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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14672
    Prev
    Next
    INDEX
    Explanations

    addresses with foreign characters

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

    The neuron fires strongly on standalone numeric tokens (integers) 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
    Ainda
    -1.47
    Além
    -1.43
    Até
    -1.41
    Também
    -1.38
    Alguns
    -1.34
    Твор
    -1.34
    Mesmo
    -1.29
    Apesar
    -1.26
    この
    -1.23
    taya
    -1.23
    POSITIVE LOGITS
     män
    1.22
     something
    1.16
     hjälper
    1.15
     necesitan
    1.15
     kän
    1.13
     siente
    1.13
    的其他
    1.12
    dali
    1.11
     ofrecen
    1.10
     novos
    1.09
    Activations Density 0.003%

    No Known Activations