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

    Neuronpedia

    INDEX
    Explanations

    Here's a breakdown

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     interference
    0.38
     upheaval
    0.38
     avoiding
    0.37
    defer
    0.36
     puissent
    0.36
     alde
    0.36
    brochen
    0.36
     puedan
    0.35
    clearInterval
    0.35
     interfered
    0.35
    POSITIVE LOGITS
    给了
    0.45
     Describes
    0.42
    дать
    0.41
     Tiene
    0.41
     включает
    0.41
    ה
    0.41
    :<
    0.41
     Są
    0.41
    ：
    0.41
    ത്ര
    0.40
    Act Density 0.005%

    No Known Activations