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

    Neuronpedia

    INDEX
    Explanations

    asking about what's happening

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    leak
    0.96
    mess
    0.90
     ruining
    0.88
     Analysts
    0.86
     Analyst
    0.85
    gleichen
    0.85
    Captcha
    0.84
     messes
    0.83
     listener
    0.83
     garr
    0.82
    POSITIVE LOGITS
    ulmonary
    1.01
     bức
    1.01
     inhalation
    0.99
     perjalanan
    0.98
     cardio
    0.97
     inhal
    0.97
     gunshot
    0.97
     આપવામાં
    0.95
     personalmente
    0.95
     exposición
    0.94
    Act Density 0.174%

    No Known Activations