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

    Neuronpedia

    INDEX
    Explanations

    predicting likelihood or next words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     endlich
    0.75
    Sometimes
    0.75
     parfois
    0.75
    unut
    0.72
    ÇÃO
    0.72
    改めて
    0.72
    ட்டும்
    0.71
     Sometimes
    0.68
    sometimes
    0.68
    有时候
    0.67
    POSITIVE LOGITS
     likelihood
    4.01
     likely
    3.85
     probability
    3.60
     Likely
    3.49
    likelihood
    3.48
    likely
    3.48
    Lik
    3.27
     probabilities
    3.20
     probable
    3.18
    probability
    3.15
    Act Density 0.484%

    No Known Activations