© 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. 22-TRANSCODER-HP
    7. 97480
    Prev
    Next
    INDEX
    Explanations

    Code/Math/Formulas

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     *</
    -0.25
    Âł↵
    -0.22
    /Add
    -0.21
    æĸĩåĮĸçļĦ
    -0.21
    arme
    -0.21
    gran
    -0.20
     ?>>↵
    -0.20
    ä¸ĭéĿ¢æĺ¯å°ı
    -0.20
    ssize
    -0.20
    eping
    -0.20
    POSITIVE LOGITS
    letic
    0.25
    0.23
    but
    0.22
    [
    0.22
     keeping
    0.22
     but
    0.22
    æ²
    0.21
    ${
    0.21
    âĢ¢
    0.21
    ><!--
    0.20
    Activations Density 0.023%

    No Known Activations