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

    Neuronpedia

    INDEX
    Explanations

    remember me or things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Graham
    -0.94
     ik
    -0.92
    hæ
    -0.91
    ]
    -0.91
    \
    -0.88
    тті
    -0.87
    ıllı
    -0.86
     копия
    -0.86
    klare
    -0.86
    ownerId
    -0.85
    POSITIVE LOGITS
     fondly
    2.17
     vividly
    1.35
     what
    1.31
     how
    1.23
     and
    1.20
    ڱ
    1.04
    vivid
    0.99
     forgot
    0.99
    インパクト
    0.98
     favourably
    0.97
    Act Density 0.021%

    No Known Activations