© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 31-TRANSCODER-HP
    7. 19600
    Prev
    Next
    INDEX
    Explanations

    :

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    xDA
    -0.29
    åĵĦ
    -0.27
     há
    -0.27
    ä¼ļ让
    -0.26
    ìĹij
    -0.26
    çľ¨çľ¼
    -0.26
    onda
    -0.25
    pyx
    -0.25
     sme
    -0.25
    _Red
    -0.24
    POSITIVE LOGITS
    ech
    0.28
    Mahon
    0.27
    æĺ¯éĿŀ
    0.27
    iali
    0.26
     non
    0.25
     grid
    0.25
     ved
    0.25
    non
    0.25
    annot
    0.25
     ferm
    0.25
    Activations Density 1.595%

    No Known Activations