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

    Neuronpedia

    INDEX
    Explanations

    actions followed by prepositions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    そして
    1.06
    그리고
    1.02
    (!)
    1.00
     그리고
    0.99
    หรือ
    0.98
    或者
    0.96
     или
    0.95
    或者是
    0.94
     وم
    0.92
    或其他
    0.92
    POSITIVE LOGITS
     via
    1.37
     with
    1.29
     through
    1.26
     from
    1.23
     on
    1.18
     within
    1.17
     alongside
    1.12
     across
    1.12
     without
    1.11
     against
    1.11
    Act Density 0.381%

    No Known Activations