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

    Neuronpedia

    INDEX
    Explanations

    break down how, what

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    storeItem
    0.55
    자리
    0.54
     가지
    0.51
    បន្ថ
    0.51
    ObjectIndex
    0.48
    SecurityMarks
    0.47
     ລັບ
    0.47
    បន្ថែម
    0.47
    㣫
    0.46
    CollectedField
    0.46
    POSITIVE LOGITS
    '
    0.74
    ’
    0.66
     clin
    0.60
     on
    0.54
    .
    0.49
     between
    0.46
    t
    0.45
    !
    0.45
     throughout
    0.44
    ↵
    0.43
    Act Density 0.004%

    No Known Activations