© 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. 59106
    Prev
    Next
    INDEX
    Explanations

    speaking in contexts after certain words

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

    This neuron detects reporting-attribution phrases specifying the context of a quote (e.g., “during a conference call,” “at a press conference,” 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_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
     These
    -1.30
    crocs
    -1.25
    ――
    -1.23
     it
    -1.23
    ūra
    -1.18
    marzo
    -1.17
    erai
    -1.13
    🤡
    -1.13
    🎶
    -1.13
     fourteen
    -1.12
    POSITIVE LOGITS
     of
    1.67
    他們
    1.38
     dévelo
    1.37
     verhind
    1.25
    ჻
    1.24
     about
    1.21
     любы
    1.16
     eignen
    1.16
     EXISTS
    1.16
     today
    1.15
    Activations Density 0.051%

    No Known Activations