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

    Neuronpedia

    INDEX
    Explanations

    3

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric tokens corresponding to patent document identifiers (e.g. the comma-separated patent numbers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Elf
    -0.07
    phrase
    -0.07
    -flag
    -0.07
    (dom
    -0.07
    (forms
    -0.07
    .fin
    -0.07
    شته
    -0.06
    Sw
    -0.06
    (sn
    -0.06
    Corn
    -0.06
    POSITIVE LOGITS
     VIII
    0.06
     Batch
    0.06
    chool
    0.06
     CPC
    0.06
    ozí
    0.06
     Releases
    0.06
    patial
    0.06
    aggable
    0.05
     pricey
    0.05
     Dairy
    0.05
    Act Density 0.005%

    No Known Activations