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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating disagreement or differing perspectives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Down
    -0.15
     DOWN
    -0.15
     Vac
    -0.14
     vac
    -0.14
     down
    -0.14
    ridor
    -0.14
     Blank
    -0.14
     McCartney
    -0.14
     tube
    -0.13
     Decompiled
    -0.13
    POSITIVE LOGITS
     outside
    0.21
    outside
    0.20
     independent
    0.20
     independ
    0.19
     recur
    0.19
    å¤ĸ
    0.19
    çĭ¬ç«ĭ
    0.18
     external
    0.18
     recurrence
    0.18
    Outside
    0.18
    Act Density 0.056%

    No Known Activations