© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 73674
    Prev
    Next
    INDEX
    Explanations

    citation chapter and verse

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

    The neuron activates on numeric tokens (especially multi‐digit numbers) that mark verses, chapters, dates, or other numbered labels.

    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
     which
    -1.11
     lalu
    -1.01
    合せ
    -0.98
     traje
    -0.96
    tı
    -0.96
     Einf
    -0.96
     alimentos
    -0.96
     just
    -0.95
     thoſe
    -0.94
    kowania
    -0.94
    POSITIVE LOGITS
    Allora
    1.12
     offiziell
    1.05
     subj
    1.04
     and
    1.02
     твое
    1.01
    Ugh
    1.00
     ぼ
    1.00
    pring
    0.96
     festge
    0.96
     эм
    0.96
    Activations Density 0.001%

    No Known Activations