diff --git "a/Hatespeech_Offensive_Classification_llama3.2-3B-instruct.ipynb" "b/Hatespeech_Offensive_Classification_llama3.2-3B-instruct.ipynb"
new file mode 100644--- /dev/null
+++ "b/Hatespeech_Offensive_Classification_llama3.2-3B-instruct.ipynb"
@@ -0,0 +1,1738 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "959444f6-bb64-49f4-b537-d764680219ca",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#!pip install -U bitsandbytes\n",
+    "#!pip install -U transformers\n",
+    "#!pip install -U accelerate\n",
+    "#!pip install -U peft\n",
+    "#!pip install -U trl"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "74ca2d22-ee78-4f31-8769-efd3dae9c46c",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "#!huggingface-cli whoami"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "682a7a96-c5b8-4595-b495-99e13b88a844",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import os\n",
+    "from tqdm import tqdm\n",
+    "import bitsandbytes as bnb\n",
+    "import torch\n",
+    "import torch.nn as nn\n",
+    "import transformers\n",
+    "from datasets import Dataset\n",
+    "from peft import LoraConfig, PeftConfig\n",
+    "from trl import SFTTrainer\n",
+    "from trl import setup_chat_format\n",
+    "from transformers import (AutoModelForCausalLM, \n",
+    "                          AutoTokenizer, \n",
+    "                          BitsAndBytesConfig, \n",
+    "                          TrainingArguments, \n",
+    "                          pipeline, \n",
+    "                          logging)\n",
+    "from sklearn.metrics import (accuracy_score, \n",
+    "                             classification_report, \n",
+    "                             confusion_matrix)\n",
+    "from sklearn.model_selection import train_test_split"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "5cbaba37-4deb-4a60-b4a7-dacd3c75c62b",
+   "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>count</th>\n",
+       "      <th>hate_speech_count</th>\n",
+       "      <th>offensive_language_count</th>\n",
+       "      <th>neither_count</th>\n",
+       "      <th>class</th>\n",
+       "      <th>tweet</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>3</td>\n",
+       "      <td>0</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>2</td>\n",
+       "      <td>!!! RT @mayasolovely: As a woman you shouldn't...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>3</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>!!!!! RT @mleew17: boy dats cold...tyga dwn ba...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>3</td>\n",
+       "      <td>0</td>\n",
+       "      <td>3</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>!!!!!!! RT @UrKindOfBrand Dawg!!!! RT @80sbaby...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>3</td>\n",
+       "      <td>0</td>\n",
+       "      <td>2</td>\n",
+       "      <td>1</td>\n",
+       "      <td>1</td>\n",
+       "      <td>!!!!!!!!! RT @C_G_Anderson: @viva_based she lo...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>6</td>\n",
+       "      <td>0</td>\n",
+       "      <td>6</td>\n",
+       "      <td>0</td>\n",
+       "      <td>1</td>\n",
+       "      <td>!!!!!!!!!!!!! RT @ShenikaRoberts: The shit you...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "   count  hate_speech_count  offensive_language_count  neither_count  class  \\\n",
+       "0      3                  0                         0              3      2   \n",
+       "1      3                  0                         3              0      1   \n",
+       "2      3                  0                         3              0      1   \n",
+       "3      3                  0                         2              1      1   \n",
+       "4      6                  0                         6              0      1   \n",
+       "\n",
+       "                                               tweet  \n",
+       "0  !!! RT @mayasolovely: As a woman you shouldn't...  \n",
+       "1  !!!!! RT @mleew17: boy dats cold...tyga dwn ba...  \n",
+       "2  !!!!!!! RT @UrKindOfBrand Dawg!!!! RT @80sbaby...  \n",
+       "3  !!!!!!!!! RT @C_G_Anderson: @viva_based she lo...  \n",
+       "4  !!!!!!!!!!!!! RT @ShenikaRoberts: The shit you...  "
+      ]
+     },
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "from datasets import load_dataset\n",
+    "\n",
+    "df = pd.read_parquet(\"hf://datasets/tdavidson/hate_speech_offensive/data/train-00000-of-00001.parquet\")\n",
+    "df.head()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "90b89c75-ab77-42ee-b50c-466d6cc9de96",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/tmp/ipykernel_11705/3716630465.py:2: FutureWarning: Setting an item of incompatible dtype is deprecated and will raise in a future error of pandas. Value '[1 1 1 ... 1 1 2]' has dtype incompatible with int64, please explicitly cast to a compatible dtype first.\n",
+      "  df.loc[:,'label'] = df.loc[:,'label'].replace(0,'Hate')\n"
+     ]
+    }
+   ],
+   "source": [
+    "df = df.rename(columns={\"class\": \"label\",\"tweet\": \"text\"}).sample(frac=1, random_state=85).reset_index(drop=True).head(3000)\n",
+    "df.loc[:,'label'] = df.loc[:,'label'].replace(0,'Hate')\n",
+    "df.loc[:,'label'] = df.loc[:,'label'].replace(1,'Offensive')\n",
+    "df.loc[:,'label'] = df.loc[:,'label'].replace(2,'Normal')\n",
+    "# Split the DataFrame\n",
+    "train_size = 0.8\n",
+    "eval_size = 0.1\n",
+    "\n",
+    "# Calculate sizes\n",
+    "train_end = int(train_size * len(df))\n",
+    "eval_end = train_end + int(eval_size * len(df))\n",
+    "\n",
+    "# Split the data\n",
+    "X_train = df[:train_end]\n",
+    "X_eval = df[train_end:eval_end]\n",
+    "X_test = df[eval_end:]\n",
+    "# Define the prompt generation functions\n",
+    "def generate_prompt(data_point):\n",
+    "    return f\"\"\"\n",
+    "            Classify the text into Hatespeech, Offensive, Normal and return the answer as the corresponding label.\n",
+    "text: {data_point[\"text\"]}\n",
+    "label: {data_point[\"label\"]}\"\"\".strip()\n",
+    "\n",
+    "def generate_test_prompt(data_point):\n",
+    "    return f\"\"\"\n",
+    "            Classify the text into Hatespeech, Offensive, Normal and return the answer as the corresponding label.\n",
+    "            text: {data_point[\"text\"]}\n",
+    "            label: \"\"\".strip()\n",
+    "\n",
+    "# Generate prompts for training and evaluation data\n",
+    "X_train.loc[:,'text'] = X_train.apply(generate_prompt, axis=1)\n",
+    "X_eval.loc[:,'text'] = X_eval.apply(generate_prompt, axis=1)\n",
+    "\n",
+    "# Generate test prompts and extract true labels\n",
+    "y_true = X_test.loc[:,'label']\n",
+    "X_test = pd.DataFrame(X_test.apply(generate_test_prompt, axis=1), columns=[\"text\"])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "25859362-af09-43fd-93ef-b981159a7dba",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "label\n",
+       "Offensive    1877\n",
+       "Normal        391\n",
+       "Hate          132\n",
+       "Name: count, dtype: int64"
+      ]
+     },
+     "execution_count": 6,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "X_train.label.value_counts()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "12010dd7-0885-4296-9d8e-dae9d2addc30",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "train_data = Dataset.from_pandas(X_train[[\"text\"]])\n",
+    "eval_data = Dataset.from_pandas(X_eval[[\"text\"]])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "e8d7feef-d881-4e90-86d6-a4b30ac040f8",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "'Classify the text into Hatespeech, Offensive, Normal and return the answer as the corresponding label.\\ntext: @kieffer_jason bitch u a thot oh fake ass nigga box up hoe u not bout nothing\\nlabel: Offensive'"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "train_data['text'][2000]"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "6e8f818f-2c29-4637-93d0-750144b3f599",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b36e9ec950eb41909ae7159673c5a774",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "base_model_name = \"meta-llama/Llama-3.2-3B-Instruct\"\n",
+    "\n",
+    "bnb_config = BitsAndBytesConfig(\n",
+    "    load_in_4bit=True,\n",
+    "    bnb_4bit_use_double_quant=False,\n",
+    "    bnb_4bit_quant_type=\"nf4\",\n",
+    "    bnb_4bit_compute_dtype=\"float16\",\n",
+    ")\n",
+    "\n",
+    "model = AutoModelForCausalLM.from_pretrained(\n",
+    "    base_model_name,\n",
+    "    device_map=\"auto\",\n",
+    "    torch_dtype=\"float16\",\n",
+    "    quantization_config=bnb_config, \n",
+    ")\n",
+    "\n",
+    "model.config.use_cache = False\n",
+    "model.config.pretraining_tp = 1\n",
+    "\n",
+    "tokenizer = AutoTokenizer.from_pretrained(base_model_name)\n",
+    "\n",
+    "tokenizer.pad_token_id = tokenizer.eos_token_id"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "e10cba81-0a98-46bc-b4ae-19c89891af67",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "  0%|          | 0/300 [00:00<?, ?it/s]Device set to use cuda:0\n",
+      "  0%|          | 1/300 [00:00<01:37,  3.06it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  1%|          | 3/300 [00:00<00:41,  7.20it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  2%|▏         | 5/300 [00:00<00:30,  9.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  2%|▏         | 7/300 [00:00<00:25, 11.41it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  3%|▎         | 9/300 [00:00<00:23, 12.39it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  4%|▎         | 11/300 [00:01<00:22, 12.92it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  4%|▍         | 13/300 [00:01<00:21, 13.45it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  5%|▌         | 15/300 [00:01<00:20, 13.64it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  6%|▌         | 17/300 [00:01<00:20, 13.89it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  6%|▋         | 19/300 [00:01<00:19, 14.21it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  7%|▋         | 21/300 [00:01<00:19, 14.03it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  8%|▊         | 23/300 [00:01<00:19, 14.19it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  8%|▊         | 25/300 [00:01<00:19, 14.38it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  9%|▉         | 27/300 [00:02<00:18, 14.57it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 10%|▉         | 29/300 [00:02<00:18, 14.71it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 10%|█         | 31/300 [00:02<00:18, 14.82it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 11%|█         | 33/300 [00:02<00:18, 14.65it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 12%|█▏        | 35/300 [00:02<00:17, 14.77it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 12%|█▏        | 37/300 [00:02<00:18, 14.25it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 13%|█▎        | 39/300 [00:02<00:18, 14.28it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 14%|█▎        | 41/300 [00:03<00:17, 14.41it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 14%|█▍        | 43/300 [00:03<00:18, 14.26it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 15%|█▌        | 45/300 [00:03<00:17, 14.31it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 16%|█▌        | 47/300 [00:03<00:17, 14.36it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 16%|█▋        | 49/300 [00:03<00:17, 14.36it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 17%|█▋        | 51/300 [00:03<00:17, 14.61it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 18%|█▊        | 53/300 [00:03<00:16, 14.69it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 18%|█▊        | 55/300 [00:04<00:16, 14.57it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 19%|█▉        | 57/300 [00:04<00:16, 14.33it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 20%|█▉        | 59/300 [00:04<00:16, 14.29it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 20%|██        | 61/300 [00:04<00:16, 14.28it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 21%|██        | 63/300 [00:04<00:16, 14.11it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 22%|██▏       | 65/300 [00:04<00:16, 14.31it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 22%|██▏       | 67/300 [00:04<00:16, 14.29it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 23%|██▎       | 69/300 [00:05<00:15, 14.48it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 24%|██▎       | 71/300 [00:05<00:15, 14.62it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 24%|██▍       | 73/300 [00:05<00:15, 14.48it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 25%|██▌       | 75/300 [00:05<00:15, 14.59it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 26%|██▌       | 77/300 [00:05<00:15, 14.71it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 26%|██▋       | 79/300 [00:05<00:15, 14.41it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 27%|██▋       | 81/300 [00:05<00:15, 14.56it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 28%|██▊       | 83/300 [00:05<00:14, 14.69it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 28%|██▊       | 85/300 [00:06<00:14, 14.74it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 29%|██▉       | 87/300 [00:06<00:14, 14.73it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 30%|██▉       | 89/300 [00:06<00:14, 14.72it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 30%|███       | 91/300 [00:06<00:14, 14.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 31%|███       | 93/300 [00:06<00:14, 14.57it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 32%|███▏      | 95/300 [00:06<00:13, 14.65it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 32%|███▏      | 97/300 [00:06<00:13, 14.72it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 33%|███▎      | 99/300 [00:07<00:13, 14.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 34%|███▎      | 101/300 [00:07<00:13, 14.52it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 34%|███▍      | 103/300 [00:07<00:13, 14.37it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 35%|███▌      | 105/300 [00:07<00:13, 14.52it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 36%|███▌      | 107/300 [00:07<00:13, 14.65it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 36%|███▋      | 109/300 [00:07<00:13, 14.67it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 37%|███▋      | 111/300 [00:07<00:12, 14.71it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 38%|███▊      | 113/300 [00:08<00:12, 14.53it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 38%|███▊      | 115/300 [00:08<00:12, 14.45it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 39%|███▉      | 117/300 [00:08<00:12, 14.59it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 40%|███▉      | 119/300 [00:08<00:12, 14.66it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 40%|████      | 121/300 [00:08<00:12, 14.72it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 41%|████      | 123/300 [00:08<00:11, 14.77it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 42%|████▏     | 125/300 [00:08<00:12, 14.57it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 42%|████▏     | 127/300 [00:09<00:11, 14.67it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 43%|████▎     | 129/300 [00:09<00:11, 14.73it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 44%|████▎     | 131/300 [00:09<00:11, 14.74it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 44%|████▍     | 133/300 [00:09<00:11, 14.65it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 45%|████▌     | 135/300 [00:09<00:11, 14.31it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 46%|████▌     | 137/300 [00:09<00:11, 14.31it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 46%|████▋     | 139/300 [00:09<00:11, 14.29it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 47%|████▋     | 141/300 [00:09<00:11, 14.23it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 48%|████▊     | 143/300 [00:10<00:10, 14.35it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 48%|████▊     | 145/300 [00:10<00:10, 14.26it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 49%|████▉     | 147/300 [00:10<00:10, 14.39it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 50%|████▉     | 149/300 [00:10<00:10, 14.60it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 50%|█████     | 151/300 [00:10<00:10, 14.46it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 51%|█████     | 153/300 [00:10<00:10, 14.18it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 52%|█████▏    | 155/300 [00:10<00:10, 14.35it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 52%|█████▏    | 157/300 [00:11<00:09, 14.36it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 53%|█████▎    | 159/300 [00:11<00:09, 14.55it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 54%|█████▎    | 161/300 [00:11<00:09, 14.73it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 54%|█████▍    | 163/300 [00:11<00:09, 14.85it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 55%|█████▌    | 165/300 [00:11<00:09, 14.73it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 56%|█████▌    | 167/300 [00:11<00:09, 14.29it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 56%|█████▋    | 169/300 [00:11<00:09, 14.13it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 57%|█████▋    | 171/300 [00:12<00:09, 14.19it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 58%|█████▊    | 173/300 [00:12<00:08, 14.47it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 58%|█████▊    | 175/300 [00:12<00:08, 14.69it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 59%|█████▉    | 177/300 [00:12<00:08, 14.60it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 60%|█████▉    | 179/300 [00:12<00:08, 14.79it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 60%|██████    | 181/300 [00:12<00:07, 14.94it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 61%|██████    | 183/300 [00:12<00:07, 15.02it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 62%|██████▏   | 185/300 [00:12<00:07, 15.02it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 62%|██████▏   | 187/300 [00:13<00:07, 14.85it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 63%|██████▎   | 189/300 [00:13<00:07, 14.67it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 64%|██████▎   | 191/300 [00:13<00:07, 14.45it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 64%|██████▍   | 193/300 [00:13<00:07, 14.65it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 65%|██████▌   | 195/300 [00:13<00:07, 14.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 66%|██████▌   | 197/300 [00:13<00:06, 14.87it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 66%|██████▋   | 199/300 [00:13<00:06, 14.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 67%|██████▋   | 201/300 [00:14<00:06, 14.87it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 68%|██████▊   | 203/300 [00:14<00:06, 14.98it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 68%|██████▊   | 205/300 [00:14<00:06, 15.06it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 69%|██████▉   | 207/300 [00:14<00:06, 14.87it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 70%|██████▉   | 209/300 [00:14<00:06, 14.74it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 70%|███████   | 211/300 [00:14<00:05, 14.84it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 71%|███████   | 213/300 [00:14<00:05, 14.90it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 72%|███████▏  | 215/300 [00:15<00:05, 14.95it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 72%|███████▏  | 217/300 [00:15<00:05, 14.83it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 73%|███████▎  | 219/300 [00:15<00:05, 14.90it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 74%|███████▎  | 221/300 [00:15<00:05, 14.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 74%|███████▍  | 223/300 [00:15<00:05, 14.61it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 75%|███████▌  | 225/300 [00:15<00:05, 14.24it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 76%|███████▌  | 227/300 [00:15<00:05, 14.44it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 76%|███████▋  | 229/300 [00:15<00:04, 14.33it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 77%|███████▋  | 231/300 [00:16<00:04, 14.27it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 78%|███████▊  | 233/300 [00:16<00:04, 14.41it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 78%|███████▊  | 235/300 [00:16<00:04, 14.64it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 79%|███████▉  | 237/300 [00:16<00:04, 14.80it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 80%|███████▉  | 239/300 [00:16<00:04, 14.91it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 80%|████████  | 241/300 [00:16<00:03, 14.97it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 81%|████████  | 243/300 [00:16<00:03, 15.04it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 82%|████████▏ | 245/300 [00:17<00:03, 15.03it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 82%|████████▏ | 247/300 [00:17<00:03, 14.59it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 83%|████████▎ | 249/300 [00:17<00:03, 14.51it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 84%|████████▎ | 251/300 [00:17<00:03, 14.42it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 84%|████████▍ | 253/300 [00:17<00:03, 14.15it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 85%|████████▌ | 255/300 [00:17<00:03, 14.43it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 86%|████████▌ | 257/300 [00:17<00:03, 14.17it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 86%|████████▋ | 259/300 [00:18<00:02, 14.17it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 87%|████████▋ | 261/300 [00:18<00:02, 14.41it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 88%|████████▊ | 263/300 [00:18<00:02, 14.31it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 88%|████████▊ | 265/300 [00:18<00:02, 14.22it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 89%|████████▉ | 267/300 [00:18<00:02, 14.22it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 90%|████████▉ | 269/300 [00:18<00:02, 14.42it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 90%|█████████ | 271/300 [00:18<00:02, 14.38it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 91%|█████████ | 273/300 [00:19<00:01, 14.56it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 92%|█████████▏| 275/300 [00:19<00:01, 14.70it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 92%|█████████▏| 277/300 [00:19<00:01, 14.82it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 93%|█████████▎| 279/300 [00:19<00:01, 14.90it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 94%|█████████▎| 281/300 [00:19<00:01, 14.93it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 94%|█████████▍| 283/300 [00:19<00:01, 14.96it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 95%|█████████▌| 285/300 [00:19<00:01, 14.97it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 96%|█████████▌| 287/300 [00:19<00:00, 14.92it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 96%|█████████▋| 289/300 [00:20<00:00, 14.72it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 97%|█████████▋| 291/300 [00:20<00:00, 14.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 98%|█████████▊| 293/300 [00:20<00:00, 14.59it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 98%|█████████▊| 295/300 [00:20<00:00, 14.69it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 99%|█████████▉| 297/300 [00:20<00:00, 14.75it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "100%|█████████▉| 299/300 [00:20<00:00, 14.81it/s]Device set to use cuda:0\n",
