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

    Neuronpedia

    INDEX
    Explanations

    words related to the positioning or placing of objects

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "position" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     darn
    -0.68
     whisk
    -0.66
    IN
    -0.64
    inav
    -0.64
     hypert
    -0.63
    aith
    -0.63
     hal
    -0.62
    GA
    -0.61
    olls
    -0.60
     damn
    -0.59
    POSITIVE LOGITS
    position
    1.33
    xual
    1.06
    ngth
    1.00
    itions
    0.90
    eering
    0.90
    eus
    0.90
    terday
    0.84
    itiveness
    0.83
    itional
    0.78
    itious
    0.78
    Act Density 0.007%

    No Known Activations