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

    Neuronpedia

    INDEX
    Explanations

    offer elaboration on any

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    :\
    0.72
    >:</
    0.71
     suivante
    0.70
    :
    0.68
     affiliated
    0.68
     möglichst
    0.68
    具体的な
    0.68
    :<
    0.67
     jeweils
    0.66
     खुशखबरी
    0.65
    POSITIVE LOGITS
     item
    1.10
     부분이
    1.05
     Item
    0.99
    Item
    0.99
     concepts
    0.99
     items
    0.98
     항목
    0.96
     Items
    0.96
     부분을
    0.95
     부분
    0.93
    Act Density 0.069%

    No Known Activations