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

    Neuronpedia

    INDEX
    Explanations

    comments

    np_max-act · gemini-2.0-flash

    Python code that uses functions to perform specific tasks, particularly related to plotting mathematical data.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on instructions and keywords related to code commenting and descriptive variable naming (e.g. “comments,” “descriptive,” “variable names”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     developed
    -0.06
    _Delay
    -0.06
     hundred
    -0.06
     activates
    -0.06
    _NAME
    -0.06
     achievement
    -0.06
     num
    -0.06
     atomic
    -0.06
    يد
    -0.06
     carrying
    -0.06
    POSITIVE LOGITS
    ******/
    0.06
     země
    0.06
    mates
    0.06
    !↵↵↵
    0.06
    bes
    0.06
    ــــ
    0.06
     isinstance
    0.06
    暂
    0.06
     Б
    0.06
       
    0.06
    Act Density 0.008%

    No Known Activations