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

    Neuronpedia

    INDEX
    Explanations

    statements related to understanding and communication

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    edback
    -0.16
    pson
    -0.14
    วà¸Ļ
    -0.14
    eselect
    -0.14
    ë¹
    -0.13
     Decomp
    -0.13
    liÄį
    -0.13
    gue
    -0.13
    lush
    -0.13
    Qed
    -0.13
    POSITIVE LOGITS
     understanding
    0.89
     understand
    0.83
     Understanding
    0.76
     understood
    0.75
     understands
    0.75
    Understanding
    0.69
     Understand
    0.68
    çIJĨ解
    0.64
    -under
    0.60
    under
    0.54
    Act Density 0.352%

    No Known Activations