+      "100%|██████████| 300/300 [00:20<00:00, 14.40it/s]\n"
+     ]
+    }
+   ],
+   "source": [
+    "def predict(test, model, tokenizer):\n",
+    "    y_pred = []\n",
+    "    labels = [\"Hate\", \"Offensive\", \"Normal\"]\n",
+    "    \n",
+    "    for i in tqdm(range(len(test))):\n",
+    "        prompt = test.iloc[i][\"text\"]\n",
+    "        pipe = pipeline(task=\"text-generation\", \n",
+    "                        model=model, \n",
+    "                        tokenizer=tokenizer, \n",
+    "                        max_new_tokens=2, \n",
+    "                        temperature=0.1)\n",
+    "        \n",
+    "        result = pipe(prompt)\n",
+    "        answer = result[0]['generated_text'].split(\"label:\")[-1].strip()\n",
+    "        \n",
+    "        # Determine the predicted category\n",
+    "        for label in labels:\n",
+    "            if label.lower() in answer.lower():\n",
+    "                y_pred.append(label)\n",
+    "                break\n",
+    "        else:\n",
+    "            y_pred.append(\"none\")\n",
+    "    \n",
+    "    return y_pred\n",
+    "\n",
+    "y_pred = predict(X_test, model, tokenizer)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "de9a621b-49af-47ce-a8ca-aa74f8da219b",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Accuracy: 0.220\n",
+      "Accuracy for label Hate: 0.588\n",
+      "Accuracy for label Offensive: 0.112\n",
+      "Accuracy for label Normal: 0.600\n",
+      "\n",
+      "Classification Report:\n",
+      "              precision    recall  f1-score   support\n",
+      "\n",
+      "        Hate       0.07      0.59      0.12        17\n",
+      "   Offensive       0.76      0.11      0.19       233\n",
+      "      Normal       0.65      0.60      0.62        50\n",
+      "\n",
+      "   micro avg       0.29      0.22      0.25       300\n",
+      "   macro avg       0.49      0.43      0.31       300\n",
+      "weighted avg       0.71      0.22      0.26       300\n",
+      "\n",
+      "\n",
+      "Confusion Matrix:\n",
+      "[[ 10   1   1]\n",
+      " [133  26  15]\n",
+      " [  8   7  30]]\n"
+     ]
+    }
+   ],
+   "source": [
+    "def evaluate(y_true, y_pred):\n",
+    "    labels = [\"Hate\", \"Offensive\", \"Normal\"]\n",
+    "    mapping = {label: idx for idx, label in enumerate(labels)}\n",
+    "    \n",
+    "    def map_func(x):\n",
+    "        return mapping.get(x, -1)  # Map to -1 if not found, but should not occur with correct data\n",
+    "    \n",
+    "    y_true_mapped = np.vectorize(map_func)(y_true)\n",
+    "    y_pred_mapped = np.vectorize(map_func)(y_pred)\n",
+    "    \n",
+    "    # Calculate accuracy\n",
+    "    accuracy = accuracy_score(y_true=y_true_mapped, y_pred=y_pred_mapped)\n",
+    "    print(f'Accuracy: {accuracy:.3f}')\n",
+    "    \n",
+    "    # Generate accuracy report\n",
+    "    unique_labels = set(y_true_mapped)  # Get unique labels\n",
+    "    \n",
+    "    for label in unique_labels:\n",
+    "        label_indices = [i for i in range(len(y_true_mapped)) if y_true_mapped[i] == label]\n",
+    "        label_y_true = [y_true_mapped[i] for i in label_indices]\n",
+    "        label_y_pred = [y_pred_mapped[i] for i in label_indices]\n",
+    "        label_accuracy = accuracy_score(label_y_true, label_y_pred)\n",
+    "        print(f'Accuracy for label {labels[label]}: {label_accuracy:.3f}')\n",
+    "        \n",
+    "    # Generate classification report\n",
+    "    class_report = classification_report(y_true=y_true_mapped, y_pred=y_pred_mapped, target_names=labels, labels=list(range(len(labels))))\n",
+    "    print('\\nClassification Report:')\n",
+    "    print(class_report)\n",
+    "    \n",
+    "    # Generate confusion matrix\n",
+    "    conf_matrix = confusion_matrix(y_true=y_true_mapped, y_pred=y_pred_mapped, labels=list(range(len(labels))))\n",
+    "    print('\\nConfusion Matrix:')\n",
+    "    print(conf_matrix)\n",
+    "\n",
+    "evaluate(y_true, y_pred)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "e5c44ef1-377b-4223-899a-6ca2a0e3d537",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['up_proj', 'q_proj', 'gate_proj', 'o_proj', 'down_proj', 'k_proj', 'v_proj']"
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import bitsandbytes as bnb\n",
+    "\n",
+    "def find_all_linear_names(model):\n",
+    "    cls = bnb.nn.Linear4bit\n",
+    "    lora_module_names = set()\n",
+    "    for name, module in model.named_modules():\n",
+    "        if isinstance(module, cls):\n",
+    "            names = name.split('.')\n",
+    "            lora_module_names.add(names[0] if len(names) == 1 else names[-1])\n",
+    "    if 'lm_head' in lora_module_names:  # needed for 16 bit\n",
+    "        lora_module_names.remove('lm_head')\n",
+    "    return list(lora_module_names)\n",
+    "modules = find_all_linear_names(model)\n",
+    "modules"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "e9820a30-e05c-46fd-b53e-b6dbe2078820",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "cf08288afc584612b257dae370b981f5",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Map:   0%|          | 0/2400 [00:00<?, ? examples/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "6992092e12594739ad1135f2a8b2b99b",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Map:   0%|          | 0/300 [00:00<?, ? examples/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "#!pip install wandb\n",
+    "output_dir=\"/home/marco/llama-3.2-3B-instruct-offensive-classification-2\"\n",
+    "\n",
+    "peft_config = LoraConfig(\n",
+    "   lora_alpha=16,\n",
+    "    lora_dropout=0,\n",
+    "    r=64,\n",
+    "    bias=\"none\",\n",
+    "   task_type=\"CAUSAL_LM\",\n",
+    "    target_modules=modules,\n",
+    ")\n",
+    "\n",
+    "training_arguments = TrainingArguments(\n",
+    "    output_dir=output_dir,                    # directory to save and repository id\n",
+    "    num_train_epochs=1,                       # number of training epochs\n",
+    "    per_device_train_batch_size=1,            # batch size per device during training\n",
+    "    gradient_accumulation_steps=8,            # number of steps before performing a backward/update pass\n",
+    "    gradient_checkpointing=True,              # use gradient checkpointing to save memory\n",
+    "    optim=\"paged_adamw_32bit\",\n",
+    "    logging_steps=1,                         \n",
+    "   learning_rate=2e-4,                       # learning rate, based on QLoRA paper\n",
+    "    weight_decay=0.001,\n",
+    "    fp16=True,\n",
+    "    bf16=False,\n",
+    "      max_grad_norm=0.3,                        # max gradient norm based on QLoRA paper\n",
+    "    max_steps=-1,\n",
+    "    warmup_ratio=0.03,                        # warmup ratio based on QLoRA paper\n",
+    "    group_by_length=False,\n",
+    "    lr_scheduler_type=\"cosine\",               # use cosine learning rate scheduler\n",
+    "    report_to=\"wandb\",                  # report metrics to w&b\n",
+    "    eval_strategy=\"steps\",              # save checkpoint every epoch\n",
+    "    eval_steps = 0.2\n",
+    ")\n",
+    "\n",
+    "trainer = SFTTrainer(\n",
+    "   model=model,\n",
+    "    args=training_arguments,\n",
+    "    train_dataset=train_data,\n",
+    "    eval_dataset=eval_data,\n",
+    "    peft_config=peft_config,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 15,
+   "id": "5f2f6464-d19c-4e0f-8189-742622d91c7b",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "\u001b[34m\u001b[1mwandb\u001b[0m: \u001b[33mWARNING\u001b[0m The `run_name` is currently set to the same value as `TrainingArguments.output_dir`. If this was not intended, please specify a different run name by setting the `TrainingArguments.run_name` parameter.\n",
+      "\u001b[34m\u001b[1mwandb\u001b[0m: Using wandb-core as the SDK backend.  Please refer to https://wandb.me/wandb-core for more information.\n",
+      "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mmarcoor\u001b[0m (\u001b[33mmarcoor-universit-t-klagenfurt\u001b[0m). Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "Tracking run with wandb version 0.19.1"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "Run data is saved locally in <code>/home/marco/wandb/run-20250112_154819-ixq2tt7q</code>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       "Syncing run <strong><a href='https://wandb.ai/marcoor-universit-t-klagenfurt/huggingface/runs/ixq2tt7q' target=\"_blank\">/home/marco/llama-3.2-3B-instruct-offensive-classification-2</a></strong> to <a href='https://wandb.ai/marcoor-universit-t-klagenfurt/huggingface' target=\"_blank\">Weights & Biases</a> (<a href='https://wandb.me/developer-guide' target=\"_blank\">docs</a>)<br>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       " View project at <a href='https://wandb.ai/marcoor-universit-t-klagenfurt/huggingface' target=\"_blank\">https://wandb.ai/marcoor-universit-t-klagenfurt/huggingface</a>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/html": [
+       " View run at <a href='https://wandb.ai/marcoor-universit-t-klagenfurt/huggingface/runs/ixq2tt7q' target=\"_blank\">https://wandb.ai/marcoor-universit-t-klagenfurt/huggingface/runs/ixq2tt7q</a>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/marco/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/torch/_dynamo/eval_frame.py:632: UserWarning: torch.utils.checkpoint: the use_reentrant parameter should be passed explicitly. In version 2.5 we will raise an exception if use_reentrant is not passed. use_reentrant=False is recommended, but if you need to preserve the current default behavior, you can pass use_reentrant=True. Refer to docs for more details on the differences between the two variants.\n",
+      "  return fn(*args, **kwargs)\n"
+     ]
+    },
+    {
+     "data": {
+      "text/html": [
+       "\n",
+       "    <div>\n",
+       "      \n",
+       "      <progress value='300' max='300' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
+       "      [300/300 09:53, Epoch 1/1]\n",
+       "    </div>\n",
+       "    <table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       " <tr style=\"text-align: left;\">\n",
+       "      <th>Step</th>\n",
+       "      <th>Training Loss</th>\n",
