© 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. 28221
    Prev
    Next
    INDEX
    Explanations

    such as / like examples

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

    The neuron responds to words that introduce or label example lists or categories—especially “such,” “examples,” and plural category nouns like “processes,” “titles,” and “players.”

    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
     désolés
    -1.03
    これは
    -1.03
    餃子
    -1.00
    rierten
    -0.98
    Erfahrung
    -0.98
    更何况
    -0.97
    Produkte
    -0.95
     очень
    -0.95
     in
    -0.94
    但这
    -0.94
    POSITIVE LOGITS
     such
    3.25
     like
    3.22
     такие
    1.86
    such
    1.86
     seperti
    1.85
     مثل
    1.80
     như
    1.66
    like
    1.64
     takich
    1.57
     likes
    1.54
    Activations Density 0.031%

    No Known Activations