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

    Neuronpedia

    INDEX
    Explanations

    Intention, purpose

    np_max-act · gemini-2.0-flash

    This neuron detects words and phrases that introduce intended purpose or design (e.g. “meant to…,” “designed to…”) in descriptions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Alcohol
    -0.07
    >b
    -0.06
     Leaves
    -0.06
     nig
    -0.06
     Pine
    -0.06
     Script
    -0.06
     uninsured
    -0.06
    $v
    -0.06
    \helpers
    -0.06
    crire
    -0.06
    POSITIVE LOGITS
     intended
    0.08
     aslında
    0.07
     Yao
    0.07
    fat
    0.07
    =headers
    0.06
    /npm
    0.06
    term
    0.06
    жень
    0.06
    ///↵
    0.06
    创建
    0.06
    Act Density 0.026%

    No Known Activations