","50%      2523.000000\n","75%      2527.000000\n","max      2649.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-mini, Shots: 5\n","count    3000.000000\n","mean     2521.595333\n","std        16.417250\n","min      2406.000000\n","25%      2521.000000\n","50%      2523.000000\n","75%      2527.000000\n","max      2649.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o, Shots: 10\n","count    3000.000000\n","mean     2699.806333\n","std        10.057595\n","min      2639.000000\n","25%      2697.000000\n","50%      2700.000000\n","75%      2703.250000\n","max      2781.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o-mini, Shots: 10\n","count    3000.000000\n","mean     2699.806333\n","std        10.057595\n","min      2639.000000\n","25%      2697.000000\n","50%      2700.000000\n","75%      2703.250000\n","max      2781.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-preview, Shots: 10\n","count    3000.000000\n","mean     4154.595333\n","std        16.417250\n","min      4039.000000\n","25%      4154.000000\n","50%      4156.000000\n","75%      4160.000000\n","max      4282.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-mini, Shots: 10\n","count    3000.000000\n","mean     4154.595333\n","std        16.417250\n","min      4039.000000\n","25%      4154.000000\n","50%      4156.000000\n","75%      4160.000000\n","max      4282.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o, Shots: 20\n","count    3000.000000\n","mean     5028.806333\n","std        10.057595\n","min      4968.000000\n","25%      5026.000000\n","50%      5029.000000\n","75%      5032.250000\n","max      5110.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o-mini, Shots: 20\n","count    3000.000000\n","mean     5028.806333\n","std        10.057595\n","min      4968.000000\n","25%      5026.000000\n","50%      5029.000000\n","75%      5032.250000\n","max      5110.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-preview, Shots: 20\n","count    3000.000000\n","mean     7653.595333\n","std        16.417250\n","min      7538.000000\n","25%      7653.000000\n","50%      7655.000000\n","75%      7659.000000\n","max      7781.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-mini, Shots: 20\n","count    3000.000000\n","mean     7653.595333\n","std        16.417250\n","min      7538.000000\n","25%      7653.000000\n","50%      7655.000000\n","75%      7659.000000\n","max      7781.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o, Shots: 30\n","count    3000.000000\n","mean     7490.806333\n","std        10.057595\n","min      7430.000000\n","25%      7488.000000\n","50%      7491.000000\n","75%      7494.250000\n","max      7572.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o-mini, Shots: 30\n","count    3000.000000\n","mean     7490.806333\n","std        10.057595\n","min      7430.000000\n","25%      7488.000000\n","50%      7491.000000\n","75%      7494.250000\n","max      7572.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-preview, Shots: 30\n","count     3000.000000\n","mean     11331.595333\n","std         16.417250\n","min      11216.000000\n","25%      11331.000000\n","50%      11333.000000\n","75%      11337.000000\n","max      11459.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-mini, Shots: 30\n","count     3000.000000\n","mean     11331.595333\n","std         16.417250\n","min      11216.000000\n","25%      11331.000000\n","50%      11333.000000\n","75%      11337.000000\n","max      11459.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o, Shots: 40\n","count    3000.000000\n","mean     9916.806333\n","std        10.057595\n","min      9856.000000\n","25%      9914.000000\n","50%      9917.000000\n","75%      9920.250000\n","max      9998.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o-mini, Shots: 40\n","count    3000.000000\n","mean     9916.806333\n","std        10.057595\n","min      9856.000000\n","25%      9914.000000\n","50%      9917.000000\n","75%      9920.250000\n","max      9998.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-preview, Shots: 40\n","count     3000.000000\n","mean     14933.595333\n","std         16.417250\n","min      14818.000000\n","25%      14933.000000\n","50%      14935.000000\n","75%      14939.000000\n","max      15061.