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

    Neuronpedia

    INDEX
    Explanations

    references to historical or cultural significance

    oai_token-act-pair · gpt-3.5-turbo

    references to heritage and its significance

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    vae
    -0.87
    kers
    -0.86
    err
    -0.76
    ched
    -0.75
    hap
    -0.74
    pher
    -0.73
    alties
    -0.71
    eah
    -0.71
    bern
    -0.68
    arters
    -0.68
    POSITIVE LOGITS
     heritage
    1.06
    conservancy
    0.92
     preservation
    0.83
    artifacts
    0.82
     relics
    0.80
     arte
    0.80
     relic
    0.72
     conserv
    0.70
     quot
    0.69
     richness
    0.69
    Act Density 0.025%

    No Known Activations