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

    Neuronpedia

    INDEX
    Explanations

    references to geographical locations, particularly lakes

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "Lake."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    nir
    -0.87
     tremend
    -0.82
    */(
    -0.81
    AAAAAAAA
    -0.77
    imental
    -0.74
    endiary
    -0.73
    CLASSIFIED
    -0.70
    iotics
    -0.70
    nels
    -0.70
    binary
    -0.69
    POSITIVE LOGITS
     Erie
    1.06
    front
    0.93
     trout
    0.89
    wood
    0.89
     Superior
    0.87
     Lake
    0.85
     Nile
    0.84
     Mead
    0.84
     Lakes
    0.82
    bone
    0.80
    Act Density 0.012%

    No Known Activations