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

    Neuronpedia

    INDEX
    Explanations

    Questions and conversational text

    np_max-act · gemini-2.0-flash

    This neuron detects numerical information—numbers and quantities (times, measurements, counts) in the text.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     وضع
    -0.08
    ausal
    -0.07
    addPreferredGap
    -0.07
    avor
    -0.07
    qd
    -0.06
    حر
    -0.06
    �
    -0.06
    ou
    -0.06
    �
    -0.06
     здоб
    -0.06
    POSITIVE LOGITS
    NULL
    0.07
     specialised
    0.07
    getNode
    0.07
    ]);↵↵
    0.06
     …↵↵
    0.06
     THREE
    0.06
    ulling
    0.06
     Node
    0.06
    els
    0.06
    ussed
    0.06
    Act Density 0.047%

    No Known Activations