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

    Neuronpedia

    INDEX
    Explanations

    numbers or dates embedded in text surrounded by special characters

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Tanz
    -0.68
     Net
    -0.65
     net
    -0.63
    onic
    -0.62
     Zup
    -0.61
     Kenyan
    -0.61
     Droid
    -0.59
     Manhattan
    -0.58
     boarding
    -0.58
     NYC
    -0.58
    POSITIVE LOGITS
    Ń
    1.22
    ħ
    1.11
    «
    1.11
    ¬
    1.10
    Ī
    1.09
    Ļ
    1.09
    Ŀ
    1.08
    ĺ
    1.06
    ķ
    1.06
    ª
    1.05
    Act Density 0.232%

    No Known Activations