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

    Neuronpedia

    INDEX
    Explanations

    emphatic positive adjectives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     zeer
    1.04
     very
    1.03
    非常に
    1.01
     весьма
    0.99
     prodigious
    0.98
     sehr
    0.98
     meget
    0.98
    very
    0.97
     vrlo
    0.97
    极为
    0.97
    POSITIVE LOGITS
     SO
    1.31
     super
    1.21
     sooo
    1.17
     Super
    1.11
     soo
    1.09
     SUCH
    1.07
     SUPER
    1.07
    Super
    1.05
    SO
    1.04
    super
    1.02
    Act Density 0.480%

    No Known Activations