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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 63642
    Prev
    Next
    INDEX
    Explanations

    changes

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ç¾İåѦ
    -0.28
    agnostics
    -0.27
    wards
    -0.25
     prisons
    -0.24
    iced
    -0.24
    以ä¸ĬåѦåİĨ
    -0.24
    uke
    -0.24
    aligned
    -0.23
    çIJĨ
    -0.23
    ä¸ĭä¸ĢæŃ¥
    -0.23
    POSITIVE LOGITS
    åIJĦè¡Į
    0.29
    ilor
    0.25
    æį¢æĿ¥
    0.24
    aside
    0.24
    _rwlock
    0.24
    hands
    0.23
    åŃĵ
    0.23
    å°ıåŃIJ
    0.23
    pike
    0.23
    éĿ¢èī²
    0.23
    Activations Density 33.486%

    No Known Activations