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

    Neuronpedia

    INDEX
    Explanations

    prepositions "to."

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating a direction or intention

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hur
    -0.47
     packets
    -0.45
    listed
    -0.42
     flagged
    -0.42
    hops
    -0.41
    calling
    -0.41
     Ended
    -0.40
    Ĥª
    -0.40
     Donation
    -0.40
     leaked
    -0.39
    POSITIVE LOGITS
     minimize
    0.89
    ggles
    0.88
     maximize
    0.87
     accommodate
    0.84
     ensure
    0.84
     achieve
    0.82
     satisfy
    0.82
     insure
    0.82
     compensate
    0.81
     conserve
    0.81
    Act Density 0.725%

    No Known Activations