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

    Neuronpedia

    INDEX
    Explanations

    statements expressing disagreement or problems with something

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing agreement or acceptance

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    igate
    -0.84
    oenix
    -0.83
    iliated
    -0.82
    kefeller
    -0.80
    veyard
    -0.77
    oided
    -0.75
    inka
    -0.73
    isance
    -0.69
    irmed
    -0.69
    iless
    -0.69
    POSITIVE LOGITS
     wording
    0.90
     characterization
    0.86
     outcome
    0.84
     arrang
    0.83
     portrayal
    0.79
     tack
    0.78
     situation
    0.78
     assumptions
    0.77
     assertion
    0.77
     proposal
    0.77
    Act Density 0.385%

    No Known Activations