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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash

    This neuron identifies occurrences of an “if” conditional statement in code (the “if (…){…}” or similar syntax).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     流
    -0.08
    -Control
    -0.07
    /site
    -0.06
     Changing
    -0.06
    状
    -0.06
    anst
    -0.06
    �
    -0.06
    linewidth
    -0.06
     cauliflower
    -0.06
     satisfies
    -0.06
    POSITIVE LOGITS
    _cons
    0.08
    най
    0.07
     stren
    0.07
    Output
    0.06
     인터
    0.06
     graf
    0.06
     :'
    0.06
     bara
    0.06
    gorithms
    0.06
    경기
    0.06
    Act Density 0.025%

    No Known Activations