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

    Neuronpedia

    INDEX
    Explanations

    more and most + adjective

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    es
    1.40
    л
    1.27
    1
    1.19
    0
    1.19
    ES
    1.18
    টিয়ে
    1.18
     sapp
    1.15
     diffère
    1.14
     parle
    1.12
    问题
    1.10
    POSITIVE LOGITS
     subdued
    1.44
     straightforward
    1.42
     sinister
    1.39
    ஸ்ட
    1.38
    и
    1.38
     oftent
    1.38
     tailored
    1.35
    𝟐
    1.30
     extensive
    1.30
     understandable
    1.28
    Act Density 0.382%

    No Known Activations