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

    Neuronpedia

    INDEX
    Explanations

    academic language related to structured analysis and discussion of healthcare and population studies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    statements in formal, analytic or scientific prose—often using first/second-person pronouns, predictive/causal language, and strong punctuation that marks structured, information-dense claims.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Negative Logits
     colours
    -0.41
    ๆ
    -0.35
     neighbours
    -0.35
     grossen
    -0.35
     colour
    -0.34
     particolari
    -0.34
    Colours
    -0.34
     colourful
    -0.34
     policemen
    -0.34
    　　
    -0.33
    POSITIVE LOGITS
     erſt
    0.73
     ddelwed
    0.70
    [@BOS@]
    0.68
    <unused52>
    0.68
    <unused41>
    0.68
    <unused14>
    0.67
    <unused43>
    0.67
    <unused8>
    0.67
    <unused79>
    0.67
    <unused3>
    0.67
    Act Density 0.328%

    No Known Activations