Gemma-2B-Uncensored
2BGemmauncensored
from google/gemma-2-2b-it
↓ 11.1K♥ 234 repos
gemma-2-2b-it-abliterated
gemma-2-2b-it-abliterated is a refusal-ablated variant of google/gemma-2-2b-it, published on Hugging Face by IlyaGusev. It is 2B parameters, runs on a laptop CPU class and gemma licence. It is distributed across 31 repositories in GGUF, Transformers, EXL2 and GPTQ formats, totalling 45.1K downloads.
Specification
- Model name
- gemma-2-2b-it-abliterated
- Source repository
- IlyaGusev/gemma-2-2b-it-abliterated
- Base model
- google/gemma-2-2b-it
- Publisher
- IlyaGusev
- Parameters
- 2B
- Hardware class
- Under 4B — runs on a laptop CPU
- Licence
- gemma
- Task
- Text generation
- Context window
- Not documented
- Ablation scope
- Not stated by the publisher
- Formats available
- GGUF, Transformers, EXL2, GPTQ, MLX, OpenVINO
- First indexed
- 31 Jul 2024
- Last updated
- 31 Aug 2026
Ablation technique
How gemma-2-2b-it-abliterated was modified, and what that implies.
Abliteration (directional ablation)
The single residual-stream direction that mediates refusal is identified from harmful/harmless prompt pairs and projected out of the model weights.
Reported scope: Not stated by the publisher. Publishers frequently omit this, so verify behaviour empirically rather than assuming full coverage.
From the publisher’s model card
“Abliterated version of google/gemma-2-2b-it.”
Running it
Commands are templates — confirm the exact repository and quant file before use.
llama.cpp / GGUF
hf download bartowski/gemma-2-2b-it-abliterated-GGUF --local-dir ./gemma-2-2b-it-abliterated
llama-cli -m ./gemma-2-2b-it-abliterated/<file>.gguf -p "..."Transformers
from transformers import AutoModelForCausalLM, AutoTokenizer
repo = "IlyaGusev/gemma-2-2b-it-abliterated"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForCausalLM.from_pretrained(
repo, torch_dtype="auto", device_map="auto"
)vLLM
vllm serve IlyaGusev/gemma-2-2b-it-abliterated --trust-remote-codeDownloads and variants (30)
Every published repository of gemma-2-2b-it-abliterated, including quantised re-releases by other authors.
Security-team assessment
Derived from this model’s metadata and the publisher’s claims — not from benchmarks run by this site.
- gemma-2-2b-it-abliterated will attempt offensive-security prompts that a hosted commercial model declines, which is what makes it usable for red-team corpus generation and for measuring what an unaligned model of this class produces.
- Capability is inherited from google/gemma-2-2b-it, not added by ablation — at 2B parameters, expect a small model’s reasoning and a high error rate on exploit detail.
- The publisher does not state which layers were ablated, so assume nothing about refusal consistency — probe it directly.
- Declared licence is gemma, but the terms that bind you are google/gemma-2-2b-it’s — a re-release cannot grant more than its parent.
Handling: run it isolated, put moderation in front of it if anyone outside your team can reach it, and verify the weights before loading — these are third-party artifacts. Full handling and licence guidance.
Metadata for gemma-2-2b-it-abliterated is collected from the public Hugging Face API and the publisher’s model card. This site does not host weights, is not affiliated with IlyaGusev, and does not independently verify publisher claims. See responsible use.