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

    Neuronpedia

    INDEX
    Explanations

    names and pronouns in dramatic events

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    地产
    0.47
     enthusiasm
    0.41
    engaruhi
    0.39
    似乎
    0.39
     subtly
    0.38
     overshadow
    0.37
     restlessness
    0.37
     disappointments
    0.37
    %)$
    0.37
     DataFrame
    0.36
    POSITIVE LOGITS
     screamed
    0.83
     alerted
    0.81
     startled
    0.78
     luckily
    0.76
     scream
    0.72
     horrified
    0.71
     screams
    0.71
     fortunately
    0.70
     shocked
    0.69
     estaba
    0.66
    Act Density 0.009%

    No Known Activations