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

    Neuronpedia

    INDEX
    Explanations

    specifies

    np_max-act · gemini-2.0-flash

    details related to chemical compound instructions and safety protocols.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on the standard boilerplate phrase describing how a generated data model class “specifies how to parse/serialize into the JSON” format.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     arcs
    -0.07
    赶
    -0.07
    ันทร
    -0.06
     hardened
    -0.06
     Live
    -0.06
    29
    -0.06
     born
    -0.06
    œur
    -0.06
     Born
    -0.06
     traveled
    -0.06
    POSITIVE LOGITS
     specify
    0.10
     specified
    0.10
     specifies
    0.09
     specification
    0.09
    Specifies
    0.09
     specifications
    0.09
     specifying
    0.09
    指定
    0.08
    specified
    0.08
    Specify
    0.08
    Act Density 0.024%

    No Known Activations