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

    Neuronpedia

    INDEX
    Explanations

    phrases involving comparison or addition

    oai_token-act-pair · gpt-3.5-turbo

    repeated phrases indicating comparison or contrast

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    "well as" in "as well as"

    Explanation Uploaded by User
    New Auto-Interp
    Negative Logits
    hr
    -0.74
    ANS
    -0.74
    eno
    -0.73
    spot
    -0.73
    arat
    -0.72
    century
    -0.72
    enium
    -0.72
    god
    -0.71
    xit
    -0.70
    awar
    -0.70
    POSITIVE LOGITS
     optionally
    0.92
     assorted
    0.85
     possibly
    0.79
     alternatively
    0.78
     occasionally
    0.76
     vice
    0.73
     occasional
    0.72
     preferably
    0.71
     whatever
    0.70
     anecdotal
    0.70
    Act Density 0.106%

    No Known Activations