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

    Neuronpedia

    INDEX
    Explanations

    Code

    np_max-act · gemini-2.0-flash

    This neuron detects C-style struct field accesses, i.e. the `->` (and related `.`) member-access operators in code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    -enable
    -0.07
    _tar
    -0.06
    	Vk
    -0.06
    <Project
    -0.06
     приня
    -0.06
     Gorgeous
    -0.06
    Matcher
    -0.06
    =val
    -0.06
    providers
    -0.06
    资源
    -0.06
    POSITIVE LOGITS
    happy
    0.07
    .empty
    0.07
     happiness
    0.07
     음악
    0.07
    زو
    0.07
    CN
    0.07
    TextField
    0.06
    .dm
    0.06
    ��
    0.06
     guint
    0.06
    Act Density 0.076%

    No Known Activations