© 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. 131117
    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
    Negative Logits
    åĬ¨æĢģ
    -0.29
     Bac
    -0.26
    è·Łéļı
    -0.25
    éĴ»çŁ³
    -0.25
    ä¸ĵåĪ©
    -0.25
    Reverse
    -0.24
    ipient
    -0.24
    讦
    -0.24
    undo
    -0.24
    è·Łä¸įä¸Ĭ
    -0.24
    POSITIVE LOGITS
    天
    0.27
    entre
    0.26
     acting
    0.25
     elev
    0.24
    Å¡e
    0.24
    ä¿®
    0.23
    家
    0.23
    tre
    0.23
    俱
    0.23
     era
    0.23
    Activations Density 37.449%

    No Known Activations

    This feature has no known activations.