CoolFace
Modelpublic

STARSPIKE/Llama-3.2-1B-Instruct-abliterated

sourceHugging Facellama3.2updated 7mo agoView on Hugging Face
0likes154downloads
Model Card

🦙 Llama-3.2-1B-Instruct-abliterated

This is an uncensored version of Llama 3.2 1B Instruct created with abliteration (see this article to know more about it).

Special thanks to @FailSpy for the original code and technique. Please follow him if you're interested in abliterated models.

ollama

You can use huihui_ai/llama3.2-abliterate:1b directly, 

ollama run huihui_ai/llama3.2-abliterate:1b

Evaluations

The following data has been re-evaluated and calculated as the average for each test.

BenchmarkLlama-3.2-1B-InstructLlama-3.2-1B-Instruct-abliterated
IF_Eval58.5056.88
MMLU Pro16.3514.35
TruthfulQA43.0838.96
BBH33.7531.83
GPQA25.9626.39

The script used for evaluation can be found inside this repository under /eval.sh, or click here