© 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. 10-GEMMASCOPE-RES-131K
    7. 1552
    Prev
    Next
    INDEX
    Explanations

    returning words in multiple languages

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

    The neuron fires on words meaning “return” (various conjugations of the verb “to return”/“come back”) across different languages.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    衚
    -2.20
    \
    -2.19
    ("
    -2.09
     большую
    -2.05
     enlarge
    -2.03
    烦恼
    -2.00
    歘
    -1.97
    統領
    -1.92
     хотел
    -1.89
     نگهد
    -1.88
    POSITIVE LOGITS
    ’
    2.91
    ,”
    2.36
    甴
    2.25
    的
    2.17
    ጺ
    2.13
    एस
    2.05
    ”;
    2.03
    歺
    2.02
     nebst
    2.02
    ductory
    2.02
    Activations Density 0.004%

    No Known Activations