tamil_llama3-SFT_test_existing_tokenizer.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "4b43de97-6a76-463e-a10d-2071fd563f0a",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting protobuf\n",
      "  Downloading protobuf-5.26.1-cp37-abi3-manylinux2014_x86_64.whl.metadata (592 bytes)\n",
      "Collecting scikit-learn\n",
      "  Downloading scikit_learn-1.4.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (11 kB)\n",
      "Collecting scipy\n",
      "  Downloading scipy-1.13.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (60 kB)\n",
      "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m60.6/60.6 kB\u001b[0m \u001b[31m2.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hRequirement already satisfied: numpy>=1.19.5 in /opt/conda/lib/python3.10/site-packages (from scikit-learn) (1.26.3)\n",
      "Collecting joblib>=1.2.0 (from scikit-learn)\n",
      "  Downloading joblib-1.4.0-py3-none-any.whl.metadata (5.4 kB)\n",
      "Collecting threadpoolctl>=2.0.0 (from scikit-learn)\n",
      "  Downloading threadpoolctl-3.4.0-py3-none-any.whl.metadata (13 kB)\n",
      "Downloading protobuf-5.26.1-cp37-abi3-manylinux2014_x86_64.whl (302 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m302.8/302.8 kB\u001b[0m \u001b[31m10.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hDownloading scikit_learn-1.4.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (12.1 MB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m12.1/12.1 MB\u001b[0m \u001b[31m107.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m \u001b[36m0:00:01\u001b[0m\n",
      "\u001b[?25hDownloading scipy-1.13.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (38.6 MB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m38.6/38.6 MB\u001b[0m \u001b[31m92.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m:00:01\u001b[0m00:01\u001b[0m\n",
      "\u001b[?25hDownloading joblib-1.4.0-py3-none-any.whl (301 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m301.2/301.2 kB\u001b[0m \u001b[31m50.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hDownloading threadpoolctl-3.4.0-py3-none-any.whl (17 kB)\n",
      "Installing collected packages: threadpoolctl, scipy, protobuf, joblib, scikit-learn\n",
      "Successfully installed joblib-1.4.0 protobuf-5.26.1 scikit-learn-1.4.2 scipy-1.13.0 threadpoolctl-3.4.0\n",
      "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
      "\u001b[0mCollecting wandb\n",
      "  Downloading wandb-0.16.6-py3-none-any.whl.metadata (10 kB)\n",
      "Requirement already satisfied: Click!=8.0.0,>=7.1 in /opt/conda/lib/python3.10/site-packages (from wandb) (8.1.7)\n",
      "Collecting GitPython!=3.1.29,>=1.0.0 (from wandb)\n",
      "  Downloading GitPython-3.1.43-py3-none-any.whl.metadata (13 kB)\n",
      "Requirement already satisfied: requests<3,>=2.0.0 in /opt/conda/lib/python3.10/site-packages (from wandb) (2.31.0)\n",
      "Requirement already satisfied: psutil>=5.0.0 in /opt/conda/lib/python3.10/site-packages (from wandb) (5.9.0)\n",
      "Collecting sentry-sdk>=1.0.0 (from wandb)\n",
      "  Downloading sentry_sdk-1.45.0-py2.py3-none-any.whl.metadata (9.9 kB)\n",
      "Collecting docker-pycreds>=0.4.0 (from wandb)\n",
      "  Downloading docker_pycreds-0.4.0-py2.py3-none-any.whl.metadata (1.8 kB)\n",
      "Requirement already satisfied: PyYAML in /opt/conda/lib/python3.10/site-packages (from wandb) (6.0.1)\n",
      "Collecting setproctitle (from wandb)\n",
      "  Downloading setproctitle-1.3.3-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (9.9 kB)\n",
      "Requirement already satisfied: setuptools in /opt/conda/lib/python3.10/site-packages (from wandb) (68.2.2)\n",
      "Collecting appdirs>=1.4.3 (from wandb)\n",
      "  Downloading appdirs-1.4.4-py2.py3-none-any.whl.metadata (9.0 kB)\n",
      "Collecting protobuf!=4.21.0,<5,>=3.19.0 (from wandb)\n",
      "  Downloading protobuf-4.25.3-cp37-abi3-manylinux2014_x86_64.whl.metadata (541 bytes)\n",
      "Requirement already satisfied: six>=1.4.0 in /opt/conda/lib/python3.10/site-packages (from docker-pycreds>=0.4.0->wandb) (1.16.0)\n",
      "Collecting gitdb<5,>=4.0.1 (from GitPython!=3.1.29,>=1.0.0->wandb)\n",
      "  Downloading gitdb-4.0.11-py3-none-any.whl.metadata (1.2 kB)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /opt/conda/lib/python3.10/site-packages (from requests<3,>=2.0.0->wandb) (2.0.4)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.10/site-packages (from requests<3,>=2.0.0->wandb) (3.4)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /opt/conda/lib/python3.10/site-packages (from requests<3,>=2.0.0->wandb) (2.1.0)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.10/site-packages (from requests<3,>=2.0.0->wandb) (2024.2.2)\n",
      "Collecting smmap<6,>=3.0.1 (from gitdb<5,>=4.0.1->GitPython!=3.1.29,>=1.0.0->wandb)\n",
      "  Downloading smmap-5.0.1-py3-none-any.whl.metadata (4.3 kB)\n",
      "Downloading wandb-0.16.6-py3-none-any.whl (2.2 MB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.2/2.2 MB\u001b[0m \u001b[31m19.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m00:01\u001b[0m0:01\u001b[0m\n",
      "\u001b[?25hDownloading appdirs-1.4.4-py2.py3-none-any.whl (9.6 kB)\n",
      "Downloading docker_pycreds-0.4.0-py2.py3-none-any.whl (9.0 kB)\n",
      "Downloading GitPython-3.1.43-py3-none-any.whl (207 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m207.3/207.3 kB\u001b[0m \u001b[31m44.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hDownloading protobuf-4.25.3-cp37-abi3-manylinux2014_x86_64.whl (294 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m294.6/294.6 kB\u001b[0m \u001b[31m74.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hDownloading sentry_sdk-1.45.0-py2.py3-none-any.whl (267 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m267.1/267.1 kB\u001b[0m \u001b[31m51.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hDownloading setproctitle-1.3.3-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (30 kB)\n",
      "Downloading gitdb-4.0.11-py3-none-any.whl (62 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m62.7/62.7 kB\u001b[0m \u001b[31m17.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hDownloading smmap-5.0.1-py3-none-any.whl (24 kB)\n",
      "Installing collected packages: appdirs, smmap, setproctitle, sentry-sdk, protobuf, docker-pycreds, gitdb, GitPython, wandb\n",
      "  Attempting uninstall: protobuf\n",
      "    Found existing installation: protobuf 5.26.1\n",
      "    Uninstalling protobuf-5.26.1:\n",
      "      Successfully uninstalled protobuf-5.26.1\n",
      "Successfully installed GitPython-3.1.43 appdirs-1.4.4 docker-pycreds-0.4.0 gitdb-4.0.11 protobuf-4.25.3 sentry-sdk-1.45.0 setproctitle-1.3.3 smmap-5.0.1 wandb-0.16.6\n",
      "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
      "\u001b[0m\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
      "\u001b[0mToken has not been saved to git credential helper. Pass `add_to_git_credential=True` if you want to set the git credential as well.\n",
      "Token is valid (permission: write).\n",
      "Your token has been saved to /root/.cache/huggingface/token\n",
      "Login successful\n",
      "Requirement already satisfied: transformers[sentencepiece] in /opt/conda/lib/python3.10/site-packages (4.40.0)\n",
      "Requirement already satisfied: filelock in /opt/conda/lib/python3.10/site-packages (from transformers[sentencepiece]) (3.13.1)\n",
      "Requirement already satisfied: huggingface-hub<1.0,>=0.19.3 in /opt/conda/lib/python3.10/site-packages (from transformers[sentencepiece]) (0.22.2)\n",
      "Requirement already satisfied: numpy>=1.17 in /opt/conda/lib/python3.10/site-packages (from transformers[sentencepiece]) (1.26.3)\n",
      "Requirement already satisfied: packaging>=20.0 in /opt/conda/lib/python3.10/site-packages (from transformers[sentencepiece]) (23.1)\n",
      "Requirement already satisfied: pyyaml>=5.1 in /opt/conda/lib/python3.10/site-packages (from transformers[sentencepiece]) (6.0.1)\n",
      "Requirement already satisfied: regex!=2019.12.17 in /opt/conda/lib/python3.10/site-packages (from transformers[sentencepiece]) (2024.4.16)\n",
      "Requirement already satisfied: requests in /opt/conda/lib/python3.10/site-packages (from transformers[sentencepiece]) (2.31.0)\n",
      "Requirement already satisfied: tokenizers<0.20,>=0.19 in /opt/conda/lib/python3.10/site-packages (from transformers[sentencepiece]) (0.19.1)\n",
      "Requirement already satisfied: safetensors>=0.4.1 in /opt/conda/lib/python3.10/site-packages (from transformers[sentencepiece]) (0.4.3)\n",
      "Requirement already satisfied: tqdm>=4.27 in /opt/conda/lib/python3.10/site-packages (from transformers[sentencepiece]) (4.65.0)\n",
      "Collecting sentencepiece!=0.1.92,>=0.1.91 (from transformers[sentencepiece])\n",
      "  Downloading sentencepiece-0.2.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (7.7 kB)\n",
      "Requirement already satisfied: protobuf in /opt/conda/lib/python3.10/site-packages (from transformers[sentencepiece]) (4.25.3)\n",
      "Requirement already satisfied: fsspec>=2023.5.0 in /opt/conda/lib/python3.10/site-packages (from huggingface-hub<1.0,>=0.19.3->transformers[sentencepiece]) (2024.2.0)\n",
      "Requirement already satisfied: typing-extensions>=3.7.4.3 in /opt/conda/lib/python3.10/site-packages (from huggingface-hub<1.0,>=0.19.3->transformers[sentencepiece]) (4.9.0)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /opt/conda/lib/python3.10/site-packages (from requests->transformers[sentencepiece]) (2.0.4)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.10/site-packages (from requests->transformers[sentencepiece]) (3.4)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /opt/conda/lib/python3.10/site-packages (from requests->transformers[sentencepiece]) (2.1.0)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.10/site-packages (from requests->transformers[sentencepiece]) (2024.2.2)\n",
      "Downloading sentencepiece-0.2.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.3 MB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.3/1.3 MB\u001b[0m \u001b[31m12.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m \u001b[36m0:00:01\u001b[0m\n",
      "\u001b[?25hInstalling collected packages: sentencepiece\n",
      "Successfully installed sentencepiece-0.2.0\n",
      "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
