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

    Neuronpedia

    INDEX
    Explanations

    curly brace

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     lite
    0.41
     Amazonas
    0.41
     hall
    0.37
     আজি
    0.37
     tubes
    0.36
     halls
    0.36
     homologous
    0.36
    ائو
    0.35
     cousins
    0.35
     ইমন
    0.35
    POSITIVE LOGITS
    各种
    0.66
    various
    0.65
    Various
    0.59
     Various
    0.59
     various
    0.55
    各種
    0.55
     berbagai
    0.51
     각종
    0.49
     çeşitli
    0.47
    さまざまな
    0.43
    Act Density 0.002%

    No Known Activations