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

    Neuronpedia

    INDEX
    Explanations

    extracting file types or data entities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     quarts
    0.37
    tene
    0.35
    सेने
    0.35
     realidad
    0.33
    னமாக
    0.33
     treas
    0.33
     garde
    0.32
    时间
    0.32
     quoi
    0.32
     كي
    0.32
    POSITIVE LOGITS
     extract
    0.51
    Extract
    0.46
     extracts
    0.45
     Extract
    0.42
    extract
    0.41
     Extracts
    0.41
     extracted
    0.40
     extraction
    0.40
    Extractor
    0.38
    INITIAL
    0.38
    Act Density 0.013%

    No Known Activations