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

    Neuronpedia

    INDEX
    Explanations

    instances of chaotic or erratic behavior and their implications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     identical
    -0.16
     Sweep
    -0.15
     hvis
    -0.15
     Twist
    -0.14
    _iff
    -0.14
     WithEvents
    -0.14
    è«
    -0.13
     unexpectedly
    -0.13
     eskort
    -0.13
    706
    -0.13
    POSITIVE LOGITS
     random
    0.32
    /random
    0.27
     randomly
    0.27
     Random
    0.26
    -random
    0.24
    Random
    0.24
    	random
    0.24
    random
    0.23
     chaotic
    0.23
    _random
    0.23
    Act Density 0.269%

    No Known Activations