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

    Neuronpedia

    INDEX
    Explanations

    vision transformer or black box

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    чів
    1.29
    אר
    1.06
    ر
    1.05
    asys
    1.04
    ом
    1.03
    してる
    1.03
    ਮ
    1.02
    mios
    1.02
    нется
    1.00
    е
    1.00
    POSITIVE LOGITS
     পালা
    1.29
    "_
    1.28
     splenic
    1.20
     genitals
    1.12
     deterioro
    1.10
     osoby
    1.10
    ąg
    1.08
     constituents
    1.07
    严重
    1.07
     spleen
    1.06
    Act Density 0.001%

    No Known Activations