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

    Neuronpedia

    INDEX
    Explanations

    say hello or goodbye

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    মিল
    0.42
    هي
    0.40
    आईएस
    0.40
     punches
    0.39
    ต้า
    0.39
     CIS
    0.38
     Manufact
    0.38
     excitations
    0.38
     dystrophy
    0.38
     Risks
    0.38
    POSITIVE LOGITS
     goodbye
    1.04
     hello
    0.86
     aloud
    0.81
     bonjour
    0.74
     Goodbye
    0.74
     अलविदा
    0.72
    Goodbye
    0.70
     спасибо
    0.62
     farewell
    0.60
    实话
    0.59
    Act Density 0.020%

    No Known Activations