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

    Neuronpedia

    INDEX
    Explanations

    direct

    np_max-act · gemini-2.0-flash

    The neuron activates on the term “Direct,” particularly when it appears as the leading word in technical headings or titles.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Baum
    -0.07
     lun
    -0.06
    anan
    -0.06
     Wow
    -0.06
    �
    -0.06
    ampaign
    -0.06
     enthusiasm
    -0.06
     стоя
    -0.06
    ảm
    -0.06
     Yao
    -0.06
    POSITIVE LOGITS
    Direct
    0.12
     Direct
    0.12
     direct
    0.11
    direct
    0.09
     secret
    0.08
     přím
    0.08
    -direct
    0.08
     RT
    0.08
    Quick
    0.07
    duct
    0.07
    Act Density 0.013%

    No Known Activations