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

    Neuronpedia

    INDEX
    Explanations

    tongue twister, cognitions, geographical names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    -
    0.52
    /
    0.44
    ë
    0.43
    нен
    0.42
    积累
    0.42
    рье
    0.41
    集体
    0.41
    ύν
    0.40
     юриди
    0.40
    と呼ば
    0.40
    POSITIVE LOGITS
    GON
    0.54
     Zheng
    0.52
     DER
    0.52
     China
    0.52
    തി
    0.51
     Ain
    0.51
     Iran
    0.51
    State
    0.50
     Gonz
    0.50
     Bech
    0.50
    Act Density 0.001%

    No Known Activations