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

    Neuronpedia

    INDEX
    Explanations

    introducing today's topic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     každ
    0.39
     každý
    0.38
    各
    0.38
     każdy
    0.37
    INCLUDE
    0.36
    ्यारह
    0.36
    歓
    0.35
    丁寧
    0.35
    াহীন
    0.35
    ዎቹ
    0.35
    POSITIVE LOGITS
     Topic
    0.57
     topic
    0.56
     theme
    0.55
     tenemos
    0.54
    今天要
    0.54
    Topic
    0.53
     موضوع
    0.52
     Thema
    0.52
     Theme
    0.52
     tratto
    0.50
    Act Density 0.028%

    No Known Activations