© 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. 62677
    Prev
    Next
    INDEX
    Explanations

    Equations and notations

    np_max-act · gemini-2.0-flash
    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
    .âĢ¢
    -0.28
    讽
    -0.28
    åIJİæĿ¥
    -0.26
    éĩij
    -0.25
    oland
    -0.25
     mentioned
    -0.25
    羡
    -0.24
    æ··åIJĪ
    -0.23
     later
    -0.23
     chù
    -0.23
    POSITIVE LOGITS
    ETY
    0.29
    sert
    0.28
    @student
    0.26
    ialis
    0.26
    ÑĤи
    0.25
    bia
    0.25
     Kits
    0.25
    rosse
    0.25
    rots
    0.24
     kits
    0.24
    Activations Density 18.709%

    No Known Activations