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

    Neuronpedia

    INDEX
    Explanations

    cut and following words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    hind
    -0.11
    oice
    -0.10
    íĮ
    -0.09
    eur
    -0.09
    izzo
    -0.09
    apolis
    -0.09
    esco
    -0.09
    edBy
    -0.09
    oled
    -0.09
    eum
    -0.09
    POSITIVE LOGITS
    aneous
    0.24
    ters
    0.23
    throat
    0.22
    tings
    0.21
    ting
    0.20
    tle
    0.18
    scenes
    0.18
    omers
    0.18
    cut
    0.17
    eness
    0.17
    Act Density 0.026%

    No Known Activations