© 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. 108811
    Prev
    Next
    INDEX
    Explanations

    Code, software, and journal excerpts

    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
    产éĩı
    -0.30
    ters
    -0.30
    ices
    -0.27
    éĿ¢åIJij
    -0.26
    个交æĺĵæĹ¥
    -0.25
    ogg
    -0.25
    -loop
    -0.25
    [:
    -0.25
    ocom
    -0.25
    /LICENSE
    -0.24
    POSITIVE LOGITS
    åľ¨è¿Ļç§į
    0.28
     nameof
    0.28
    ionate
    0.26
     descendant
    0.24
    antha
    0.24
    åŀł
    0.24
    zÄħd
    0.24
     Absolute
    0.23
    çªľ
    0.23
    æł¹æºIJ
    0.23
    Activations Density 0.885%

    No Known Activations