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

    Neuronpedia

    INDEX
    Explanations

    options and advice

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     geometries
    0.46
     protected
    0.41
     chảy
    0.40
     Protected
    0.40
     যখন
    0.40
    だったので
    0.39
     geometry
    0.39
     History
    0.38
     cancerous
    0.38
    SourceObject
    0.38
    POSITIVE LOGITS
     you
    0.56
     youll
    0.55
     તમારે
    0.55
    you
    0.54
    leyebilirsiniz
    0.54
    无疑
    0.54
     valam
    0.50
     вам
    0.49
     olacaktır
    0.48
    かもしれません
    0.48
    Act Density 0.080%

    No Known Activations