© 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. 25-TRANSCODER-HP
    7. 108113
    Prev
    Next
    INDEX
    Explanations

    file paths and code

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.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¨Ģè¡Į
    -0.30
    ulaire
    -0.28
    ç§ij
    -0.26
    Xã
    -0.26
    á»Ļc
    -0.25
    åĨħå¤ĸ
    -0.25
    åıĮåıĮ
    -0.24
    £
    -0.24
    ousse
    -0.24
     fully
    -0.24
    POSITIVE LOGITS
    .fil
    0.26
    Planet
    0.24
     Prescott
    0.24
    PIC
    0.24
     Shelter
    0.24
     `[
    0.23
    riere
    0.23
    .isdigit
    0.23
    _FEED
    0.23
    Foto
    0.23
    Activations Density 0.929%

    No Known Activations