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

    Neuronpedia

    INDEX
    Explanations

    words with strong emotional valence

    oai_token-act-pair · gemini-2.0-flash

    words related to public themes

    np_token-act-pair-logits · gpt-4o-mini

    something gone wrong/public

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     oprot
    -0.58
    AddHtmlAttribute
    -0.57
     nahilalakip
    -0.56
    ัติ
    -0.54
    Sucesor
    -0.53
    adpleegd
    -0.51
    sibilité
    -0.51
    SourceChecksum
    -0.50
     stabilisation
    -0.50
     Référence
    -0.50
    POSITIVE LOGITS
     public
    0.73
    úblico
    0.61
     green
    0.60
     wild
    0.60
    wild
    0.56
    public
    0.54
    nuts
    0.54
     öffentlichen
    0.54
     lenker
    0.54
    ública
    0.53
    Act Density 1.621%

    No Known Activations