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

    Neuronpedia

    INDEX
    Explanations

    abstract or summary section

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     keep
    0.61
     onlyOwner
    0.61
    quels
    0.59
     generosity
    0.59
     journeys
    0.59
     Helpers
    0.59
     ruin
    0.58
     friendships
    0.57
     }]);
    0.57
    </body>
    0.57
    POSITIVE LOGITS
    ABSTRACT
    1.33
    Abstract
    1.29
     Abstract
    1.20
     ABSTRACT
    1.18
    abstract
    1.03
     abstract
    1.00
    摘要
    0.88
     SUMMARY
    0.88
    Synopsis
    0.85
    简介
    0.84
    Act Density 0.009%

    No Known Activations