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. 15082
    Prev
    Next
    INDEX
    Explanations

    Mixed and random text

    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
    éķĴ
    -0.28
    kle
    -0.28
    itin
    -0.25
     Lamar
    -0.24
    bout
    -0.24
    ordion
    -0.24
     tint
    -0.24
    lider
    -0.24
    .mj
    -0.23
    ãģŀ
    -0.23
    POSITIVE LOGITS
     wave
    0.28
    pc
    0.26
    æĿ¥çļĦ
    0.26
    çIJĨä¼ļ
    0.25
     pair
    0.25
    åĸĩåıŃ
    0.25
    olv
    0.25
    åΤæĸŃ
    0.24
     AC
    0.24
     speaker
    0.24
    Activations Density 0.001%

    No Known Activations