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

    Neuronpedia

    INDEX
    Explanations

    conversational language

    np_max-act · gemini-2.0-flash

    The neuron detects phrases about media or public reaction (e.g. mentions of “media,” “outlets,” “react,” “report,” “would”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     MAP
    -0.07
     Lion
    -0.07
     все
    -0.07
     bowed
    -0.06
     nested
    -0.06
    ườ
    -0.06
     Versions
    -0.06
    	reload
    -0.06
     Salon
    -0.06
    kara
    -0.06
    POSITIVE LOGITS
     看
    0.07
    ======↵
    0.06
     figura
    0.06
    (find
    0.06
    胡
    0.06
     skins
    0.06
     neighbours
    0.06
    尽管
    0.06
    Aus
    0.06
     downfall
    0.06
    Act Density 0.052%

    No Known Activations