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

    Neuronpedia

    INDEX
    Explanations

    declarative statements or sentences ending with a period

    oai_token-act-pair · gpt-3.5-turbo

    periods and statements that reflect concluding thoughts or summaries

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     glim
    -1.23
     volunte
    -0.93
     carbohyd
    -0.82
     babys
    -0.80
     nude
    -0.79
     rabbit
    -0.78
     extinct
    -0.78
     creek
    -0.78
     bunny
    -0.78
    ancies
    -0.77
    POSITIVE LOGITS
     Moreover
    1.42
     Nevertheless
    1.40
     Meanwhile
    1.38
     Whereas
    1.37
     Consequently
    1.35
     Nonetheless
    1.34
     Yet
    1.33
     Hence
    1.30
     Indeed
    1.27
     However
    1.27
    Act Density 0.450%

    No Known Activations