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

    Neuronpedia

    INDEX
    Explanations

    context and boundaries

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Arizona
    0.24
    0.23
     девя
    0.23
    ت
    0.21
     five
    0.21
     Ending
    0.21
     Га
    0.21
    Alternative
    0.21
    ffel
    0.21
    6
    0.20
    POSITIVE LOGITS
    不管是
    0.35
    不仅仅
    0.34
     without
    0.34
     سواء
    0.34
     irrespective
    0.33
     whether
    0.32
    whether
    0.32
     sebagaimana
    0.32
    不论
    0.32
     включая
    0.32
    Act Density 0.056%

    No Known Activations