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

    Neuronpedia

    INDEX
    Explanations

    to extremes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     آہستہ
    0.79
     Preferred
    0.78
    گرم
    0.76
     আস্তে
    0.75
     आसानी
    0.74
     als
    0.73
     Chocolate
    0.72
     ACM
    0.72
     راحت
    0.70
     Favorite
    0.69
    POSITIVE LOGITS
     extreme
    1.36
     extremes
    1.35
    extreme
    1.19
     extent
    1.15
     death
    1.15
     exclusion
    1.12
    极致
    1.12
    Extreme
    1.08
    death
    1.07
     extrême
    1.06
    Act Density 0.083%

    No Known Activations