© 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. 128256
    Prev
    Next
    INDEX
    Explanations

    This neuron detects date‐ and number‐style metadata tokens (e.g. months, days, years, file identifiers) in the document header or byline.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    dates and publications

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    ≍
    -1.11
     jScrollPane
    -1.03
     caricatura
    -1.02
    litz
    -0.99
    utters
    -0.98
    来到了
    -0.97
     қара
    -0.96
    culously
    -0.96
     spirito
    -0.96
    Jeden
    -0.95
    POSITIVE LOGITS
     Gründung
    0.91
     recently
    0.89
    ちゃった
    0.88
     
    0.88
    度に
    0.86
    americanas
    0.82
    られない
    0.82
    нін
    0.81
    ₗ
    0.81
     current
    0.79
    Activations Density 0.023%

    No Known Activations