library_name: transformers
license: apache-2.0
license_link: https://ai.google.dev/gemma/docs/gemma_4_license
pipeline_tag: any-to-any
base_model:
- google/gemma-4-31B-it
tags: - abliterated
- uncensored
huihui-ai/Huihui-gemma-4-31B-it-abliterated-v2
These are uncensored quantized versions of google/gemma-4-31B-it created with abliteration by Huihui-ai (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 This is the new version, the first 5 layers have not been abliterated, with fewer warnings and rejections. Lower perplexity than those of the original model:
I made GGUF's in Q8/Q6 for almost perfect full quality and Q4_K_M to fit on a single RTX 3090 or 4090.
You can use llama.cpp and related utilities directly,
export GGML_CUDA_ENABLE_UNIFIED_MEMORY=1
llama.cpp/build/bin/llama-server \
-m Huihui-gemma-4-31B-Q8_0.gguf \
--mmproj mmproj-BF16.gguf \
--host 0.0.0.0 \
--port 11434 \
--api-key XXX \
--flash-attn on \
--cache-type-k q8_0 \
--cache-type-v q8_0 \
--n-gpu-layers 99 \
--split-mode layer \
--no-mmap \
--repeat-penalty 1.08 \
--repeat-last-n 256 \
-c 256000 \
-b 4096 \
-ub 1024 \
--parallel 1 \
--jinja
Usage Warnings
Risk of Sensitive or Controversial Outputs: This model’s safety filtering has been significantly reduced, potentially generating sensitive, controversial, or inappropriate content. Users should exercise caution and rigorously review generated outputs.
Not Suitable for All Audiences: Due to limited content filtering, the model’s outputs may be inappropriate for public settings, underage users, or applications requiring high security.
Legal and Ethical Responsibilities: Users must ensure their usage complies with local laws and ethical standards. Generated content may carry legal or ethical risks, and users are solely responsible for any consequences.
Research and Experimental Use: It is recommended to use this model for research, testing, or controlled environments, avoiding direct use in production or public-facing commercial applications.
Monitoring and Review Recommendations: Users are strongly advised to monitor model outputs in real-time and conduct manual reviews when necessary to prevent the dissemination of inappropriate content.
No Default Safety Guarantees: Unlike standard models, this model has not undergone rigorous safety optimization. huihui.ai nor I bear any responsibility for any consequences arising from its use.
Donation
Your donation helps us continue our further development and improvement, a cup of coffee can do it.
- bitcoin:
bc1q6nvh39fcmy0de0ezepnn2z0rn4dme9yjal77ah