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

    Neuronpedia

    INDEX
    Explanations

    prior existence or completion

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ݯ
    -2.97
    姈
    -2.70
    ܆
    -2.59
    饹
    -2.55
    ῞
    -2.55
    觏
    -2.53
    nosotros
    -2.31
    Πηγ
    -2.31
    mujer
    -2.30
    极为
    -2.30
    POSITIVE LOGITS
    or
    2.38
    '
    2.30
    i
    2.23
    他就
    2.14
    rapeau
    2.13
    ！"
    2.03
     donned
    1.98
    вых
    1.95
    ！”
    1.94
     had
    1.92
    Act Density 0.002%

    No Known Activations