© 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. 64618
    Prev
    Next
    INDEX
    Explanations

    digits

    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
    chers
    -0.29
    ilers
    -0.28
    otas
    -0.27
     Guarantee
    -0.27
    pch
    -0.26
    æģ°å½ĵ
    -0.26
    rium
    -0.26
    eper
    -0.26
    ols
    -0.26
    rates
    -0.25
    POSITIVE LOGITS
    æľĢåIJİä¸Ģ个
    0.27
     loose
    0.27
     lÆ°ng
    0.26
    æľĢåIJİä¸Ģ次
    0.25
     ÙħجÙĦ
    0.25
    ->$
    0.25
     VOID
    0.25
    ढ
    0.24
    éŀ
    0.24
    ä¸įåĩºæĿ¥
    0.23
    Activations Density 2.567%

    No Known Activations