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

    Neuronpedia

    INDEX
    Explanations

    phrases related to conditions, requests, or possibilities using the word "if"

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Clicker
    -1.14
    axy
    -1.05
     Deity
    -1.04
    versive
    -1.03
    FTWARE
    -0.96
    ohydrate
    -0.96
    ultane
    -0.96
    advertisement
    -0.94
    STAR
    -0.93
    Poké
    -0.93
    POSITIVE LOGITS
     anyone
    1.23
     someone
    1.18
     somebody
    1.16
     anybody
    1.15
    fy
    1.13
     landlords
    1.09
     policymakers
    1.08
     they
    1.04
     McKin
    1.03
     someday
    1.01
    Act Density 0.405%

    No Known Activations