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

    Neuronpedia

    INDEX
    Explanations

    positive attributes or qualities

    oai_token-act-pair · gpt-3.5-turbo

    various conjunctions and phrases that qualify or describe characteristics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    onica
    -0.88
    atives
    -0.78
    prise
    -0.75
    udence
    -0.74
    mma
    -0.74
    olicy
    -0.72
    inar
    -0.71
    acas
    -0.71
     Else
    -0.71
    ctuary
    -0.71
    POSITIVE LOGITS
     furthermore
    1.11
     moreover
    1.09
     secondly
    1.07
     boasts
    1.03
     contains
    1.03
     possesses
    1.00
     consequently
    1.00
     therefore
    0.99
     ooz
    0.97
     hence
    0.96
    Act Density 0.377%

    No Known Activations