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

    Neuronpedia

    INDEX
    Explanations

    phrases expressing accomplishment or action

    oai_token-act-pair · gpt-3.5-turbo

    descriptive verbs or phrases pertaining to actions or states of being

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    CLUD
    -0.68
    ston
    -0.64
    unal
    -0.62
    ãĤ¹
    -0.61
    rones
    -0.61
    igor
    -0.61
    urations
    -0.60
    *=-
    -0.60
    atus
    -0.60
    locked
    -0.59
    POSITIVE LOGITS
     create
    1.14
     revise
    1.14
     educate
    1.10
     introduce
    1.10
     imitate
    1.07
     legalize
    1.06
     emulate
    1.05
     prepare
    1.05
     nominate
    1.05
     organize
    1.04
    Act Density 0.170%

    No Known Activations