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

    Neuronpedia

    INDEX
    Explanations

    open weights model platform

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     caregivers
    0.52
     maksimal
    0.49
     quantas
    0.48
     тъ
    0.46
     attraverso
    0.45
     cikin
    0.43
     liés
    0.43
     blames
    0.42
     melalui
    0.42
     quantidade
    0.42
    POSITIVE LOGITS
    ！！
    0.48
     ！
    0.44
    cb
    0.43
    坚定
    0.43
    하니
    0.41
    今日
    0.40
    ！（
    0.40
     mmf
    0.40
     重要
    0.40
    未知
    0.40
    Act Density 0.011%

    No Known Activations