© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 25279
    Prev
    Next
    INDEX
    Explanations

    image tags and citations

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

    The neuron is primarily detecting runs of numeric digits (i.e. sequences of numbers).

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

    numeric page or image identifiers in scanned document metadata.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    document metadata and scanned-archive artifacts, such as markdown image links, journal citation headers, identifiers, and page/issue numbering strings.

    oai_token-act-pair · gpt-5Triggered 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
    mohon
    -1.23
    eada
    -1.18
     berbicara
    -1.18
    fible
    -1.04
    ẖ
    -1.03
    jorge
    -1.00
     menjalankan
    -0.99
    jours
    -0.98
    historie
    -0.98
     kui
    -0.97
    POSITIVE LOGITS
     نیز
    1.11
     телефона
    1.04
     also
    1.00
     mẽ
    0.97
    emia
    0.94
    ats
    0.94
     certamen
    0.94
    foria
    0.93
    en
    0.93
     captioned
    0.92
    Activations Density 0.020%

    No Known Activations