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-mini, Shots: 40\n","count     3000.000000\n","mean     14933.595333\n","std         16.417250\n","min      14818.000000\n","25%      14933.000000\n","50%      14935.000000\n","75%      14939.000000\n","max      15061.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o, Shots: 50\n","count     3000.000000\n","mean     12394.806333\n","std         10.057595\n","min      12334.000000\n","25%      12392.000000\n","50%      12395.000000\n","75%      12398.250000\n","max      12476.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: gpt-4o-mini, Shots: 50\n","count     3000.000000\n","mean     12394.806333\n","std         10.057595\n","min      12334.000000\n","25%      12392.000000\n","50%      12395.000000\n","75%      12398.250000\n","max      12476.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-preview, Shots: 50\n","count     3000.000000\n","mean     18632.595333\n","std         16.417250\n","min      18517.000000\n","25%      18632.000000\n","50%      18634.000000\n","75%      18638.000000\n","max      18760.000000\n","Name: num_tokens, dtype: float64\n","loading train/test data files\n","DatasetDict({\n","    train: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 25000\n","    })\n","    test: Dataset({\n","        features: ['text', 'label', 'title', 'puzzle', 'truth'],\n","        num_rows: 3000\n","    })\n","})\n","Model: o1-mini, Shots: 50\n","count     3000.000000\n","mean     18632.595333\n","std         16.417250\n","min      18517.000000\n","25%      18632.000000\n","50%      18634.000000\n","75%      18638.000000\n","max      18760.000000\n","Name: num_tokens, dtype: float64\n"]}],"source":["for num_shots in [0, 5, 10, 20, 30, 40, 50]:\n","    for model_name in [\"gpt-4o\", \"gpt-4o-mini\", \"o1-preview\", \"o1-mini\"]:\n","\n","        datasets = load_logical_reasoning_dataset(\n","            data_path,\n","        )\n","        print(f\"Model: {model_name}, Shots: {num_shots}\")\n","        test_df = datasets[\"test\"].to_pandas()\n","        # print_row_details(test_df)\n","        test_df[\"num_tokens\"] = test_df.apply(\n","            lambda x: num_tokens_from_row(x, num_shots, datasets[\"train\"].to_pandas(), model=model_name), axis=1\n","        )\n","        print(test_df[\"num_tokens\"].describe())\n","\n","        model_test_dfs[(model_name, num_shots)] = test_df"]},{"cell_type":"code","execution_count":17,"metadata":{},"outputs":[{"data":{"text/html":["<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>model_name</th>\n","      <th>num_shots</th>\n","      <th>max</th>\n","      <th>min</th>\n","      <th>mean</th>\n","      <th>std</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>Llama3.1-8B-Chinese-Chat</td>\n","      <td>0</td>\n","      <td>652</td>\n","      <td>512</td>\n","      <td>571.091000</td>\n","      <td>9.115687</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>Llama3.1-70B-Chinese-Chat</td>\n","      <td>0</td>\n","      <td>652</td>\n","      <td>512</td>\n","      <td>571.091000</td>\n","      <td>9.115687</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>Mistral-7B-v0.3-Chinese-Chat</td>\n","      <td>0</td>\n","      <td>928</td>\n","      <td>694</td>\n","      <td>799.354000</td>\n","      <td>15.567385</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>internlm2_5-7b-chat</td>\n","      <td>0</td>\n","      <td>511</td>\n","      <td>426</td>\n","      <td>461.917667</td>\n","      <td>7.767732</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>internlm2_5-7b-chat-1m</td>\n","      <td>0</td>\n","      <td>511</td>\n","      <td>426</td>\n","      <td>461.917667</td>\n","      <td>7.767732</td>\n","    </tr>\n","    <tr>\n","      <th>5</th>\n","      <td>internlm2_5-20b-chat</td>\n","      <td>0</td>\n","      <td>511</td>\n","      <td>426</td>\n","      <td>461.917667</td>\n","      <td>7.767732</td>\n","    </tr>\n","    <tr>\n","      <th>6</th>\n","      <td>Qwen2.5-0.5B-Instruct</td>\n","      <td>0</td>\n","      <td>517</td>\n","      <td>426</td>\n","      <td>465.338667</td>\n","      <td>8.617118</td>\n","    </tr>\n","    <tr>\n","      <th>7</th>\n","      <td>Qwen2.5-1.5B-Instruct</td>\n","      <td>0</td>\n","      <td>517</td>\n","      <td>426</td>\n","      <td>465.338667</td>\n","      <td>8.617118</td>\n","    </tr>\n","    <tr>\n","      <th>8</th>\n","      <td>Qwen2.5-3B-Instruct</td>\n","      <td>0</td>\n","      <td>517</td>\n","      <td>426</td>\n","      <td>465.338667</td>\n","      <td>8.617118</td>\n","    </tr>\n","    <tr>\n","      <th>9</th>\n","      <td>Qwen2.5-7B-Instruct</td>\n","      <td>0</td>\n","      <td>517</td>\n","      <td>426</td>\n","      <td>465.338667</td>\n","      <td>8.617118</td>\n","    </tr>\n","  </tbody>\n","</table>\n","</div>"],"text/plain":["                     model_name  num_shots  max  min        mean        std\n","0      Llama3.1-8B-Chinese-Chat          0  652  512  571.091000   9.115687\n","1     Llama3.1-70B-Chinese-Chat          0  652  512  571.091000   9.115687\n","2  Mistral-7B-v0.3-Chinese-Chat          0  928  694  799.354000  15.567385\n","3           internlm2_5-7b-chat          0  511  426  461.917667   7.767732\n","4        internlm2_5-7b-chat-1m          0  511  426  461.917667   7.767732\n","5          internlm2_5-20b-chat          0  511  426  461.917667   7.767732\n","6         Qwen2.5-0.5B-Instruct          0  517  426  465.338667   8.617118\n","7         Qwen2.5-1.5B-Instruct          0  517  426  465.338667   8.617118\n","8           Qwen2.5-3B-Instruct          0  517  426  465.338667   8.617118\n","9           Qwen2.5-7B-Instruct          0  517  426  465.338667   8.617118"]},"execution_count":17,"metadata":{},"output_type":"execute_result"}],"source":["df_token_counts = pd.DataFrame(\n","    model_test_dfs.keys(), columns=[\"model_name\", \"num_shots\"]\n",")\n","\n","max = []\n","min = []\n","mean = []\n","std = []\n","\n","for model_name, num_shots in model_test_dfs.keys():\n","    test_df = model_test_dfs[(model_name, num_shots)]\n","    max.append(test_df[\"num_tokens\"].max())\n","    min.append(test_df[\"num_tokens\"].min())\n","    mean.append(test_df[\"num_tokens\"].mean())\n","    std.append(test_df[\"num_tokens\"].std())\n","\n","df_token_counts[\"max\"] = max\n","df_token_counts[\"min\"] = min\n","df_token_counts[\"mean\"] = mean\n","df_token_counts[\"std\"] = std\n","\n","df_token_counts.head(10)"]},{"cell_type":"code","execution_count":18,"metadata":{},"outputs":[{"data":{"text/html":["<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>model_name</th>\n","      <th>num_shots</th>\n","      <th>max</th>\n","      <th>min</th>\n","      <th>mean</th>\n","      <th>std</th>\n","    </tr>\n","  </thead>\n","  <tbody>\n","    <tr>\n","      <th>0</th>\n","      <td>Llama3.1-8B-Chinese-Chat</td>\n","      <td>0</td>\n","      <td>652</td>\n","      <td>512</td>\n","      <td>571.091000</td>\n","      <td>9.115687</td>\n","    </tr>\n","    <tr>\n","      <th>1</th>\n","      <td>Llama3.1-70B-Chinese-Chat</td>\n","      <td>0</td>\n","      <td>652</td>\n","      <td>512</td>\n","      <td>571.091000</td>\n","      <td>9.115687</td>\n","    </tr>\n","    <tr>\n","      <th>2</th>\n","      <td>Mistral-7B-v0.3-Chinese-Chat</td>\n","      <td>0</td>\n","      <td>928</td>\n","      <td>694</td>\n","      <td>799.354000</td>\n","      <td>15.567385</td>\n","    </tr>\n","    <tr>\n","      <th>3</th>\n","      <td>internlm2_5-7b-chat</td>\n","      <td>0</td>\n","      <td>511</td>\n","      <td>426</td>\n","      <td>461.917667</td>\n","      <td>7.767732</td>\n","    </tr>\n","    <tr>\n","      <th>4</th>\n","      <td>internlm2_5-7b-chat-1m</td>\n","      <td>0</td>\n","      <td>511</td>\n","      <td>426</td>\n","      <td>461.917667</td>\n","      <td>7.767732</td>\n","    </tr>\n","    <tr>\n","      <th>...</th>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","      <td>...</td>\n","    </tr>\n","    <tr>\n","      <th>100</th>\n","      <td>o1-mini</td>\n","      <td>40</td>\n","      <td>15061</td>\n","      <td>14818</td>\n","      <td>14933.595333</td>\n","      <td>16.417250</td>\n","    </tr>\n","    <tr>\n","      <th>101</th>\n","      <td>gpt-4o</td>\n","      <td>50</td>\n","      <td>12476</td>\n","      <td>12334</td>\n","      <td>12394.806333</td>\n","      <td>10.057595</td>\n","    </tr>\n","    <tr>\n","      <th>102</th>\n","      <td>gpt-4o-mini</td>\n","      <td>50</td>\n","      <td>12476</td>\n","      <td>12334</td>\n","      <td>12394.806333</td>\n","      <td>10.057595</td>\n","    </tr>\n","    <tr>\n","      <th>103</th>\n","      <td>o1-preview</td>\n","      <td>50</td>\n","      <td>18760</td>\n","      <td>18517</td>\n","      <td>18632.595333</td>\n","      <td>16.417250</td>\n","    </tr>\n","    <tr>\n","      <th>104</th>\n","      <td>o1-mini</td>\n","      <td>50</td>\n","      <td>18760</td>\n","      <td>18517</td>\n","      <td>18632.595333</td>\n","      <td>16.417250</td>\n","    </tr>\n","  </tbody>\n","</table>\n","<p>105 rows × 6 columns</p>\n","</div>"],"text/plain":["                       model_name  num_shots    max    min          mean  \\\n","0        Llama3.1-8B-Chinese-Chat          0    652    512    571.091000   \n","1       Llama3.1-70B-Chinese-Chat          0    652    512    571.091000   \n","2    Mistral-7B-v0.3-Chinese-Chat          0    928    694    799.354000   \n","3             internlm2_5-7b-chat          0    511    426    461.917667   \n","4          internlm2_5-7b-chat-1m          0    511    426    461.917667   \n","..                            ...        ...    ...    ...           ...   \n","100                       o1-mini         40  15061  14818  14933.595333   \n","101                        gpt-4o         50  12476  12334  12394.806333   \n","102                   gpt-4o-mini         50  12476  12334  12394.806333   \n","103                    o1-preview         50  18760  18517  18632.595333   \n","104                       o1-mini         50  18760  18517  18632.595333   \n","\n","           std  \n","0     9.115687  \n","1     9.115687  \n","2    15.567385  \n","3     7.767732  \n","4     7.767732  \n","..         ...  \n","100  16.417250  \n","101  10.057595  \n","102  10.057595  \n","103  16.417250  \n","104  16.417250  \n","\n","[105 rows x 6 columns]"]},"execution_count":18,"metadata":{},"output_type":"execute_result"}],"source":["df_token_counts"]},{"cell_type":"code","execution_count":28,"metadata":{},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 1200x600 with 1 Axes>"]},"metadata":{},"output_type":"display_data"}],"source":["# plot mean/std of token counts vs shots for each model\n","fig, ax = plt.subplots(1, 1, figsize=(12, 6))\n","\n","for model_name in model_orders.keys():\n","    model_df = df_token_counts[df_token_counts[\"model_name\"] == model_name]\n","    # ax.errorbar(\n","    #     model_df[\"num_shots\"],\n","    #     model_df[\"mean\"],\n","    #     yerr=model_df[\"std\"],\n","    #     marker=model_markers[model_name],\n","    #     label=model_name,\n","    #     fmt='o',  # Format of the marker\n","    #     capsize=5,  # Length of the error bar caps\n","    #     capthick=2,  # Thickness of the error bar caps\n","    #     elinewidth=2,  # Thickness of the error bars\n","    # )\n","\n","    # Connect mean dots with lines\n","    ax.plot(\n","        model_df[\"num_shots\"],\n","        model_df[\"mean\"],\n","        marker=model_markers[model_name],\n","        label=model_name,\n","    )\n","\n","# Add title and labels\n","ax.set_xlabel(\"Number of Shots\")\n","ax.set_ylabel(\"Mean Token Count\")\n","ax.set_title(\"Mean Token Count vs Shots for Models\")\n","\n","# Set x-axis grid spacing to 5\n","ax.xaxis.set_major_locator(MultipleLocator(5))\n","\n","# Rotate x labels\n","plt.xticks(rotation=0)\n","plt.grid(True)\n","# plt.tight_layout()\n","\n","# Set legend at the right to avoid overlapping with lines\n","plt.legend(loc=\"center left\", bbox_to_anchor=(1.0, 0.5))\n","\n","plt.show()"]},{"cell_type":"code","execution_count":20,"metadata":{},"outputs":[],"source":["df_token_counts.to_csv(\"data/all_model_token_counts.csv\", index=False)"]}],"metadata":{"accelerator":"GPU","application/vnd.databricks.v1+notebook":{"dashboards":[],"environmentMetadata":null,"language":"python","notebookMetadata":{"pythonIndentUnit":4},"notebookName":"07_MAC_+_Qwen2-7B-Instructi_Unsloth_train","widgets":{}},"colab":{"gpuType":"T4","provenance":[]},"kernelspec":{"display_name":"Python 3","name":"python3"},"language_info":{"codemirror_mode":{"name":"ipython","version":3},"file_extension":".py","mimetype":"text/x-python","name":"python","nbconvert_exporter":"python","pygments_lexer":"ipython3","version":"3.11.9"}},"nbformat":4,"nbformat_minor":0}
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8744ee475803e9996107b2826fcf4de51a243900e22107861787163d394ea0e
+size 419879