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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating finality or conclusion

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Slate
    -0.61
    Spread
    -0.61
    maid
    -0.60
    Lago
    -0.59
     Maxim
    -0.59
     Scrib
    -0.58
     Plum
    -0.55
    folk
    -0.54
     exagger
    -0.54
     varied
    -0.53
    POSITIVE LOGITS
    asio
    0.72
     reconciliation
    0.69
    aterasu
    0.68
    ileaks
    0.66
    awaru
    0.65
     reckoning
    0.64
    iflower
    0.63
     completes
    0.63
     exoner
    0.60
    mony
    0.60
    Act Density 10.288%

    No Known Activations