© 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. 131041
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on standalone numeric tokens (digits or numbers), i.e. it detects numeric identifiers in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    legal document codes

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    ...
    -1.15
     incredibly
    -1.14
    A
    -1.09
     beautifully
    -1.09
    の本
    -1.08
     extremely
    -1.08
    ~~
    -1.07
    D
    -1.06
     dreadful
    -1.06
     .
    -1.06
    POSITIVE LOGITS
    かれ
    1.32
    Warna
    1.31
     fourrure
    1.29
    ญี่ป
    1.27
    پاسخ
    1.25
     reklama
    1.23
     salonu
    1.22
    tehen
    1.21
    してます
    1.20
    Lokasi
    1.19
    Activations Density 0.067%

    No Known Activations