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

    Neuronpedia

    INDEX
    Explanations

    kill all, most e, return none, from yoga

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    modification
    0.40
    language
    0.39
    ো
    0.38
    有问题
    0.38
    [.
    0.38
    environment
    0.37
     tecnológicos
    0.36
    Landscape
    0.36
    landscape
    0.36
    modified
    0.36
    POSITIVE LOGITS
     breath
    0.46
     ball
    0.41
     breaths
    0.41
     fale
    0.41
     spot
    0.41
     guts
    0.41
     neck
    0.40
     hals
    0.40
     slider
    0.40
    ዳል
    0.40
    Act Density 0.002%

    No Known Activations