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

    Neuronpedia

    INDEX
    Explanations

    references to geographical locations, specifically villages

    oai_token-act-pair · gpt-3.5-turbo

    references to villages

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    bol
    -0.76
    ————————
    -0.70
    ————————————————
    -0.67
    cer
    -0.67
    rium
    -0.67
    reme
    -0.65
    role
    -0.64
    sem
    -0.63
    given
    -0.62
    Entry
    -0.62
    POSITIVE LOGITS
     villages
    1.19
     villagers
    0.84
    folk
    0.82
     village
    0.78
    ushi
    0.77
     dwellings
    0.77
     countryside
    0.75
    ongs
    0.74
     towns
    0.74
     valleys
    0.73
    Act Density 0.007%

    No Known Activations