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

    Neuronpedia

    INDEX
    Explanations

    links followed by parentheses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    .')
    0.70
    !')
    0.67
    ...')
    0.67
    ･
    0.66
     incongru
    0.65
    .~\
    0.65
     pres
    0.65
    仃
    0.64
    ល់
    0.64
    ;')
    0.64
    POSITIVE LOGITS
    #:
    0.97
     Opens
    0.95
    Этот
    0.93
    #
    0.90
    Opens
    0.88
     అనేది
    0.87
    这个
    0.85
     Этот
    0.84
    ​​
    0.81
     에서
    0.81
    Act Density 0.284%

    No Known Activations