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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 40390
    Prev
    Next
    INDEX
    Explanations

    connecting clauses or list items

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron consistently activates on the coordinating conjunction “and.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     kinds
    -1.04
    そうで
    -1.02
    尔的
    -1.02
     lorsque
    -0.97
    あらゆる
    -0.97
    こうして
    -0.95
    ストレート
    -0.95
    そういった
    -0.94
    ねぇ
    -0.94
     other
    -0.94
    POSITIVE LOGITS
    ()),
    1.20
    \}$
    1.12
    wek
    1.06
    NEY
    1.04
    мить
    1.01
    ()],
    0.98
    PHYS
    0.97
    toDouble
    0.96
    ()])
    0.94
    "},
    0.94
    Activations Density 0.115%

    No Known Activations