© 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. 129971
    Prev
    Next
    INDEX
    Explanations

    News articles

    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
    è¿Ļ个èģĮä¸ļ
    -0.27
    è°¢è°¢ä½ł
    -0.26
    kart
    -0.26
    yme
    -0.26
     commute
    -0.25
    ä¸įåĩºæĿ¥
    -0.25
    åģĥ
    -0.25
    EMS
    -0.24
     Martial
    -0.24
    åIJĪçIJĨæĢ§
    -0.24
    POSITIVE LOGITS
    ÃŁen
    0.29
    rex
    0.28
    åľº
    0.27
    åªĴ
    0.26
    /env
    0.26
    atories
    0.26
    rez
    0.26
     seeding
    0.26
    (prompt
    0.25
    UCH
    0.24
    Activations Density 1.902%

    No Known Activations