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

    Neuronpedia

    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens used to praise someone’s character or professional virtues—e.g. words like “responsible,” “reliable,” “integrity,” “compassion,” and similar descriptors of trustworthiness and professionalism.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    otine
    -0.07
     grandchildren
    -0.06
    optimizer
    -0.06
    审
    -0.06
    оч
    -0.06
     hen
    -0.06
    shouldReceive
    -0.06
    垂
    -0.06
    <article
    -0.06
    isOpen
    -0.06
    POSITIVE LOGITS
    boa
    0.07
    .General
    0.07
     Only
    0.07
     الى
    0.06
    /ca
    0.06
     cerr
    0.06
     hodiny
    0.06
     conveying
    0.06
    หาก
    0.06
     dims
    0.06
    Act Density 0.089%

    No Known Activations