© 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. 10-GEMMASCOPE-RES-131K
    7. 8998
    Prev
    Next
    INDEX
    Explanations

    years since

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

    The neuron fires on standalone numeric tokens—especially four-digit numbers used as years.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     his
    -2.09
     teils
    -1.79
     points
    -1.63
     trzyma
    -1.63
     eigenes
    -1.62
     vastly
    -1.60
     Vermögen
    -1.59
     own
    -1.58
    等は
    -1.57
     crucial
    -1.56
    POSITIVE LOGITS
    skrä
    1.66
     daire
    1.63
     Musée
    1.61
     Congrès
    1.61
     projet
    1.58
     noviembre
    1.55
     müh
    1.55
    ᵗ
    1.53
     gelen
    1.52
     cilt
    1.50
    Activations Density 0.154%

    No Known Activations