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

    Neuronpedia

    INDEX
    Explanations

    like a simile

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     reportedly
    0.62
    ?
    0.61
     apparently
    0.57
     displays
    0.55
     referred
    0.53
     (
    0.52
     that
    0.52
     often
    0.51
     usually
    0.50
     कथित
    0.50
    POSITIVE LOGITS
     isang
    0.75
    adiator
    0.69
    enteuer
    0.64
    orkeling
    0.64
    誂
    0.64
     будто
    0.61
    một
    0.61
     unto
    0.60
    atedral
    0.60
     സിനിമ
    0.60
    Act Density 0.028%

    No Known Activations