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

    Neuronpedia

    INDEX
    Explanations

    units and specific terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    cmt
    0.41
     puis
    0.40
    ইন
    0.39
     end
    0.38
     Alta
    0.38
     然后
    0.38
    varlak
    0.37
     so
    0.36
     Invite
    0.36
     End
    0.36
    POSITIVE LOGITS
     ಪ್ರತ
    0.39
    𒊩
    0.39
    CHEAT
    0.38
    amentul
    0.38
    Goal
    0.37
    飲食店
    0.37
     Representative
    0.37
    髌
    0.37
    Representative
    0.36
     representative
    0.36
    Act Density 0.000%

    No Known Activations