© 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. 20-TRANSCODER-HP
    7. 142430
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
     Associ
    -0.31
    é»ijèī²
    -0.28
    o
    -0.27
     associ
    -0.26
    Associ
    -0.25
    ç»Ŀç¼ĺ
    -0.25
     short
    -0.24
    Young
    -0.24
    wise
    -0.24
     Schwartz
    -0.24
    POSITIVE LOGITS
    RAP
    0.27
    æŀķ头
    0.27
     Minimum
    0.26
    asio
    0.26
    ipur
    0.26
    æĭ¿åĩº
    0.25
    arta
    0.25
    ä¸įå¼Ģ
    0.25
    uur
    0.25
     OSI
    0.24
    Activations Density 78.589%

    No Known Activations

    This feature has no known activations.