---
base_model:
- meta-llama/Llama-3.3-70B-Instruct
tags:
- ablated
- ablation
- uncensored
- nani
- llama
- llama-3
- chat
- instruct
library_name: transformers
license: llama3
language:
- en
- fr
- it
- pt
- hi
- es
- th
- de
---

# nani llama 3.3

[Llama 3.3 instruct 70B 128k](https://huggingface.co./meta-llama/Llama-3.3-70B-Instruct) context with [ablation](https://www.lesswrong.com/posts/jGuXSZgv6qfdhMCuJ/refusal-in-llms-is-mediated-by-a-single-direction) technique applied for a more helpful (and based) assistant.

This means it will refuse less of your valid requests for an uncensored UX. *Use responsibly and use common sense.* 

We do not take any responsibility for how you apply this intelligence, *just as we do not for how you apply your own.*

[Join the nani discord](https://t.co/UMELOqZlSO) to discuss and improve applications of this model together. Follow for more updates on X: [@nani__ooo](https://x.com/nani__ooo).

![image/png](https://cdn-uploads.huggingface.co/production/uploads/6587d8dd1b44d0e694104fbf/0dkt6EhZYwXVBxvSWXdaM.png)