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

    Neuronpedia

    INDEX
    Explanations

    racial and derogatory terms

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

    racial slurs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     initial
    -0.29
    INITIAL
    -0.26
     Initial
    -0.26
     suitability
    -0.25
     mix
    -0.25
    line
    -0.25
    ][
    -0.24
    ?
    -0.24
     calon
    -0.24
     herself
    -0.23
    POSITIVE LOGITS
    ConstraintMaker
    1.11
     nigga
    1.09
     nigger
    1.05
    UserScript
    0.98
     AssemblyTitle
    0.94
    Nig
    0.91
     Nig
    0.90
     niggas
    0.90
    richTextPanel
    0.87
    findpost
    0.84
    Act Density 0.070%

    No Known Activations