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

    Neuronpedia

    INDEX
    Explanations

    words ending in -ment, -ation, -or, -izer

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ğından
    0.40
    ový
    0.39
    ovou
    0.38
    igraf
    0.38
    olini
    0.38
    ugeot
    0.38
    ko
    0.36
    ombo
    0.36
    kom
    0.36
    roz
    0.36
    POSITIVE LOGITS
    ations
    0.55
    ation
    0.51
    ating
    0.46
    ators
    0.46
    েশন
    0.46
    ization
    0.44
    ments
    0.43
    atory
    0.43
    ative
    0.42
    মেন্ট
    0.42
    Act Density 0.280%

    No Known Activations