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

    Neuronpedia

    INDEX
    Explanations

    quoted text

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    €¦
    0.99
    €“
    0.86
    \...
    0.82
     \...
    0.78
    »),
    0.76
    ）」
    0.75
    \]
    0.75
    ）、
    0.73
    ());
    0.71
     \|
    0.70
    POSITIVE LOGITS
     "
    5.83
     “
    5.08
    “
    4.03
     "'
    3.91
     "[
    3.75
     "(
    3.74
     "¿
    3.72
     „
    3.71
     "...
    3.63
    「
    3.59
    Act Density 5.993%

    No Known Activations