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

    Neuronpedia

    INDEX
    Explanations

    the verb "compelled" followed by a relevant context or action

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "compelled" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Liber
    -0.71
     Chop
    -0.69
     Parad
    -0.68
     TBD
    -0.66
     Stam
    -0.65
     Trojan
    -0.64
     Grail
    -0.64
     Purpose
    -0.63
     Piece
    -0.62
    tan
    -0.62
    POSITIVE LOGITS
     compelled
    1.30
     obliged
    1.13
     compel
    1.13
     destro
    0.97
    ĸļ
    0.96
    untarily
    0.95
     confir
    0.94
     enthusi
    0.89
     compe
    0.87
    cffff
    0.85
    Act Density 0.008%

    No Known Activations