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

    Neuronpedia

    INDEX
    Explanations

    | delimited text

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    痣
    0.33
    遂
    0.33
    নায়
    0.32
    𒊒
    0.32
    BufOffset
    0.31
     prioritise
    0.31
    ков
    0.30
    elier
    0.30
     catwalk
    0.30
    кування
    0.30
    POSITIVE LOGITS
     unless
    0.37
    unless
    0.36
    嗯
    0.36
    incorrect
    0.34
    urion
    0.33
     uro
    0.32
     বাদ
    0.32
    blah
    0.32
     역시
    0.32
     말고
    0.32
    Act Density 0.003%

    No Known Activations