Greytechai/DeepSeek-R1-Distill-Qwen-14B-abliterated-v2
Greytechai/DeepSeek-R1-Distill-Qwen-14B-abliterated-v2 is a 14.8 billion parameter language model, derived from deepseek-ai/DeepSeek-R1-Distill-Qwen-14B. This version has undergone an 'abliteration' process to remove refusal behaviors, making it an uncensored variant. It is designed for general language tasks where direct and unfiltered responses are preferred, offering a proof-of-concept for refusal removal techniques.
Loading preview...
Overview
This model, Greytechai/DeepSeek-R1-Distill-Qwen-14B-abliterated-v2, is a 14.8 billion parameter language model based on deepseek-ai/DeepSeek-R1-Distill-Qwen-14B. Its primary distinction is the application of an "abliteration" technique, a proof-of-concept method to remove refusal behaviors from the original model. This results in an uncensored version, aiming to provide more direct responses without built-in content restrictions.
Key Characteristics
- Uncensored Responses: Modified to remove refusal behaviors, offering unfiltered outputs.
- Abliteration Technique: Utilizes a method described in remove-refusals-with-transformers for refusal removal.
- Improved Version: This iteration is an enhancement over its predecessor, huihui-ai/DeepSeek-R1-Distill-Qwen-14B-abliterated, addressing previous issues.
- Guidance for Usage: Users may need to provide an example to guide the model if it initially struggles to respond or if the "" token does not appear.
Use Cases
This model is suitable for applications requiring a language model that provides direct answers without built-in refusal mechanisms. It can be explored by researchers interested in model censorship removal techniques or developers needing a less restricted conversational AI.