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

    Neuronpedia

    INDEX
    Explanations

    words related to characters revealing their identities and intentions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Wicidata
    -0.71
    ScopeManager
    -0.69
     sometimes
    -0.68
     meines
    -0.68
     ibland
    -0.65
    sometimes
    -0.64
     kadang
    -0.62
     soms
    -0.61
     quelquefois
    -0.60
    kadang
    -0.59
    POSITIVE LOGITS
     overheard
    0.80
     reveals
    0.78
     Meanwhile
    0.75
     convinces
    0.74
     confesses
    0.74
     revealed
    0.73
     blackmail
    0.73
     flashback
    0.73
     disguised
    0.72
     threatens
    0.71
    Act Density 0.406%

    No Known Activations