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

    Neuronpedia

    INDEX
    Explanations

    adjectives conveying strong negative sentiment

    oai_token-act-pair · gpt-3.5-turbo

    descriptions of negative experiences or qualities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    VO
    -0.90
    tein
    -0.84
    cius
    -0.83
    trl
    -0.83
    PT
    -0.78
    incinn
    -0.77
    venants
    -0.77
    xtap
    -0.77
    zinski
    -0.73
    Lago
    -0.72
    POSITIVE LOGITS
     awful
    0.96
    ness
    0.87
    nesses
    0.87
     adolesc
    0.84
     lot
    0.83
     darn
    0.82
     smelling
    0.80
     traged
    0.79
     metic
    0.78
     crap
    0.78
    Act Density 0.009%

    No Known Activations