© 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. 10-GEMMASCOPE-RES-131K
    7. 71345
    Prev
    Next
    INDEX
    Explanations

    reporting speech or statements

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

    The neuron fires on reporting or attribution language—speaker names, quotation markers, and verbs like “said” or “explains” that introduce reported speech.

    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
    的气息
    -1.09
     of
    -1.09
    </i>
    -1.01
    ढ़
    -0.99
     DIGEST
    -0.99
    </em>
    -0.98
     '.';
    -0.98
    ≻
    -0.94
     for
    -0.94
    jón
    -0.94
    POSITIVE LOGITS
     zawsze
    1.45
     Sometimes
    1.16
     dispozici
    1.14
    されていて
    1.14
     nhưng
    1.09
     sometimes
    1.09
     expériment
    1.07
    Regarding
    1.07
     proprietario
    1.07
    Since
    1.06
    Activations Density 0.106%

    No Known Activations