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

    Neuronpedia

    INDEX
    Explanations

    adjectives or adverbs describing actions or qualities in a negative context

    oai_token-act-pair · gpt-3.5-turbo

    instances of "large" creatures in descriptions, particularly related to Pokémon

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    anders
    -0.70
    ategory
    -0.67
    conservancy
    -0.61
    anners
    -0.60
    ulhu
    -0.59
    annis
    -0.59
    illon
    -0.59
    REL
    -0.57
    aterial
    -0.56
     mosqu
    -0.56
    POSITIVE LOGITS
    ly
    3.50
    LY
    2.36
    fully
    1.57
    lys
    1.57
    lies
    1.49
    edly
    1.48
    liness
    1.46
    ELY
    1.41
    ously
    1.36
    lly
    1.36
    Act Density 0.408%

    No Known Activations