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

    Neuronpedia

    INDEX
    Explanations

    file path and title

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    сро
    1.83
     schemes
    1.65
     selectors
    1.65
    夫妇
    1.65
    ματο
    1.55
    unct
    1.54
     organisation
    1.52
     scheme
    1.52
     one
    1.50
    囝
    1.50
    POSITIVE LOGITS
    7
    3.74
    6
    3.52
    8
    3.49
    5
    3.48
    9
    3.39
    3
    3.12
    4
    3.04
    2
    2.53
    0
    2.51
    rinsic
    2.30
    Act Density 0.198%

    No Known Activations