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

    Neuronpedia

    INDEX
    Explanations

    words

    np_max-act · gemini-2.0-flash

    This neuron responds to mentions of feature‐vector representation methods (e.g. bag‐of‐words, TF-IDF, vectorization, embeddings).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    /backend
    -0.07
    小学
    -0.06
     getSize
    -0.06
     επί
    -0.06
    Is
    -0.06
    ssp
    -0.06
     arrest
    -0.06
      ↵  ↵
    -0.06
    _ME
    -0.06
     replication
    -0.05
    POSITIVE LOGITS
    xxxxxxxx
    0.07
    (position
    0.07
    Feb
    0.07
     Transitional
    0.07
     relied
    0.07
     xlink
    0.06
     topl
    0.06
     pai
    0.06
    ，我
    0.06
    $i
    0.06
    Act Density 0.006%

    No Known Activations