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

    Neuronpedia

    INDEX
    Explanations

    pure

    np_max-act · gemini-2.0-flash

    This neuron detects the adjective “pure” when used to describe a technology (e.g. “pure HTML,” “pure JavaScript,” “pure Cordova”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    oyo
    -0.07
     AssemblyDescription
    -0.07
     bem
    -0.07
    .manual
    -0.07
    .pageSize
    -0.07
    GenericType
    -0.07
    mayacak
    -0.07
    -chan
    -0.07
    .Intent
    -0.07
    .on
    -0.06
    POSITIVE LOGITS
    �
    0.06
     pure
    0.06
    升
    0.06
     perpetrated
    0.06
     fundamentals
    0.06
     straight
    0.06
    agnostics
    0.05
     PURE
    0.05
     outright
    0.05
    accessible
    0.05
    Act Density 0.007%

    No Known Activations