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

    Neuronpedia

    INDEX
    Explanations

    Bloom

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    с
    1.30
    ного
    1.26
    ed
    1.20
    s
    1.13
    ,
    1.10
    id
    1.09
    d
    1.05
    ure
    1.03
    unters
    1.02
    eg
    1.02
    POSITIVE LOGITS
    다
    1.95
    ために
    1.50
    ため
    1.33
    ない
    1.21
    ки
    1.20
    き
    1.17
    м
    1.16
    তার
    1.13
    ための
    1.12
    자
    1.11
    Act Density 0.007%

    No Known Activations