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

    Neuronpedia

    INDEX
    Explanations

    potential harm or injury

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    at
    0.63
    an
    0.50
     Studio
    0.49
     بیمار
    0.44
     map
    0.44
     hade
    0.44
     mour
    0.43
    ir
    0.43
     lange
    0.42
    in
    0.42
    POSITIVE LOGITS
     உள்ப
    0.53
    ünk
    0.48
    +="
    0.45
    центри
    0.44
    вре
    0.44
     STEELS
    0.43
     вклю
    0.43
    Includes
    0.43
    мо
    0.42
     préférable
    0.42
    Act Density 0.002%

    No Known Activations