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

    Neuronpedia

    INDEX
    Explanations

    what is being

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    categories
    0.39
     utilisent
    0.39
     пункт
    0.39
     Focused
    0.38
    対応
    0.38
    Focused
    0.38
     formatting
    0.38
    εργ
    0.38
    Used
    0.38
     によ
    0.37
    POSITIVE LOGITS
     trans
    0.45
    onitrile
    0.40
    يج
    0.39
     transpired
    0.38
     Polonia
    0.38
     transp
    0.37
    岂
    0.37
    নকে
    0.36
     disampaikan
    0.36
     expressed
    0.36
    Act Density 0.007%

    No Known Activations