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

    Neuronpedia

    INDEX
    Explanations

    string startswith or delete

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     "-";\n
    -0.12
     '-';\n
    -0.12
    ('.');\n
    -0.12
    ('_',
    -0.11
     "[%
    -0.10
     '/');\n
    -0.10
    ('/');\n
    -0.09
     proverb
    -0.09
     GI
    -0.08
    inx
    -0.08
    POSITIVE LOGITS
    ("
    0.12
     "
    0.11
     `"
    0.11
    _("
    0.11
    azzi
    0.10
     str
    0.10
     "-
    0.09
    оÑĤÑĮ
    0.09
    anda
    0.09
    azo
    0.09
    Act Density 0.065%

    No Known Activations