DavidAU/gemma-3-1b-it-heretic-abliterated-uncensored

DavidAU/gemma-3-1b-it-heretic-abliterated-uncensored is a 1 billion parameter Gemma-based instruction-tuned language model, modified by Heretic v1.0.1 to significantly reduce refusal rates while maintaining model integrity. With a 32k context length, this model achieves a refusal rate of 17/100 and a KL divergence of 0.09, making it suitable for applications requiring less censored content generation without significant performance degradation.