© 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. 90345
    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
    身份
    -0.30
    XT
    -0.28
     Xml
    -0.28
    (Xml
    -0.26
    åŃļ
    -0.24
    -Bar
    -0.24
    Xd
    -0.24
    /xml
    -0.24
    ophil
    -0.24
    çļĦ身份
    -0.23
    POSITIVE LOGITS
    æĭ©
    0.30
    zent
    0.25
     tend
    0.24
     clipping
    0.24
    dong
    0.24
    .Embed
    0.23
     //{↵
    0.23
    ä¹ĭæĪĺ
    0.23
    rising
    0.23
    discard
    0.23
    Activations Density 0.109%

    No Known Activations

    This feature has no known activations.