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 configuration metadata keys and related terms (e.g. platform, host, port, token, description) in structured YAML/metadata sections.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    	My
    -0.07
    	fr
    -0.07
    dict
    -0.07
     sona
    -0.06
    alers
    -0.06
    	div
    -0.06
    ones
    -0.06
    unda
    -0.06
    OWER
    -0.06
    工程
    -0.06
    POSITIVE LOGITS
     aforementioned
    0.06
     taj
    0.06
    ample
    0.06
    .userdetails
    0.06
     清
    0.06
     twin
    0.06
     �
    0.06
     SQ
    0.06
    304
    0.06
     الداخ
    0.06
    Act Density 0.010%

    No Known Activations