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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating agreement or alignment with a specific idea or plan

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating disagreement or division among groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     emergencies
    -0.69
     harvested
    -0.63
    ess
    -0.61
     datas
    -0.61
     lookup
    -0.61
     Resources
    -0.59
    ãĥĺ
    -0.59
    stress
    -0.59
     dinners
    -0.58
     Detail
    -0.58
    POSITIVE LOGITS
     disagree
    0.83
     dismissing
    0.83
     disappro
    0.83
     recommending
    0.82
    essim
    0.81
    Rate
    0.80
     disagrees
    0.80
     approving
    0.79
     endorsing
    0.79
    agree
    0.78
    Act Density 0.434%

    No Known Activations