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

    Neuronpedia

    INDEX
    Explanations

    performance, development, testing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     enzym
    0.57
     is
    0.52
     order
    0.48
     rows
    0.47
     Enzymes
    0.46
     items
    0.46
     jib
    0.45
     Players
    0.45
     apply
    0.45
     ORDER
    0.45
    POSITIVE LOGITS
    지와
    0.48
    사와
    0.45
     વિશ્વ
    0.44
    ວິ
    0.43
    swith
    0.42
    marco
    0.41
    buffalo
    0.41
    Vladimir
    0.41
     прибли
    0.40
     венти
    0.40
    Act Density 0.000%

    No Known Activations