Wizard-Vicuna-7B-Uncensored HF

This is a float16 HF repo of Eric Hartford's 'uncensored' training of Wizard-Vicuna 7B.

It is the result of converting Eric's float32 repo to float16 for easier storage.

Repositories available

• 4-bit GPTQ models for GPU inference.
• 4-bit, 5-bit and 8-bit GGML models for CPU (+CUDA) inference.
• float16 HF format model for GPU inference and further conversions.

Oh sorry, do you mean like an original Wizard Vicuna 7B, not Eric's uncensored version? If so then no, Wizard Vicuna was only done in 13B. It was Eric who extended it to 7B using his uncensored/filtered dataset

Ah, that makes sense. Yes, I meant original wizard vicuna, not uncensored version.