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

    Neuronpedia

    INDEX
    Explanations

    how things were created or trained

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    readFile
    0.46
    arci
    0.45
    ix
    0.43
     wrongly
    0.43
    onna
    0.42
    omach
    0.42
    ując
    0.42
    घव
    0.42
     objecting
    0.42
    loadFile
    0.41
    POSITIVE LOGITS
    撕
    0.47
     univers
    0.45
     då
    0.43
    」（
    0.43
     Contreras
    0.42
    hä
    0.42
     etern
    0.41
    കളും
    0.41
     enz
    0.40
     biel
    0.40
    Act Density 0.003%

    No Known Activations