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

    Neuronpedia

    INDEX
    Explanations

    watching

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    >
    0.51
    of
    0.41
     одежды
    0.40
    in
    0.38
     of
    0.38
     weren
    0.36
    стые
    0.36
    viä
    0.36
     сеть
    0.35
    boli
    0.35
    POSITIVE LOGITS
     helplessly
    0.67
     attentively
    0.66
     Watching
    0.60
     watching
    0.59
    Watching
    0.57
     watch
    0.57
     beobachten
    0.56
     WATCH
    0.51
     наблю
    0.51
     atent
    0.51
    Act Density 0.049%

    No Known Activations