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

    Neuronpedia

    INDEX
    Explanations

    performance and efficiency issues

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    āya
    0.50
    ective
    0.44
    лів
    0.44
    absence
    0.42
    epin
    0.42
    🟠
    0.41
    ਜ਼
    0.41
    unce
    0.40
     काढ
    0.40
    ূট
    0.40
    POSITIVE LOGITS
    م
    0.56
    ِ
    0.48
    \!\
    0.48
     сохрани
    0.48
     coisa
    0.46
     Unreal
    0.46
    िक
    0.45
     bandwidth
    0.45
     calent
    0.45
     dakika
    0.45
    Act Density 0.723%

    No Known Activations