# GPT-J-6B - Fine Tuning

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fgpt-j-6b-fine-tuning%2F10218%3E%5C
**Category:** 🤗Transformers\
**Created:** [September 22, 2021, 12:35pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fgpt-j-6b-fine-tuning%2F10218 "2021-09-22T12:35:50Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![BalajiAJ](https://avatars.discourse-cdn.com/v4/letter/b/67e7ee/32.png) [@BalajiAJ](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FBalajiAJ)%5C
**Post date:** [September 22, 2021, 12:35pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fgpt-j-6b-fine-tuning%2F10218%2F1 "2021-09-22T12:35:50Z")

</div>

Hello,

I am trying to understand the GPT-J architecture, am having a question on the fine tuning code.  
While we are doing fine tuning, are we unfreezing any layers specifically in the network or does it happen internally. Could anyone briefly explain how fine tuning woks for GPT-J. Thank you

Regards,  
Balaji