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

    Neuronpedia

    INDEX
    Explanations

    gradually add

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Margins
    0.76
     (!_
    0.75
    Funny
    0.72
    File
    0.65
    Unsupported
    0.65
    কুল
    0.64
     *_
    0.64
    isNaN
    0.63
    平凡
    0.63
     Document
    0.62
    POSITIVE LOGITS
     aware
    0.84
    她们
    0.82
     conscientes
    0.77
     entrees
    0.76
    ὥ
    0.76
    had
    0.76
     raging
    0.76
     종료
    0.75
     commencé
    0.75
     answered
    0.75
    Act Density 0.002%

    No Known Activations