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

    Neuronpedia

    INDEX
    Explanations

    warmth and positivity in descriptions or interactions

    oai_token-act-pair · gpt-3.5-turbo

    warm adjectives or phrases describing comfort and friendliness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    IBLE
    -0.67
     Chaff
    -0.64
     unlaw
    -0.63
    arge
    -0.62
     bankrupt
    -0.62
    $$
    -0.62
    issors
    -0.61
     legalized
    -0.61
     Goff
    -0.61
    gur
    -0.61
    POSITIVE LOGITS
    achine
    1.47
    est
    1.08
     fuzz
    1.02
     fuzzy
    1.02
    hearted
    0.96
    estone
    0.90
    blooded
    0.89
    ening
    0.88
    ests
    0.85
     welcome
    0.84
    Act Density 0.026%

    No Known Activations