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

    Neuronpedia

    INDEX
    Explanations

    positive adjectives related to experiences or sensations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "pleasant" and its variations, along with contextually related terms like "unpleasant."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    NUM
    -0.67
    aucus
    -0.65
    ULE
    -0.64
    ithing
    -0.63
     helic
    -0.63
    FIL
    -0.62
     HELP
    -0.62
    flex
    -0.62
    ARS
    -0.62
     Estimates
    -0.62
    POSITIVE LOGITS
    ries
    1.16
     surprises
    0.96
     pleasant
    0.89
    ties
    0.89
    lihood
    0.89
    ness
    0.88
     smelling
    0.80
    terness
    0.77
    istic
    0.75
     unpleasant
    0.75
    Act Density 0.020%

    No Known Activations