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

    Neuronpedia

    INDEX
    Explanations

    references to cleanliness and related concepts

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

    clean state or action

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     فيما
    -0.44
     postData
    -0.43
     kaarangay
    -0.43
     whom
    -0.42
     groupBox
    -0.42
    nextSibling
    -0.41
    maktadır
    -0.40
     period
    -0.40
     Caso
    -0.40
    iconque
    -0.39
    POSITIVE LOGITS
     clean
    1.90
    clean
    1.84
    Clean
    1.80
     Clean
    1.79
     CLEAN
    1.63
    CLEAN
    1.47
     limpio
    1.16
     cleanest
    1.15
     pulito
    1.13
     cleanly
    1.08
    Act Density 0.004%

    No Known Activations