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

    Neuronpedia

    INDEX
    Explanations

    references to specific geographic locations, specifically islands

    oai_token-act-pair · gpt-3.5-turbo

    references to specific islands and islanders

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     referen
    -0.71
     clen
    -0.69
     ther
    -0.66
    bers
    -0.65
     MEM
    -0.64
    pher
    -0.64
     vacc
    -0.63
     lig
    -0.61
    aspers
    -0.61
     buckle
    -0.59
    POSITIVE LOGITS
     Island
    3.98
     island
    2.42
     Islands
    2.38
     Islanders
    2.13
     islands
    1.89
     Isle
    1.88
     Isles
    1.70
     Beach
    1.44
     Reef
    1.39
     Harbour
    1.30
    Act Density 0.016%

    No Known Activations