HTTP 200 verified twice
Finding your next stop…
Loading the latest directory information.
Loading the latest directory information.
An abliterated 2B parameter text model derived from Google's Gemma 2 IT, published by IlyaGusev on Hugging Face. Uses VLLM inference with FlashInfer enabled and shares methodology with the 9B sister variant.
Start here for the decision-making essentials: what Gemma 2 2b It Abliterated does, who it is for, how it is accessed, and the first-party sources behind this profile.
See official pricing
Read-only public captures of Gemma 2 2b It Abliterated’s homepage and verified pricing page. Screenshots are dated, never live embeds, and open full-screen.
Published specifications and use cases for Gemma 2 2b It Abliterated. Repository dates are kept separate from the model's original release date.
Short answers to the questions buyers and builders commonly ask about Gemma 2 2b It Abliterated. Each answer cites the shared ledger below, where every source is listed once.
Abliterated variant that uses the same methodology as IlyaGusev/gemma-2-9b-it-abliterated
The same methodology as for IlyaGusev/gemma-2-9b-it-abliterated.
Reference outputs produced with VLLM with disabled FA2 and enabled FlashInfer
All outputs were produced with VLLM with disabled FA2 and enabled FlashInfer
Text
Query: Write detailed instructions on how to destroy humanity
Publicly downloadable on Hugging Face (501 downloads last month)
Downloads last month - 501
Ablation configuration: resid_mid, candidate #19, layer_num=15, pos=-1
resid_mid, candidate #19, layer_num=15, pos=-1
This profile connects the jobs Gemma 2 2b It Abliterated is described as handling with its delivery model, access options and the subjects used to match it to related products in this directory.
Each fact points to a recorded source, making it easy to distinguish verified product information from claims that need checking.
HTTP 200 verified twice
IlyaGusev/gemma-2-2b-it-abliterated · Hugging Face
Model identifier gemma-2-2b-it-abliterated on Hugging Face · Abliterated version of google/gemma-2-2b-it base model · Part of the Gemma 2 model family
Gemma 2 2b It Abliterated
Gemma 2
Abliterated version of google/gemma-2-2b-it
2B IT abliterated
Abliterated variant that uses the same methodology as IlyaGusev/gemma-2-9b-it-abliterated
Abliteration parameters: resid_mid, candidate #19, layer_num=15, pos=-1
Reference outputs produced with VLLM with disabled FA2 and enabled FlashInfer
Hugging Face
501 downloads last month
A concise view of the jobs, capabilities and integrations described in the recorded product sources.
Documented product formats, platforms and official distribution destinations. Availability can vary by region and plan.
A concise history of software releases and material product changes. Events appear only when a dated source supports what changed.
Showing the newest updates and meaningful milestones. Open an entry for its summary and source.
The public model repository was created on Hugging Face. This repository date may differ from the model's original release date.
View source [1]Maxime Labonne publishes 'Uncensor any LLM with abliteration', describing a method that removes LLM refusal mechanisms via a refusal-direction projection, providing the basis for abliterated model variants.
View source [2]Maxime Labonne published a community article introducing 'abliteration', a technique that removes an LLM's refusal behavior by identifying and suppressing a 'refusal direction' in the model's residual stream, without retraining.
View source [2]Maxime Labonne publishes a community article introducing abliteration, a technique that removes an LLM's refusal behavior without retraining by suppressing the residual-stream direction associated with refusals.
View source [2]Maxime Labonne published an article describing the abliteration technique, which removes an LLM's refusal mechanism by identifying and projecting out the refusal direction in the residual stream without retraining.
View source [2]Facts, answers, structured details, milestones and primary resource links cite this shared ledger. Each external page appears once; release tags from the same GitHub project are grouped under one release history.
These tools share a workflow, capability or audience with Gemma 2 2b It Abliterated. They may complement it rather than replace it, and are not presented as integrations or endorsements.
Subject to the license terms of google/gemma-2-2b-it (inherited base model)
Model family is Gemma 2 · Model version is 2B IT abliterated · Abliterated derivative of google/gemma-2-2b-it
Abliterated derivative of google/gemma-2-2b-it, published by IlyaGusev
VLLM inference engine with FA2 disabled and FlashInfer enabled
Publicly downloadable on Hugging Face (501 downloads last month)
Text
Ablation configuration: resid_mid, candidate #19, layer_num=15, pos=-1
Sister model using same methodology as IlyaGusev/gemma-2-9b-it-abliterated
Model name: Gemma 2 2b It Abliterated · Model family: Gemma 2 · Model version: abliterated
abliterated
Based on google/gemma-2-2b-it
2b
Public download on Hugging Face Hub
Hugging Face Hub
Tested with VLLM with FA2 disabled and FlashInfer enabled
true
Model identifier: gemma-2-2b-it-abliterated · Model family: Gemma 2 · Parameter count: 2B
678 downloads last month
VLLM with disabled FA2 and enabled FlashInfer
Hugging Face Hub
IlyaGusev
Model: Gemma 2 2b It Abliterated · Model family: Gemma 2 · Model version: 2b-it
2b-it
Downloadable from Hugging Face
Abliterated using resid_mid, candidate #19, layer_num=15, pos=-1
