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

    Neuronpedia

    INDEX
    Explanations

    'd would

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ua
    0.98
    u
    0.95
    흙
    0.89
     вашей
    0.88
    م
    0.86
    quele
    0.86
     globo
    0.85
    ao
    0.84
     недоста
    0.84
    能
    0.83
    POSITIVE LOGITS
     arrange
    1.04
     potuto
    1.01
     proclaim
    0.99
     llegado
    0.98
     threaten
    0.96
     describe
    0.95
    ногда
    0.94
     reorganize
    0.93
     spesso
    0.93
     linger
    0.93
    Act Density 0.013%

    No Known Activations