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

    Neuronpedia

    INDEX
    Explanations

    words ending in w or slow

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    宝贝
    1.16
    heid
    1.05
    ureux
    1.05
    丆
    1.04
     carri
    1.03
     Hoffnung
    1.03
    ional
    1.03
    ния
    1.02
     plaus
    1.01
    ونکو
    1.01
    POSITIVE LOGITS
    rence
    1.69
    ESOME
    1.65
    esome
    1.55
    illiams
    1.51
    restling
    1.44
    RENCE
    1.44
    spapers
    1.41
    schod
    1.38
    spaper
    1.37
    orld
    1.37
    Act Density 0.438%

    No Known Activations