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

    Neuronpedia

    INDEX
    Explanations

    verifying

    np_max-act · gemini-2.0-flash

    The neuron fires on polite, procedural agent prompts/offers of help (e.g. “May I have…,” “Let me check…,” “I’m sorry…”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    venient
    -0.07
    -play
    -0.07
     Hast
    -0.06
     Fin
    -0.06
    808
    -0.06
    .writeInt
    -0.06
     padding
    -0.06
    .x
    -0.06
     会
    -0.06
    LOGY
    -0.06
    POSITIVE LOGITS
    isinde
    0.07
    ocre
    0.06
    entario
    0.06
    enser
    0.06
    0.06
    rå
    0.06
     Boyle
    0.06
    _finished
    0.06
    	dir
    0.06
    (summary
    0.06
    Act Density 0.022%

    No Known Activations