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

    Neuronpedia

    INDEX
    Explanations

    two or both

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    rifice
    1.79
    x
    1.72
    ıl
    1.66
    un
    1.65
    d
    1.63
    ação
    1.52
    v
    1.52
    ö
    1.52
    bres
    1.45
    iting
    1.43
    POSITIVE LOGITS
    thirds
    2.27
     sexes
    2.13
    sides
    1.98
     dozen
    1.95
     thirds
    1.91
     तरह
    1.90
     hemispheres
    1.76
     sides
    1.73
     halves
    1.70
     দুই
    1.65
    Act Density 0.170%

    No Known Activations