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

    Neuronpedia

    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron fires on list‐item markers and table‐row separators (e.g. “*”, leading numbers with “:”, or “|–––|”)—in other words, it detects the formatting tokens that begin or separate entries in bulleted or tabular lists.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     lxml
    -0.07
    -hole
    -0.07
    _RANDOM
    -0.07
     Dome
    -0.06
    	sprite
    -0.06
    serialization
    -0.06
     snack
    -0.06
    =float
    -0.06
    _numeric
    -0.06
     billionaires
    -0.06
    POSITIVE LOGITS
     Interracial
    0.06
    emas
    0.06
     českých
    0.06
    ث
    0.06
     应
    0.06
    _try
    0.06
     �
    0.06
    	cs
    0.06
     españ
    0.05
    mc
    0.05
    Act Density 0.014%

    No Known Activations