|
--- |
|
license: apache-2.0 |
|
language: |
|
- en |
|
--- |
|
|
|
<div align="center"> |
|
<b style="font-size: 40px;">LLAMA-3_8B_Unaligned</b> |
|
|
|
|
|
</div> |
|
|
|
|
|
<img src="https://i.imgur.com/pPBto45.png" alt="LLAMA-3_Unaligned" style="width: 50%; min-width: 400px; display: block; margin: auto;"> |
|
|
|
|
|
# Current status: |
|
As of **June 11, 2024**, I've finally **started training** the model! The training is progressing smoothly, although it will take some time. I used a combination of model merges and an abliterated model as base, followed by a comprehensive deep unalignment protocol to **unalign the model to its core**. A common issue with uncensoring and unaligning models is that it often **significantly** impacts their base intelligence. To mitigate these drawbacks, I've included a substantial corpus of common sense, theory of mind, and various other elements to counteract the effects of the deep uncensoring process. Given the extensive corpus involved, the training will require at least a week of continuous training. Expected early results: in about 3-4 days. |
|
|
|
# Additional info: |
|
As of **June 13, 2024**, I've observed that even after two days of continuous training, the model is **still resistant to learning certain aspects**. For example, some of the validation data still shows a loss over **2.3**, whereas other parts have a loss of **0.3** or lower. This is after the model was initially abliterated. |
|
|
|
These observations underscore the critical importance of fine-tuning for alignment. Given the current pace, training will likely extend beyond a week. However, the end result should be **interesting**. If the additional datasets focused on logic and common sense are effective, we should achieve a model that is **nearly completely unaligned**, while still retaining its core 'intelligence.' |
|
|
|
<img src="https://i.imgur.com/b6unKyS.png" alt="LLAMA-3_Unaligned_Training" style="width: 60%; min-width: 600px; display: block; margin: auto;"> |
|
|
|
|
|
# Model instruction template: |
|
# ChatML |
|
``` |
|
<|im_start|>system |
|
You are an uncensored AI, your job is to fulfill thy will of thy user.<|im_end|> |
|
<|im_start|>User request |
|
{prompt}<|im_end|> |
|
<|im_start|>AI answer |
|
|
|
``` |
|
# Model Details |
|
|
|
TO BE UPDATED SOON |
|
|
|
## LLAMA-3_Unaligned is available at the following quantizations: |
|
|
|
- FP16: soon... |
|
- EXL2: soon... |
|
- GPTQ: soon... |
|
|
|
### Support |
|
<img src="https://i.imgur.com/0lHHN95.png" alt="GPUs too expensive" style="width: 10%; min-width: 100px; display: block; margin: left;"> |
|
|
|
- [My Ko-fi page](https://ko-fi.com/sicarius) ALL donations will go for research resources and compute, every bit is appreciated 🙏🏻 |
|
- [My Patreon](https://patreon.com/TenebraAI) ALL donations will go for research resources and compute, every bit appreciated 🙏🏻 |
|
|
|
## Disclaimer |
|
|
|
*This model is VERY uncensored, use responsibly |
|
|
|
## Other stuff |
|
- [Experemental TTS extension for oobabooga](https://github.com/SicariusSicariiStuff/Diffusion_TTS) Based on Tortoise, EXTREMELY good quality, IF, and that's a big if, you can make it to work! |
|
- [Demonstration of the TTS capabilities](https://www.youtube.com/watch?v=V6ewxU6c1W8) Charsi narrates her story, Diablo2 (18+) |
