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

    Neuronpedia

    INDEX
    Explanations

    well-being and related phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    鰐
    0.81
     belong
    0.75
     grove
    0.74
     resemble
    0.71
     flatten
    0.70
     discriminate
    0.68
     deserve
    0.68
    峮
    0.67
     vary
    0.67
     truncate
    0.67
    POSITIVE LOGITS
    bee
    1.36
     Bee
    1.36
     b
    1.35
    Bee
    1.31
     BEE
    1.30
     би
    1.29
    Be
    1.28
     bean
    1.24
     bi
    1.21
    BE
    1.20
    Act Density 0.120%

    No Known Activations