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

    Neuronpedia

    INDEX
    Explanations

    positive outcomes from inputs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    rets
    0.38
    versing
    0.37
    占据
    0.36
     provoca
    0.35
    Differences
    0.35
    失去了
    0.35
    apshot
    0.35
     esc
    0.35
     introductions
    0.35
     discouraging
    0.34
    POSITIVE LOGITS
     benefitting
    2.00
     benefiting
    1.95
     benefit
    1.76
     beneficio
    1.66
     benefited
    1.62
    benefit
    1.61
     benefitted
    1.55
     beneficia
    1.53
    benef
    1.46
     Benefit
    1.43
    Act Density 0.012%

    No Known Activations