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

    Neuronpedia

    INDEX
    Explanations

    learning from human feedback

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     intrig
    0.58
     Jep
    0.57
    ti
    0.55
    뭉
    0.55
    ાઇ
    0.54
     Intermediate
    0.54
     ti
    0.54
     transformative
    0.52
    cione
    0.52
    坭
    0.52
    POSITIVE LOGITS
    ſh
    0.74
    hess
    0.70
    havam
    0.64
    Hab
    0.63
    暐
    0.63
     ஹ
    0.63
    ჰ
    0.62
     Há
    0.61
    وبا
    0.61
    шили
    0.60
    Act Density 0.200%

    No Known Activations