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

    Neuronpedia

    INDEX
    Explanations

    python code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     (
    1.09
    t
    0.96
     be
    0.90
     -
    0.79
     It
    0.77
     </
    0.75
              
    0.73
     This
    0.73
     Biografie
    0.70
            
    0.70
    POSITIVE LOGITS
    न
    1.23
    नी
    1.10
    ن
    0.97
    ب
    0.92
    מ
    0.91
    م
    0.86
    ສ່ວນ
    0.85
    ނ
    0.85
    ন
    0.84
    ຜະລິດຕະພັນ
    0.82
    Act Density 0.015%

    No Known Activations