      "\u001b[0mRequirement already satisfied: sentencepiece in /opt/conda/lib/python3.10/site-packages (0.2.0)\n",
      "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
      "Reading package lists... Done\n",
      "Building dependency tree... Done\n",
      "Reading state information... Done\n",
      "The following NEW packages will be installed:\n",
      "  git-lfs\n",
      "0 upgraded, 1 newly installed, 0 to remove and 32 not upgraded.\n",
      "Need to get 3503 kB of archives.\n",
      "After this operation, 10.4 MB of additional disk space will be used.\n",
      "Get:1 http://archive.ubuntu.com/ubuntu jammy-updates/universe amd64 git-lfs amd64 3.0.2-1ubuntu0.2 [3503 kB]\n",
      "Fetched 3503 kB in 1s (3173 kB/s)\n",
      "debconf: delaying package configuration, since apt-utils is not installed\n",
      "Selecting previously unselected package git-lfs.\n",
      "(Reading database ... 16755 files and directories currently installed.)\n",
      "Preparing to unpack .../git-lfs_3.0.2-1ubuntu0.2_amd64.deb ...\n",
      "Unpacking git-lfs (3.0.2-1ubuntu0.2) ...\n",
      "Setting up git-lfs (3.0.2-1ubuntu0.2) ...\n",
      "Collecting tensorboardX\n",
      "  Downloading tensorboardX-2.6.2.2-py2.py3-none-any.whl.metadata (5.8 kB)\n",
      "Requirement already satisfied: numpy in /opt/conda/lib/python3.10/site-packages (from tensorboardX) (1.26.3)\n",
      "Requirement already satisfied: packaging in /opt/conda/lib/python3.10/site-packages (from tensorboardX) (23.1)\n",
      "Requirement already satisfied: protobuf>=3.20 in /opt/conda/lib/python3.10/site-packages (from tensorboardX) (4.25.3)\n",
      "Downloading tensorboardX-2.6.2.2-py2.py3-none-any.whl (101 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m101.7/101.7 kB\u001b[0m \u001b[31m3.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hInstalling collected packages: tensorboardX\n",
      "Successfully installed tensorboardX-2.6.2.2\n",
      "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
      "\u001b[0m"
     ]
    }
   ],
   "source": [
    "!pip install protobuf scikit-learn scipy\n",
    "!export CUDA_HOME=cuda-12.2\n",
    "!pip install wandb\n",
    "!pip install -q accelerate  bitsandbytes  trl==0.4.7\n",
    "!huggingface-cli login --token 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\n",
    "!pip install  transformers[sentencepiece]\n",
    "!pip install sentencepiece\n",
    "!apt-get install git-lfs\n",
    "!pip install tensorboardX"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "19bd94ce-13fa-4115-8f04-11ad3bb3aba2",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: accelerate==0.29.3 in /opt/conda/lib/python3.10/site-packages (0.29.3)\n",
      "Collecting peft==0.10.0\n",
      "  Downloading peft-0.10.0-py3-none-any.whl.metadata (13 kB)\n",
      "Collecting bitsandbytes==0.40.2\n",
      "  Downloading bitsandbytes-0.40.2-py3-none-any.whl.metadata (9.8 kB)\n",
      "Requirement already satisfied: transformers==4.40.0 in /opt/conda/lib/python3.10/site-packages (4.40.0)\n",
      "Collecting trl==0.8.5\n",
      "  Downloading trl-0.8.5-py3-none-any.whl.metadata (11 kB)\n",
      "Requirement already satisfied: numpy>=1.17 in /opt/conda/lib/python3.10/site-packages (from accelerate==0.29.3) (1.26.3)\n",
      "Requirement already satisfied: packaging>=20.0 in /opt/conda/lib/python3.10/site-packages (from accelerate==0.29.3) (23.1)\n",
      "Requirement already satisfied: psutil in /opt/conda/lib/python3.10/site-packages (from accelerate==0.29.3) (5.9.0)\n",
      "Requirement already satisfied: pyyaml in /opt/conda/lib/python3.10/site-packages (from accelerate==0.29.3) (6.0.1)\n",
      "Requirement already satisfied: torch>=1.10.0 in /opt/conda/lib/python3.10/site-packages (from accelerate==0.29.3) (2.2.1)\n",
      "Requirement already satisfied: huggingface-hub in /opt/conda/lib/python3.10/site-packages (from accelerate==0.29.3) (0.22.2)\n",
      "Requirement already satisfied: safetensors>=0.3.1 in /opt/conda/lib/python3.10/site-packages (from accelerate==0.29.3) (0.4.3)\n",
      "Requirement already satisfied: tqdm in /opt/conda/lib/python3.10/site-packages (from peft==0.10.0) (4.65.0)\n",
      "Requirement already satisfied: filelock in /opt/conda/lib/python3.10/site-packages (from transformers==4.40.0) (3.13.1)\n",
      "Requirement already satisfied: regex!=2019.12.17 in /opt/conda/lib/python3.10/site-packages (from transformers==4.40.0) (2024.4.16)\n",
      "Requirement already satisfied: requests in /opt/conda/lib/python3.10/site-packages (from transformers==4.40.0) (2.31.0)\n",
      "Requirement already satisfied: tokenizers<0.20,>=0.19 in /opt/conda/lib/python3.10/site-packages (from transformers==4.40.0) (0.19.1)\n",
      "Requirement already satisfied: datasets in /opt/conda/lib/python3.10/site-packages (from trl==0.8.5) (2.19.0)\n",
      "Collecting tyro>=0.5.11 (from trl==0.8.5)\n",
      "  Downloading tyro-0.8.3-py3-none-any.whl.metadata (7.9 kB)\n",
      "Requirement already satisfied: fsspec>=2023.5.0 in /opt/conda/lib/python3.10/site-packages (from huggingface-hub->accelerate==0.29.3) (2024.2.0)\n",
      "Requirement already satisfied: typing-extensions>=3.7.4.3 in /opt/conda/lib/python3.10/site-packages (from huggingface-hub->accelerate==0.29.3) (4.9.0)\n",
      "Requirement already satisfied: sympy in /opt/conda/lib/python3.10/site-packages (from torch>=1.10.0->accelerate==0.29.3) (1.12)\n",
      "Requirement already satisfied: networkx in /opt/conda/lib/python3.10/site-packages (from torch>=1.10.0->accelerate==0.29.3) (3.1)\n",
      "Requirement already satisfied: jinja2 in /opt/conda/lib/python3.10/site-packages (from torch>=1.10.0->accelerate==0.29.3) (3.1.3)\n",
      "Collecting docstring-parser>=0.14.1 (from tyro>=0.5.11->trl==0.8.5)\n",
      "  Downloading docstring_parser-0.16-py3-none-any.whl.metadata (3.0 kB)\n",
      "Collecting rich>=11.1.0 (from tyro>=0.5.11->trl==0.8.5)\n",
      "  Downloading rich-13.7.1-py3-none-any.whl.metadata (18 kB)\n",
      "Collecting shtab>=1.5.6 (from tyro>=0.5.11->trl==0.8.5)\n",
      "  Downloading shtab-1.7.1-py3-none-any.whl.metadata (7.3 kB)\n",
      "Requirement already satisfied: pyarrow>=12.0.0 in /opt/conda/lib/python3.10/site-packages (from datasets->trl==0.8.5) (16.0.0)\n",
      "Requirement already satisfied: pyarrow-hotfix in /opt/conda/lib/python3.10/site-packages (from datasets->trl==0.8.5) (0.6)\n",
      "Requirement already satisfied: dill<0.3.9,>=0.3.0 in /opt/conda/lib/python3.10/site-packages (from datasets->trl==0.8.5) (0.3.8)\n",
      "Requirement already satisfied: pandas in /opt/conda/lib/python3.10/site-packages (from datasets->trl==0.8.5) (2.2.2)\n",
      "Requirement already satisfied: xxhash in /opt/conda/lib/python3.10/site-packages (from datasets->trl==0.8.5) (3.4.1)\n",
      "Requirement already satisfied: multiprocess in /opt/conda/lib/python3.10/site-packages (from datasets->trl==0.8.5) (0.70.16)\n",
      "Requirement already satisfied: aiohttp in /opt/conda/lib/python3.10/site-packages (from datasets->trl==0.8.5) (3.9.5)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /opt/conda/lib/python3.10/site-packages (from requests->transformers==4.40.0) (2.0.4)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.10/site-packages (from requests->transformers==4.40.0) (3.4)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /opt/conda/lib/python3.10/site-packages (from requests->transformers==4.40.0) (2.1.0)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.10/site-packages (from requests->transformers==4.40.0) (2024.2.2)\n",
      "Requirement already satisfied: aiosignal>=1.1.2 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets->trl==0.8.5) (1.3.1)\n",
      "Requirement already satisfied: attrs>=17.3.0 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets->trl==0.8.5) (23.1.0)\n",
      "Requirement already satisfied: frozenlist>=1.1.1 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets->trl==0.8.5) (1.4.1)\n",
      "Requirement already satisfied: multidict<7.0,>=4.5 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets->trl==0.8.5) (6.0.5)\n",
      "Requirement already satisfied: yarl<2.0,>=1.0 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets->trl==0.8.5) (1.9.4)\n",
      "Requirement already satisfied: async-timeout<5.0,>=4.0 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets->trl==0.8.5) (4.0.3)\n",
      "Collecting markdown-it-py>=2.2.0 (from rich>=11.1.0->tyro>=0.5.11->trl==0.8.5)\n",
      "  Downloading markdown_it_py-3.0.0-py3-none-any.whl.metadata (6.9 kB)\n",
      "Requirement already satisfied: pygments<3.0.0,>=2.13.0 in /opt/conda/lib/python3.10/site-packages (from rich>=11.1.0->tyro>=0.5.11->trl==0.8.5) (2.15.1)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /opt/conda/lib/python3.10/site-packages (from jinja2->torch>=1.10.0->accelerate==0.29.3) (2.1.3)\n",
      "Requirement already satisfied: python-dateutil>=2.8.2 in /opt/conda/lib/python3.10/site-packages (from pandas->datasets->trl==0.8.5) (2.8.2)\n",
      "Requirement already satisfied: pytz>=2020.1 in /opt/conda/lib/python3.10/site-packages (from pandas->datasets->trl==0.8.5) (2023.3.post1)\n",
      "Requirement already satisfied: tzdata>=2022.7 in /opt/conda/lib/python3.10/site-packages (from pandas->datasets->trl==0.8.5) (2024.1)\n",
      "Requirement already satisfied: mpmath>=0.19 in /opt/conda/lib/python3.10/site-packages (from sympy->torch>=1.10.0->accelerate==0.29.3) (1.3.0)\n",
      "Collecting mdurl~=0.1 (from markdown-it-py>=2.2.0->rich>=11.1.0->tyro>=0.5.11->trl==0.8.5)\n",
      "  Downloading mdurl-0.1.2-py3-none-any.whl.metadata (1.6 kB)\n",
      "Requirement already satisfied: six>=1.5 in /opt/conda/lib/python3.10/site-packages (from python-dateutil>=2.8.2->pandas->datasets->trl==0.8.5) (1.16.0)\n",
      "Downloading peft-0.10.0-py3-none-any.whl (199 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m199.1/199.1 kB\u001b[0m \u001b[31m4.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
      "\u001b[?25hDownloading bitsandbytes-0.40.2-py3-none-any.whl (92.5 MB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m92.5/92.5 MB\u001b[0m \u001b[31m68.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m:00:01\u001b[0m00:01\u001b[0m\n",
