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

    Neuronpedia

    INDEX
    Explanations

    some or quality

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Components
    0.45
     Components
    0.45
    Factors
    0.43
    Everything
    0.43
     Elements
    0.42
    Features
    0.42
     Factors
    0.42
    Technologies
    0.41
    Categories
    0.39
     Features
    0.38
    POSITIVE LOGITS
     some
    0.75
     specific
    0.66
     meaningful
    0.64
     occasional
    0.60
     некоторые
    0.59
     thoughtful
    0.57
     bazı
    0.57
     nogle
    0.57
    一些
    0.56
     sporadic
    0.55
    Act Density 4.506%

    No Known Activations