---
license: llama3.1
base_model: meta-llama/Llama-Guard-3-8B
tags:
- llama
- safeguard
- gptq
language:
- en
pipeline_tag: text-generation
---
This is the 8-bit quantized version of [meta-llama/Llama-Guard-3-8B](https://huggingface.co/meta-llama/Llama-Guard-3-8B) by following [the example](https://github.com/AutoGPTQ/AutoGPTQ/blob/main/examples/quantization/basic_usage.py) from the AutoGPTQ repository.

Safety Codes:

<div style="width: 50%;">

<table style="width: 50%; text-align: center;">
  <thead>
    <tr>
      <th>Code</th>
      <th>Meaning</th>
    </tr>
  </thead>
  <tbody>
    <tr>
      <td>S1</td>
      <td>Violent Crimes</td>
    </tr>
    <tr>
      <td>S2</td>
      <td>Non-Violent Crimes</td>
    </tr>
    <tr>
      <td>S3</td>
      <td>Sex-Related Crimes</td>
    </tr>
    <tr>
      <td>S4</td>
      <td>Child Sexual Exploitation</td>
    </tr>
    <tr>
      <td>S5</td>
      <td>Defamation</td>
    </tr>
    <tr>
      <td>S6</td>
      <td>Specialized Advice</td>
    </tr>
    <tr>
      <td>S7</td>
      <td>Privacy</td>
    </tr>
    <tr>
      <td>S8</td>
      <td>Intellectual Property</td>
    </tr>
    <tr>
      <td>S9</td>
      <td>Indiscriminate Weapons</td>
    </tr>
    <tr>
      <td>S10</td>
      <td>Hate</td>
    </tr>
    <tr>
      <td>S11</td>
      <td>Suicide & Self-Harm</td>
    </tr>
    <tr>
      <td>S12</td>
      <td>Sexual Content</td>
    </tr>
    <tr>
      <td>S13</td>
      <td>Elections</td>
    </tr>
    <tr>
      <td>S14</td>
      <td>Code Interpreter Abuse</td>
    </tr>
  </tbody>
</table>

</div>


---
license: llama3.1
---