© 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. 89
    Prev
    Next
    INDEX
    Explanations

    Q: and page titles

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

    The neuron fires on digit‐heavy strings and numeric identifiers—e.g. DOIs, article numbers, volume/issues, years, URLs—found in scholarly‐article metadata.

    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
    情況
    -1.03
     devenus
    -1.02
     lemari
    -1.01
     définiti
    -1.01
    ниях
    -1.00
     signifikan
    -0.97
     kafe
    -0.95
    onya
    -0.94
     beban
    -0.94
    viš
    -0.94
    POSITIVE LOGITS
    ीस
    1.02
     at
    0.98
     rev
    0.93
     partners
    0.88
    azure
    0.88
    ),
    0.87
    等到
    0.87
     link
    0.86
     glowing
    0.86
     siquiera
    0.85
    Activations Density 0.003%

    No Known Activations