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

    Neuronpedia

    INDEX
    Explanations

    formal declarations or endorsements

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "formally."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rug
    -0.70
     Belt
    -0.67
     Values
    -0.67
     heart
    -0.66
     Eden
    -0.65
     Mile
    -0.65
     Daily
    -0.64
     Gar
    -0.64
     Cris
    -0.64
     Gore
    -0.63
    POSITIVE LOGITS
     exting
    0.89
     formally
    0.86
    idable
    0.85
     addressed
    0.81
    dated
    0.80
     challenged
    0.79
     lodged
    0.78
     contradicted
    0.78
    ntil
    0.77
    lishes
    0.76
    Act Density 0.014%

    No Known Activations