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

    Neuronpedia

    INDEX
    Explanations

    cache

    np_max-act · gemini-2.0-flash

    This neuron detects terms related to cache hierarchy and coherence protocols (e.g., “cache,” “coherency,” “cache hierarchy,” “memory coherence”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     secondo
    -0.07
     BUILD
    -0.07
    acht
    -0.06
     proceeds
    -0.06
     numbered
    -0.06
     dicts
    -0.06
    _bundle
    -0.06
     cleaning
    -0.06
     binaries
    -0.06
     参数
    -0.06
    POSITIVE LOGITS
     같이
    0.07
     투
    0.07
     sup
    0.07
    Ш
    0.06
    Txt
    0.06
     Lau
    0.06
     Closet
    0.06
    +N
    0.06
    (H
    0.06
    topic
    0.06
    Act Density 0.004%

    No Known Activations