© 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. 64757
    Prev
    Next
    INDEX
    Explanations

    et

    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
    åŃĺåľ¨çļĦ
    -0.30
    elow
    -0.28
    —we
    -0.28
    ä¸ŃåŃĺåľ¨çļĦ
    -0.28
     existing
    -0.27
    peria
    -0.26
    existing
    -0.26
    åŃĺåľ¨
    -0.26
    andin
    -0.25
    åŃĺåľ¨çĿĢ
    -0.25
    POSITIVE LOGITS
     Folk
    0.28
    oque
    0.27
    @Module
    0.25
    lat
    0.25
    -oper
    0.24
    åĨµ
    0.24
    weet
    0.24
    ék
    0.24
    izzy
    0.24
    èĿĻ
    0.24
    Activations Density 1.473%

    No Known Activations