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

    Neuronpedia

    INDEX
    Explanations

    explaining how or why

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Whatever
    0.14
    io
    0.13
     designate
    0.13
     Wherever
    0.13
     स्वीकृत
    0.13
    јединачна
    0.12
     trusty
    0.12
    erà
    0.12
    rdquo
    0.12
     lengthen
    0.12
    POSITIVE LOGITS
     how
    0.44
     bahwa
    0.38
     why
    0.38
     cómo
    0.35
    why
    0.33
     bahawa
    0.33
     adanya
    0.32
     bagaimana
    0.32
     hvordan
    0.32
     rằng
    0.31
    Act Density 0.242%

    No Known Activations