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

    Neuronpedia

    INDEX
    Explanations

    whales and watching

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ertes
    2.73
    д
    2.69
    これも
    2.54
    arbeiten
    2.50
    心思
    2.46
     maison
    2.46
     combien
    2.45
    leri
    2.44
     estar
    2.41
    2.39
    POSITIVE LOGITS
    𝐒
    3.14
    𝐃
    2.84
    ɑ
    2.83
    ム
    2.76
     textView
    2.75
    ことになる
    2.74
     dashed
    2.74
    𝐎
    2.73
    ことにより
    2.72
    𝐄
    2.67
    Act Density 0.135%

    No Known Activations