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

    Neuronpedia

    INDEX
    Explanations

    lucky numbers, charms, superstitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    かで
    0.50
    écoute
    0.43
    ভাষা
    0.43
    Borderless
    0.42
    இய
    0.39
    luxury
    0.39
    amaran
    0.38
     revolutionize
    0.38
    Atlantic
    0.38
     பங்கு
    0.38
    POSITIVE LOGITS
     charms
    0.92
     charm
    0.89
     Charm
    0.83
    Charm
    0.64
     superst
    0.61
     Strike
    0.60
     strike
    0.58
    charm
    0.58
     charmed
    0.57
     pennies
    0.57
    Act Density 0.003%

    No Known Activations