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

    Neuronpedia

    INDEX
    Explanations

    variable or attribute assignment

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     страниц
    0.42
     இவர்
    0.41
    aspetto
    0.41
    adhesive
    0.39
     этим
    0.39
     شماره
    0.39
    comers
    0.39
     آورد
    0.38
    친
    0.38
    に伴
    0.38
    POSITIVE LOGITS
    ="
    0.46
    ='
    0.46
    =
    0.41
    ↵
    0.40
     devour
    0.39
     bills
    0.38
     =
    0.38
     uang
    0.38
     constable
    0.38
     म
    0.38
    Act Density 0.011%

    No Known Activations