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

    Neuronpedia

    INDEX
    Explanations

    exciting new features

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     molto
    1.00
     meget
    0.91
     foarte
    0.90
     매우
    0.88
     sehr
    0.88
    Very
    0.83
     очень
    0.81
     Very
    0.81
    かなり
    0.79
     veldig
    0.79
    POSITIVE LOGITS
     stunning
    1.73
     vibrant
    1.65
     captivating
    1.64
     breathtaking
    1.59
     luscious
    1.57
     lively
    1.57
     thrilling
    1.54
     lush
    1.52
     delectable
    1.52
     dazzling
    1.50
    Act Density 1.164%

    No Known Activations