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

    Neuronpedia

    INDEX
    Explanations

    commencing actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     geworden
    0.31
     показателей
    0.29
    дах
    0.29
     sırasında
    0.28
     Starting
    0.28
     товары
    0.28
     посредством
    0.27
    Starting
    0.27
     Complet
    0.27
     poprzez
    0.27
    POSITIVE LOGITS
     experimenting
    0.58
     exploring
    0.56
     collecting
    0.56
     playing
    0.55
     accruing
    0.53
     chipping
    0.52
     taking
    0.52
     assembling
    0.52
     investigating
    0.52
     talking
    0.50
    Act Density 0.065%

    No Known Activations