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

    Neuronpedia

    INDEX
    Explanations

    sparked widespread

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Children
    0.49
     Palestinians
    0.45
     be
    0.43
     Women
    0.42
     Cuando
    0.41
     Democrats
    0.40
     derniers
    0.40
     Poison
    0.40
    响应
    0.40
     Humans
    0.40
    POSITIVE LOGITS
     transmits
    0.49
     transmit
    0.46
     transmis
    0.46
     fiable
    0.46
     timings
    0.46
     transmite
    0.45
    tional
    0.45
     capire
    0.44
    GROUP
    0.44
    බ්
    0.44
    Act Density 0.007%

    No Known Activations