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

    Neuronpedia

    INDEX
    Explanations

    still / yet / more

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ate
    1.25
    is
    1.19
    ov
    1.18
    0
    1.10
    os
    1.08
    ong
    1.00
    othe
    0.97
    ok
    0.96
    ick
    0.96
    amed
    0.96
    POSITIVE LOGITS
    ول
    0.98
    ي
    0.95
    ר
    0.94
    ל
    0.91
    י
    0.87
    р
    0.84
    тие
    0.78
     dividend
    0.78
    ل
    0.77
     mattress
    0.76
    Act Density 0.000%

    No Known Activations