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

    Neuronpedia

    INDEX
    Explanations

    Okay, followed by comma

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    न
    2.02
    ের
    1.88
    ни
    1.87
    ن
    1.85
    πο
    1.79
    ek
    1.66
    ार
    1.64
    டன்
    1.59
     metabolites
    1.55
    ю
    1.55
    POSITIVE LOGITS
     asalamualaikum
    1.95
    the
    1.90
     Besonders
    1.86
    erweise
    1.83
    taining
    1.79
    swing
    1.75
    tained
    1.73
    ttä
    1.71
     appunto
    1.70
     ولك
    1.68
    Act Density 0.325%

    No Known Activations