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

    Neuronpedia

    INDEX
    Explanations

    phrases related to matters of influence or containment

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "conf" in various forms and contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    DAY
    -0.78
    hyde
    -0.78
    Ô
    -0.73
    senal
    -0.72
    ï¸ı
    -0.69
    terday
    -0.64
    WARE
    -0.59
     ende
    -0.59
    bial
    -0.59
     millenn
    -0.58
    POSITIVE LOGITS
    ederation
    1.29
    essional
    1.20
    eder
    1.16
    luence
    1.12
    usions
    1.03
    idences
    1.02
    essor
    0.95
    eren
    0.93
    ining
    0.93
    erences
    0.91
    Act Density 0.016%

    No Known Activations