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

    Neuronpedia

    INDEX
    Explanations

    Foundation and Association names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ש
    1.05
    ח
    0.99
     respectable
    0.97
    지가
    0.93
     traders
    0.91
    殲
    0.89
     respecter
    0.88
    রক্ষ
    0.88
    вших
    0.88
    হি
    0.88
    POSITIVE LOGITS
    drawable
    1.01
    d
    1.00
    CISE
    0.98
    STRING
    0.96
    ObjectModel
    0.95
    .}$
    0.94
    Jeg
    0.93
    笑顔
    0.93
     Такие
    0.93
    ্স
    0.92
    Act Density 0.022%

    No Known Activations