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

    Neuronpedia

    INDEX
    Explanations

    extract information

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     açıkl
    0.45
    摻
    0.38
    ambahan
    0.38
    ائو
    0.38
     adição
    0.37
     disprove
    0.37
    俢
    0.37
    ckenridge
    0.36
    充足
    0.36
    संदर्भ
    0.35
    POSITIVE LOGITS
     extraction
    2.34
     extracted
    2.30
     extracting
    2.28
     extract
    2.19
    提取
    2.13
     Extract
    2.09
     Extraction
    2.06
    Extract
    2.02
    extract
    1.99
    extracted
    1.96
    Act Density 0.063%

    No Known Activations