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

    Neuronpedia

    INDEX
    Explanations

    command

    np_max-act · gemini-2.0-flash

    This neuron activates on inline code tokens and command‐line references—e.g. backticks, “pip,” “command,” and similar code or shell keywords.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     transcend
    -0.07
     COS
    -0.06
     surged
    -0.06
    (CON
    -0.06
    749
    -0.06
    .HCM
    -0.06
    ury
    -0.05
     Sony
    -0.05
    Epoch
    -0.05
     esports
    -0.05
    POSITIVE LOGITS
    ेड
    0.07
    loi
    0.07
     الاست
    0.07
     España
    0.07
    estro
    0.07
    gw
    0.06
     '_',
    0.06
    (),
    0.06
    	pass
    0.06
    има
    0.06
    Act Density 0.013%

    No Known Activations