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

    Neuronpedia

    INDEX
    Explanations

    following punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    (
    0.55
    =
    0.54
    "
    0.50
    '
    0.49
    K
    0.48
    C
    0.47
    L
    0.47
    A
    0.47
    \
    0.46
    An
    0.45
    POSITIVE LOGITS
     thậm
    0.56
     tinkering
    0.54
     minimalism
    0.54
     tweaks
    0.53
     misinformation
    0.52
     quirky
    0.51
     quirks
    0.51
     pitfalls
    0.51
     scams
    0.50
     hardships
    0.50
    Act Density 1.350%

    No Known Activations