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

    Neuronpedia

    INDEX
    Explanations

    verbs followed by "to" indicating an action or intention

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating a tendency or habit

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    listed
    -0.86
    rieved
    -0.66
    irez
    -0.66
    anooga
    -0.66
    went
    -0.65
    opened
    -0.65
     Registry
    -0.64
    Letter
    -0.63
    FILE
    -0.63
    posted
    -0.63
    POSITIVE LOGITS
     stick
    0.98
     rely
    0.89
     pick
    0.88
     resemble
    0.88
     attract
    0.88
     accumulate
    0.88
     prioritize
    0.87
     improve
    0.86
     find
    0.84
     lose
    0.84
    Act Density 0.074%

    No Known Activations