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

    Neuronpedia

    INDEX
    Explanations

    phrases expressing reluctance or unwillingness

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing desires or intentions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Compass
    -0.80
    grounds
    -0.78
    hooting
    -0.75
    metadata
    -0.67
     Needs
    -0.64
     Kinnikuman
    -0.63
    herer
    -0.62
    Kings
    -0.61
    agonists
    -0.61
     Returning
    -0.60
    POSITIVE LOGITS
     anymore
    1.03
     bother
    1.01
     spoil
    0.92
     offend
    0.92
     celebrate
    0.92
     capitalize
    0.89
     interfere
    0.88
     compete
    0.88
     participate
    0.87
     hear
    0.86
    Act Density 0.081%

    No Known Activations