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

    Neuronpedia

    INDEX
    Explanations

    Youtube videos

    np_max-act · gemini-2.0-flash

    This neuron activates on fragments of YouTube video URLs—particularly the alphanumeric video‐ID tokens that follow “v=” in the link.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     glyphs
    -0.07
    _white
    -0.07
     kultur
    -0.06
    _PARSER
    -0.06
    keterangan
    -0.06
    ’de
    -0.06
    izzo
    -0.06
    Leon
    -0.06
    ρούν
    -0.06
    LTR
    -0.06
    POSITIVE LOGITS
     ])->
    0.07
    plist
    0.06
    оти
    0.06
     incapac
    0.06
     fis
    0.06
     qs
    0.06
    Buying
    0.06
    .Widget
    0.06
     simplify
    0.06
    packages
    0.05
    Act Density 0.002%

    No Known Activations