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

    Neuronpedia

    INDEX
    Explanations

    structural keywords like 'or' and 'where'

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Championship
    0.52
    os
    0.41
    in
    0.39
    idi
    0.39
    cknowled
    0.38
    ed
    0.38
    مول
    0.38
    ]{
    0.37
    arik
    0.37
    OCE
    0.36
    POSITIVE LOGITS
     arcs
    0.50
     biện
    0.49
    𒅗
    0.47
    자
    0.47
     manipulations
    0.47
    ክት
    0.46
     nasıl
    0.46
    리
    0.46
     wikipagina
    0.46
     knobs
    0.46
    Act Density 0.295%

    No Known Activations