      "\u001b[?25hDownloading trl-0.8.5-py3-none-any.whl (245 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m245.1/245.1 kB\u001b[0m \u001b[31m75.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hDownloading tyro-0.8.3-py3-none-any.whl (102 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m102.0/102.0 kB\u001b[0m \u001b[31m25.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hDownloading docstring_parser-0.16-py3-none-any.whl (36 kB)\n",
      "Downloading rich-13.7.1-py3-none-any.whl (240 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m240.7/240.7 kB\u001b[0m \u001b[31m50.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hDownloading shtab-1.7.1-py3-none-any.whl (14 kB)\n",
      "Downloading markdown_it_py-3.0.0-py3-none-any.whl (87 kB)\n",
      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m87.5/87.5 kB\u001b[0m \u001b[31m24.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
      "\u001b[?25hDownloading mdurl-0.1.2-py3-none-any.whl (10.0 kB)\n",
      "Installing collected packages: bitsandbytes, shtab, mdurl, docstring-parser, markdown-it-py, rich, tyro, trl, peft\n",
      "  Attempting uninstall: bitsandbytes\n",
      "    Found existing installation: bitsandbytes 0.43.1\n",
      "    Uninstalling bitsandbytes-0.43.1:\n",
      "      Successfully uninstalled bitsandbytes-0.43.1\n",
      "  Attempting uninstall: trl\n",
      "    Found existing installation: trl 0.4.7\n",
      "    Uninstalling trl-0.4.7:\n",
      "      Successfully uninstalled trl-0.4.7\n",
      "Successfully installed bitsandbytes-0.40.2 docstring-parser-0.16 markdown-it-py-3.0.0 mdurl-0.1.2 peft-0.10.0 rich-13.7.1 shtab-1.7.1 trl-0.8.5 tyro-0.8.3\n",
      "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
      "\u001b[0m"
     ]
    }
   ],
   "source": [
    "!pip install -U accelerate==0.29.3 peft==0.10.0 bitsandbytes==0.40.2 transformers==4.40.0 trl==0.8.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "4526cde9-8b9b-4917-bb50-d1dd284de4dd",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import torch\n",
    "from datasets import load_dataset\n",
    "from transformers import (\n",
    "    AutoModelForCausalLM,\n",
    "    AutoTokenizer,\n",
    "    BitsAndBytesConfig,\n",
    "    HfArgumentParser,\n",
    "    TrainingArguments,\n",
    "    pipeline,\n",
    "    logging,\n",
    ")\n",
    "from peft import LoraConfig, PeftModel\n",
    "from trl import SFTTrainer\n",
    "hf_token = \"hf_...\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "eaa6d9b2-9d4d-42b2-aba6-cc1a5d71c551",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GPU name: NVIDIA L40\n",
      "GPU model: NVIDIA L40\n"
     ]
    }
   ],
   "source": [
    "import torch\n",
    "\n",
    "if torch.cuda.is_available():\n",
    "    print(\"GPU name:\", torch.cuda.get_device_name(0))\n",
    "    print(\"GPU model:\", torch.cuda.get_device_name(0))\n",
    "else:\n",
    "    print(\"No GPU available\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "d9985edf-667a-4326-81c1-f5526bf34375",
   "metadata": {},
   "outputs": [],
   "source": [
    "# The model that you want to train from the Hugging Face hub\n",
    "model_name = \"NousResearch/Meta-Llama-3-8B\"\n",
    "\n",
    "# The instruction dataset to use\n",
    "dataset_name = \"Hemanth-thunder/tamil-open-instruct-v1\"\n",
    "\n",
    "# Fine-tuned model name\n",
    "new_model = \"llama-3-8b-tamil-open-instruct-v1\"\n",
    "\n",
    "################################################################################\n",
    "# QLoRA parameters\n",
    "################################################################################\n",
    "\n",
    "# LoRA attention dimension\n",
    "lora_r = 64\n",
    "\n",
    "# Alpha parameter for LoRA scaling\n",
    "lora_alpha = 16\n",
    "\n",
    "# Dropout probability for LoRA layers\n",
    "lora_dropout = 0\n",
    "\n",
    "################################################################################\n",
    "# bitsandbytes parameters\n",
    "################################################################################\n",
    "\n",
    "# Activate 4-bit precision base model loading\n",
    "use_4bit = True\n",
    "\n",
    "# Compute dtype for 4-bit base models\n",
    "bnb_4bit_compute_dtype = \"float16\"\n",
    "\n",
    "# Quantization type (fp4 or nf4)\n",
    "bnb_4bit_quant_type = \"nf4\"\n",
    "\n",
    "# Activate nested quantization for 4-bit base models (double quantization)\n",
    "use_nested_quant = False\n",
    "\n",
    "################################################################################\n",
    "# TrainingArguments parameters\n",
    "################################################################################\n",
    "\n",
    "# Output directory where the model predictions and checkpoints will be stored\n",
    "output_dir = \"./results\"\n",
    "\n",
    "# Number of training epochs\n",
    "num_train_epochs = 1\n",
    "\n",
    "# Enable fp16/bf16 training (set bf16 to True with an A100)\n",
    "fp16 = False\n",
    "bf16 = False\n",
    "\n",
    "# Batch size per GPU for training\n",
    "per_device_train_batch_size = 2\n",
    "\n",
    "# Batch size per GPU for evaluation\n",
    "per_device_eval_batch_size = 2\n",
    "\n",
    "# Number of update steps to accumulate the gradients for\n",
    "gradient_accumulation_steps = 1\n",
    "\n",
    "# Enable gradient checkpointing\n",
    "gradient_checkpointing = True\n",
    "\n",
    "# Maximum gradient normal (gradient clipping)\n",
    "max_grad_norm = 0.3\n",
    "\n",
    "# Initial learning rate (AdamW optimizer)\n",
    "learning_rate = 2e-4\n",
    "\n",
    "# Weight decay to apply to all layers except bias/LayerNorm weights\n",
    "weight_decay = 0.001\n",
    "\n",
    "# Optimizer to use\n",
    "optim = \"paged_adamw_32bit\"\n",
    "\n",
    "# Learning rate schedule\n",
    "lr_scheduler_type = \"cosine\"\n",
    "\n",
    "# Number of training steps (overrides num_train_epochs)\n",
    "max_steps = -1\n",
    "\n",
    "# Ratio of steps for a linear warmup (from 0 to learning rate)\n",
    "warmup_ratio = 0.03\n",
    "\n",
    "# Group sequences into batches with same length\n",
    "# Saves memory and speeds up training considerably\n",
    "group_by_length = True\n",
    "\n",
    "# Save checkpoint every X updates steps\n",
    "save_steps = 0\n",
    "\n",
    "# Log every X updates steps\n",
    "logging_steps = 25\n",
    "\n",
    "################################################################################\n",
    "# SFT parameters\n",
    "################################################################################\n",
    "\n",
    "# Maximum sequence length to use\n",
    "max_seq_length = None\n",
    "\n",
    "# Pack multiple short examples in the same input sequence to increase efficiency\n",
    "packing = False\n",
    "\n",
    "# Load the entire model on the GPU 0\n",
    "device_map = {\"\": 0}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f3b93ec1-47a0-45e9-926b-0bcae1c0fa12",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "df26de72bbcc4db1a2ad6d3474d4125f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading readme:   0%|          | 0.00/769 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "da8431769ed64422ae82d9ee13c7ad51",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading data:   0%|          | 0.00/95.1M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b66f74bd0afc42c3bcde1a24eba910f6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading data:   0%|          | 0.00/109M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b2b3b573b6644d91b075e4d6e8408f76",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading data:   0%|          | 0.00/140M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "8a467545b82f47a994b9f73fe8f8f7ea",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading data:   0%|          | 0.00/204M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3e22588f59524f19bec79cecb6ed7c6e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Generating train split:   0%|          | 0/493813 [00:00<?, ? examples/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Load dataset (you can process it here)\n",
    "dataset = load_dataset(dataset_name, split=\"train\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "adcefe75-8f08-4bad-8a9d-14a2a8a26507",
   "metadata": {},
   "outputs": [],
   "source": [
    "sample_dataset = dataset.shuffle().select(range(150000))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "6dd0c0e5-d31b-469e-b6b4-1da7358d991b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dataset({\n",
       "    features: ['instruction', 'output', 'input', 'text'],\n",
       "    num_rows: 150000\n",
       "})"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample_dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "1bf647af-06a1-4934-a27e-df9b3547389c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "சரியான பதிலுடன் வேலையை வெற்றிகரமாக முடிக்க. தேவையான தகவலை உள்ளிடவும்.\n",
      "\n",
      "### Instruction:\n",
      "பூனை என்ற வார்த்தையில் எத்தனை எழுத்துக்கள் உள்ளன?\n",
      "\n",
      "### Response:\n",
      "பூனை என்ற சொல்லுக்கு ஒரே ஒரு எழுத்து மட்டுமே உண்டு.\n"
     ]
    }
   ],
   "source": [
    "print(sample_dataset['text'][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "0bfbb931-68dc-42fc-bcfc-e2e0fd9277fc",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5872add8f5964d24960d641ad185df38",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "config.json:   0%|          | 0.00/654 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "b224503ca8794ca6aa1fc4a07f89a1f4",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model.safetensors.index.json:   0%|          | 0.00/23.9k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4f7980c459d14c2aadaab3fc59a8247b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading shards:   0%|          | 0/4 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "513b5081256a4ada88c3bc8f875bd598",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model-00001-of-00004.safetensors:   0%|          | 0.00/4.98G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0694d9fab375422eb3d1a99920b335e0",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model-00002-of-00004.safetensors:   0%|          | 0.00/5.00G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "acfb5204ee164be88938c6da128cca4b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model-00003-of-00004.safetensors:   0%|          | 0.00/4.92G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1ea19400ae1c44dcb923763eefe42833",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model-00004-of-00004.safetensors:   0%|          | 0.00/1.17G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f3ab2434c6c64e6b93f26d57e1c82774",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Loading checkpoint shards:   0%|          | 0/4 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ea1e3d2ef7d44099a8174fd47581e956",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "generation_config.json:   0%|          | 0.00/126 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "You are calling `save_pretrained` to a 4-bit converted model, but your `bitsandbytes` version doesn't support it. If you want to save 4-bit models, make sure to have `bitsandbytes>=0.41.3` installed.\n"
