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

    Neuronpedia

    INDEX
    Explanations

    experts and likely outcomes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     misrepresented
    0.86
     fictitious
    0.84
     fictional
    0.83
     fict
    0.73
     fabricated
    0.72
     fake
    0.71
     appalling
    0.69
     फर्जी
    0.69
     pretended
    0.68
     topics
    0.67
    POSITIVE LOGITS
     analyst
    2.39
     analysts
    2.38
    Analysts
    2.04
     Analysts
    2.04
     Analyst
    2.02
     experts
    1.94
     экспер
    1.92
     expert
    1.92
    Experts
    1.79
     expertos
    1.77
    Act Density 0.145%

    No Known Activations