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

    Neuronpedia

    INDEX
    Explanations

    code explanations or code blocks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mobilizing
    0.39
     particuliers
    0.39
    ამის
    0.39
     вследствие
    0.37
    الك
    0.37
    𝒞
    0.36
     नजदी
    0.36
     Deutscher
    0.36
    Ƙ
    0.35
    humans
    0.35
    POSITIVE LOGITS
     script
    0.70
     revised
    0.66
     original
    0.60
     updated
    0.59
     code
    0.59
     inclusion
    0.55
     improved
    0.54
     ORIGINAL
    0.54
    original
    0.52
     revisions
    0.52
    Act Density 0.013%

    No Known Activations