© 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. 33517
    Prev
    Next
    INDEX
    Explanations

    domain followed by identifier

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

    The neuron reliably fires on Wikipedia article section headings (e.g. “Biography,” “Academic life,” “Career,” “Background,” etc.).

    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
     tässä
    -1.20
    MANUEL
    -1.11
     a
    -1.09
    ınıza
    -1.04
     seront
    -1.02
     vielä
    -1.01
    这
    -1.00
     induk
    -1.00
    ℗
    -0.99
     geringe
    -0.99
    POSITIVE LOGITS
    </h2>
    1.52
    </h1>
    1.50
     and
    1.45
    </h3>
    1.20
     estekak
    1.16
     ediciones
    1.16
     khác
    1.16
    𝐽
    1.15
    しま
    1.13
    ÈNE
    1.11
    Activations Density 0.073%

    No Known Activations