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

    Neuronpedia

    INDEX
    Explanations

    listing examples

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ्स
    2.22
    מ
    2.08
    ви
    1.84
    م
    1.77
    க
    1.77
    ಲ್ಲಿ
    1.69
    ৬০
    1.69
    м
    1.68
    ని
    1.64
    اں
    1.62
    POSITIVE LOGITS
    ah
    2.11
    iz
    2.03
    let
    2.00
    gence
    1.85
    teilung
    1.84
    qu
    1.79
    ar
    1.76
    a
    1.76
    いた
    1.73
    ena
    1.72
    Act Density 0.119%

    No Known Activations