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

    Neuronpedia

    INDEX
    Explanations

    sentence break transitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    =
    0.42
     only
    0.41
     they
    0.40
    They
    0.39
     likely
    0.37
    Only
    0.37
     They
    0.37
    把
    0.36
     because
    0.36
     maka
    0.36
    POSITIVE LOGITS
     beispielsweise
    0.54
     Nevertheless
    0.48
    比如说
    0.47
     Anyways
    0.46
     Anyway
    0.46
     misalnya
    0.45
    Nonetheless
    0.45
     например
    0.44
    Anyway
    0.43
    比如說
    0.43
    Act Density 0.000%

    No Known Activations