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

    Neuronpedia

    INDEX
    Explanations

    often used as greeting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     is
    0.40
     sincerely
    0.39
     people
    0.38
     focusing
    0.38
     inherit
    0.38
     insist
    0.38
     inception
    0.38
     suicide
    0.37
     can
    0.36
    inscription
    0.35
    POSITIVE LOGITS
     recouvert
    0.44
    Decorative
    0.43
    濺
    0.40
    Multicolored
    0.40
     الزمن
    0.40
     cirrh
    0.39
     rø
    0.39
    檐
    0.39
    砶
    0.39
     کھیلیں
    0.38
    Act Density 0.001%

    No Known Activations