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

    Neuronpedia

    INDEX
    Explanations

    discussing possibilities and hypotheticals

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Probably
    0.52
     probablement
    0.51
     wahrscheinlich
    0.50
     probablemente
    0.48
    probably
    0.48
     provavelmente
    0.47
    Can
    0.46
    Probably
    0.46
     Presumably
    0.46
     probabilmente
    0.45
    POSITIVE LOGITS
     conceivably
    0.86
     have
    0.78
     might
    0.74
     be
    0.72
     may
    0.64
     possibly
    0.60
     inadvertently
    0.59
     könnte
    0.58
    might
    0.58
     require
    0.57
    Act Density 0.098%

    No Known Activations