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

    Neuronpedia

    INDEX
    Explanations

    numerical details and technical information in a financial or statistical context

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    advertisement
    -1.18
    /#
    -1.14
    lift
    -1.13
    MAP
    -1.09
    named
    -1.08
    ACA
    -1.07
    MAS
    -1.07
    WHO
    -1.06
    Recomm
    -1.06
    CHAT
    -1.02
    POSITIVE LOGITS
     practicable
    1.35
     possible
    1.32
     distances
    1.15
    afort
    1.10
     regards
    1.09
    iversal
    1.08
     latitude
    1.00
     outsiders
    0.99
     permissible
    0.99
    ylum
    0.95
    Act Density 0.138%

    No Known Activations