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

    Neuronpedia

    INDEX
    Explanations

    maximizing adjectives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Muito
    0.16
     بہت
    0.16
    çok
    0.15
     Muito
    0.15
     slags
    0.14
    ("
    0.14
    很好
    0.14
    very
    0.14
     muito
    0.14
     очень
    0.14
    POSITIVE LOGITS
     intriguing
    0.25
     impactful
    0.23
     enigmatic
    0.23
     poignant
    0.22
     insidious
    0.22
     insightful
    0.22
     unsettling
    0.22
     alluring
    0.22
     daunting
    0.22
     perplexing
    0.22
    Act Density 0.193%

    No Known Activations