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

    Neuronpedia

    INDEX
    Explanations

    при получении

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     NFTs
    1.62
     serien
    1.58
     leftover
    1.57
     Romania
    1.55
     relational
    1.54
     intense
    1.53
     Discuss
    1.52
     giver
    1.51
     misog
    1.51
     esports
    1.50
    POSITIVE LOGITS
     не
    3.09
     при
    2.92
     на
    2.89
     за
    2.86
     до
    2.85
     под
    2.85
     по
    2.84
     и
    2.76
     так
    2.76
     из
    2.74
    Act Density 0.314%

    No Known Activations