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

    Neuronpedia

    INDEX
    Explanations

    phrases or words related to power or influence

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of the word "powerful."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    chal
    -0.86
    forth
    -0.84
    eday
    -0.83
    cision
    -0.77
    ajor
    -0.76
    kay
    -0.73
    den
    -0.72
    ©¶æ
    -0.71
    ey
    -0.70
    aked
    -0.70
    POSITIVE LOGITS
     chords
    0.96
     enough
    0.95
     vested
    0.87
     chord
    0.86
     motiv
    0.82
     impulses
    0.81
     machinery
    0.80
     emotions
    0.79
     emotion
    0.77
     adversary
    0.77
    Act Density 0.038%

    No Known Activations