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

    Neuronpedia

    INDEX
    Explanations

    references to specific numeric or coded values in diverse contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    âĢ¦↵↵
    -0.15
    âĢ¦)↵↵
    -0.15
    illard
    -0.14
    neau
    -0.14
    rint
    -0.14
    âĢĬ
    -0.14
     Geb
    -0.13
    virt
    -0.13
     “âĢ¦
    -0.13
    ῶ
    -0.13
    POSITIVE LOGITS
     Bul
    0.23
     Iraq
    0.20
     War
    0.20
     --↵
    0.19
     Ava
    0.18
     grasp
    0.17
     Mosul
    0.17
     --
    0.17
     Iraqi
    0.17
     '--
    0.16
    Act Density 0.001%

    No Known Activations