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

    Neuronpedia

    INDEX
    Explanations

    appearance and deception

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Predicting
    0.73
     noncommutative
    0.69
     unfore
    0.68
     Prediction
    0.67
    Prediction
    0.66
    )(-
    0.65
     Yath
    0.65
    Wildlife
    0.65
    単純
    0.65
     ungroup
    0.64
    POSITIVE LOGITS
     facade
    2.05
     façade
    2.03
     facades
    1.77
     veneer
    1.73
     fachada
    1.63
     фаса
    1.47
     outwardly
    1.45
     carefully
    1.44
     pretending
    1.40
     masking
    1.40
    Act Density 0.379%

    No Known Activations