© 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. 90135
    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
     Politics
    -0.27
     suddenly
    -0.26
    æľĢåŁºæľ¬
    -0.24
    çªģçĦ¶
    -0.24
    ogi
    -0.24
    诡
    -0.23
    รà¸Ńà¸ĩ
    -0.23
    isoner
    -0.23
     hl
    -0.23
    æ°ijå¿ĥ
    -0.23
    POSITIVE LOGITS
     fill
    0.27
    abela
    0.26
    ãģĤãĤĬãģ¾ãģĻ
    0.25
     shots
    0.25
    pite
    0.25
     dub
    0.24
    Skeleton
    0.24
    .Qt
    0.24
    (da
    0.24
     trần
    0.24
    Activations Density 9.266%

    No Known Activations

    This feature has no known activations.