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

    Neuronpedia

    INDEX
    Explanations

    references to power-related terms in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    powered

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Gruß
    -0.40
     vell
    -0.38
     Recife
    -0.37
     []:
    -0.37
    ,:)
    -0.36
     Jérusalem
    -0.36
     mistakes
    -0.35
    ▌
    -0.35
     pytest
    -0.35
    EDEFAULT
    -0.35
    POSITIVE LOGITS
     powered
    2.25
     Powered
    2.19
    powered
    2.16
    Powered
    2.11
     fueled
    1.06
    powering
    1.05
     fuelled
    1.03
     powering
    1.02
     powdered
    0.91
     propelled
    0.89
    Act Density 0.008%

    No Known Activations