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

    Neuronpedia

    INDEX
    Explanations

    self. attribute access

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     lighting
    0.95
     light
    0.86
     Kawasaki
    0.79
     tuition
    0.79
     Light
    0.77
     Lighting
    0.76
     luz
    0.71
     Daylight
    0.71
     energized
    0.70
     exclaimed
    0.70
    POSITIVE LOGITS
     _
    2.25
     !_
    1.49
    -_
    1.45
    ._
    1.39
    :_
    1.36
     (!_
    1.34
    __
    1.34
     _.
    1.33
     (_
    1.31
    (_
    1.30
    Act Density 0.130%

    No Known Activations