© 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. 19608
    Prev
    Next
    INDEX
    Explanations

    Locations and place names

    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
    ä¸ĩ
    -0.29
    vore
    -0.28
    å¹ħ
    -0.28
    åİ
    -0.27
    ovid
    -0.25
    fore
    -0.25
    pañ
    -0.25
    azz
    -0.25
    org
    -0.24
    HAVE
    -0.24
    POSITIVE LOGITS
    eng
    0.31
    iller
    0.28
    ä¸¤å¼ł
    0.27
    ä¼łç»ŁçļĦ
    0.27
    heel
    0.27
    ENTE
    0.26
    (Me
    0.26
    rica
    0.26
    unan
    0.25
    robat
    0.25
    Activations Density 0.004%

    No Known Activations