+       "      <th>Validation Loss</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>60</td>\n",
+       "      <td>2.011600</td>\n",
+       "      <td>2.002830</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>120</td>\n",
+       "      <td>1.853900</td>\n",
+       "      <td>1.961909</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>180</td>\n",
+       "      <td>2.088800</td>\n",
+       "      <td>1.939240</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>240</td>\n",
+       "      <td>1.923100</td>\n",
+       "      <td>1.927367</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>300</td>\n",
+       "      <td>2.089000</td>\n",
+       "      <td>1.924164</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table><p>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/marco/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/peft/utils/other.py:716: UserWarning: Unable to fetch remote file due to the following error (ProtocolError('Connection aborted.', RemoteDisconnected('Remote end closed connection without response')), '(Request ID: ebbfa21b-df77-43ef-bddd-b95ec07a63f8)') - silently ignoring the lookup for the file config.json in meta-llama/Llama-3.2-3B-Instruct.\n",
+      "  warnings.warn(\n",
+      "/home/marco/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/peft/utils/save_and_load.py:246: UserWarning: Could not find a config file in meta-llama/Llama-3.2-3B-Instruct - will assume that the vocabulary was not modified.\n",
+      "  warnings.warn(\n"
+     ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "TrainOutput(global_step=300, training_loss=2.0926066251595814, metrics={'train_runtime': 596.9063, 'train_samples_per_second': 4.021, 'train_steps_per_second': 0.503, 'total_flos': 2216727844706304.0, 'train_loss': 2.0926066251595814, 'epoch': 1.0})"
+      ]
+     },
+     "execution_count": 15,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "trainer.train()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 16,
+   "id": "d50bf803-dedd-44bf-8016-6d80b5726803",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "('/home/marco/llama-3.2-3B-instruct-offensive-classification-2/tokenizer_config.json',\n",
+       " '/home/marco/llama-3.2-3B-instruct-offensive-classification-2/special_tokens_map.json',\n",
+       " '/home/marco/llama-3.2-3B-instruct-offensive-classification-2/tokenizer.json')"
+      ]
+     },
+     "execution_count": 16,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "trainer.save_model(output_dir)\n",
+    "tokenizer.save_pretrained(output_dir)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 17,
+   "id": "37e30c0e-dd90-4400-aaab-b3526ac4a7ab",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "  0%|          | 0/300 [00:00<?, ?it/s]Device set to use cuda:0\n",
+      "  0%|          | 1/300 [00:00<00:32,  9.15it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  1%|          | 3/300 [00:00<00:30,  9.80it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  2%|▏         | 5/300 [00:00<00:28, 10.20it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  2%|▏         | 7/300 [00:00<00:27, 10.48it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  3%|▎         | 9/300 [00:00<00:27, 10.58it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  4%|▎         | 11/300 [00:01<00:27, 10.46it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  4%|▍         | 13/300 [00:01<00:26, 10.69it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  5%|▌         | 15/300 [00:01<00:26, 10.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  6%|▌         | 17/300 [00:01<00:26, 10.83it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  6%|▋         | 19/300 [00:01<00:25, 10.93it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  7%|▋         | 21/300 [00:01<00:25, 10.78it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  8%|▊         | 23/300 [00:02<00:25, 10.87it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  8%|▊         | 25/300 [00:02<00:25, 10.92it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "  9%|▉         | 27/300 [00:02<00:24, 10.98it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 10%|▉         | 29/300 [00:02<00:24, 11.04it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 10%|█         | 31/300 [00:02<00:24, 11.06it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 11%|█         | 33/300 [00:03<00:24, 10.87it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 12%|█▏        | 35/300 [00:03<00:24, 10.94it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 12%|█▏        | 37/300 [00:03<00:24, 10.68it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 13%|█▎        | 39/300 [00:03<00:24, 10.77it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 14%|█▎        | 41/300 [00:03<00:23, 10.87it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 14%|█▍        | 43/300 [00:03<00:23, 10.77it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 15%|█▌        | 45/300 [00:04<00:23, 10.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 16%|█▌        | 47/300 [00:04<00:23, 10.71it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 16%|█▋        | 49/300 [00:04<00:23, 10.67it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 17%|█▋        | 51/300 [00:04<00:23, 10.69it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 18%|█▊        | 53/300 [00:04<00:23, 10.70it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 18%|█▊        | 55/300 [00:05<00:22, 10.83it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 19%|█▉        | 57/300 [00:05<00:22, 10.70it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 20%|█▉        | 59/300 [00:05<00:22, 10.65it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 20%|██        | 61/300 [00:05<00:22, 10.57it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 21%|██        | 63/300 [00:05<00:23, 10.26it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 22%|██▏       | 65/300 [00:06<00:22, 10.33it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 22%|██▏       | 67/300 [00:06<00:22, 10.39it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 23%|██▎       | 69/300 [00:06<00:21, 10.57it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 24%|██▎       | 71/300 [00:06<00:21, 10.58it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 24%|██▍       | 73/300 [00:06<00:21, 10.46it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 25%|██▌       | 75/300 [00:07<00:21, 10.63it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 26%|██▌       | 77/300 [00:07<00:20, 10.77it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 26%|██▋       | 79/300 [00:07<00:21, 10.52it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 27%|██▋       | 81/300 [00:07<00:20, 10.69it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 28%|██▊       | 83/300 [00:07<00:20, 10.74it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 28%|██▊       | 85/300 [00:07<00:19, 10.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 29%|██▉       | 87/300 [00:08<00:20, 10.65it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 30%|██▉       | 89/300 [00:08<00:19, 10.69it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 30%|███       | 91/300 [00:08<00:19, 10.70it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 31%|███       | 93/300 [00:08<00:19, 10.46it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 32%|███▏      | 95/300 [00:08<00:19, 10.59it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 32%|███▏      | 97/300 [00:09<00:18, 10.72it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 33%|███▎      | 99/300 [00:09<00:18, 10.78it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 34%|███▎      | 101/300 [00:09<00:18, 10.51it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 34%|███▍      | 103/300 [00:09<00:18, 10.37it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 35%|███▌      | 105/300 [00:09<00:18, 10.56it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 36%|███▌      | 107/300 [00:10<00:18, 10.69it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 36%|███▋      | 109/300 [00:10<00:17, 10.70it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 37%|███▋      | 111/300 [00:10<00:17, 10.67it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 38%|███▊      | 113/300 [00:10<00:17, 10.55it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 38%|███▊      | 115/300 [00:10<00:17, 10.51it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 39%|███▉      | 117/300 [00:10<00:17, 10.65it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 40%|███▉      | 119/300 [00:11<00:16, 10.74it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 40%|████      | 121/300 [00:11<00:16, 10.84it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 41%|████      | 123/300 [00:11<00:16, 10.91it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 42%|████▏     | 125/300 [00:11<00:16, 10.73it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 42%|████▏     | 127/300 [00:11<00:15, 10.85it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 43%|████▎     | 129/300 [00:12<00:15, 10.93it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 44%|████▎     | 131/300 [00:12<00:15, 11.01it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 44%|████▍     | 133/300 [00:12<00:15, 11.04it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 45%|████▌     | 135/300 [00:12<00:15, 10.86it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 46%|████▌     | 137/300 [00:12<00:14, 10.88it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 46%|████▋     | 139/300 [00:12<00:14, 10.96it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 47%|████▋     | 141/300 [00:13<00:14, 10.81it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 48%|████▊     | 143/300 [00:13<00:14, 10.87it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 48%|████▊     | 145/300 [00:13<00:14, 10.84it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 49%|████▉     | 147/300 [00:13<00:14, 10.90it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 50%|████▉     | 149/300 [00:13<00:13, 10.94it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 50%|██���██     | 151/300 [00:14<00:13, 10.96it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 51%|█████     | 153/300 [00:14<00:13, 10.79it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 52%|█████▏    | 155/300 [00:14<00:13, 10.80it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 52%|█████▏    | 157/300 [00:14<00:13, 10.65it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 53%|█████▎    | 159/300 [00:14<00:13, 10.74it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 54%|█████▎    | 161/300 [00:15<00:12, 10.74it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 54%|█████▍    | 163/300 [00:15<00:12, 10.78it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 55%|█████▌    | 165/300 [00:15<00:12, 10.63it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 56%|█████▌    | 167/300 [00:15<00:12, 10.36it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 56%|█████▋    | 169/300 [00:15<00:12, 10.25it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 57%|█████▋    | 171/300 [00:16<00:12, 10.27it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 58%|█████▊    | 173/300 [00:16<00:12, 10.48it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 58%|█████▊    | 175/300 [00:16<00:11, 10.61it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 59%|█████▉    | 177/300 [00:16<00:11, 10.54it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 60%|█████▉    | 179/300 [00:16<00:11, 10.68it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 60%|██████    | 181/300 [00:16<00:11, 10.