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

    Neuronpedia

    INDEX
    Explanations

    time commitment and availability

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Finance
    0.39
     Comes
    0.39
     Sometimes
    0.39
    bians
    0.39
    finance
    0.38
     rewards
    0.38
    Muk
    0.37
     Zell
    0.37
     sometimes
    0.37
     monks
    0.36
    POSITIVE LOGITS
     одном
    0.47
    夸
    0.43
     exaggerated
    0.41
    achos
    0.41
     comenzamos
    0.41
    /**/*
    0.40
    尽快
    0.40
     کنس
    0.40
     ಪ್ರಕರಣ
    0.40
     raggiunto
    0.40
    Act Density 0.001%

    No Known Activations