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

    Neuronpedia

    INDEX
    Explanations

    smoke-free

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     accompany
    0.41
     bezahlen
    0.40
     tabs
    0.40
     cited
    0.38
     Scripts
    0.38
     Psychology
    0.38
     Availability
    0.38
     War
    0.38
     Chat
    0.38
     intimated
    0.38
    POSITIVE LOGITS
    clean
    0.60
     dusty
    0.57
     odor
    0.55
     clean
    0.54
    Clean
    0.53
     சுத்த
    0.53
     ماحول
    0.53
    環境
    0.51
     environnement
    0.50
    清洁
    0.50
    Act Density 0.004%

    No Known Activations