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

    Neuronpedia

    INDEX
    Explanations

    important role

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     foundational
    1.24
     eben
    1.15
     wieder
    1.13
     ethos
    1.11
     Māori
    1.09
     mensch
    1.09
     Venn
    1.08
     mL
    1.08
     langsam
    1.05
     twee
    1.05
    POSITIVE LOGITS
     не
    2.82
     под
    2.72
     без
    2.70
     при
    2.69
     з
    2.66
     на
    2.65
     много
    2.63
     ди
    2.57
     и
    2.57
     ре
    2.56
    Act Density 0.215%

    No Known Activations