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

    Neuronpedia

    INDEX
    Explanations

    I will not

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     enormes
    0.52
     trillions
    0.51
     billions
    0.47
     conduits
    0.47
     GaAs
    0.45
     exoskeleton
    0.45
     silos
    0.44
    なのに
    0.44
     reactors
    0.44
     centaines
    0.43
    POSITIVE LOGITS
     будет
    0.76
    我会
    0.73
     будут
    0.68
     иногда
    0.64
     will
    0.63
    Sometimes
    0.63
     będzie
    0.62
     sometimes
    0.61
     bude
    0.60
    จะไม่
    0.60
    Act Density 0.130%

    No Known Activations