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

    Neuronpedia

    INDEX
    Explanations

    section headings like "i."

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     usages
    0.38
     cliques
    0.36
     sacc
    0.36
     bitmaps
    0.34
     operands
    0.34
     mesons
    0.33
     atoms
    0.33
     rations
    0.33
     isos
    0.32
     servants
    0.32
    POSITIVE LOGITS
    The
    0.45
    This
    0.40
    1
    0.38
    There
    0.37
    Our
    0.36
    ↵↵
    0.36
    that
    0.36
    この
    0.35
    さて
    0.35
    Before
    0.35
    Act Density 3.260%

    No Known Activations