© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 27929
    Prev
    Next
    INDEX
    Explanations

    kinds of information

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

    This neuron fires on structural/document‐layout markers—section titles, headings, enumerations, and other labels (e.g. numbered and capitalized headings, colons, parentheses), rather than ordinary running 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
    verständnis
    -1.20
     TMPro
    -1.20
    čki
    -1.13
     nagyon
    -1.12
     Ε
    -1.11
     enää
    -1.09
    ␣
    -1.09
     très
    -1.09
     ø
    -1.05
     Ł
    -1.05
    POSITIVE LOGITS
     Buchung
    1.10
     chrétien
    1.06
     seldom
    1.05
    アドレス
    1.03
     kinds
    1.02
     through
    1.00
     RISERV
    0.98
     many
    0.97
    お部屋
    0.95
     більш
    0.94
    Activations Density 0.006%

    No Known Activations