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

    Neuronpedia

    INDEX
    Explanations

    understanding limitations and offering safe help

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     (
    0.37
     If
    0.31
     Data
    0.31
     Review
    0.31
     A
    0.30
    .
    0.29
     Laser
    0.29
     Grove
    0.29
     K
    0.29
    K
    0.29
    POSITIVE LOGITS
    ErrorClazz
    0.30
    袠
    0.29
    ेन
    0.28
     ulterior
    0.28
    あえず
    0.28
    iduci
    0.28
     hypocrisy
    0.27
    funcion
    0.27
    गति
    0.27
     Фурга
    0.27
    Act Density 0.063%

    No Known Activations