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

    Neuronpedia

    INDEX
    Explanations

    thoroughness and detailed explanations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ן
    1.51
    ле
    1.13
    ний
    1.10
    ர்
    1.05
    ка
    1.03
    ING
    1.01
    ни
    0.98
    ми
    0.96
    nych
    0.96
    ко
    0.93
    POSITIVE LOGITS
    د
    1.19
    '
    1.09
     oxid
    1.03
    H
    1.00
    将
    0.97
    ב
    0.97
     밝혔
    0.89
    ل
    0.89
    ↵
    0.89
    at
    0.87
    Act Density 0.008%

    No Known Activations