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

    Neuronpedia

    INDEX
    Explanations

    descriptive adjectives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     intang
    0.67
     fleeting
    0.64
     Produits
    0.63
     Produkte
    0.61
     dwindling
    0.60
     soreness
    0.60
     Atmosphäre
    0.60
    制度
    0.59
     wrongdoing
    0.58
     criminality
    0.58
    POSITIVE LOGITS
     triangular
    0.56
     (!)
    0.53
     bilingual
    0.53
    Ash
    0.51
    ,
    0.50
     (?)
    0.49
    ASH
    0.47
    Arabic
    0.46
    (!)
    0.46
     Ash
    0.46
    Act Density 0.228%

    No Known Activations