Juju-sxkxi/Meta-Llama-3-70B-Instruct-abliterated-v3.5
TEXT GENERATIONConcurrency Cost:4Model Size:70BQuant:FP8Ctx Length:8kPublished:Mar 30, 2026License:llama3Architecture:Transformer Cold

Juju-sxkxi/Meta-Llama-3-70B-Instruct-abliterated-v3.5 is a 70 billion parameter instruction-tuned causal language model based on Meta-Llama-3-70B-Instruct. Developed by Juju-sxkxi, this model features orthogonalized bfloat16 safetensor weights, specifically manipulated to inhibit refusal behaviors. It maintains the original model's knowledge and training while surgically removing the tendency to generate moralizing disclaimers, making it suitable for applications requiring direct responses without ethical lecturing.

Loading preview...