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

    Neuronpedia

    INDEX
    Explanations

    rejecting Hero

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ዐይን
    0.53
     recommends
    0.48
     impeded
    0.47
     preds
    0.47
     tattooed
    0.46
     directed
    0.45
     grievous
    0.45
    istency
    0.44
     meaty
    0.44
     directs
    0.44
    POSITIVE LOGITS
    cena
    0.54
    ter
    0.51
    c
    0.51
    pre
    0.50
    sta
    0.50
    muc
    0.49
    to
    0.48
    ta
    0.48
    soldiers
    0.47
    arquía
    0.47
    Act Density 0.002%

    No Known Activations