Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Qwen3-4B
    3. 30-TRANSCODER-HP
    4. 15193
    Prev
    Next
    INDEX
    Explanations

    a

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     squid
    -0.28
    ÑĮе
    -0.27
    proxy
    -0.25
    æĻļ
    -0.25
    êu
    -0.25
    OfString
    -0.24
    Made
    -0.24
    egg
    -0.24
    çļĦåĽłç´ł
    -0.23
    yat
    -0.23
    POSITIVE LOGITS
    asers
    0.27
    ldr
    0.26
    åĩºè½¨
    0.26
    ÑģÑĤи
    0.24
    Unsafe
    0.23
    Fund
    0.23
    cki
    0.23
    ãİĿ
    0.23
     approved
    0.23
     Stud
    0.23
    Activations Density 0.919%

    No Known Activations