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

    Neuronpedia

    INDEX
    Explanations

    phrases related to love and romance

    oai_token-act-pair · gpt-3.5-turbo

    references to famous television shows and movies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     stewards
    -0.84
     podium
    -0.77
     sidew
    -0.75
     consolidation
    -0.75
     boosters
    -0.71
     redundancy
    -0.71
    aciously
    -0.70
     immersion
    -0.69
     hunger
    -0.69
     popul
    -0.68
    POSITIVE LOGITS
     Lonely
    1.05
     Murder
    0.97
     Love
    0.97
     Wonderful
    0.96
     Feeling
    0.96
     Fairy
    0.93
     Pretty
    0.93
     Certain
    0.92
     Seasons
    0.92
     Witches
    0.91
    Act Density 0.142%

    No Known Activations