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

    Neuronpedia

    INDEX
    Explanations

    phrases related to appointments or assignments

    oai_token-act-pair · gpt-3.5-turbo

    instances of the preposition "to" to indicate purpose or direction

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     flakes
    -0.79
     packets
    -0.71
     wastes
    -0.68
     gaps
    -0.67
    hops
    -0.67
    pointers
    -0.63
     scales
    -0.61
     glitches
    -0.61
     casc
    -0.61
    chains
    -0.61
    POSITIVE LOGITS
     commemorate
    1.05
     celebrate
    1.00
     accommodate
    1.00
     promote
    0.98
     preserve
    0.97
    pload
    0.96
     provide
    0.95
     protect
    0.93
     minimize
    0.92
     investigate
    0.92
    Act Density 0.315%

    No Known Activations