© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 26422
    Prev
    Next
    INDEX
    Explanations

    numbers in citations and references

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

    The neuron activates on standalone numeric tokens (especially multi‐digit numbers) that look like page, section, or reference 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
    為に
    -1.14
     mivel
    -1.11
    is
    -1.07
    imanapun
    -1.05
    あー
    -1.05
    マジで
    -1.02
    ことも
    -1.02
    えっ
    -1.01
    ILES
    -1.00
     ebenfalls
    -0.99
    POSITIVE LOGITS
    Zutaten
    1.04
    ograms
    1.02
     piscinas
    0.99
    ressee
    0.99
    かで
    0.98
    Ingredi
    0.97
    caping
    0.97
    dresser
    0.96
    angerschaft
    0.96
     floresta
    0.96
    Activations Density 0.005%

    No Known Activations