© 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. 71448
    Prev
    Next
    INDEX
    Explanations

    Chinese names, dynasties, and dishes

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

    This neuron responds to very high-frequency tokens—especially common function words (e.g. “is,” “to,” “the,” “as”) and punctuation like parentheses—marking the most frequent, boilerplate elements in the text.

    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
     povezave
    -1.21
     pilotes
    -1.09
    ########.
    -1.08
    zember
    -1.07
     sagesse
    -1.05
    Ofic
    -1.05
     redor
    -1.04
    Lycka
    -1.01
    Jeśli
    -1.01
    でも
    -1.01
    POSITIVE LOGITS
    &
    1.07
    0
    1.02
    .
    1.00
     notes
    1.00
     &
    1.00
     осно
    0.93
    </h2>
    0.92
    2
    0.92
    ."
    0.91
    1
    0.91
    Activations Density 0.017%

    No Known Activations