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

    Neuronpedia

    INDEX
    Explanations

    particular characters within the text ('Ċ')

    oai_token-act-pair · gpt-3.5-turbo

    phrases highlighting socio-political critiques or contradictions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     nab
    -0.79
     aquarium
    -0.79
     beverage
    -0.76
     amph
    -0.74
     hatch
    -0.73
     paddle
    -0.71
     fres
    -0.71
     lookout
    -0.70
    ogly
    -0.70
     diving
    -0.70
    POSITIVE LOGITS
    Moreover
    1.26
    Therefore
    1.25
    Furthermore
    1.20
    Thus
    1.18
    Consider
    1.14
    ³³³³³³³³
    1.10
    Such
    1.10
    Yet
    1.10
    Indeed
    1.10
    Conclusion
    1.08
    Act Density 0.628%

    No Known Activations