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

    Neuronpedia

    INDEX
    Explanations

    deeply impressed or captivated

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    รอ
    0.44
     incapable
    0.42
     capable
    0.40
     تجاه
    0.40
     hypoc
    0.39
    ปร
    0.39
     wrongfully
    0.39
     angesehen
    0.38
    Capable
    0.38
    aternion
    0.37
    POSITIVE LOGITS
     impressed
    0.74
     impress
    0.67
     beeindruck
    0.62
     impresses
    0.62
    震撼
    0.57
     immersing
    0.55
     ấn
    0.54
     immerse
    0.53
     captivated
    0.53
     immers
    0.52
    Act Density 0.002%

    No Known Activations