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

    Neuronpedia

    INDEX
    Explanations

    highlights something relevant

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    i
    1.70
    ל
    1.67
    ه
    1.57
    ل
    1.55
    ll
    1.53
    d
    1.53
    ’
    1.52
    dni
    1.45
    t
    1.45
    id
    1.45
    POSITIVE LOGITS
     prominently
    1.61
     shining
    1.32
     shine
    1.19
     plight
    1.13
     marshmallows
    1.10
    ্কার
    1.08
     restorative
    1.07
     prominent
    1.06
     prophetic
    1.06
     prerogative
    1.06
    Act Density 0.055%

    No Known Activations