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

    Neuronpedia

    INDEX
    Explanations

    words related to ignoring or being ignored

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    isher
    -0.80
    uster
    -0.72
    seed
    -0.70
    gans
    -0.67
    gain
    -0.67
    Finish
    -0.66
    ikuman
    -0.65
    gars
    -0.64
    kaya
    -0.64
    iov
    -0.63
    POSITIVE LOGITS
     warnings
    1.14
     inconvenient
    1.10
     pleas
    0.95
     obvious
    0.89
     glaring
    0.89
     cues
    0.89
     caveats
    0.87
     concerns
    0.83
     altogether
    0.81
     objections
    0.81
    Act Density 0.113%

    No Known Activations