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

    Neuronpedia

    INDEX
    Explanations

    purpose or goal related phrases

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express desires or goals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    antha
    -0.73
     transcripts
    -0.64
    average
    -0.61
    guards
    -0.60
    ensis
    -0.60
    mint
    -0.60
     Guards
    -0.60
    notations
    -0.60
     clauses
    -0.59
     Leban
    -0.59
    POSITIVE LOGITS
     vengeance
    1.26
     revenge
    1.09
     pursuit
    1.09
     Purpose
    1.06
     endeavour
    1.05
     goal
    0.99
     endeavor
    0.98
     crusade
    0.97
     endeav
    0.95
     aim
    0.95
    Act Density 0.963%

    No Known Activations