     ]
    }
   ],
   "source": [
    "# Load tokenizer and model with QLoRA configuration\n",
    "compute_dtype = getattr(torch, bnb_4bit_compute_dtype)\n",
    "\n",
    "bnb_config = BitsAndBytesConfig(\n",
    "    load_in_4bit=use_4bit,\n",
    "    bnb_4bit_quant_type=bnb_4bit_quant_type,\n",
    "    bnb_4bit_compute_dtype=compute_dtype,\n",
    "    bnb_4bit_use_double_quant=use_nested_quant,\n",
    ")\n",
    "\n",
    " \n",
    "\n",
    "# Load base model\n",
    "model = AutoModelForCausalLM.from_pretrained(\n",
    "    model_name,\n",
    "    quantization_config=bnb_config,\n",
    "    device_map=device_map,token=hf_token\n",
    ")\n",
    "model.config.use_cache = False\n",
    "model.config.pretraining_tp = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "afafb0bd-29bc-4800-ae50-1bf7bcbeff97",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "445a7cc0f85e40b5852a26b46d1a49af",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer_config.json:   0%|          | 0.00/50.9k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "1cae5de71fb24189a97ac781c0341e0a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "tokenizer.json:   0%|          | 0.00/9.08M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ded9cbe3b02846d4aeb5dca3eed9fe91",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "special_tokens_map.json:   0%|          | 0.00/73.0 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n"
     ]
    }
   ],
   "source": [
    "# Load LLaMA tokenizer\n",
    "tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True,token=hf_token)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "1da6f461-1cec-4500-9022-11ed50875f93",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7473192ced96421a955e0317041cfd79",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Map:   0%|          | 0/150000 [00:00<?, ? examples/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Must add EOS_TOKEN at response last line\n",
    "tokenizer.pad_token = tokenizer.eos_token\n",
    "EOS_TOKEN = tokenizer.eos_token \n",
    "def prompt_eos(sample):\n",
    "    sample['text'] = sample['text']+EOS_TOKEN\n",
    "    return sample\n",
    "sample_dataset = sample_dataset.map(prompt_eos)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "c197b380-2c70-4c6a-a4a9-97fbe62d7b65",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load LoRA configuration\n",
    "peft_config = LoraConfig(\n",
    "    lora_alpha=lora_alpha,\n",
    "    lora_dropout=lora_dropout,\n",
    "    r=lora_r,\n",
    "    bias=\"none\",\n",
    "    task_type=\"CAUSAL_LM\",\n",
    ")\n",
    "\n",
    "# Set training parameters\n",
    "training_arguments = TrainingArguments(\n",
    "    output_dir=output_dir,\n",
    "    num_train_epochs=num_train_epochs,\n",
    "    per_device_train_batch_size=per_device_train_batch_size,\n",
    "    gradient_accumulation_steps=gradient_accumulation_steps,\n",
    "    optim=optim,\n",
    "    save_steps=save_steps,\n",
    "    logging_steps=logging_steps,\n",
    "    learning_rate=learning_rate,\n",
    "    weight_decay=weight_decay,\n",
    "    fp16=fp16,\n",
    "    bf16=bf16,\n",
    "    max_grad_norm=max_grad_norm,\n",
    "    max_steps=max_steps,\n",
    "    warmup_ratio=warmup_ratio,\n",
    "    group_by_length=group_by_length,\n",
    "    lr_scheduler_type=lr_scheduler_type,\n",
    "    report_to=\"tensorboard\"\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "c6d21207-8de4-4b98-bf1b-66f2fa43456d",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/conda/lib/python3.10/site-packages/trl/trainer/sft_trainer.py:246: UserWarning: You didn't pass a `max_seq_length` argument to the SFTTrainer, this will default to 1024\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "55425f4a1c5a4ce7bff8d12023427e15",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Map:   0%|          | 0/150000 [00:00<?, ? examples/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Detected kernel version 5.4.0, which is below the recommended minimum of 5.5.0; this can cause the process to hang. It is recommended to upgrade the kernel to the minimum version or higher.\n"
     ]
    }
   ],
   "source": [
    "# Set supervised fine-tuning parameters\n",
    "trainer = SFTTrainer(\n",
    "    model=model,\n",
    "    train_dataset=sample_dataset,\n",
    "    peft_config=peft_config,\n",
    "    dataset_text_field=\"text\",\n",
    "    max_seq_length=max_seq_length,\n",
    "    tokenizer=tokenizer,\n",
    "    args=training_arguments,\n",
    "    packing=packing,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "12168034-3c16-40f9-9144-29cb0890ab50",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "You are using 8-bit optimizers with a version of `bitsandbytes` < 0.41.1. It is recommended to update your version as a major bug has been fixed in 8-bit optimizers.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "      \n",
       "      <progress value='7002' max='75000' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      [ 7002/75000 1:21:20 < 13:10:06, 1.43 it/s, Epoch 0.09/1]\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",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>25</td>\n",
       "      <td>0.660800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>0.827500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>75</td>\n",
       "      <td>0.659400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>100</td>\n",
       "      <td>0.765900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>125</td>\n",
       "      <td>0.607800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>150</td>\n",
       "      <td>0.671900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>175</td>\n",
       "      <td>0.561000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>200</td>\n",
       "      <td>0.534100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>225</td>\n",
       "      <td>0.482900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>250</td>\n",
       "      <td>0.433000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>275</td>\n",
       "      <td>0.467000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>300</td>\n",
       "      <td>0.433500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>325</td>\n",
       "      <td>0.451000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>350</td>\n",
       "      <td>0.384200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>375</td>\n",
       "      <td>0.458800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>400</td>\n",
       "      <td>0.389400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>425</td>\n",
       "      <td>0.455700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>450</td>\n",
       "      <td>0.429300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>475</td>\n",
       "      <td>0.449300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>500</td>\n",
       "      <td>0.399000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>525</td>\n",
       "      <td>0.425200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>550</td>\n",
       "      <td>0.339400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>575</td>\n",
       "      <td>0.428600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>600</td>\n",
       "      <td>0.380800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>625</td>\n",
       "      <td>0.443500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>650</td>\n",
       "      <td>0.364600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>675</td>\n",
       "      <td>0.437700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>700</td>\n",
       "      <td>0.389400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>725</td>\n",
       "      <td>0.427500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>750</td>\n",
       "      <td>0.349100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>775</td>\n",
       "      <td>0.430100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>800</td>\n",
       "      <td>0.381000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>825</td>\n",
       "      <td>0.432200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>850</td>\n",
       "      <td>0.378200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>875</td>\n",
       "      <td>0.440000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>900</td>\n",
       "      <td>0.366400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>925</td>\n",
       "      <td>0.423000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>950</td>\n",
       "      <td>0.358800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>975</td>\n",
       "      <td>0.438000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1000</td>\n",
       "      <td>0.345700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1025</td>\n",
       "      <td>0.410000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1050</td>\n",
       "      <td>0.329900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1075</td>\n",
       "      <td>0.418200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1100</td>\n",
       "      <td>0.351900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1125</td>\n",
       "      <td>0.399200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1150</td>\n",
       "      <td>0.362400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1175</td>\n",
       "      <td>0.423600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1200</td>\n",
