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

    Neuronpedia

    INDEX
    Explanations

    action verbs followed by an infinitive phrase denoting purpose or goal

    oai_token-act-pair · gpt-3.5-turbo

    statements of goals and objectives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    assian
    -0.70
    Guard
    -0.69
     trib
    -0.69
    fe
    -0.67
    ustomed
    -0.66
    yes
    -0.66
     Leban
    -0.64
    assed
    -0.63
    fax
    -0.61
    pite
    -0.61
    POSITIVE LOGITS
    imize
    0.72
     SPACE
    0.69
     KILL
    0.68
     SOM
    0.66
     optimizing
    0.65
     educating
    0.65
    ongevity
    0.64
     maximizing
    0.64
     conserve
    0.64
     Humanity
    0.64
    Act Density 0.267%

    No Known Activations