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

    Neuronpedia

    INDEX
    Explanations

    past conditional structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     bumpers
    0.75
     hum
    0.72
    Hum
    0.72
     counting
    0.71
    hum
    0.70
     Hum
    0.70
    塑料
    0.69
     kole
    0.69
     itemCount
    0.69
     stereo
    0.69
    POSITIVE LOGITS
     past
    1.69
     Past
    1.69
    Past
    1.63
    past
    1.62
    过去
    1.55
     Vergangenheit
    1.53
    過去
    1.48
     passato
    1.48
     PAST
    1.39
     passado
    1.32
    Act Density 0.330%

    No Known Activations