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 61%|██████    | 183/300 [00:17<00:10, 10.82it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 62%|██████▏   | 185/300 [00:17<00:10, 10.81it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 62%|██████▏   | 187/300 [00:17<00:10, 10.71it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 63%|██████▎   | 189/300 [00:17<00:10, 10.63it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 64%|██████▎   | 191/300 [00:17<00:10, 10.54it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 64%|██████▍   | 193/300 [00:18<00:10, 10.67it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 65%|██████▌   | 195/300 [00:18<00:09, 10.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 66%|██████▌   | 197/300 [00:18<00:09, 10.86it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 66%|██████▋   | 199/300 [00:18<00:09, 10.75it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 67%|██████▋   | 201/300 [00:18<00:09, 10.82it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 68%|██████▊   | 203/300 [00:18<00:08, 10.87it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 68%|██████▊   | 205/300 [00:19<00:08, 10.88it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 69%|██████▉   | 207/300 [00:19<00:08, 10.73it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 70%|██████▉   | 209/300 [00:19<00:08, 10.65it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 70%|███████   | 211/300 [00:19<00:08, 10.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 71%|███████   | 213/300 [00:19<00:08, 10.79it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 72%|███████▏  | 215/300 [00:20<00:07, 10.75it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 72%|███████▏  | 217/300 [00:20<00:07, 10.76it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 73%|███████▎  | 219/300 [00:20<00:07, 10.79it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 74%|███████▎  | 221/300 [00:20<00:07, 10.67it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 74%|███████▍  | 223/300 [00:20<00:07, 10.58it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 75%|███████▌  | 225/300 [00:21<00:07, 10.36it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 76%|███████▌  | 227/300 [00:21<00:06, 10.54it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 76%|███████▋  | 229/300 [00:21<00:06, 10.41it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 77%|███████▋  | 231/300 [00:21<00:06, 10.22it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 78%|███████▊  | 233/300 [00:21<00:06, 10.40it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 78%|███████▊  | 235/300 [00:21<00:06, 10.56it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 79%|███████▉  | 237/300 [00:22<00:05, 10.65it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 80%|███████▉  | 239/300 [00:22<00:05, 10.72it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 80%|████████  | 241/300 [00:22<00:05, 10.75it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 81%|████████  | 243/300 [00:22<00:05, 10.85it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 82%|████████▏ | 245/300 [00:22<00:05, 10.90it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 82%|████████▏ | 247/300 [00:23<00:04, 10.69it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 83%|████████▎ | 249/300 [00:23<00:04, 10.53it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 84%|████████▎ | 251/300 [00:23<00:04, 10.45it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 84%|████████▍ | 253/300 [00:23<00:04, 10.36it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 85%|████████▌ | 255/300 [00:23<00:04, 10.52it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 86%|████████▌ | 257/300 [00:24<00:04, 10.18it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 86%|████████▋ | 259/300 [00:24<00:04, 10.05it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 87%|████████▋ | 261/300 [00:24<00:03, 10.24it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 88%|████████▊ | 263/300 [00:24<00:03, 10.28it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 88%|████████▊ | 265/300 [00:24<00:03, 10.21it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 89%|████████▉ | 267/300 [00:25<00:03, 10.18it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 90%|████████▉ | 269/300 [00:25<00:02, 10.35it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 90%|█████████ | 271/300 [00:25<00:02, 10.28it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 91%|█████████ | 273/300 [00:25<00:02, 10.36it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 92%|█████████▏| 275/300 [00:25<00:02, 10.37it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 92%|█████████▏| 277/300 [00:26<00:02, 10.51it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 93%|█████████▎| 279/300 [00:26<00:01, 10.59it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 94%|█████████▎| 281/300 [00:26<00:01, 10.53it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 94%|█████████▍| 283/300 [00:26<00:01, 10.46it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 95%|█████████▌| 285/300 [00:26<00:01, 10.43it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 96%|█████████▌| 287/300 [00:26<00:01, 10.44it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 96%|█████████▋| 289/300 [00:27<00:01, 10.35it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 97%|█████████▋| 291/300 [00:27<00:00, 10.49it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 98%|█████████▊| 293/300 [00:27<00:00, 10.44it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 98%|█████████▊| 295/300 [00:27<00:00, 10.52it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      " 99%|█████████▉| 297/300 [00:27<00:00, 10.53it/s]Device set to use cuda:0\n",
+      "Device set to use cuda:0\n",
+      "100%|█████████▉| 299/300 [00:28<00:00, 10.48it/s]Device set to use cuda:0\n",
+      "100%|██████████| 300/300 [00:28<00:00, 10.63it/s]\n"
+     ]
+    }
+   ],
+   "source": [
+    "y_pred = predict(X_test, model, tokenizer)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "id": "3449c759-b590-4455-80d7-a1535e1b8f9f",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Accuracy: 0.903\n",
+      "Accuracy for label Hate: 0.176\n",
+      "Accuracy for label Offensive: 0.961\n",
+      "Accuracy for label Normal: 0.880\n",
+      "\n",
+      "Classification Report:\n",
+      "              precision    recall  f1-score   support\n",
+      "\n",
+      "        Hate       0.43      0.18      0.25        17\n",
+      "   Offensive       0.92      0.96      0.94       233\n",
+      "      Normal       0.88      0.88      0.88        50\n",
+      "\n",
+      "    accuracy                           0.90       300\n",
+      "   macro avg       0.74      0.67      0.69       300\n",
+      "weighted avg       0.89      0.90      0.89       300\n",
+      "\n",
+      "\n",
+      "Confusion Matrix:\n",
+      "[[  3  13   1]\n",
+      " [  4 224   5]\n",
+      " [  0   6  44]]\n"
+     ]
+    }
+   ],
+   "source": [
+    "evaluate(y_true, y_pred)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 19,
+   "id": "591f2b8c-3cb1-448d-b70a-e26b0dcfbafd",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "base_model = \"meta-llama/Llama-3.2-3B-Instruct\"\n",
+    "fine_tuned_model = \"/home/marco/llama-3.2-3B-instruct-offensive-classification-2\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 20,
+   "id": "64a3591f-8c98-41f9-8326-a9ed7d0da461",
+   "metadata": {
+    "scrolled": true
+   },
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "62a6fde13e574c21ae5c482ae9c6d9dc",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline\n",
+    "from peft import PeftModel\n",
+    "import torch\n",
+    "\n",
+    "\n",
+    "# Reload tokenizer and model\n",
+    "tokenizer = AutoTokenizer.from_pretrained(base_model)\n",
+    "\n",
+    "base_model_reload = AutoModelForCausalLM.from_pretrained(\n",
+    "        base_model,\n",
+    "        return_dict=True,\n",
+    "        low_cpu_mem_usage=True,\n",
+    "        torch_dtype=torch.float16,\n",
+    "        device_map=\"auto\",\n",
+    "        trust_remote_code=True,\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 21,
+   "id": "37e19428-9786-429f-b119-61ae7b7355ab",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "model = PeftModel.from_pretrained(base_model_reload, fine_tuned_model)\n",
+    "model = model.merge_and_unload()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 22,
+   "id": "9e448406-60a1-4c76-a827-0284f859843e",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Device set to use cpu\n",
+      "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Offensive\n"
+     ]
+    }
+   ],
+   "source": [
+    "text = \"You are an asshole!\"\n",
+    "prompt = f\"\"\"Classify the text into Hatespeech, Offensive, Normal and return the answer as the corresponding label.\n",
+    "text: {text}\n",
+    "label: \"\"\".strip()\n",
+    "\n",
+    "pipe = pipeline(\n",
+    "    \"text-generation\",\n",
+    "    model=model,\n",
+    "    tokenizer=tokenizer,\n",
+    "    torch_dtype=torch.float16,\n",
+    "    device_map=\"auto\",\n",
+    ")\n",
+    "\n",
+    "outputs = pipe(prompt, max_new_tokens=2, do_sample=True, temperature=0.1)\n",
+    "print(outputs[0][\"generated_text\"].split(\"label: \")[-1].strip())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 23,
+   "id": "8f479001-338c-4a06-b4e5-6d874a6d9b30",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Device set to use cpu\n",
+      "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Normal\n"
+     ]
+    }
+   ],
+   "source": [
+    "text = \"I walk in the park!\"\n",
+    "prompt = f\"\"\"Classify the text into Hatespeech, Offensive, Normal and return the answer as the corresponding label.\n",
+    "text: {text}\n",
+    "label: \"\"\".strip()\n",
+    "\n",
+    "pipe = pipeline(\n",
+    "    \"text-generation\",\n",
+    "    model=model,\n",
+    "    tokenizer=tokenizer,\n",
+    "    torch_dtype=torch.float16,\n",
+    "    device_map=\"auto\",\n",
+    ")\n",
+    "\n",
+    "outputs = pipe(prompt, max_new_tokens=2, do_sample=True, temperature=0.1)\n",
+    "print(outputs[0][\"generated_text\"].split(\"label: \")[-1].strip())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 27,
+   "id": "b182f963-ce02-465a-87dd-312fe782451d",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Device set to use cpu\n",
+      "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Hate Speech\n"
+     ]
+    }
+   ],
+   "source": [
+    "text = \"Jews are conspiratorial, devious, treacherous, sadistic, child killers, and subversive\"\n",
+    "prompt = f\"\"\"Classify the text into Hatespeech, Offensive, Normal and return the answer as the corresponding label.\n",
+    "text: {text}\n",
+    "label: \"\"\".strip()\n",
+    "pipe = pipeline(\n",
+    "    \"text-generation\",\n",
+    "    model=model,\n",
+    "    tokenizer=tokenizer,\n",
+    "    torch_dtype=torch.float16,\n",
+    "    device_map=\"auto\",\n",
+    ")\n",
+    "\n",
+    "outputs = pipe(prompt, max_new_tokens=2, do_sample=True, temperature=0.1)\n",
