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

    Neuronpedia

    INDEX
    Explanations

    incomplete, cut, truncated

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    fron
    0.45
    kitchen
    0.44
    Kitchen
    0.43
     vanes
    0.41
    beeld
    0.40
    lech
    0.40
    laces
    0.39
     iron
    0.38
    villa
    0.38
    BackButton
    0.38
    POSITIVE LOGITS
     incomplete
    0.60
     cut
    0.51
     incomple
    0.51
     truncated
    0.49
     Cut
    0.48
    incomplete
    0.47
     truncate
    0.46
     interrupted
    0.45
     Interrupt
    0.45
     CUT
    0.44
    Act Density 0.011%

    No Known Activations