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

    Neuronpedia

    INDEX
    Explanations

    prefixes and suffixes

    np_acts-logits-general · gemini-2.5-flash-lite

    deficiency

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ࣪
    0.45
    fue
    0.43
    tiden
    0.41
    म्
    0.41
    fiber
    0.41
    fill
    0.39
    рк
    0.39
    झा
    0.39
    tte
    0.39
    fans
    0.39
    POSITIVE LOGITS
    iciency
    1.02
    initely
    0.96
    requent
    0.96
    icionado
    0.92
    requently
    0.91
    iciencies
    0.91
    USION
    0.86
    riends
    0.86
    essional
    0.84
    amiliar
    0.84
    Act Density 0.583%

    No Known Activations