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

    Neuronpedia

    INDEX
    Explanations

    states of falsehood

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    7
    1.74
    5
    1.61
    6
    1.56
    9
    1.55
    3
    1.52
    2
    1.51
    4
    1.48
    8
    1.42
    1
    1.39
    0
    1.30
    POSITIVE LOGITS
    ோ
    1.19
    特征
    1.12
    дің
    1.11
    いは
    1.11
    徝
    1.11
    ipore
    1.08
     величина
    1.07
     보이는
    1.06
     উহা
    1.05
    梸
    1.05
    Act Density 0.500%

    No Known Activations