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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "attend"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "attend."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    smart
    -0.66
    irus
    -0.64
     fatally
    -0.60
     hybrid
    -0.58
     bipolar
    -0.57
    PRESS
    -0.57
    EStreamFrame
    -0.56
     dri
    -0.56
    slice
    -0.56
    cold
    -0.56
    POSITIVE LOGITS
     Attend
    1.10
     attend
    1.07
     attends
    1.00
    eret
    0.93
     RTX
    0.91
     attendance
    0.89
    enance
    0.85
     attended
    0.84
    uates
    0.82
    hips
    0.81
    Act Density 0.009%

    No Known Activations