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

    Neuronpedia

    INDEX
    Explanations

    references to physical holes or openings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     radio
    -0.41
    
    -0.41
     hydro
    -0.39
     pumpkin
    -0.39
     Veen
    -0.39
     ancient
    -0.36
     primary
    -0.36
     Afro
    -0.35
     extro
    -0.35
     super
    -0.35
    POSITIVE LOGITS
     hole
    1.38
     Hole
    1.32
     HOLE
    1.28
     holes
    1.23
    hole
    1.22
    Hole
    1.22
    holes
    1.17
    Holes
    1.16
    HOLE
    1.15
     Holes
    1.07
    Act Density 0.077%

    No Known Activations