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

    Neuronpedia

    INDEX
    Explanations

    increase and growth trends

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Can
    0.55
    Safety
    0.54
     möglichen
    0.54
    Impl
    0.53
     または
    0.53
     Suggest
    0.52
    Allows
    0.52
     Safety
    0.51
    Couldn
    0.51
    implies
    0.51
    POSITIVE LOGITS
     soared
    0.93
     continues
    0.93
     steadily
    0.92
     বেড়েছে
    0.88
     has
    0.86
     continue
    0.84
     увеличи
    0.84
     telah
    0.83
     meningkat
    0.83
     đã
    0.79
    Act Density 0.004%

    No Known Activations