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

    Neuronpedia

    INDEX
    Explanations

    concepts related to correctness and appropriateness in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    omanip
    -0.16
     Ðijи
    -0.15
    wig
    -0.15
    γκÏĮ
    -0.14
     Maiden
    -0.14
    anki
    -0.14
    _MA
    -0.14
    -haspopup
    -0.14
    Reuse
    -0.14
    ée
    -0.14
    POSITIVE LOGITS
     proper
    0.23
     Proper
    0.20
    proper
    0.19
     appropriate
    0.18
    appropriate
    0.18
     correct
    0.18
    æŃ£ç¡®
    0.16
     mix
    0.16
     correctly
    0.15
    uji
    0.15
    Act Density 0.132%

    No Known Activations