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

    Neuronpedia

    INDEX
    Explanations

    stop loss orders

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    give
    0.67
     دیے
    0.64
    elected
    0.63
    cane
    0.58
    B
    0.58
    degenerate
    0.57
     υπε
    0.56
    ing
    0.53
    single
    0.53
    commutative
    0.53
    POSITIVE LOGITS
     premières
    0.55
    ict
    0.54
    icted
    0.54
     not
    0.53
     percepción
    0.52
    š
    0.52
    ž
    0.52
     Lebens
    0.51
     Ведь
    0.50
    的
    0.49
    Act Density 0.002%

    No Known Activations