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

    Neuronpedia

    INDEX
    Explanations

    intent and purpose

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     serão
    0.61
    odds
    0.61
    membrane
    0.59
    SELF
    0.57
    নিউ
    0.55
     でしょう
    0.55
    zahl
    0.55
    UNCIL
    0.55
     terão
    0.55
     වර්ග
    0.55
    POSITIVE LOGITS
     intent
    0.82
     intentions
    0.75
    Intent
    0.69
     Intent
    0.69
    intent
    0.61
     intentional
    0.58
    앙
    0.57
     Badger
    0.54
     intention
    0.54
     purposeful
    0.54
    Act Density 0.133%

    No Known Activations