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

    Neuronpedia

    INDEX
    Explanations

    that [noun]

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     kinds
    1.54
     kind
    1.22
     types
    1.21
     sorts
    1.16
     kinda
    1.15
     caveats
    1.05
     concepts
    1.04
     are
    0.97
     distinctions
    0.95
    kinds
    0.95
    POSITIVE LOGITS
     чему
    0.80
    Hö
    0.79
    their
    0.74
    applicable
    0.70
     друз
    0.69
    recated
    0.69
    Required
    0.69
     calitate
    0.68
     fois
    0.68
     celui
    0.68
    Act Density 0.097%

    No Known Activations