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

    Neuronpedia

    INDEX
    Explanations

    often treated, within contexts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    iaire
    0.46
     پردا
    0.45
    ille
    0.45
     دارند
    0.45
    ea
    0.44
     exhibited
    0.44
     تب
    0.44
    ués
    0.43
     ብዙውን
    0.43
    èves
    0.43
    POSITIVE LOGITS
    髗
    0.48
     pressing
    0.46
     기본적인
    0.41
     garantia
    0.41
    친구
    0.40
     compromising
    0.40
    mannschaft
    0.39
    Para
    0.39
    football
    0.39
    人を
    0.38
    Act Density 0.002%

    No Known Activations