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

    Neuronpedia

    INDEX
    Explanations

    layer mask thumbnail

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    for
    0.48
    ir
    0.46
    at
    0.45
    net
    0.45
    riving
    0.44
    ty
    0.44
    ni
    0.44
    importance
    0.43
    uk
    0.43
    into
    0.43
    POSITIVE LOGITS
     pets
    0.42
     Según
    0.42
    狲
    0.41
     informática
    0.41
     datasets
    0.40
     várias
    0.39
     dajj
    0.39
     перио
    0.38
     వివిధ
    0.38
     fútbol
    0.38
    Act Density 0.001%

    No Known Activations