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

    Neuronpedia

    INDEX
    Explanations

    qualifications or exceptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     их
    0.43
     всички
    0.43
    пер
    0.42
     его
    0.40
    他的
    0.40
     mortal
    0.40
     империи
    0.39
     his
    0.39
    வார்கள்
    0.39
    থর
    0.39
    POSITIVE LOGITS
     contexto
    0.47
     contextual
    0.46
     terasa
    0.45
     sedang
    0.41
     context
    0.41
     контек
    0.40
     Context
    0.40
     feels
    0.39
     please
    0.39
     adverb
    0.38
    Act Density 0.015%

    No Known Activations