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

    Neuronpedia

    INDEX
    Explanations

    numbers and repetition

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     leaving
    3.63
     celebrating
    3.44
     strengthening
    3.39
     entries
    3.32
     refreshed
    3.27
     announcements
    3.21
     commented
    3.20
     older
    3.17
     planned
    3.14
     celebrated
    3.14
    POSITIVE LOGITS
    7
    2.96
    8
    2.77
    3
    2.76
    6
    2.75
    4
    2.68
    5
    2.46
    2
    2.37
    9
    2.27
    1
    2.26
    0
    2.02
    Act Density 0.578%

    No Known Activations