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

    Neuronpedia

    INDEX
    Explanations

    questions asking for reasons or explanations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     believed
    1.10
     diketahui
    1.05
     observed
    0.98
     understood
    0.96
     seen
    0.90
     known
    0.89
     assumed
    0.84
    observed
    0.84
     suspected
    0.84
     noted
    0.82
    POSITIVE LOGITS
    QS
    0.40
    पट्टी
    0.37
     அம்மா
    0.36
     IANS
    0.36
     性
    0.35
     pavattati
    0.34
    wpi
    0.34
    landers
    0.34
    Torrent
    0.34
     пытается
    0.34
    Act Density 0.134%

    No Known Activations