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

    Neuronpedia

    INDEX
    Explanations

    injustice

    np_max-act · gemini-2.0-flash

    The neuron activates on words referring to broad social groups or the general populace (e.g. “people,” “everyone,” “citizen,” “public”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    115
    -0.07
    16
    -0.07
     BitConverter
    -0.07
     convers
    -0.06
    iga
    -0.06
     swung
    -0.06
     blog
    -0.06
    ())↵
    -0.06
     coleg
    -0.06
     SID
    -0.06
    POSITIVE LOGITS
    }$/
    0.07
     iCloud
    0.06
     narrowing
    0.06
     neut
    0.06
    吉
    0.06
    /de
    0.06
    �
    0.06
    Return
    0.06
     варто
    0.06
    kın
    0.06
    Act Density 0.152%

    No Known Activations