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

    Neuronpedia

    INDEX
    Explanations

    specific roles or tools

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     más
    0.50
     de
    0.49
    ：
    0.47
     và
    0.46
    EM
    0.45
    ﻔ
    0.45
     método
    0.45
    с
    0.44
    і
    0.43
     varía
    0.42
    POSITIVE LOGITS
     Chihuahua
    0.46
     Democrat
    0.45
     Afrika
    0.45
     instability
    0.44
     disguised
    0.43
     ಅದು
    0.43
     Indiana
    0.42
     Death
    0.42
     Aaron
    0.42
     Elephant
    0.41
    Act Density 0.677%

    No Known Activations