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

    Neuronpedia

    INDEX
    Explanations

    phrases related to past and established practices or situations

    oai_token-act-pair · gpt-3.5-turbo

    references to longstanding practices or topics that have been previously established

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ogie
    -0.79
     saddle
    -0.68
    illas
    -0.67
    owed
    -0.65
     pies
    -0.63
    emo
    -0.63
    ibles
    -0.63
    aredevil
    -0.62
     flock
    -0.62
    addons
    -0.61
    POSITIVE LOGITS
    ifact
    0.77
    soType
    0.75
     CrossRef
    0.70
    ortium
    0.70
     Reincarn
    0.70
     redacted
    0.68
    âĢ¢âĢ¢
    0.68
     alas
    0.67
     rewritten
    0.66
    APTER
    0.66
    Act Density 0.862%

    No Known Activations