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

    Neuronpedia

    INDEX
    Explanations

    phrases related to a specific type of solution or approach

    oai_token-act-pair · gpt-3.5-turbo

    phrases referring to a specific category or type of subject matter

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    WOR
    -0.71
     Bars
    -0.70
    ModLoader
    -0.70
    adden
    -0.70
     Balls
    -0.70
     Doors
    -0.68
    agnar
    -0.66
    estern
    -0.64
    YR
    -0.64
    ampions
    -0.63
    POSITIVE LOGITS
    lihood
    0.95
    ilege
    0.82
    hearted
    0.79
    liest
    0.79
    etting
    0.77
    lier
    0.76
    ileged
    0.76
    liness
    0.71
    ship
    0.71
    face
    0.70
    Act Density 0.028%

    No Known Activations