       "      <td>0.313800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1225</td>\n",
       "      <td>0.433900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1250</td>\n",
       "      <td>0.347900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1275</td>\n",
       "      <td>0.406900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1300</td>\n",
       "      <td>0.334100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1325</td>\n",
       "      <td>0.430900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1350</td>\n",
       "      <td>0.337800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1375</td>\n",
       "      <td>0.404100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1400</td>\n",
       "      <td>0.320800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1425</td>\n",
       "      <td>0.410300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1450</td>\n",
       "      <td>0.356000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1475</td>\n",
       "      <td>0.418500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1500</td>\n",
       "      <td>0.324800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1525</td>\n",
       "      <td>0.410600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1550</td>\n",
       "      <td>0.322100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1575</td>\n",
       "      <td>0.393000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1600</td>\n",
       "      <td>0.325300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1625</td>\n",
       "      <td>0.419100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1650</td>\n",
       "      <td>0.337400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1675</td>\n",
       "      <td>0.435600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1700</td>\n",
       "      <td>0.339800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1725</td>\n",
       "      <td>0.407300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1750</td>\n",
       "      <td>0.320400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1775</td>\n",
       "      <td>0.408600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1800</td>\n",
       "      <td>0.330300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1825</td>\n",
       "      <td>0.408300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1850</td>\n",
       "      <td>0.320600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1875</td>\n",
       "      <td>0.424500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1900</td>\n",
       "      <td>0.351000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1925</td>\n",
       "      <td>0.390200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1950</td>\n",
       "      <td>0.367000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1975</td>\n",
       "      <td>0.424100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2000</td>\n",
       "      <td>0.338800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2025</td>\n",
       "      <td>0.426700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2050</td>\n",
       "      <td>0.310100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2075</td>\n",
       "      <td>0.418100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2100</td>\n",
       "      <td>0.344800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2125</td>\n",
       "      <td>0.401700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2150</td>\n",
       "      <td>0.337100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2175</td>\n",
       "      <td>0.419400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2200</td>\n",
       "      <td>0.377800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2225</td>\n",
       "      <td>0.448300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2250</td>\n",
       "      <td>0.334100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2275</td>\n",
       "      <td>0.417900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2300</td>\n",
       "      <td>0.331900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2325</td>\n",
       "      <td>0.398100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2350</td>\n",
       "      <td>0.302400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2375</td>\n",
       "      <td>0.390300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2400</td>\n",
       "      <td>0.323700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2425</td>\n",
       "      <td>0.392400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2450</td>\n",
       "      <td>0.331700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2475</td>\n",
       "      <td>0.433100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2500</td>\n",
       "      <td>0.326300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2525</td>\n",
       "      <td>0.406900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2550</td>\n",
       "      <td>0.340300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2575</td>\n",
       "      <td>0.409400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2600</td>\n",
       "      <td>0.324500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2625</td>\n",
       "      <td>0.404300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2650</td>\n",
       "      <td>0.323700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2675</td>\n",
       "      <td>0.413300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2700</td>\n",
       "      <td>0.323700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2725</td>\n",
       "      <td>0.389200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2750</td>\n",
       "      <td>0.331100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2775</td>\n",
       "      <td>0.387900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2800</td>\n",
       "      <td>0.306800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2825</td>\n",
       "      <td>0.409300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2850</td>\n",
       "      <td>0.326500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2875</td>\n",
       "      <td>0.381500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2900</td>\n",
       "      <td>0.315800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2925</td>\n",
       "      <td>0.393200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2950</td>\n",
       "      <td>0.342200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2975</td>\n",
       "      <td>0.414900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3000</td>\n",
       "      <td>0.302400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3025</td>\n",
       "      <td>0.420300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3050</td>\n",
       "      <td>0.311700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3075</td>\n",
       "      <td>0.401700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3100</td>\n",
       "      <td>0.310500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3125</td>\n",
       "      <td>0.394500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3150</td>\n",
       "      <td>0.324800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3175</td>\n",
       "      <td>0.397100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3200</td>\n",
       "      <td>0.336000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3225</td>\n",
       "      <td>0.400700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3250</td>\n",
       "      <td>0.310000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3275</td>\n",
       "      <td>0.395600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3300</td>\n",
       "      <td>0.314200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3325</td>\n",
       "      <td>0.416000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3350</td>\n",
       "      <td>0.290200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3375</td>\n",
       "      <td>0.392600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3400</td>\n",
       "      <td>0.324200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3425</td>\n",
       "      <td>0.397000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3450</td>\n",
       "      <td>0.310600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3475</td>\n",
       "      <td>0.394000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3500</td>\n",
       "      <td>0.340900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3525</td>\n",
       "      <td>0.376300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3550</td>\n",
       "      <td>0.329500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3575</td>\n",
       "      <td>0.407400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3600</td>\n",
       "      <td>0.304400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3625</td>\n",
       "      <td>0.399000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3650</td>\n",
       "      <td>0.298000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3675</td>\n",
       "      <td>0.382700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3700</td>\n",
       "      <td>0.333000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3725</td>\n",
       "      <td>0.407600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3750</td>\n",
       "      <td>0.296700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3775</td>\n",
       "      <td>0.399300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3800</td>\n",
       "      <td>0.283200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3825</td>\n",
       "      <td>0.391400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3850</td>\n",
       "      <td>0.310800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3875</td>\n",
       "      <td>0.402200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3900</td>\n",
       "      <td>0.318700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3925</td>\n",
       "      <td>0.391900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3950</td>\n",
       "      <td>0.295400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3975</td>\n",
       "      <td>0.388300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4000</td>\n",
