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

    Neuronpedia

    INDEX
    Explanations

    facts or biographical information about notable individuals

    oai_token-act-pair · gpt-3.5-turbo

    sentences that describe individuals or entities in a noteworthy or unique manner

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     conclud
    -0.80
     unnecess
    -0.80
     attrition
    -0.78
     reversible
    -0.78
     tremend
    -0.78
     destro
    -0.76
     feasible
    -0.76
     consolation
    -0.76
     challeng
    -0.76
    ascus
    -0.74
    POSITIVE LOGITS
     Born
    1.68
     He
    1.55
     His
    1.54
     Known
    1.53
    Born
    1.44
     Unlike
    1.34
     Though
    1.32
     Recently
    1.30
     Serving
    1.30
     Whenever
    1.30
    Act Density 0.294%

    No Known Activations