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

    Neuronpedia

    INDEX
    Explanations

    statements made in professional contexts

    oai_token-act-pair · gpt-3.5-turbo

    the frequent use of the word "in" to indicate location or context within statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    76561
    -0.77
     %%
    -0.77
     combatants
    -0.64
     snipers
    -0.64
     therein
    -0.64
     needs
    -0.62
     ens
    -0.61
     heirs
    -0.59
     learners
    -0.59
     distur
    -0.59
    POSITIVE LOGITS
     lieu
    1.19
     unison
    1.06
    aug
    0.97
     advance
    0.91
    humane
    0.90
    clus
    0.89
     conjunction
    0.89
    cluded
    0.87
     accordance
    0.86
     order
    0.86
    Act Density 0.095%

    No Known Activations