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

    Neuronpedia

    INDEX
    Explanations

    phrases related to healthcare or medical procedures

    oai_token-act-pair · gpt-3.5-turbo

    elements related to financial transactions and evaluations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Defense
    -1.21
    defense
    -1.10
     realize
    -0.99
     realized
    -0.98
     afterward
    -0.98
     recognize
    -0.96
     someday
    -0.96
    avored
    -0.95
     favorably
    -0.94
     honored
    -0.93
    POSITIVE LOGITS
     tyres
    1.68
     £
    1.60
     tyre
    1.58
     petrol
    1.53
     aluminium
    1.49
     licence
    1.44
     authorised
    1.44
     metre
    1.43
     licences
    1.42
     rubbish
    1.42
    Act Density 0.574%

    No Known Activations