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

    Neuronpedia

    INDEX
    Explanations

    bail, cash, ring, spreads, evils

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ascertaining
    0.31
     beho
    0.29
     deception
    0.28
    Akt
    0.28
    obtenir
    0.28
    useful
    0.27
     satisfactory
    0.27
     competence
    0.27
     ascertain
    0.27
    satisf
    0.27
    POSITIVE LOGITS
    ™.
    0.34
     trùng
    0.31
     Ramona
    0.30
     ہے۔
    0.30
     frantically
    0.30
    约
    0.30
    ®.
    0.29
     around
    0.29
    uldron
    0.29
     banjo
    0.29
    Act Density 0.001%

    No Known Activations