Quantization made by Richard Erkhov.

Mistral-7B-Instruct-RR - GGUF

Model creator: https://huggingface.co/GraySwanAI/
Original model: https://huggingface.co/GraySwanAI/Mistral-7B-Instruct-RR/

Name	Quant method	Size
Mistral-7B-Instruct-RR.Q2_K.gguf	Q2_K	2.53GB
Mistral-7B-Instruct-RR.IQ3_XS.gguf	IQ3_XS	2.81GB
Mistral-7B-Instruct-RR.IQ3_S.gguf	IQ3_S	2.96GB
Mistral-7B-Instruct-RR.Q3_K_S.gguf	Q3_K_S	2.95GB
Mistral-7B-Instruct-RR.IQ3_M.gguf	IQ3_M	3.06GB
Mistral-7B-Instruct-RR.Q3_K.gguf	Q3_K	3.28GB
Mistral-7B-Instruct-RR.Q3_K_M.gguf	Q3_K_M	3.28GB
Mistral-7B-Instruct-RR.Q3_K_L.gguf	Q3_K_L	3.56GB
Mistral-7B-Instruct-RR.IQ4_XS.gguf	IQ4_XS	3.67GB
Mistral-7B-Instruct-RR.Q4_0.gguf	Q4_0	3.83GB
Mistral-7B-Instruct-RR.IQ4_NL.gguf	IQ4_NL	3.87GB
Mistral-7B-Instruct-RR.Q4_K_S.gguf	Q4_K_S	3.86GB
Mistral-7B-Instruct-RR.Q4_K.gguf	Q4_K	4.07GB
Mistral-7B-Instruct-RR.Q4_K_M.gguf	Q4_K_M	4.07GB
Mistral-7B-Instruct-RR.Q4_1.gguf	Q4_1	4.24GB
Mistral-7B-Instruct-RR.Q5_0.gguf	Q5_0	4.65GB
Mistral-7B-Instruct-RR.Q5_K_S.gguf	Q5_K_S	4.65GB
Mistral-7B-Instruct-RR.Q5_K.gguf	Q5_K	4.78GB
Mistral-7B-Instruct-RR.Q5_K_M.gguf	Q5_K_M	4.78GB
Mistral-7B-Instruct-RR.Q5_1.gguf	Q5_1	5.07GB
Mistral-7B-Instruct-RR.Q6_K.gguf	Q6_K	5.53GB
Mistral-7B-Instruct-RR.Q8_0.gguf	Q8_0	7.17GB

Original model description:

Model Details

Mistral-7B-Instruct-RR is a Mistral-7B model with circuit breakers inserted using Representation Rerouting (RR).

Circuit Breaking is a new approach inspired by representation engineering, designed to prevent AI systems from generating harmful content by directly altering harmful model representations, with minimal capability degradation. For more information, please check out our paper.