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

    Neuronpedia

    INDEX
    Explanations

    programming code

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens that are names of software packages or libraries (i.e. dependency names) in build or requirement lists.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     accommodation
    -0.08
    (),"
    -0.07
     cafes
    -0.07
    helpers
    -0.06
     Seç
    -0.06
    BIG
    -0.06
    -code
    -0.06
     které
    -0.06
     ####
    -0.06
     Gry
    -0.06
    POSITIVE LOGITS
    oub
    0.07
    мотря
    0.07
    ेहतर
    0.06
    BUTTONDOWN
    0.06
    ΟΥΛ
    0.06
    implicitly
    0.06
    ій
    0.06
     sensing
    0.06
    -coordinate
    0.06
     Oswald
    0.06
    Act Density 0.016%

    No Known Activations