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

    Neuronpedia

    INDEX
    Explanations

    phrases related to specific locations or buildings

    oai_token-act-pair · gpt-3.5-turbo

    relative clauses referring to specific places or structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    azeera
    -0.73
    obin
    -0.70
     guessing
    -0.69
    Lear
    -0.68
    playing
    -0.68
     Checking
    -0.64
    berman
    -0.64
    calling
    -0.62
    redo
    -0.62
    ulz
    -0.62
    POSITIVE LOGITS
     housed
    1.23
     overlook
    1.14
     reopened
    1.11
     belonged
    1.10
     neighb
    1.08
     occupies
    1.07
     sits
    1.02
    Located
    1.02
     empt
    0.99
     houses
    0.99
    Act Density 0.335%

    No Known Activations