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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron fires on capitalized identifier tokens (e.g. Java class/type names) in code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    )");↵↵
    -0.07
     MHz
    -0.07
    MHz
    -0.07
    	Thread
    -0.06
    ,row
    -0.06
    "]))↵
    -0.06
    мини
    -0.06
    .wp
    -0.06
    :frame
    -0.06
     presumption
    -0.06
    POSITIVE LOGITS
    31
    0.07
    út
    0.07
     fostering
    0.07
     कव
    0.06
     František
    0.06
     equipe
    0.06
    .sulake
    0.06
     있어서
    0.06
    ательных
    0.06
    cling
    0.06
    Act Density 0.063%

    No Known Activations