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

    Neuronpedia

    INDEX
    Explanations

    facial features

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Legs
    -1.14
     legs
    -1.09
    Knee
    -1.08
     thighs
    -1.06
    footer
    -1.05
    knee
    -1.03
    legs
    -1.02
     limbs
    -1.00
     Limb
    -0.99
     Legs
    -0.99
    POSITIVE LOGITS
     facial
    2.73
     face
    2.25
     Facial
    2.14
    facial
    2.05
     Face
    1.84
    Facial
    1.84
    Face
    1.73
     cheek
    1.70
     forehead
    1.69
     nose
    1.63
    Act Density 0.111%

    No Known Activations