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

    Neuronpedia

    INDEX
    Explanations

    explains purpose or subtle nature

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     conducting
    0.48
     telah
    0.46
     has
    0.43
     seeking
    0.42
     Biomedical
    0.42
     pursuant
    0.42
     Mycobacterium
    0.42
     pursuing
    0.42
     biomedical
    0.41
     undertook
    0.40
    POSITIVE LOGITS
     subtleties
    0.58
     иногда
    0.54
     fáciles
    0.53
     subtlety
    0.52
     nuance
    0.51
     diferença
    0.50
     czasem
    0.50
     když
    0.50
     разными
    0.50
     parfois
    0.50
    Act Density 2.124%

    No Known Activations