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

    Neuronpedia

    INDEX
    Explanations

    sample bubble

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     speakers
    0.58
    곶
    0.58
    かう
    0.56
     trump
    0.54
    centralwidget
    0.53
     speaker
    0.52
     resuming
    0.52
     troops
    0.51
     pf
    0.51
     பேசு
    0.51
    POSITIVE LOGITS
    拉克
    0.70
     Bubble
    0.63
     끌
    0.63
     בת
    0.63
    Bubble
    0.62
    bubble
    0.61
     Samples
    0.60
     Ultrasonic
    0.59
     Н
    0.58
    hydrogen
    0.58
    Act Density 0.000%

    No Known Activations