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

    Neuronpedia

    INDEX
    Explanations

    phrases that describe actions or behaviors in a specific manner

    oai_token-act-pair · gpt-3.5-turbo

    phrases that describe conditions or actions involving "that" followed by a specific outcome or manner

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Ruk
    -0.84
    yne
    -0.70
     Breast
    -0.67
     Corpus
    -0.64
     Hunting
    -0.64
     Wife
    -0.64
    MQ
    -0.64
    AY
    -0.63
     Bott
    -0.63
     Thoughts
    -0.62
    POSITIVE LOGITS
     extends
    0.85
     resembles
    0.84
    ertodd
    0.83
     consumes
    0.83
     enables
    0.82
     enhances
    0.78
     satisfies
    0.77
     allows
    0.76
     involves
    0.76
     suits
    0.75
    Act Density 0.218%

    No Known Activations