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

    Neuronpedia

    INDEX
    Explanations

    symbols, numbers, punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    pans
    0.67
     calculation
    0.65
     Epson
    0.63
    pli
    0.61
    PSG
    0.61
     computation
    0.61
     palladium
    0.61
    pson
    0.61
     Pd
    0.61
    Pd
    0.60
    POSITIVE LOGITS
     الج
    0.62
    ட்டும்
    0.60
    noj
    0.59
    滀
    0.58
    জ
    0.57
     tarde
    0.55
     nö
    0.55
    Ꮖ
    0.55
    ট
    0.55
    LOC
    0.55
    Act Density 0.156%

    No Known Activations