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

    Neuronpedia

    INDEX
    Explanations

    how to use or how it works

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     soort
    0.90
     amount
    0.84
    amount
    0.76
     kind
    0.76
    !”
    0.74
    ताच
    0.74
     kinds
    0.73
     !”
    0.73
     happening
    0.71
     نوع
    0.69
    POSITIVE LOGITS
     did
    0.83
     does
    0.80
    ceso
    0.78
    тэй
    0.75
    的事
    0.74
    ulfate
    0.71
     funcionan
    0.71
     funciona
    0.71
     old
    0.71
     kasar
    0.71
    Act Density 0.176%

    No Known Activations