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

    Neuronpedia

    INDEX
    Explanations

    unwittingly or deliberately disrupt

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ≤
    0.47
     wahrscheinlich
    0.46
     পেলাম
    0.45
     probablement
    0.45
     Tabelle
    0.45
     sagte
    0.45
     verwenden
    0.45
     وقال
    0.44
    you
    0.44
     컴퓨
    0.43
    POSITIVE LOGITS
     flashbacks
    0.71
     unwittingly
    0.70
     mysterious
    0.68
     reluctantly
    0.68
     secretly
    0.66
     haunted
    0.65
     unorthodox
    0.65
     discovers
    0.63
     secrets
    0.63
     mysteriously
    0.63
    Act Density 0.128%

    No Known Activations