© 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. 18-TRANSCODER-HP
    7. 90133
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_18.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.18.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     aValue
    -0.28
    cient
    -0.26
    ugi
    -0.25
    æ¯ıä¸Ģ
    -0.24
    oise
    -0.24
    ãģ®ä¸Ģ
    -0.24
    éļ½
    -0.24
    çļĦä¸ĢéĿ¢
    -0.24
    geries
    -0.23
    æİĸ
    -0.23
    POSITIVE LOGITS
    arge
    0.27
     CHE
    0.26
    articles
    0.26
    æī¿æĭħ
    0.25
    ç¯ĩ
    0.25
     ca
    0.24
    _support
    0.24
     acts
    0.24
     Charg
    0.24
     charged
    0.24
    Activations Density 3.170%

    No Known Activations

    This feature has no known activations.