© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128643
    Prev
    Next
    INDEX
    Explanations

    The neuron detects the pattern “) is a” that typically appears at the start of a Wikipedia‐style biographical definition sentence.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    nationalities of people

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     each
    -1.62
     their
    -1.49
     this
    -1.42
     new
    -1.30
     those
    -1.15
     before
    -1.13
     all
    -1.10
     when
    -1.10
     these
    -1.09
     similar
    -1.09
    POSITIVE LOGITS
     siè
    1.23
     estadounid
    1.17
     hongos
    1.12
     devraient
    1.09
     marquées
    1.08
     mê
    1.07
     apparaissent
    1.05
     ọ
    1.04
     essayer
    1.03
    えっ
    1.03
    Activations Density 0.192%

    No Known Activations