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

    Neuronpedia

    INDEX
    Explanations

    absence and negation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     not
    -1.38
     biraz
    -1.24
     också
    -1.18
    でない
    -1.15
    じゃなくて
    -1.13
    ところに
    -1.12
     bolj
    -1.08
     não
    -1.05
     számára
    -1.05
     meilleurs
    -1.04
    POSITIVE LOGITS
     anymore
    1.88
     nor
    1.55
     any
    1.46
     даже
    1.42
     anyone
    1.36
    t
    1.33
     anything
    1.30
     even
    1.30
     unless
    1.30
     except
    1.28
    Act Density 0.020%

    No Known Activations