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

    Neuronpedia

    INDEX
    Explanations

    key decrease availability weight changes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    (
    0.52
    pek
    0.43
     espect
    0.42
    iqu
    0.40
     torres
    0.40
     bietet
    0.40
    scrollTop
    0.40
    不是
    0.40
     título
    0.39
    łoż
    0.39
    POSITIVE LOGITS
    drivers
    0.46
    уса
    0.46
    พัฒ
    0.45
     EMEA
    0.45
     SSD
    0.44
    जलवायु
    0.44
    тину
    0.44
    тове
    0.44
    喲
    0.44
     ORNL
    0.44
    Act Density 0.001%

    No Known Activations