© 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. 36533
    Prev
    Next
    INDEX
    Explanations

    OMIM numbers

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

    numeric identifiers or codes in scientific and financial documents.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron selectively activates on numeric tokens—particularly multi‐digit identifiers, section numbers, and decimal or code‐style numbers.

    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
    jured
    -1.23
     chers
    -1.21
     like
    -1.18
    suzuki
    -1.14
     in
    -1.12
     cuivre
    -1.12
    enseignements
    -1.10
    閾
    -1.10
     kwi
    -1.10
     you
    -1.10
    POSITIVE LOGITS
    DAFTAR
    1.36
    7
    1.30
     edilen
    1.24
    Élet
    1.23
     trunc
    1.22
    معرفی
    1.22
    مسلسل
    1.20
     Vereinig
    1.20
    5
    1.16
     sélectionner
    1.14
    Activations Density 0.052%

    No Known Activations