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

    Neuronpedia

    INDEX
    Explanations

    pauses followed by commas

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    hlen
    1.64
    可以
    1.58
    hindi
    1.52
    ﻥ
    1.50
    rq
    1.48
    tti
    1.45
     prezzi
    1.44
    ﺀ
    1.44
    堭
    1.43
     necessari
    1.42
    POSITIVE LOGITS
     prejudice
    1.41
    습니다
    1.36
     sturdy
    1.36
     jeopardy
    1.30
     tights
    1.27
    THING
    1.27
    Y
    1.27
    ே
    1.27
     snacks
    1.26
     timeouts
    1.25
    Act Density 0.003%

    No Known Activations