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

    Neuronpedia

    INDEX
    Explanations

    improved positive qualities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ewalk
    0.74
    带有
    0.72
    Strawberry
    0.71
    ොර
    0.70
    imeric
    0.69
     मास्क
    0.69
    བ
    0.67
     корректи
    0.67
    的一种
    0.67
     അരി
    0.66
    POSITIVE LOGITS
     usability
    1.97
     durability
    1.87
     reliability
    1.85
     performance
    1.76
     aesthetics
    1.76
     ease
    1.68
     성능
    1.66
     functionality
    1.64
     ergonomics
    1.62
    性能
    1.61
    Act Density 1.740%

    No Known Activations