# Finetuning llm model using Trainer and reusing it

**URL:** <https://discuss.huggingface.co/t/finetuning-llm-model-using-trainer-and-reusing-it/48334>\
**Category:** Beginners\
**Created:** [July 27, 2023, 5:48am UTC](https://discuss.huggingface.co/t/finetuning-llm-model-using-trainer-and-reusing-it/48334 "2023-07-27T05:48:38Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![abd007](https://avatars.discourse-cdn.com/v4/letter/a/c57346/32.png) [@abd007](https://discuss.huggingface.co/u/abd007)\
**Post date:** [July 27, 2023, 5:48am UTC](https://discuss.huggingface.co/t/finetuning-llm-model-using-trainer-and-reusing-it/48334/1 "2023-07-27T05:48:38Z")

</div>

Hi , I am new to llm finetuning . Recently I finetuned a llama-7b model on custom data for question answering .  
After training I saved the model using **“model.save\_pretrained(OUTPUT\_DIR)”** command . But in the OUTPUT\_DIR I got two files **adapter\_config.json** and **“adapter\_model.bin”** file . I dont know how to use it to reload the model . Also finding difficulty in further using the model for text generation . Can anyone help here ?

---

<div class="post-metadata">

**Author:** ![Pranavagrl](https://avatars.discourse-cdn.com/v4/letter/p/73ab20/32.png) [@Pranavagrl](https://discuss.huggingface.co/u/Pranavagrl)\
**Post date:** [July 27, 2023, 6:17am UTC](https://discuss.huggingface.co/t/finetuning-llm-model-using-trainer-and-reusing-it/48334/2 "2023-07-27T06:17:20Z")

</div>

Hello, It will be a great help if you will provide your notebook

---

<div class="post-metadata">

**Author:** ![abd007](https://avatars.discourse-cdn.com/v4/letter/a/c57346/32.png) [@abd007](https://discuss.huggingface.co/u/abd007)\
**Post date:** [July 27, 2023, 6:52am UTC](https://discuss.huggingface.co/t/finetuning-llm-model-using-trainer-and-reusing-it/48334/3 "2023-07-27T06:52:54Z")

</div>

> **[Google Colaboratory](https://colab.research.google.com/drive/1dvxOpH7CLZl1CYv5zWQVAzjs8f2h1yfp?usp=sharing)**

---

<div class="post-metadata">

**Author:** ![shezanbaig](https://avatars.discourse-cdn.com/v4/letter/s/d07c76/32.png) [@shezanbaig](https://discuss.huggingface.co/u/shezanbaig)\
**Post date:** [August 17, 2023, 12:30pm UTC](https://discuss.huggingface.co/t/finetuning-llm-model-using-trainer-and-reusing-it/48334/4 "2023-08-17T12:30:45Z")

</div>

Can someone pls help here? I am getting the same dilema.