© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    5. Qwen2.5-7B-IT
    qwen2.5-7b-it
    Alibaba
    Jump to Source/SAE
    Jump to Feature
    INDEX

    Releases

    Qwen2.5 7B IT Matryoshka SAE
    March 2026
    David Chanin
    qwen2.5-7b-it-matryoshka
    Finding Misaligned Persona Features in Open-Weight Models
    September 2025
    Andy Arditi
    misaligned-persona

    Attention Visualizer

    HeadVis (Luger, Kamath et al.)
    Find Head By Metric
    Metric & Number of Heads
    Top 8
    Click head to select
    Layer 2
    Layer 5
    Layer 14
    Layer 22
    Layer 23
    Select Head Manually
    Layer
    Head Index

    Activation Distribution & Metrics

    Q-K Distance Distribution

    Top Query Tokens

    Top Key Tokens

    Top Activating Sequences

    Search Explanations

    Browse

    Features in QWEN2.5-7B-IT@20-matryoshka-65k
    1. Hover over a feature on the left to preview its details.
    2. Click a feature to lock it and interact with it.