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

    Neuronpedia

    INDEX
    Explanations

    code referencing array/list indices

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ’
    -3.13
     After
    -2.88
    These
    -2.50
     from
    -2.45
    o
    -2.33
     for
    -2.23
     in
    -2.19
    -2.17
    独特
    -2.17
    ↵
    -2.16
    POSITIVE LOGITS
    鑁
    3.34
    籀
    2.84
    すっ
    2.84
    簓
    2.78
    饪
    2.72
    のでしょう
    2.67
     dreary
    2.64
    釤
    2.63
    樒
    2.63
    シャレ
    2.61
    Act Density 0.014%

    No Known Activations