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

    Neuronpedia

    INDEX
    Explanations

    character names followed by dialogue

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    stä
    0.60
    多くの
    0.60
    しかも
    0.60
     mondiale
    0.57
     모든
    0.56
    is
    0.55
    通常の
    0.55
     exced
    0.54
    0.54
     대부분
    0.53
    POSITIVE LOGITS
     ascribe
    0.74
     menjelaskan
    0.73
     recommends
    0.72
     suggested
    0.71
     textView
    0.71
     quantifying
    0.71
     TestAvg
    0.69
     caseworker
    0.66
    臾
    0.66
    쾺
    0.66
    Act Density 0.072%

    No Known Activations