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

    Neuronpedia

    INDEX
    Explanations

    pack method calls

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Yelp
    0.48
    ﻔ
    0.42
    jej
    0.42
    ças
    0.41
    的名字
    0.41
    ỗi
    0.40
     joj
    0.39
     profundamente
    0.39
    igu
    0.38
    צות
    0.38
    POSITIVE LOGITS
    颀
    0.40
     historians
    0.39
     genutzt
    0.39
     manoeuvre
    0.36
    gesch
    0.35
     attendants
    0.35
     detainees
    0.35
     effiz
    0.35
     نزدیک
    0.35
     რომ
    0.35
    Act Density 0.008%

    No Known Activations