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

    Neuronpedia

    INDEX
    Explanations

    conjunctions indicating contrast or exception

    oai_token-act-pair · gpt-3.5-turbo

    discourse markers that introduce contrasting information

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     segment
    -0.77
     Cathedral
    -0.60
     oriented
    -0.59
     dedicated
    -0.57
     mound
    -0.57
     perceived
    -0.57
     trad
    -0.56
     segments
    -0.56
     commentary
    -0.56
     hall
    -0.55
    POSITIVE LOGITS
    tons
    1.80
    tery
    1.18
    ts
    1.15
    chers
    1.05
    tern
    1.02
    chery
    1.02
     alas
    1.01
    ylene
    0.98
    tle
    0.98
    cher
    0.94
    Act Density 0.103%

    No Known Activations