© 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. 129855
    Prev
    Next
    INDEX
    Explanations

    The neuron activates on standalone numeric strings (e.g. years, version numbers, IDs)—i.e. any token composed entirely of digits.

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

    quantitative image reconstructionQ: question prefixadvertisements prefixDhoni

    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
     ۲۰
    -1.38
    或
    -1.38
     ${
    -1.21
    AttributeSet
    -1.20
    крой
    -1.20
     voor
    -1.18
     WhatsApp
    -1.18
    靭
    -1.16
     теп
    -1.16
     sediment
    -1.16
    POSITIVE LOGITS
    んじゃ
    1.34
     combatir
    1.30
     різних
    1.29
     constituyen
    1.28
     recuperação
    1.27
    嘢
    1.23
     resaltar
    1.22
     reforzar
    1.21
     sufren
    1.21
     almacenar
    1.21
    Activations Density 0.001%

    No Known Activations