       "      <td>0.336100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4025</td>\n",
       "      <td>0.391000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4050</td>\n",
       "      <td>0.303600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4075</td>\n",
       "      <td>0.402400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4100</td>\n",
       "      <td>0.291200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4125</td>\n",
       "      <td>0.412700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4150</td>\n",
       "      <td>0.298800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4175</td>\n",
       "      <td>0.388300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4200</td>\n",
       "      <td>0.330300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4225</td>\n",
       "      <td>0.363700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4250</td>\n",
       "      <td>0.291500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4275</td>\n",
       "      <td>0.405800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4300</td>\n",
       "      <td>0.314700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4325</td>\n",
       "      <td>0.373100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4350</td>\n",
       "      <td>0.284500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4375</td>\n",
       "      <td>0.403400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4400</td>\n",
       "      <td>0.296500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4425</td>\n",
       "      <td>0.417300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4450</td>\n",
       "      <td>0.320200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4475</td>\n",
       "      <td>0.371700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4500</td>\n",
       "      <td>0.307300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4525</td>\n",
       "      <td>0.403500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4550</td>\n",
       "      <td>0.328900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4575</td>\n",
       "      <td>0.403600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4600</td>\n",
       "      <td>0.299200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4625</td>\n",
       "      <td>0.409600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4650</td>\n",
       "      <td>0.326400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4675</td>\n",
       "      <td>0.368000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4700</td>\n",
       "      <td>0.301600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4725</td>\n",
       "      <td>0.380600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4750</td>\n",
       "      <td>0.296600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4775</td>\n",
       "      <td>0.400200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4800</td>\n",
       "      <td>0.279800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4825</td>\n",
       "      <td>0.389100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4850</td>\n",
       "      <td>0.310300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4875</td>\n",
       "      <td>0.393700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4900</td>\n",
       "      <td>0.316800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4925</td>\n",
       "      <td>0.399000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4950</td>\n",
       "      <td>0.280100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4975</td>\n",
       "      <td>0.377800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5000</td>\n",
       "      <td>0.302500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5025</td>\n",
       "      <td>0.417900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5050</td>\n",
       "      <td>0.275300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5075</td>\n",
       "      <td>0.384600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5100</td>\n",
       "      <td>0.288700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5125</td>\n",
       "      <td>0.392600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5150</td>\n",
       "      <td>0.332600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5175</td>\n",
       "      <td>0.385300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5200</td>\n",
       "      <td>0.274900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5225</td>\n",
       "      <td>0.377400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5250</td>\n",
       "      <td>0.312300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5275</td>\n",
       "      <td>0.378900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5300</td>\n",
       "      <td>0.284900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5325</td>\n",
       "      <td>0.386700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5350</td>\n",
       "      <td>0.310100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5375</td>\n",
       "      <td>0.380300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5400</td>\n",
       "      <td>0.296900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5425</td>\n",
       "      <td>0.361400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5450</td>\n",
       "      <td>0.321700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5475</td>\n",
       "      <td>0.367400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5500</td>\n",
       "      <td>0.305400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5525</td>\n",
       "      <td>0.390800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5550</td>\n",
       "      <td>0.291600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5575</td>\n",
       "      <td>0.376000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5600</td>\n",
       "      <td>0.290500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5625</td>\n",
       "      <td>0.390400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5650</td>\n",
       "      <td>0.293600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5675</td>\n",
       "      <td>0.388400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5700</td>\n",
       "      <td>0.301800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5725</td>\n",
       "      <td>0.410500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5750</td>\n",
       "      <td>0.312400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5775</td>\n",
       "      <td>0.361300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5800</td>\n",
       "      <td>0.290500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5825</td>\n",
       "      <td>0.394600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5850</td>\n",
       "      <td>0.299200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5875</td>\n",
       "      <td>0.383400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5900</td>\n",
       "      <td>0.326300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5925</td>\n",
       "      <td>0.411100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5950</td>\n",
       "      <td>0.300900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5975</td>\n",
       "      <td>0.402300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6000</td>\n",
       "      <td>0.286000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6025</td>\n",
       "      <td>0.388700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6050</td>\n",
       "      <td>0.311300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6075</td>\n",
       "      <td>0.383500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6100</td>\n",
       "      <td>0.299200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6125</td>\n",
       "      <td>0.388200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6150</td>\n",
       "      <td>0.310300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6175</td>\n",
       "      <td>0.381600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6200</td>\n",
       "      <td>0.274100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6225</td>\n",
       "      <td>0.359400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6250</td>\n",
       "      <td>0.303800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6275</td>\n",
       "      <td>0.372100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6300</td>\n",
       "      <td>0.285300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6325</td>\n",
       "      <td>0.373400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6350</td>\n",
       "      <td>0.275800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6375</td>\n",
       "      <td>0.361600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6400</td>\n",
       "      <td>0.287500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6425</td>\n",
       "      <td>0.394100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6450</td>\n",
       "      <td>0.276800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6475</td>\n",
       "      <td>0.364900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6500</td>\n",
       "      <td>0.270900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6525</td>\n",
       "      <td>0.374100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6550</td>\n",
       "      <td>0.334700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6575</td>\n",
       "      <td>0.383000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6600</td>\n",
       "      <td>0.306100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6625</td>\n",
       "      <td>0.385300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6650</td>\n",
       "      <td>0.287100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6675</td>\n",
       "      <td>0.355800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6700</td>\n",
       "      <td>0.281700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6725</td>\n",
       "      <td>0.373800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6750</td>\n",
       "      <td>0.294100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6775</td>\n",
       "      <td>0.391800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6800</td>\n",
