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

    Neuronpedia

    INDEX
    Explanations

    phrases related to making suggestions

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "suggest" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ccording
    -0.71
    percent
    -0.70
    cler
    -0.70
    pes
    -0.68
    fare
    -0.67
    gart
    -0.65
    Boy
    -0.65
    wig
    -0.65
    respective
    -0.64
    estine
    -0.64
    POSITIVE LOGITS
     suggestions
    0.76
    ively
    0.76
    ibility
    0.73
    ible
    0.71
     hypot
    0.70
    ibilities
    0.66
    eele
    0.65
    ibly
    0.64
     otherwise
    0.64
    esson
    0.64
    Act Density 0.047%

    No Known Activations