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

    Neuronpedia

    INDEX
    Explanations

    strategies and code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    色
    0.34
    引き続き
    0.32
     cameo
    0.32
     Yav
    0.32
    ؍
    0.32
    !।
    0.32
     পাদ
    0.31
    essex
    0.31
    хь
    0.31
    がん
    0.31
    POSITIVE LOGITS
    चा
    0.42
     mesmerizing
    0.37
     resolvers
    0.34
     crafting
    0.33
    ्रम
    0.32
     तिलावत
    0.32
    मग
    0.32
    UID
    0.31
    Gram
    0.31
     Gram
    0.31
    Act Density 0.005%

    No Known Activations