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

    Neuronpedia

    INDEX
    Explanations

    verbs in the infinitive form

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "to be" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Immunity
    -0.71
    antry
    -0.70
     Moose
    -0.69
    orum
    -0.65
     Stain
    -0.65
    osa
    -0.63
     Signal
    -0.61
     fray
    -0.60
    vas
    -0.59
     Corpus
    -0.58
    POSITIVE LOGITS
     reckoned
    1.15
     enjoyed
    1.00
     avoided
    0.99
     explored
    0.97
     depended
    0.97
     regretted
    0.97
     admired
    0.97
     gained
    0.93
     desired
    0.93
     found
    0.93
    Act Density 0.067%

    No Known Activations