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

    Neuronpedia

    INDEX
    Explanations

    introductions or definitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     yada
    1.35
     etc
    1.20
     inoltre
    1.20
    なども
    1.14
     plz
    1.11
     др
    1.10
     bonus
    1.09
     ebenfalls
    1.09
     برضو
    1.07
    etc
    1.07
    POSITIVE LOGITS
     The
    1.42
    The
    1.26
    :
    1.23
     What
    1.22
     By
    1.15
     For
    1.15
     And
    1.13
    What
    1.12
    ?
    1.10
     –
    1.09
    Act Density 0.594%

    No Known Activations