+    "print(outputs[0][\"generated_text\"].split(\"label: \")[-1].strip())"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 28,
+   "id": "dfb5394b-8be3-4e25-9417-165f08ff0fad",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "('/home/marco/llama-3.2-3B-instruct-offensive-classification-2/tokenizer_config.json',\n",
+       " '/home/marco/llama-3.2-3B-instruct-offensive-classification-2/special_tokens_map.json',\n",
+       " '/home/marco/llama-3.2-3B-instruct-offensive-classification-2/tokenizer.json')"
+      ]
+     },
+     "execution_count": 28,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "model_dir = \"/home/marco/llama-3.2-3B-instruct-offensive-classification-2\"\n",
+    "model.save_pretrained(model_dir)\n",
+    "tokenizer.save_pretrained(model_dir)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "id": "aa0ffb80-4d69-4099-b279-712c3dfac2fb",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "a2267a17809c48c9a0a310197a828c6a",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Upload 2 LFS files:   0%|          | 0/2 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "8c57390c36df4d8aa597180b831a63bf",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "model-00002-of-00002.safetensors:   0%|          | 0.00/1.46G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "86df050df82b40a59c3c6f46594153a0",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "model-00001-of-00002.safetensors:   0%|          | 0.00/4.97G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "ename": "ConnectionError",
+     "evalue": "(ProtocolError('Connection aborted.', RemoteDisconnected('Remote end closed connection without response')), '(Request ID: 4f29ef98-416f-4d91-bd52-997d447a0e2a)')",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mRemoteDisconnected\u001b[0m                        Traceback (most recent call last)",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/urllib3/connectionpool.py:789\u001b[0m, in \u001b[0;36mHTTPConnectionPool.urlopen\u001b[0;34m(self, method, url, body, headers, retries, redirect, assert_same_host, timeout, pool_timeout, release_conn, chunked, body_pos, preload_content, decode_content, **response_kw)\u001b[0m\n\u001b[1;32m    788\u001b[0m \u001b[38;5;66;03m# Make the request on the HTTPConnection object\u001b[39;00m\n\u001b[0;32m--> 789\u001b[0m response \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_make_request\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    790\u001b[0m \u001b[43m    \u001b[49m\u001b[43mconn\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    791\u001b[0m \u001b[43m    \u001b[49m\u001b[43mmethod\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    792\u001b[0m \u001b[43m    \u001b[49m\u001b[43murl\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    793\u001b[0m \u001b[43m    \u001b[49m\u001b[43mtimeout\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtimeout_obj\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    794\u001b[0m \u001b[43m    \u001b[49m\u001b[43mbody\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mbody\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    795\u001b[0m \u001b[43m    \u001b[49m\u001b[43mheaders\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mheaders\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    796\u001b[0m \u001b[43m    \u001b[49m\u001b[43mchunked\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mchunked\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    797\u001b[0m \u001b[43m    \u001b[49m\u001b[43mretries\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mretries\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    798\u001b[0m \u001b[43m    \u001b[49m\u001b[43mresponse_conn\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mresponse_conn\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    799\u001b[0m \u001b[43m    \u001b[49m\u001b[43mpreload_content\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mpreload_content\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    800\u001b[0m \u001b[43m    \u001b[49m\u001b[43mdecode_content\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdecode_content\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    801\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mresponse_kw\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    802\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    804\u001b[0m \u001b[38;5;66;03m# Everything went great!\u001b[39;00m\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/urllib3/connectionpool.py:536\u001b[0m, in \u001b[0;36mHTTPConnectionPool._make_request\u001b[0;34m(self, conn, method, url, body, headers, retries, timeout, chunked, response_conn, preload_content, decode_content, enforce_content_length)\u001b[0m\n\u001b[1;32m    535\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 536\u001b[0m     response \u001b[38;5;241m=\u001b[39m \u001b[43mconn\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mgetresponse\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    537\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m (BaseSSLError, \u001b[38;5;167;01mOSError\u001b[39;00m) \u001b[38;5;28;01mas\u001b[39;00m e:\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/urllib3/connection.py:464\u001b[0m, in \u001b[0;36mHTTPConnection.getresponse\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    463\u001b[0m \u001b[38;5;66;03m# Get the response from http.client.HTTPConnection\u001b[39;00m\n\u001b[0;32m--> 464\u001b[0m httplib_response \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43msuper\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mgetresponse\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    466\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/http/client.py:1428\u001b[0m, in \u001b[0;36mHTTPConnection.getresponse\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m   1427\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m-> 1428\u001b[0m     \u001b[43mresponse\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbegin\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1429\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mConnectionError\u001b[39;00m:\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/http/client.py:331\u001b[0m, in \u001b[0;36mHTTPResponse.begin\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    330\u001b[0m \u001b[38;5;28;01mwhile\u001b[39;00m \u001b[38;5;28;01mTrue\u001b[39;00m:\n\u001b[0;32m--> 331\u001b[0m     version, status, reason \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_read_status\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    332\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m status \u001b[38;5;241m!=\u001b[39m CONTINUE:\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/http/client.py:300\u001b[0m, in \u001b[0;36mHTTPResponse._read_status\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    297\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m line:\n\u001b[1;32m    298\u001b[0m     \u001b[38;5;66;03m# Presumably, the server closed the connection before\u001b[39;00m\n\u001b[1;32m    299\u001b[0m     \u001b[38;5;66;03m# sending a valid response.\u001b[39;00m\n\u001b[0;32m--> 300\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m RemoteDisconnected(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mRemote end closed connection without\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    301\u001b[0m                              \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m response\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m    302\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n",
+      "\u001b[0;31mRemoteDisconnected\u001b[0m: Remote end closed connection without response",
+      "\nDuring handling of the above exception, another exception occurred:\n",
+      "\u001b[0;31mProtocolError\u001b[0m                             Traceback (most recent call last)",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/requests/adapters.py:667\u001b[0m, in \u001b[0;36mHTTPAdapter.send\u001b[0;34m(self, request, stream, timeout, verify, cert, proxies)\u001b[0m\n\u001b[1;32m    666\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 667\u001b[0m     resp \u001b[38;5;241m=\u001b[39m \u001b[43mconn\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43murlopen\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    668\u001b[0m \u001b[43m        \u001b[49m\u001b[43mmethod\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mrequest\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmethod\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    669\u001b[0m \u001b[43m        \u001b[49m\u001b[43murl\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43murl\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    670\u001b[0m \u001b[43m        \u001b[49m\u001b[43mbody\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mrequest\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbody\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    671\u001b[0m \u001b[43m        \u001b[49m\u001b[43mheaders\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mrequest\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mheaders\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    672\u001b[0m \u001b[43m        \u001b[49m\u001b[43mredirect\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m    673\u001b[0m \u001b[43m        \u001b[49m\u001b[43massert_same_host\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m    674\u001b[0m \u001b[43m        \u001b[49m\u001b[43mpreload_content\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m    675\u001b[0m \u001b[43m        \u001b[49m\u001b[43mdecode_content\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[1;32m    676\u001b[0m \u001b[43m        \u001b[49m\u001b[43mretries\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmax_retries\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    677\u001b[0m \u001b[43m        \u001b[49m\u001b[43mtimeout\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtimeout\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    678\u001b[0m \u001b[43m        \u001b[49m\u001b[43mchunked\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mchunked\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    679\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    681\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m (ProtocolError, \u001b[38;5;167;01mOSError\u001b[39;00m) \u001b[38;5;28;01mas\u001b[39;00m err:\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/urllib3/connectionpool.py:843\u001b[0m, in \u001b[0;36mHTTPConnectionPool.urlopen\u001b[0;34m(self, method, url, body, headers, retries, redirect, assert_same_host, timeout, pool_timeout, release_conn, chunked, body_pos, preload_content, decode_content, **response_kw)\u001b[0m\n\u001b[1;32m    841\u001b[0m     new_e \u001b[38;5;241m=\u001b[39m ProtocolError(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mConnection aborted.