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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "spark" and its variations, indicating moments of ignition or inspiration

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

    spark, sparks, sparked

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    classnames
    -0.44
    CDATA
    -0.44
    ########.
    -0.41
     overridden
    -0.40
     noqa
    -0.40
     censiti
    -0.38
    hyrchwyd
    -0.38
     vie
    -0.38
     Niv
    -0.38
    dieu
    -0.38
    POSITIVE LOGITS
     spark
    1.65
     sparks
    1.55
    spark
    1.46
     sparked
    1.41
     Sparks
    1.39
    Sparks
    1.38
     sparking
    1.34
     Spark
    1.34
     SPARK
    1.33
    Spark
    1.28
    Act Density 0.007%

    No Known Activations