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

    Neuronpedia

    INDEX
    Explanations

    revised response or answer

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     persecution
    0.49
     horrors
    0.48
     abhor
    0.47
     raged
    0.46
    近年
    0.45
     atrocities
    0.44
     horrifying
    0.42
     gente
    0.41
     taboo
    0.41
     totalitarian
    0.41
    POSITIVE LOGITS
     avvic
    0.47
     allocations
    0.46
     thoughtfully
    0.43
    깆
    0.43
     satisfactorily
    0.42
     ampio
    0.42
     webhook
    0.42
     thoroughly
    0.41
     lineup
    0.41
     completes
    0.41
    Act Density 0.016%

    No Known Activations