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

    Neuronpedia

    INDEX
    Explanations

    words related to groundskeeping or specific locations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "grounds" and its various uses in context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    apt
    -0.70
     Sins
    -0.67
    thal
    -0.65
    arus
    -0.63
     Scotia
    -0.62
     Sachs
    -0.62
     sadd
    -0.62
     barrel
    -0.61
     Abyssal
    -0.60
     calib
    -0.60
    POSITIVE LOGITS
    keepers
    1.07
    keeper
    1.03
     grounds
    0.97
    grounds
    0.94
    ground
    0.89
    keeping
    0.86
    hog
    0.81
    fare
    0.77
    piece
    0.75
    GROUND
    0.74
    Act Density 0.026%

    No Known Activations