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

    Neuronpedia

    INDEX
    Explanations

    how it works

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     played
    0.50
     created
    0.49
     doma
    0.48
     dataset
    0.46
     represented
    0.45
     increased
    0.44
     hvis
    0.44
     mileage
    0.43
     topic
    0.42
     applied
    0.42
    POSITIVE LOGITS
     ሲ
    0.45
    Embed
    0.45
    $:
    0.45
     एक्शन
    0.44
     संकट
    0.44
    }-(\
    0.44
    踏
    0.43
    变革
    0.43
    IRQHandler
    0.42
    ebvre
    0.42
    Act Density 0.001%

    No Known Activations