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

    Neuronpedia

    INDEX
    Explanations

    multilingual conjunctions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Prior
    0.66
     Stats
    0.57
     See
    0.56
     Logic
    0.54
    ผม
    0.54
     Mental
    0.54
     Software
    0.54
     Globally
    0.53
     Admin
    0.53
     Political
    0.52
    POSITIVE LOGITS
     που
    1.09
     και
    0.97
     ਅਤੇ
    0.97
     અને
    0.97
     மற்றும்
    0.96
     и
    0.91
     ಮತ್ತು
    0.91
     которые
    0.90
     וה
    0.89
     और
    0.89
    Act Density 0.584%

    No Known Activations