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

    Neuronpedia

    INDEX
    Explanations

    or and

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    が
    -1.66
    で
    -1.59
    ですが
    -1.44
    を
    -1.41
    では
    -1.34
    は
    -1.32
    に
    -1.29
     Appetit
    -1.24
    ということで
    -1.21
    </h4>
    -1.20
    POSITIVE LOGITS
     or
    2.09
     and
    1.81
    為に
    1.22
     which
    1.21
     difficult
    1.14
     different
    1.14
     not
    1.14
     no
    1.10
    際は
    1.09
    際には
    1.09
    Act Density 0.035%

    No Known Activations