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

    Neuronpedia

    INDEX
    Explanations

    options

    np_max-act · gemini-2.0-flash

    This neuron activates on words meaning “options” (i.e. nouns ending in –ion/–ione/–ción/–ção) across several languages.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    妮
    -0.07
    	buffer
    -0.07
    address
    -0.07
    Frequency
    -0.07
    acomment
    -0.06
     broadcast
    -0.06
    NSString
    -0.06
     charity
    -0.06
     Hotel
    -0.06
     благод
    -0.06
    POSITIVE LOGITS
     options
    0.12
     option
    0.12
     alternatives
    0.08
     OPTIONS
    0.08
     Options
    0.08
     opciones
    0.08
     possibilities
    0.08
     happily
    0.07
     Option
    0.07
     choice
    0.07
    Act Density 0.036%

    No Known Activations