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

    Neuronpedia

    INDEX
    Explanations

    phrases related to geographical locations being home to a specific entity or community

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating a sense of ownership or being home to entities, populations, or characteristics within a place

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iew
    -0.72
    iev
    -0.70
    iron
    -0.67
    direction
    -0.66
    nir
    -0.64
    RB
    -0.63
    isms
    -0.63
    doi
    -0.63
    Cause
    -0.63
    iry
    -0.62
    POSITIVE LOGITS
     dozens
    0.84
     many
    0.83
     innumerable
    0.81
     countless
    0.81
     numerous
    0.80
     several
    0.79
     plenty
    0.74
     hundreds
    0.73
     thousands
    0.72
     both
    0.69
    Act Density 0.103%

    No Known Activations