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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating reported speech or citations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    gab
    -0.55
     pers
    -0.54
     Wer
    -0.50
    Wer
    -0.48
     off
    -0.46
    mittel
    -0.45
    isc
    -0.45
     Department
    -0.45
     bau
    -0.45
    ork
    -0.44
    POSITIVE LOGITS
    saying
    1.18
     saying
    1.17
     Saying
    1.14
    Saying
    1.11
    say
    1.10
    SAY
    1.08
     say
    1.08
     SAY
    1.06
     says
    1.04
     Says
    1.00
    Act Density 0.171%

    No Known Activations