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

    Neuronpedia

    INDEX
    Explanations

    Technical procedures

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens that are part of Python code snippets—particularly library names, import statements, and code keywords.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    <::
    -0.07
    -prof
    -0.06
     Arist
    -0.06
    =Integer
    -0.06
    Mos
    -0.06
    .selectAll
    -0.06
    (writer
    -0.06
    -ac
    -0.06
    Threads
    -0.06
     gian
    -0.06
    POSITIVE LOGITS
     subsidized
    0.06
     plentiful
    0.06
     solar
    0.06
     ideological
    0.06
     lain
    0.06
    lopen
    0.06
    .Fat
    0.06
    licative
    0.06
     vite
    0.06
    .Ş
    0.06
    Act Density 0.379%

    No Known Activations