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

    Neuronpedia

    INDEX
    Explanations

    phrases related to physical lifting or raising objects or people

    oai_token-act-pair · gpt-3.5-turbo

    references to lifting and physical effort

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    915
    -0.68
    erker
    -0.66
    llah
    -0.64
    sg
    -0.61
    pps
    -0.61
    essor
    -0.60
    ucci
    -0.59
    erity
    -0.59
    ymes
    -0.59
     Forever
    -0.59
    POSITIVE LOGITS
     weights
    1.45
    weight
    0.97
    lift
    0.90
     lift
    0.88
     weight
    0.87
    weights
    0.84
     lid
    0.83
     curtain
    0.82
     lifted
    0.82
     curtains
    0.81
    Act Density 0.034%

    No Known Activations