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

    Neuronpedia

    INDEX
    Explanations

    generic collections

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused30>
    0.73
     turning
    0.71
     mesh
    0.68
     water
    0.68
     life
    0.68
    ICKET
    0.67
     boule
    0.67
    water
    0.67
     উপল
    0.67
     вода
    0.66
    POSITIVE LOGITS
    Generic
    1.35
     Generic
    1.24
    generic
    1.17
     generic
    1.08
    Concurrent
    1.05
     GÉN
    0.95
    ObjectModel
    0.93
    Gener
    0.93
    Immutable
    0.92
    Special
    0.88
    Act Density 0.053%

    No Known Activations