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

    Neuronpedia

    INDEX
    Explanations

    magnets

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     स्मार्ट
    -0.08
     droog
    -0.08
    едель
    -0.08
    웨어
    -0.08
     trocken
    -0.07
     smartwatch
    -0.07
     erk
    -0.07
     välja
    -0.07
    ánu
    -0.07
     تازه
    -0.07
    POSITIVE LOGITS
     attraction
    0.14
     aantrekk
    0.13
     attracted
    0.13
     Attraction
    0.12
     आकर्ष
    0.12
     attracting
    0.12
     attracts
    0.12
     attract
    0.11
     attractions
    0.10
     affinity
    0.10
    Act Density 0.012%

    No Known Activations