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

    Neuronpedia

    INDEX
    Explanations

    words related to historical or cultural significance

    oai_token-act-pair · gpt-3.5-turbo

    references to cultural heritage

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    vae
    -0.94
    kers
    -0.78
    alties
    -0.69
    utsu
    -0.68
    err
    -0.67
    plet
    -0.66
    ules
    -0.66
    secut
    -0.65
    ching
    -0.65
    pher
    -0.65
    POSITIVE LOGITS
     heritage
    1.10
    conservancy
    1.01
     preservation
    0.85
    artifacts
    0.76
     arte
    0.73
    stones
    0.72
     relics
    0.71
     corpus
    0.70
     legacy
    0.68
     richness
    0.68
    Act Density 0.010%

    No Known Activations