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

    Neuronpedia

    INDEX
    Explanations

    URL redirects and web access

    np_max-act · gemini-2.0-flash

    The neuron activates on URLs and domain‐style tokens (e.g. “http”, “://”, “.com” and hostnames).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     gấp
    -0.08
    眾
    -0.06
    有效
    -0.06
    Creative
    -0.06
     sponge
    -0.06
    PointXYZ
    -0.06
     NC
    -0.06
     všechny
    -0.06
    atatype
    -0.06
     Sponge
    -0.06
    POSITIVE LOGITS
    Wh
    0.08
     overt
    0.07
    -j
    0.07
    _ste
    0.06
    рощ
    0.06
    らの
    0.06
    ’app
    0.06
     이
    0.06
     ज
    0.06
    éo
    0.06
    Act Density 0.015%

    No Known Activations