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

    Neuronpedia

    INDEX
    Explanations

    desert, processing, accuracy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     V
    0.92
    log
    0.90
    7
    0.87
     O
    0.86
     S
    0.86
    ↵↵
    0.85
    1
    0.83
    les
    0.82
    age
    0.82
    2
    0.81
    POSITIVE LOGITS
     asymptotics
    1.26
    eae
    1.23
    thisTrack
    1.21
     Tjiwarl
    1.19
     inactivació
    1.18
    dayspecial
    1.17
     derivs
    1.16
     lâng
    1.15
    GTUIKit
    1.15
    ucers
    1.14
    Act Density 0.000%

    No Known Activations