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

    Neuronpedia

    INDEX
    Explanations

    how specific entities are handled

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     mengenai
    0.20
     eski
    0.20
     mnist
    0.20
     zuletzt
    0.20
     MonoBehaviour
    0.20
     acquainted
    0.19
    یی
    0.19
    باز
    0.19
     letz
    0.19
     trotz
    0.19
    POSITIVE LOGITS
     separately
    0.37
     самостоятельно
    0.35
     intelligently
    0.33
     directly
    0.32
     differently
    0.32
     cleanly
    0.31
    directly
    0.31
     afresh
    0.31
     securely
    0.30
     cheaply
    0.30
    Act Density 0.359%

    No Known Activations