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

    Neuronpedia

    INDEX
    Explanations

    negations and emphatic assertions in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    CUL
    -0.49
     CER
    -0.49
     
    -0.48
      
    -0.48
    PY
    -0.48
     
    -0.47
     BIB
    -0.46
     Gees
    -0.46
     GAM
    -0.46
     PY
    -0.46
    POSITIVE LOGITS
     BOTH
    1.57
     ALWAYS
    1.56
     REALLY
    1.56
     VERY
    1.55
     ANYTHING
    1.52
     ANY
    1.52
     NEVER
    1.51
     MANY
    1.51
     ONLY
    1.50
     EVERY
    1.50
    Act Density 0.498%

    No Known Activations