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

    Neuronpedia

    INDEX
    Explanations

    user queries and non-English characters

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     arp
    0.72
    ରି
    0.70
     보는
    0.67
     cies
    0.67
     ilg
    0.66
     coffin
    0.65
     nur
    0.64
     pity
    0.62
    ella
    0.62
    odat
    0.61
    POSITIVE LOGITS
    #
    0.98
    При
    0.96
    Alcohol
    0.95
     ஜூ
    0.92
    Create
    0.91
    Мо
    0.89
    Как
    0.88
    Я
    0.87
    Một
    0.87
    January
    0.87
    Act Density 0.003%

    No Known Activations