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

    Neuronpedia

    INDEX
    Explanations

    technical language

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Neighbor
    1.43
     Neighbor
    1.36
    Neighbors
    1.29
    Behavior
    1.28
    Defense
    1.26
    Gray
    1.20
    Aluminum
    1.19
    Hemat
    1.18
    Colors
    1.18
     flavors
    1.16
    POSITIVE LOGITS
     apple
    0.71
     ஃப
    0.70
    嗥
    0.67
     SER
    0.67
     сви
    0.66
     ва
    0.66
     rabbit
    0.64
     Ф
    0.63
     ფ
    0.63
    嶷
    0.63
    Act Density 0.000%

    No Known Activations