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

    Neuronpedia

    INDEX
    Explanations

    safe followed by opening parenthesis

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    +#+#
    0.43
     vVertex
    0.42
    Semit
    0.40
    arquía
    0.39
     necesita
    0.39
     plufieurs
    0.39
    による
    0.38
    При
    0.38
    wir
    0.37
    С
    0.37
    POSITIVE LOGITS
    (
    0.74
     (
    0.66
    （
    0.57
    (_
    0.53
    (~
    0.51
    (_)
    0.50
    ([
    0.50
    ((
    0.50
    (\
    0.49
     $(
    0.49
    Act Density 0.004%

    No Known Activations