       "      <td>0.288100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6825</td>\n",
       "      <td>0.420100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6850</td>\n",
       "      <td>0.301100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6875</td>\n",
       "      <td>0.391900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6900</td>\n",
       "      <td>0.300900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6925</td>\n",
       "      <td>0.365600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6950</td>\n",
       "      <td>0.291900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6975</td>\n",
       "      <td>0.358800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7000</td>\n",
       "      <td>0.300300</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table><p>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Cell \u001b[0;32mIn[14], line 1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[43mtrainer\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtrain\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m/opt/conda/lib/python3.10/site-packages/trl/trainer/sft_trainer.py:361\u001b[0m, in \u001b[0;36mSFTTrainer.train\u001b[0;34m(self, *args, **kwargs)\u001b[0m\n\u001b[1;32m    358\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mneftune_noise_alpha \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_trainer_supports_neftune:\n\u001b[1;32m    359\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmodel \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_trl_activate_neftune(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmodel)\n\u001b[0;32m--> 361\u001b[0m output \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[43mtrain\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    363\u001b[0m \u001b[38;5;66;03m# After training we make sure to retrieve back the original forward pass method\u001b[39;00m\n\u001b[1;32m    364\u001b[0m \u001b[38;5;66;03m# for the embedding layer by removing the forward post hook.\u001b[39;00m\n\u001b[1;32m    365\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mneftune_noise_alpha \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_trainer_supports_neftune:\n",
      "File \u001b[0;32m/opt/conda/lib/python3.10/site-packages/transformers/trainer.py:1859\u001b[0m, in \u001b[0;36mTrainer.train\u001b[0;34m(self, resume_from_checkpoint, trial, ignore_keys_for_eval, **kwargs)\u001b[0m\n\u001b[1;32m   1857\u001b[0m         hf_hub_utils\u001b[38;5;241m.\u001b[39menable_progress_bars()\n\u001b[1;32m   1858\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 1859\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43minner_training_loop\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1860\u001b[0m \u001b[43m        \u001b[49m\u001b[43margs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1861\u001b[0m \u001b[43m        \u001b[49m\u001b[43mresume_from_checkpoint\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mresume_from_checkpoint\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1862\u001b[0m \u001b[43m        \u001b[49m\u001b[43mtrial\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtrial\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1863\u001b[0m \u001b[43m        \u001b[49m\u001b[43mignore_keys_for_eval\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mignore_keys_for_eval\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1864\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m/opt/conda/lib/python3.10/site-packages/transformers/trainer.py:2203\u001b[0m, in \u001b[0;36mTrainer._inner_training_loop\u001b[0;34m(self, batch_size, args, resume_from_checkpoint, trial, ignore_keys_for_eval)\u001b[0m\n\u001b[1;32m   2200\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcontrol \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcallback_handler\u001b[38;5;241m.\u001b[39mon_step_begin(args, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstate, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcontrol)\n\u001b[1;32m   2202\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39maccelerator\u001b[38;5;241m.\u001b[39maccumulate(model):\n\u001b[0;32m-> 2203\u001b[0m     tr_loss_step \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtraining_step\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmodel\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43minputs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   2205\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m (\n\u001b[1;32m   2206\u001b[0m     args\u001b[38;5;241m.\u001b[39mlogging_nan_inf_filter\n\u001b[1;32m   2207\u001b[0m     \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m is_torch_xla_available()\n\u001b[1;32m   2208\u001b[0m     \u001b[38;5;129;01mand\u001b[39;00m (torch\u001b[38;5;241m.\u001b[39misnan(tr_loss_step) \u001b[38;5;129;01mor\u001b[39;00m torch\u001b[38;5;241m.\u001b[39misinf(tr_loss_step))\n\u001b[1;32m   2209\u001b[0m ):\n\u001b[1;32m   2210\u001b[0m     \u001b[38;5;66;03m# if loss is nan or inf simply add the average of previous logged losses\u001b[39;00m\n\u001b[1;32m   2211\u001b[0m     tr_loss \u001b[38;5;241m+\u001b[39m\u001b[38;5;241m=\u001b[39m tr_loss \u001b[38;5;241m/\u001b[39m (\u001b[38;5;241m1\u001b[39m \u001b[38;5;241m+\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstate\u001b[38;5;241m.\u001b[39mglobal_step \u001b[38;5;241m-\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_globalstep_last_logged)\n",
      "File \u001b[0;32m/opt/conda/lib/python3.10/site-packages/transformers/trainer.py:3147\u001b[0m, in \u001b[0;36mTrainer.training_step\u001b[0;34m(self, model, inputs)\u001b[0m\n\u001b[1;32m   3145\u001b[0m         scaled_loss\u001b[38;5;241m.\u001b[39mbackward()\n\u001b[1;32m   3146\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 3147\u001b[0m     \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43maccelerator\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbackward\u001b[49m\u001b[43m(\u001b[49m\u001b[43mloss\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   3149\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m loss\u001b[38;5;241m.\u001b[39mdetach() \u001b[38;5;241m/\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39margs\u001b[38;5;241m.\u001b[39mgradient_accumulation_steps\n",
      "File \u001b[0;32m/opt/conda/lib/python3.10/site-packages/accelerate/accelerator.py:2013\u001b[0m, in \u001b[0;36mAccelerator.backward\u001b[0;34m(self, loss, **kwargs)\u001b[0m\n\u001b[1;32m   2011\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mscaler\u001b[38;5;241m.\u001b[39mscale(loss)\u001b[38;5;241m.\u001b[39mbackward(\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[1;32m   2012\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 2013\u001b[0m     \u001b[43mloss\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbackward\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/opt/conda/lib/python3.10/site-packages/torch/_tensor.py:522\u001b[0m, in \u001b[0;36mTensor.backward\u001b[0;34m(self, gradient, retain_graph, create_graph, inputs)\u001b[0m\n\u001b[1;32m    512\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m has_torch_function_unary(\u001b[38;5;28mself\u001b[39m):\n\u001b[1;32m    513\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m handle_torch_function(\n\u001b[1;32m    514\u001b[0m         Tensor\u001b[38;5;241m.\u001b[39mbackward,\n\u001b[1;32m    515\u001b[0m         (\u001b[38;5;28mself\u001b[39m,),\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m    520\u001b[0m         inputs\u001b[38;5;241m=\u001b[39minputs,\n\u001b[1;32m    521\u001b[0m     )\n\u001b[0;32m--> 522\u001b[0m \u001b[43mtorch\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mautograd\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbackward\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m    523\u001b[0m \u001b[43m    \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mgradient\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mretain_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcreate_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43minputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43minputs\u001b[49m\n\u001b[1;32m    524\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[0;32m/opt/conda/lib/python3.10/site-packages/torch/autograd/__init__.py:266\u001b[0m, in \u001b[0;36mbackward\u001b[0;34m(tensors, grad_tensors, retain_graph, create_graph, grad_variables, inputs)\u001b[0m\n\u001b[1;32m    261\u001b[0m     retain_graph \u001b[38;5;241m=\u001b[39m create_graph\n\u001b[1;32m    263\u001b[0m \u001b[38;5;66;03m# The reason we repeat the same comment below is that\u001b[39;00m\n\u001b[1;32m    264\u001b[0m \u001b[38;5;66;03m# some Python versions print out the first line of a multi-line function\u001b[39;00m\n\u001b[1;32m    265\u001b[0m \u001b[38;5;66;03m# calls in the traceback and some print out the last line\u001b[39;00m\n\u001b[0;32m--> 266\u001b[0m \u001b[43mVariable\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_execution_engine\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun_backward\u001b[49m\u001b[43m(\u001b[49m\u001b[43m  \u001b[49m\u001b[38;5;66;43;03m# Calls into the C++ engine to run the backward pass\u001b[39;49;00m\n\u001b[1;32m    267\u001b[0m \u001b[43m    \u001b[49m\u001b[43mtensors\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    268\u001b[0m \u001b[43m    \u001b[49m\u001b[43mgrad_tensors_\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    269\u001b[0m \u001b[43m    \u001b[49m\u001b[43mretain_graph\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    270\u001b[0m \u001b[43m    \u001b[49m\u001b[43mcreate_graph\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    271\u001b[0m \u001b[43m    \u001b[49m\u001b[43minputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m    272\u001b[0m \u001b[43m    \u001b[49m\u001b[43mallow_unreachable\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\u001b[1;32m    273\u001b[0m \u001b[43m    \u001b[49m\u001b[43maccumulate_grad\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\u001b[1;32m    274\u001b[0m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "trainer.train()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "369ca860-5669-47dd-ba34-f3808713afa7",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "7d1fa166-3fd6-4f8e-8da2-23c210c5e343",
   "metadata": {},
   "outputs": [],
