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

    Neuronpedia

    INDEX
    Explanations

    looks at potential

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    lare
    0.57
     seront
    0.53
     forme
    0.52
     elixir
    0.52
     quedan
    0.51
     olmak
    0.50
     révèle
    0.50
    ć
    0.50
    file
    0.50
     alchemy
    0.49
    POSITIVE LOGITS
    ра
    0.63
     TEX
    0.55
     цих
    0.54
    した
    0.54
    тан
    0.50
     Umar
    0.49
     :}
    0.49
     GX
    0.49
    🌤
    0.48
     Kupp
    0.48
    Act Density 0.000%

    No Known Activations