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

    Neuronpedia

    INDEX
    Explanations

    descriptive settings and ambiance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     reckless
    0.75
     weakens
    0.67
     vergessen
    0.66
     melts
    0.66
     weak
    0.66
     trial
    0.65
     효과
    0.65
     yếu
    0.63
     feeble
    0.63
     टाकी
    0.63
    POSITIVE LOGITS
     beautifully
    2.06
     nicely
    1.95
     brightly
    1.75
     elegantly
    1.65
     well
    1.65
     stylish
    1.63
     tastefully
    1.62
     spacious
    1.59
     neatly
    1.53
     красиво
    1.50
    Act Density 0.307%

    No Known Activations