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

    Neuronpedia

    INDEX
    Explanations

    to address

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    م
    1.38
    在
    1.34
    رو
    1.21
    b
    1.10
    ur
    1.00
    من
    0.93
    مس
    0.93
    L
    0.93
    بو
    0.92
    도
    0.90
    POSITIVE LOGITS
     are
    1.52
     isoniazid
    0.96
    lar
    0.95
    li
    0.91
     ﺍ
    0.89
     შორის
    0.88
    la
    0.86
    ite
    0.85
    ih
    0.84
     liquidated
    0.84
    Act Density 0.000%

    No Known Activations