# Continual Training on my own checkpoint

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcontinual-training-on-my-own-checkpoint%2F94408%3E%5C
**Category:** 🤗Transformers\
**Created:** [June 27, 2024, 2:18pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcontinual-training-on-my-own-checkpoint%2F94408 "2024-06-27T14:18:12Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![taric49](https://avatars.discourse-cdn.com/v4/letter/t/c77e96/32.png) [@taric49](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Ftaric49)%5C
**Post date:** [June 27, 2024, 2:18pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcontinual-training-on-my-own-checkpoint%2F94408%2F1 "2024-06-27T14:18:12Z")

</div>

I trained llama 3 8b using unsloth libirary with QLoRA I load adapaters and merged model to huggingface hub can I train on top of that model ? if yes how ?

---

<div class="post-metadata">

**Author:** ![taric49](https://avatars.discourse-cdn.com/v4/letter/t/c77e96/32.png) [@taric49](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Ftaric49)%5C
**Post date:** [June 27, 2024, 2:22pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fcontinual-training-on-my-own-checkpoint%2F94408%2F2 "2024-06-27T14:22:40Z")

</div>

I tried “load\_apaters” function then “get\_peft\_model” again but it dosent seem right