© 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. 12-TRANSCODER-HP
    7. 22051
    Prev
    Next
    INDEX
    Explanations

    those

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    æħ¢æħ¢åľ°
    -0.30
     grate
    -0.29
    è§Ĩ线
    -0.28
    kie
    -0.26
     Fres
    -0.26
    æİ¥
    -0.26
    ãģ¡ãģ¯
    -0.26
    æħ¢æħ¢çļĦ
    -0.25
    åį³ä½¿æĺ¯
    -0.25
    profil
    -0.25
    POSITIVE LOGITS
    åij¨æľ«
    0.25
     split
    0.25
    inox
    0.24
    .Enums
    0.24
    alker
    0.24
     entirety
    0.23
     Norton
    0.23
    amon
    0.23
     restraining
    0.23
     unconstitutional
    0.23
    Activations Density 0.228%

    No Known Activations