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

    Neuronpedia

    INDEX
    Explanations

    starts explanatory phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0.91
    0.91
    0.77
    0.75
     surfact
    0.74
     biti
    0.74
     puede
    0.73
    𝐚
    0.73
    𝗦
    0.73
    =(\
    0.73
    POSITIVE LOGITS
    ଖ
    0.91
    ୌ
    0.91
    ື່ອງ
    0.90
    л
    0.88
    <unused1741>
    0.86
    <unused398>
    0.86
    dbjc
    0.85
    ्रम
    0.84
     Bücher
    0.83
    őség
    0.83
    Act Density 0.345%

    No Known Activations