\u001b[39m\u001b[38;5;124m\"\u001b[39m, new_e)\n\u001b[0;32m--> 843\u001b[0m retries \u001b[38;5;241m=\u001b[39m \u001b[43mretries\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mincrement\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    844\u001b[0m \u001b[43m    \u001b[49m\u001b[43mmethod\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43murl\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43merror\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mnew_e\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m_pool\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m_stacktrace\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43msys\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mexc_info\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m[\u001b[49m\u001b[38;5;241;43m2\u001b[39;49m\u001b[43m]\u001b[49m\n\u001b[1;32m    845\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    846\u001b[0m retries\u001b[38;5;241m.\u001b[39msleep()\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/urllib3/util/retry.py:474\u001b[0m, in \u001b[0;36mRetry.increment\u001b[0;34m(self, method, url, response, error, _pool, _stacktrace)\u001b[0m\n\u001b[1;32m    473\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m read \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mFalse\u001b[39;00m \u001b[38;5;129;01mor\u001b[39;00m method \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_is_method_retryable(method):\n\u001b[0;32m--> 474\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[43mreraise\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mtype\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43merror\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43merror\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m_stacktrace\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    475\u001b[0m \u001b[38;5;28;01melif\u001b[39;00m read \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/urllib3/util/util.py:38\u001b[0m, in \u001b[0;36mreraise\u001b[0;34m(tp, value, tb)\u001b[0m\n\u001b[1;32m     37\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m value\u001b[38;5;241m.\u001b[39m__traceback__ \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m tb:\n\u001b[0;32m---> 38\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m value\u001b[38;5;241m.\u001b[39mwith_traceback(tb)\n\u001b[1;32m     39\u001b[0m \u001b[38;5;28;01mraise\u001b[39;00m value\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/urllib3/connectionpool.py:789\u001b[0m, in \u001b[0;36mHTTPConnectionPool.urlopen\u001b[0;34m(self, method, url, body, headers, retries, redirect, assert_same_host, timeout, pool_timeout, release_conn, chunked, body_pos, preload_content, decode_content, **response_kw)\u001b[0m\n\u001b[1;32m    788\u001b[0m \u001b[38;5;66;03m# Make the request on the HTTPConnection object\u001b[39;00m\n\u001b[0;32m--> 789\u001b[0m response \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_make_request\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    790\u001b[0m \u001b[43m    \u001b[49m\u001b[43mconn\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    791\u001b[0m \u001b[43m    \u001b[49m\u001b[43mmethod\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    792\u001b[0m \u001b[43m    \u001b[49m\u001b[43murl\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    793\u001b[0m \u001b[43m    \u001b[49m\u001b[43mtimeout\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtimeout_obj\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    794\u001b[0m \u001b[43m    \u001b[49m\u001b[43mbody\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mbody\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    795\u001b[0m \u001b[43m    \u001b[49m\u001b[43mheaders\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mheaders\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    796\u001b[0m \u001b[43m    \u001b[49m\u001b[43mchunked\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mchunked\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    797\u001b[0m \u001b[43m    \u001b[49m\u001b[43mretries\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mretries\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    798\u001b[0m \u001b[43m    \u001b[49m\u001b[43mresponse_conn\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mresponse_conn\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    799\u001b[0m \u001b[43m    \u001b[49m\u001b[43mpreload_content\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mpreload_content\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    800\u001b[0m \u001b[43m    \u001b[49m\u001b[43mdecode_content\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdecode_content\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    801\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mresponse_kw\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    802\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    804\u001b[0m \u001b[38;5;66;03m# Everything went great!\u001b[39;00m\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/urllib3/connectionpool.py:536\u001b[0m, in \u001b[0;36mHTTPConnectionPool._make_request\u001b[0;34m(self, conn, method, url, body, headers, retries, timeout, chunked, response_conn, preload_content, decode_content, enforce_content_length)\u001b[0m\n\u001b[1;32m    535\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m--> 536\u001b[0m     response \u001b[38;5;241m=\u001b[39m \u001b[43mconn\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mgetresponse\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    537\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m (BaseSSLError, \u001b[38;5;167;01mOSError\u001b[39;00m) \u001b[38;5;28;01mas\u001b[39;00m e:\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/urllib3/connection.py:464\u001b[0m, in \u001b[0;36mHTTPConnection.getresponse\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    463\u001b[0m \u001b[38;5;66;03m# Get the response from http.client.HTTPConnection\u001b[39;00m\n\u001b[0;32m--> 464\u001b[0m httplib_response \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43msuper\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mgetresponse\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    466\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/http/client.py:1428\u001b[0m, in \u001b[0;36mHTTPConnection.getresponse\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m   1427\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m-> 1428\u001b[0m     \u001b[43mresponse\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbegin\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1429\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mConnectionError\u001b[39;00m:\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/http/client.py:331\u001b[0m, in \u001b[0;36mHTTPResponse.begin\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    330\u001b[0m \u001b[38;5;28;01mwhile\u001b[39;00m \u001b[38;5;28;01mTrue\u001b[39;00m:\n\u001b[0;32m--> 331\u001b[0m     version, status, reason \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_read_status\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    332\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m status \u001b[38;5;241m!=\u001b[39m CONTINUE:\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/http/client.py:300\u001b[0m, in \u001b[0;36mHTTPResponse._read_status\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m    297\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m line:\n\u001b[1;32m    298\u001b[0m     \u001b[38;5;66;03m# Presumably, the server closed the connection before\u001b[39;00m\n\u001b[1;32m    299\u001b[0m     \u001b[38;5;66;03m# sending a valid response.\u001b[39;00m\n\u001b[0;32m--> 300\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m RemoteDisconnected(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mRemote end closed connection without\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m    301\u001b[0m                              \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m response\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m    302\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n",
+      "\u001b[0;31mProtocolError\u001b[0m: ('Connection aborted.', RemoteDisconnected('Remote end closed connection without response'))",
+      "\nDuring handling of the above exception, another exception occurred:\n",
+      "\u001b[0;31mConnectionError\u001b[0m                           Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[30], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mmodel\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpush_to_hub\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mmarcoorasch/llama-3.2-3B-instruct-hatespeech-offensive-classification\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43muse_temp_dir\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/transformers/modeling_utils.py:3087\u001b[0m, in \u001b[0;36mPreTrainedModel.push_to_hub\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   3085\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m tags:\n\u001b[1;32m   3086\u001b[0m     kwargs[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtags\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m tags\n\u001b[0;32m-> 3087\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43msuper\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpush_to_hub\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/transformers/utils/hub.py:954\u001b[0m, in \u001b[0;36mPushToHubMixin.push_to_hub\u001b[0;34m(self, repo_id, use_temp_dir, commit_message, private, token, max_shard_size, create_pr, safe_serialization, revision, commit_description, tags, **deprecated_kwargs)\u001b[0m\n\u001b[1;32m    951\u001b[0m \u001b[38;5;66;03m# Update model card if needed:\u001b[39;00m\n\u001b[1;32m    952\u001b[0m model_card\u001b[38;5;241m.\u001b[39msave(os\u001b[38;5;241m.\u001b[39mpath\u001b[38;5;241m.\u001b[39mjoin(work_dir, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mREADME.md\u001b[39m\u001b[38;5;124m\"\u001b[39m))\n\u001b[0;32m--> 954\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_upload_modified_files\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    955\u001b[0m \u001b[43m    \u001b[49m\u001b[43mwork_dir\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    956\u001b[0m \u001b[43m    \u001b[49m\u001b[43mrepo_id\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    957\u001b[0m \u001b[43m    \u001b[49m\u001b[43mfiles_timestamps\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    958\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcommit_message\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcommit_message\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    959\u001b[0m \u001b[43m    \u001b[49m\u001b[43mtoken\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtoken\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    960\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcreate_pr\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcreate_pr\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    961\u001b[0m \u001b[43m    \u001b[49m\u001b[43mrevision\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mrevision\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    962\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcommit_description\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcommit_description\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    963\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/transformers/utils/hub.py:818\u001b[0m, in \u001b[0;36mPushToHubMixin._upload_modified_files\u001b[0;34m(self, working_dir, repo_id, files_timestamps, commit_message, token, create_pr, revision, commit_description)\u001b[0m\n\u001b[1;32m    815\u001b[0m             \u001b[38;5;28;01mraise\u001b[39;00m\n\u001b[1;32m    817\u001b[0m logger\u001b[38;5;241m.