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

    Neuronpedia

    INDEX
    Explanations

    personal details or characteristics of individuals

    oai_token-act-pair · gpt-3.5-turbo

    pronouns and indicative verbs suggesting individuals and their actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Adjust
    -0.69
    mast
    -0.69
    cible
    -0.63
    grid
    -0.63
    verse
    -0.63
    sweet
    -0.61
    pract
    -0.60
    Repeat
    -0.59
    ventory
    -0.59
    Herm
    -0.59
    POSITIVE LOGITS
    reditary
    1.00
    pherd
    0.84
    enegger
    0.79
    zek
    0.77
    'll
    0.75
     specialized
    0.74
     wrote
    0.74
    lain
    0.74
     specializes
    0.72
     chairs
    0.72
    Act Density 0.331%

    No Known Activations