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

    Neuronpedia

    INDEX
    Explanations

    all-domain, belong

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ollary
    0.52
    ող
    0.49
    اریخ
    0.46
     rôle
    0.46
    噹
    0.46
     stipend
    0.46
    ্বরের
    0.46
     jude
    0.46
     मायणी
    0.46
     entice
    0.45
    POSITIVE LOGITS
    t
    0.52
    자
    0.52
     in
    0.51
    s
    0.50
    스
    0.50
    하
    0.50
    \
    0.49
    ।
    0.49
    클
    0.48
    선
    0.48
    Act Density 0.000%

    No Known Activations