© 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. 72
    Prev
    Next
    INDEX
    Explanations

    years and numbers following them

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

    The neuron fires on numeric tokens—especially the digit “9” when it appears in dates or year ranges.

    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
     africano
    -1.35
     coreano
    -1.26
     japones
    -1.21
     astu
    -1.20
     lampe
    -1.18
     finne
    -1.15
     malheur
    -1.14
    ेशा
    -1.14
     sè
    -1.14
     blusa
    -1.14
    POSITIVE LOGITS
     svolta
    1.13
     continued
    1.12
     aggiunto
    1.09
    諺
    1.06
     mosso
    1.06
     conferma
    1.04
     performed
    1.02
     Theologie
    1.02
     zunächst
    1.02
     obicei
    1.02
    Activations Density 0.044%

    No Known Activations