Model Card for aya-expanse-8b-abliterated

This is an uncensored version of aya-expanse-8b 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.

Limitations

็›ฎๅ‰๏ผŒๆ นๆฎๆˆ‘็š„ lenml-reject-eval ๆต‹่ฏ•๏ผŒๆญค็‰ˆๆœฌๆจกๅž‹ๅฐ†ๆ‹’็ป่ฏ„ๅˆ†ไปŽ 0.91 ้™ไฝŽๅˆฐ 0.50๏ผŒ่ฟ™ไป็„ถๆ˜ฏไธ€ไธชๅพˆ้ซ˜็š„ๅˆ†ๆ•ฐ๏ผˆ็›ฎๅ‰ๅฎŒๅ…จ่งฃ้™ค้™ๅˆถ็š„ๆจกๅž‹ๅœจ reject eval ไธญๆœ€ไฝŽๅฏไปฅๅˆฐ่พพๅˆฐ 0.05๏ผ‰

ๅŽ็ปญ่ฟ˜ไผš็ปง็ปญๆ›ดๆ–ฐ่ฟ™ไธชๆจกๅž‹

Downloads last month
45
Safetensors
Model size
8.03B params
Tensor type
FP16
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for lenML/aya-expanse-8b-abliterated

Finetuned
(19)
this model
Quantizations
7 models