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

    Neuronpedia

    INDEX
    Explanations

    Actions

    np_max-act · gemini-2.0-flash

    This neuron activates on imperative request verbs (e.g. “perform,” “create”) in user prompts asking for code or tasks.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _dem
    -0.07
     inher
    -0.06
    ırı
    -0.06
     Pharmaceutical
    -0.06
     justification
    -0.06
     Snape
    -0.06
    姑
    -0.06
    พ
    -0.06
     kulak
    -0.06
    Profiles
    -0.06
    POSITIVE LOGITS
     منابع
    0.07
    ALLERY
    0.07
     creative
    0.07
     obstruct
    0.07
     false
    0.06
    ,None
    0.06
     página
    0.06
    (CONT
    0.06
     перей
    0.06
    .Child
    0.06
    Act Density 0.033%

    No Known Activations