   "source": [
    " def generate_response(prompt, model):\n",
    "    encoded_input = tokenizer(prompt,  return_tensors=\"pt\", add_special_tokens=True)\n",
    "    model_inputs = encoded_input.to('cuda')\n",
    "    \n",
    "    generated_ids = model.generate(**model_inputs, max_new_tokens=512, do_sample=True, pad_token_id=tokenizer.eos_token_id)\n",
    "    \n",
    "    decoded_output = tokenizer.batch_decode(generated_ids)\n",
    "    \n",
    "    return decoded_output[0].replace(prompt, \"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "098c5a4c-5de1-4062-8ec2-dad8e5784613",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' வரலாற்றின் முக்கியத்துவத்தைப் பற்றி விவாதிக்கும் இரண்டு நண்பர்களிடையே ஒரு உரையாடலை உருவாக்கவும்.'"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample_dataset['instruction'][5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "0d233579-fefe-4234-ba46-a5c165bdc300",
   "metadata": {},
   "outputs": [],
   "source": [
    "prompt_template_wo_input=\"\"\"சரியான பதிலுடன் வேலையை வெற்றிகரமாக முடிக்க, தேவையான தகவலை உள்ளிடவும்.\n",
    "\n",
    "### Instruction:\n",
    "{}\n",
    "\n",
    "### Response:\"\"\"\n",
    "\n",
    "# with Input\n",
    "prompt_template_input=\"\"\"சரியான பதிலுடன் வேலையை வெற்றிகரமாக முடிக்க, வழங்கப்பட்ட வழிகாட்டுதல்களைப் பின்பற்றி, தேவையான தகவலை உள்ளிடவும்.\n",
    "\n",
    "### Instruction:\n",
    "{}\n",
    "\n",
    "### Input:\n",
    "{}\n",
    "\n",
    "### Response:\"\"\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "fb03cea4-c1a2-4842-8f74-f70749687936",
   "metadata": {},
   "outputs": [],
   "source": [
    "# def create_prompt(sample):\n",
    "#     if 'nan' not in sample['input']:\n",
    "#       prompt = prompt_template_input.format(sample['instruction'][-1].strip(),sample['input'])\n",
    "#     else:\n",
    "#      prompt = prompt_template_wo_input.format(sample['instruction'][-1].strip())\n",
    "#     return prompt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "d4abc647-ecfe-4189-a036-7bd0dfeb49b1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' இந்தப் பொருளை நீங்கள் எப்படி பயன்படுத்த விரும்புகிறீர்கள் என்பதை விவரிக்கவும். தன்னை கோடிட வஹப்பதை நான் கடைக்கோடிப்பேன். நான் என் தங்கியபடி, என் நண்பர், அவளும் நல்லது மேலும் பெரிதும் ஆம்! சாக். இது எனது மிகவும் பிரியப்பட்ட உணவு என்பதை என்னால் நான் என்னால் எப்படி விவரிக்க முடியும்? இது மிக சுவையாக, மிக இனிமையானது மற்றும் பஸ்தாக் மென்மையானது. இதில் நான் சர்க்கரையைக் கடைக்கோடினால்'"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt_template_wo_input=\"\"\"சரியான பதிலுடன் வேலையை வெற்றிகரமாக முடிக்க, தேவையான தகவலை உள்ளிடவும்.\n",
    "\n",
    "### Instruction:\n",
    "வணக்கம்\n",
    "\n",
    "### Response:\"\"\"\n",
    "generate_response(prompt_template_wo_input, model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "e670ec48-466f-4f23-ba9b-0139e20d1fd0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' டெல்லி. இந்தியாவின் தலைநகரம் டெல்லி. அதன் மக்கள்தொகை 2906404 ஆகும். பின்வருபவை நகரில் அமைந்துள்ள சிறந்த 10 இடங்களில் சிலவுடன், அது உலகின் சிறந்த நகரங்களில் ஒன்றாகும். அருகிலுள்ள பிரதேசங்கள் சிலவும் இதில் அடங்கும், உதாரணமாக வைசாகபாத், ஹரியானா மற்றும் மகராஷ்டிரா. இந்தியாவில் பெரிய நகரங்களில் பெங்களூர், ஹைடராபாத், மும்பை மற்றும் புவனேசுவர் ஆகியவை சில. அரசாங்கத்தின் ஆட்சிப்பிரிவுகள் ம'"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt_template_wo_input=\"\"\"சரியான பதிலுடன் வேலையை வெற்றிகரமாக முடிக்க, தேவையான தகவலை உள்ளிடவும்.\n",
    "\n",
    "### Instruction:\n",
    "இந்தியாவின் தலைநகரம் என்ன?\n",
    "\n",
    "### Response:\"\"\"\n",
    "generate_response(prompt_template_wo_input, model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "e8fdb161-00bc-4c66-81d6-576e21f17fbe",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' இரண்டாவது அறிமையான திரைப்படம் லீ டி ஃபியூச்சியின் திரைப்படம்; பிக்லின் புதியது என்றும் வாழ்ந்ததில்லை என்னும் கல்லூரிக்காரரின் நோக்கையும் ஆட்கொள்ளும் இது காதல். அவரது உறவைச் செய்து, ஆசிரியரில் ஆசிரியரைக் கற்றுக்கொள்ள மாணவர்களிடையே ஆற்றல்மிக்க ஈர்ப்பை வைத்திருக்கிறது. வியாதி செயலாற்றுவதில் முதலாவது ஆற்றல்மிக்க செய்தி விசித்திரமான அபயத்தின் அருகில் தடுப்பது, வியாதியில் மகிழ்ச்சியைத் தேடுவ'"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt_template_wo_input=\"\"\"சரியான பதிலுடன் வேலையை வெற்றிகரமாக முடிக்க, தேவையான தகவலை உள்ளிடவும்.\n",
    "\n",
    "### Instruction:\n",
    "யார் நீ\n",
    "\n",
    "### Response:\"\"\"\n",
    "generate_response(prompt_template_wo_input, model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "aec4efd4-e44f-47b9-8ba7-6d3610b6bcef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' இ-காமர்ஸ் இணையதளம் உருவாக்குவதற்கான செலவு சராசரியாக $3,000 இல் தொடங்குகிறது, ஆனால் பொதுவாக பெரிய தயாரிப்பு நிறுவனங்களுக்கு சண்டையாக இருக்கும், ஏனெனில் அவர்கள் தங்கள் பள்ளிவாசல் கட்டுமானத்தை விவகாரமாக்க விரும்புகிறார்கள். Shopify, Wix மற்றும் WooCommerce போன்ற பல அணுகல்கள் இருந்தாலும், கடைசி வழியில் கொள்ளுங்கள் என்றால், தொனியான சிஐ பலகைகளுக்கு $100 ஆக அல்லது கூடுதல் பாதிக்கப்பட்ட பணத்திற்கு $500 இல் �'"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt_template_wo_input=\"\"\"சரியான பதிலுடன் வேலையை வெற்றிகரமாக முடிக்க, தேவையான தகவலை உள்ளிடவும்.\n",
    "\n",
    "### Instruction:\n",
    "இ-காமர்ஸ் இணையதளத்தை உருவாக்குவதற்கான செலவை மதிப்பிடவும்.\n",
    "\n",
    "### Response:\"\"\"\n",
    "generate_response(prompt_template_wo_input, model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "d485e3a8-0596-4c9c-a8b9-c12004f9ea55",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' தசவதமாண்டில் இராஜாமண்டிலால் அமுல்படுத்தப்பட்ட மரம் அழிப்பு வழியாக, ஐரோப்பிய கண்காணிப்புக் காலத்தில் ஐரோப்பிய வரலாறு கைவிடுமாறு நாட்டில் பல செய்திகள் வழியாக பாதுகாக்கப்பட்டு வருகின்றன. வரலாற்று சுயமயப்படுத்துதல், இந்த நிலப்பரப்புகளில் தெரிந்து கொள்ளும் வரலாற்றாசிரியர்களுடன் பகுதியில் கபின்காலத்திற்கு அப்பாலாக உலக அமைப்புகளில் அதிகப்படியான பங்கைக் கொண்டுள்ளது. கற்பனை படைப்ப�'"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "prompt_template_wo_input=\"\"\"சரியான பதிலுடன் வேலையை வெற்றிகரமாக முடிக்க, தேவையான தகவலை உள்ளிடவும்.\n",
    "\n",
    "### Instruction:\n",
    "நீங்கள் ஒரு வரலாற்றாசிரியர் என்று கற்பனை செய்து பாருங்கள். மறுமலர்ச்சிக் காலத்தில் அச்சகத்தின் முக்கியத்துவத்தையும் சமூகத்தில் அதன் தாக்கத்தையும் விளக்கவும்.\n",
    "\n",
    "### Response:\"\"\"\n",
    "generate_response(prompt_template_wo_input, model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "cc82b03a-847a-4753-b0ba-b09c119990ae",
   "metadata": {},
   "outputs": [],
   "source": [
    "trainer.save_model(\"Hemanth-thunder/tamil-llama3-8B-test-open-instruct-v1-peft\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "7369daf5-001b-45ec-956b-d69d4e0fa742",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CommitInfo(commit_url='https://huggingface.co/Hemanth-thunder/tamil-llama3-8B-test-open-instruct-v1-peft/commit/ea5c1c774c652147e088e0f496f20df9cbdf124c', commit_message='Upload tokenizer', commit_description='', oid='ea5c1c774c652147e088e0f496f20df9cbdf124c', pr_url=None, pr_revision=None, pr_num=None)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# model.push_to_hub(\"Hemanth-thunder/tamil-llama3-8B-test-open-instruct-v1-peft\")\n",
    "tokenizer.push_to_hub(\"Hemanth-thunder/tamil-llama3-8B-test-open-instruct-v1-peft\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "b8abb033-eeec-4c90-8b9e-f6198163c8d8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Flush memory\n",
    "del trainer, model\n",
    "torch.cuda.empty_cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "567fc4b9-bf29-4164-a522-bc8a6de24206",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "11dd2650bc8e468eb76b060742de0f5e",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Loading checkpoint shards:   0%|          | 0/4 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "tokenizer = AutoTokenizer.from_pretrained(model_name)\n",
    "fp16_model = AutoModelForCausalLM.from_pretrained(\n",
    "    model_name,\n",
    "    low_cpu_mem_usage=True,\n",
    "    return_dict=True,\n",
    "    torch_dtype=torch.float16,\n",
    "    device_map=\"auto\",\n",
    ")\n",
    "# fp16_model, tokenizer = setup_chat_format(fp16_model, tokenizer)\n",
    "\n",
    "# Merge adapter with base model\n",
    "model = PeftModel.from_pretrained(fp16_model, \"Hemanth-thunder/tamil-llama3-8B-test-open-instruct-v1-peft\")\n",
    "model = model.merge_and_unload()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "a81fe514-4af8-425d-be3f-317ec19477c1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a42987f24cfb4d5a85a51d2691cb91c2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model-00003-of-00004.safetensors:   0%|          | 0.00/4.92G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f69bf406daab47dea0772d877672684b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Upload 4 LFS files:   0%|          | 0/4 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "96c60b7f34234563a92645c66fe2ebed",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model-00001-of-00004.safetensors:   0%|          | 0.00/4.98G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "5c4fe90fd6564f7b8b7db6c4204515ee",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model-00002-of-00004.safetensors:   0%|          | 0.00/5.00G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "79b00804cc6a46099c5af8073142fe01",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model-00004-of-00004.safetensors:   0%|          | 0.00/1.17G [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a63616e559bc4b7392300ac59053c156",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "README.md:   0%|          | 0.00/5.17k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "CommitInfo(commit_url='https://huggingface.co/Hemanth-thunder/tamil-llama3-8B-open-instruct-v1-SFT-test/commit/d17ed00330b71d459e877e7c71fd7fedc34b0f09', commit_message='Upload tokenizer', commit_description='', oid='d17ed00330b71d459e877e7c71fd7fedc34b0f09', pr_url=None, pr_revision=None, pr_num=None)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.push_to_hub(\"Hemanth-thunder/tamil-llama3-8B-open-instruct-v1-SFT-test\" ,token=hf_token)\n",
    "tokenizer.push_to_hub(\"Hemanth-thunder/tamil-llama3-8B-open-instruct-v1-SFT-test\",token=hf_token)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7eea29a2-6440-446b-a7a0-f04b4db6ed2e",
   "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": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}