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

    Neuronpedia

    INDEX
    Explanations

    JSON data structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     CONCLUSIONS
    0.24
    THEOREM
    0.23
     BULLETIN
    0.23
    CONCLUSIONS
    0.23
     CONTINUE
    0.23
    ARCHIVO
    0.22
     Cómo
    0.22
     mannit
    0.22
     SUBTOTAL
    0.22
     말미암아
    0.22
    POSITIVE LOGITS
    _
    0.37
    To
    0.36
     =
    0.36
    Of
    0.34
    Data
    0.33
    Or
    0.32
    In
    0.31
    And
    0.31
    Id
    0.31
    For
    0.31
    Act Density 0.331%

    No Known Activations