© 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. 20873
    Prev
    Next
    INDEX
    Explanations

    introductions to statements

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

    The neuron detects the discourse marker “so” when used to start or segue into a new statement.

    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
    </th>
    -1.68
    );
    -1.67
    <td>
    -1.66
    larda
    -1.52
     Projektu
    -1.52
    }
    -1.50
    ließt
    -1.48
    ";
    -1.48
    логу
    -1.46
     pelbagai
    -1.45
    POSITIVE LOGITS
    begin
    1.91
     again
    1.89
    also
    1.87
    gulation
    1.70
    for
    1.67
    first
    1.67
    their
    1.66
     were
    1.60
    these
    1.60
     today
    1.58
    Activations Density 0.004%

    No Known Activations