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

    Neuronpedia

    INDEX
    Explanations

    phrases related to personal preferences or favorites

    oai_token-act-pair · gpt-3.5-turbo

    statements that express personal preferences or favorites

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    uum
    -0.79
    iates
    -0.70
    urches
    -0.69
    iate
    -0.68
    ients
    -0.67
    verett
    -0.66
     merits
    -0.65
    ynes
    -0.64
     harms
    -0.63
    ples
    -0.62
    POSITIVE LOGITS
     undoubtedly
    1.00
     definitely
    0.84
     probably
    0.83
    ometric
    0.81
     indeed
    0.80
     called
    0.77
    ovie
    0.73
     doubtless
    0.73
     Solitaire
    0.72
    abi
    0.71
    Act Density 0.198%

    No Known Activations