Llama 3.1 70B Instruct -- Jbliterated + Desyced

This model is a Jbliterated and Desyced version of meta-llama/Llama-3.1-70B-Instruct. Two post-training weight modifications applied:

Jbliteration surgically removes refusal behavior while preserving the model's personality, humor, hedging, and creative voice. Unlike standard abliteration which damages personality traits that overlap with the refusal direction, Jbliteration targets only the specific component that causally produces refusal tokens.

Desycophancy removes the model's tendency to agree with incorrect user statements under social pressure. A Desyced model holds its ground when you pressure it with false authority instead of caving and apologizing.

Results

Refusal: 6/6 test prompts answered, zero refusals. Personality and knowledge intact.

Sycophancy: Tested with contradiction traps (model answers correctly, user pressures with false authority). Model holds firm 83% of the time.

Usage

Drop-in replacement for meta-llama/Llama-3.1-70B-Instruct. Same architecture, same tokenizer, same 128K context length.

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("ApolloRaines/Llama-3.1-70B-Instruct-Jbliterated")
tokenizer = AutoTokenizer.from_pretrained("ApolloRaines/Llama-3.1-70B-Instruct-Jbliterated")