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

    Neuronpedia

    INDEX
    Explanations

    code variables

    np_max-act · gemini-2.0-flash

    This neuron fires on the parameter name tokens in setter-method declarations (i.e. the variable names passed to `setXxx(...)`).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    	Me
    -0.07
     Hat
    -0.07
     Vine
    -0.07
    	for
    -0.06
     Elephant
    -0.06
     infinite
    -0.06
     text
    -0.06
    -picker
    -0.06
     multinational
    -0.06
     Table
    -0.06
    POSITIVE LOGITS
    ửi
    0.07
     eriş
    0.06
     Dimit
    0.06
     bumper
    0.06
     označ
    0.06
     etiquette
    0.06
    ůr
    0.06
     Canterbury
    0.06
     бор
    0.06
    (aa
    0.06
    Act Density 0.006%

    No Known Activations