\u001b[39minfo(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mUploading the following files to \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mrepo_id\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m: \u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m,\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;241m.\u001b[39mjoin(modified_files)\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m--> 818\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mcreate_commit\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    819\u001b[0m \u001b[43m    \u001b[49m\u001b[43mrepo_id\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mrepo_id\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    820\u001b[0m \u001b[43m    \u001b[49m\u001b[43moperations\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43moperations\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    821\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcommit_message\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcommit_message\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    822\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcommit_description\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcommit_description\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    823\u001b[0m \u001b[43m    \u001b[49m\u001b[43mtoken\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtoken\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    824\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcreate_pr\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcreate_pr\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    825\u001b[0m \u001b[43m    \u001b[49m\u001b[43mrevision\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mrevision\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    826\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/huggingface_hub/utils/_validators.py:114\u001b[0m, in \u001b[0;36mvalidate_hf_hub_args.<locals>._inner_fn\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m    111\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m check_use_auth_token:\n\u001b[1;32m    112\u001b[0m     kwargs \u001b[38;5;241m=\u001b[39m smoothly_deprecate_use_auth_token(fn_name\u001b[38;5;241m=\u001b[39mfn\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m, has_token\u001b[38;5;241m=\u001b[39mhas_token, kwargs\u001b[38;5;241m=\u001b[39mkwargs)\n\u001b[0;32m--> 114\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfn\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/huggingface_hub/hf_api.py:1518\u001b[0m, in \u001b[0;36mfuture_compatible.<locals>._inner\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m   1515\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mrun_as_future(fn, \u001b[38;5;28mself\u001b[39m, \u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[1;32m   1517\u001b[0m \u001b[38;5;66;03m# Otherwise, call the function normally\u001b[39;00m\n\u001b[0;32m-> 1518\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfn\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/huggingface_hub/hf_api.py:4086\u001b[0m, in \u001b[0;36mHfApi.create_commit\u001b[0;34m(self, repo_id, operations, commit_message, commit_description, token, repo_type, revision, create_pr, num_threads, parent_commit, run_as_future)\u001b[0m\n\u001b[1;32m   4083\u001b[0m params \u001b[38;5;241m=\u001b[39m {\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcreate_pr\u001b[39m\u001b[38;5;124m\"\u001b[39m: \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m1\u001b[39m\u001b[38;5;124m\"\u001b[39m} \u001b[38;5;28;01mif\u001b[39;00m create_pr \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m   4085\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m-> 4086\u001b[0m     commit_resp \u001b[38;5;241m=\u001b[39m \u001b[43mget_session\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mpost\u001b[49m\u001b[43m(\u001b[49m\u001b[43murl\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mcommit_url\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mheaders\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mheaders\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdata\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdata\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mparams\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mparams\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   4087\u001b[0m     hf_raise_for_status(commit_resp, endpoint_name\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mcommit\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m   4088\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m RepositoryNotFoundError \u001b[38;5;28;01mas\u001b[39;00m e:\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/requests/sessions.py:637\u001b[0m, in \u001b[0;36mSession.post\u001b[0;34m(self, url, data, json, **kwargs)\u001b[0m\n\u001b[1;32m    626\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mpost\u001b[39m(\u001b[38;5;28mself\u001b[39m, url, data\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, json\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs):\n\u001b[1;32m    627\u001b[0m \u001b[38;5;250m    \u001b[39m\u001b[38;5;124mr\u001b[39m\u001b[38;5;124;03m\"\"\"Sends a POST request. Returns :class:`Response` object.\u001b[39;00m\n\u001b[1;32m    628\u001b[0m \n\u001b[1;32m    629\u001b[0m \u001b[38;5;124;03m    :param url: URL for the new :class:`Request` object.\u001b[39;00m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    634\u001b[0m \u001b[38;5;124;03m    :rtype: requests.Response\u001b[39;00m\n\u001b[1;32m    635\u001b[0m \u001b[38;5;124;03m    \"\"\"\u001b[39;00m\n\u001b[0;32m--> 637\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrequest\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[38;5;124;43mPOST\u001b[39;49m\u001b[38;5;124;43m\"\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43murl\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdata\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdata\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mjson\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mjson\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/requests/sessions.py:589\u001b[0m, in \u001b[0;36mSession.request\u001b[0;34m(self, method, url, params, data, headers, cookies, files, auth, timeout, allow_redirects, proxies, hooks, stream, verify, cert, json)\u001b[0m\n\u001b[1;32m    584\u001b[0m send_kwargs \u001b[38;5;241m=\u001b[39m {\n\u001b[1;32m    585\u001b[0m     \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtimeout\u001b[39m\u001b[38;5;124m\"\u001b[39m: timeout,\n\u001b[1;32m    586\u001b[0m     \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mallow_redirects\u001b[39m\u001b[38;5;124m\"\u001b[39m: allow_redirects,\n\u001b[1;32m    587\u001b[0m }\n\u001b[1;32m    588\u001b[0m send_kwargs\u001b[38;5;241m.\u001b[39mupdate(settings)\n\u001b[0;32m--> 589\u001b[0m resp \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msend\u001b[49m\u001b[43m(\u001b[49m\u001b[43mprep\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43msend_kwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    591\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m resp\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/requests/sessions.py:703\u001b[0m, in \u001b[0;36mSession.send\u001b[0;34m(self, request, **kwargs)\u001b[0m\n\u001b[1;32m    700\u001b[0m start \u001b[38;5;241m=\u001b[39m preferred_clock()\n\u001b[1;32m    702\u001b[0m \u001b[38;5;66;03m# Send the request\u001b[39;00m\n\u001b[0;32m--> 703\u001b[0m r \u001b[38;5;241m=\u001b[39m \u001b[43madapter\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msend\u001b[49m\u001b[43m(\u001b[49m\u001b[43mrequest\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m    705\u001b[0m \u001b[38;5;66;03m# Total elapsed time of the request (approximately)\u001b[39;00m\n\u001b[1;32m    706\u001b[0m elapsed \u001b[38;5;241m=\u001b[39m preferred_clock() \u001b[38;5;241m-\u001b[39m start\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/huggingface_hub/utils/_http.py:93\u001b[0m, in \u001b[0;36mUniqueRequestIdAdapter.send\u001b[0;34m(self, request, *args, **kwargs)\u001b[0m\n\u001b[1;32m     91\u001b[0m \u001b[38;5;250m\u001b[39m\u001b[38;5;124;03m\"\"\"Catch any RequestException to append request id to the error message for debugging.\"\"\"\u001b[39;00m\n\u001b[1;32m     92\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m---> 93\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43msuper\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43msend\u001b[49m\u001b[43m(\u001b[49m\u001b[43mrequest\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[38;5;241;43m*\u001b[39;49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     94\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m requests\u001b[38;5;241m.\u001b[39mRequestException \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m     95\u001b[0m     request_id \u001b[38;5;241m=\u001b[39m request\u001b[38;5;241m.\u001b[39mheaders\u001b[38;5;241m.\u001b[39mget(X_AMZN_TRACE_ID)\n",
+      "File \u001b[0;32m~/.config/jupyterlab-desktop/jlab_server/lib/python3.12/site-packages/requests/adapters.py:682\u001b[0m, in \u001b[0;36mHTTPAdapter.send\u001b[0;34m(self, request, stream, timeout, verify, cert, proxies)\u001b[0m\n\u001b[1;32m    667\u001b[0m     resp \u001b[38;5;241m=\u001b[39m conn\u001b[38;5;241m.\u001b[39murlopen(\n\u001b[1;32m    668\u001b[0m         method\u001b[38;5;241m=\u001b[39mrequest\u001b[38;5;241m.\u001b[39mmethod,\n\u001b[1;32m    669\u001b[0m         url\u001b[38;5;241m=\u001b[39murl,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    678\u001b[0m         chunked\u001b[38;5;241m=\u001b[39mchunked,\n\u001b[1;32m    679\u001b[0m     )\n\u001b[1;32m    681\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m (ProtocolError, \u001b[38;5;167;01mOSError\u001b[39;00m) \u001b[38;5;28;01mas\u001b[39;00m err:\n\u001b[0;32m--> 682\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mConnectionError\u001b[39;00m(err, request\u001b[38;5;241m=\u001b[39mrequest)\n\u001b[1;32m    684\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m MaxRetryError \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[1;32m    685\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(e\u001b[38;5;241m.\u001b[39mreason, ConnectTimeoutError):\n\u001b[1;32m    686\u001b[0m         \u001b[38;5;66;03m# TODO: Remove this in 3.0.0: see #2811\u001b[39;00m\n",
+      "\u001b[0;31mConnectionError\u001b[0m: (ProtocolError('Connection aborted.', RemoteDisconnected('Remote end closed connection without response')), '(Request ID: 4f29ef98-416f-4d91-bd52-997d447a0e2a)')"
+     ]
+    }
+   ],
+   "source": [
+    "model.push_to_hub(\"marcoorasch/llama-3.2-3B-instruct-hatespeech-offensive-classification\",use_temp_dir=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "3d94aa30-c59d-4575-a0b6-1d6c87f1087e",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "language": "python",
+   "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": "3.12.5"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}