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

    Neuronpedia

    INDEX
    Explanations

    initializes, listing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     purely
    0.43
     vroeg
    0.43
     logically
    0.42
     varie
    0.42
     kult
    0.40
     prevede
    0.40
     literally
    0.39
     ber
    0.39
     legit
    0.39
     saa
    0.38
    POSITIVE LOGITS
    訢
    0.47
    Компания
    0.43
    PopupMenu
    0.43
    Cohort
    0.43
    EventRecorder
    0.43
    Owned
    0.42
    Ƣ
    0.41
    停
    0.40
    🗓
    0.40
    ୮
    0.39
    Act Density 0.004%

    No Known Activations