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

    Neuronpedia

    INDEX
    Explanations

    openings, curtains, doors, communication

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ু
    1.05
    在
    0.94
    ى
    0.94
    v
    0.93
     in
    0.93
     \
    0.93
    s
    0.90
    锟
    0.90
    ل
    0.90
    0.87
    POSITIVE LOGITS
    в
    0.88
     fermeture
    0.87
    is
    0.86
     aberta
    0.85
    ють
    0.84
    ?
    0.83
    ца
    0.82
     Open
    0.82
    ier
    0.81
     ouvert
    0.80
    Act Density 0.151%

    No Known Activations