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

    Neuronpedia

    INDEX
    Explanations

    specifications, attributes, types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Porque
    0.26
    で行
    0.25
     cắt
    0.25
    ꯔ
    0.24
    න
    0.23
     berikut
    0.23
     nào
    0.22
     condivid
    0.22
    つまり
    0.22
    ](./
    0.22
    POSITIVE LOGITS
     age
    0.26
    Model
    0.26
    Type
    0.25
    gender
    0.25
    Maximum
    0.25
    Theme
    0.25
    mapper
    0.24
    type
    0.24
    accent
    0.23
     accent
    0.23
    Act Density 0.000%

    No Known Activations