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

    Neuronpedia

    INDEX
    Explanations

    helpful AI assistant refusing requests

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     निर्णायक
    0.35
     layoffs
    0.34
    毎年
    0.33
    brak
    0.32
     governing
    0.32
    Closest
    0.32
    props
    0.32
     gemeente
    0.32
     classed
    0.31
    Authorization
    0.31
    POSITIVE LOGITS
     поэтому
    0.42
     Поэтому
    0.42
     což
    0.40
    Therefore
    0.35
    нома
    0.35
     derfor
    0.35
     Therefore
    0.35
    므로
    0.34
    だったので
    0.34
     harmless
    0.34
    Act Density 0.272%

    No Known Activations