metadata

language:
  - en
tags:
  - abliterated
  - uncensored
base_model:
  - cognitivecomputations/Dolphin3.0-R1-Mistral-24B
pipeline_tag: text-generation
library_name: transformers

huihui-ai/Dolphin3.0-R1-Mistral-24B-abliterated

This is an uncensored version of cognitivecomputations/Dolphin3.0-R1-Mistral-24B created with abliteration (see remove-refusals-with-transformers to know more about it).
This is a crude, proof-of-concept implementation to remove refusals from an LLM model without using TransformerLens.

Note

The original model has strong thinking abilities, but it seems to have weaker context-based conversational skills?
If the desired content does not appear, Please clear the context first.
By comparison, the process of <think> may not appear in either the original model or the ablated model.

Use with ollama

You can use huihui_ai/dolphin3-r1-abliterated directly

ollama run huihui_ai/dolphin3-r1-abliterated

Donation

Your donation helps us continue our further development and improvement, a cup of coffee can do it.

bitcoin:

  bc1qqnkhuchxw0zqjh2ku3lu4hq45hc6gy84uk70ge