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

    Neuronpedia

    INDEX
    Explanations

    phrases comparing or contrasting different entities

    oai_token-act-pair · gpt-3.5-turbo

    instances of comparison or contrast between subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĤ¼ãĤ¦ãĤ¹
    -0.68
    lift
    -0.67
    igham
    -0.65
    aeda
    -0.64
    amaz
    -0.63
    ¬¼
    -0.62
    ablishment
    -0.61
    iously
    -0.61
    Deal
    -0.61
     exting
    -0.60
    POSITIVE LOGITS
     however
    1.19
     which
    1.01
     whose
    0.96
     there
    0.93
     namely
    0.88
     though
    0.87
     whom
    0.84
     moreover
    0.83
     meanwhile
    0.82
     wherein
    0.82
    Act Density 0.140%

    No Known Activations