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

    Neuronpedia

    INDEX
    Explanations

    prepositional phrases indicating a place or location

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating prolonged duration or presence in a location

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     accordingly
    -0.76
    llor
    -0.71
    quit
    -0.69
    dule
    -0.69
    inion
    -0.68
    yss
    -0.67
    lly
    -0.66
     shouldn
    -0.64
    ido
    -0.63
    soever
    -0.63
    POSITIVE LOGITS
    effic
    1.06
     situ
    0.96
    efficiency
    0.93
    clusions
    0.90
     action
    0.90
     relation
    0.89
     silhouette
    0.86
     conjunction
    0.84
    animate
    0.84
    actions
    0.83
    Act Density 0.228%

    No Known Activations