© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 36370
    Prev
    Next
    INDEX
    Explanations

    sources with said

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

    This neuron detects journalistic attribution phrases introducing anonymous or unnamed sources (e.g., “people familiar with the matter said,” “a source said,” 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
     "
    -1.64
     even
    -1.29
     '
    -1.12
    ！（
    -1.10
     “
    -1.10
    ！(
    -1.05
     just
    -1.05
     actually
    -0.99
     particularly
    -0.98
     really
    -0.96
    POSITIVE LOGITS
    igree
    1.06
    攙
    1.05
    ixin
    1.04
     vostre
    1.02
    对她
    1.00
    igth
    1.00
    あぁ
    1.00
     obicei
    1.00
    還會
    1.00
    還沒
    0.99
    Activations Density 0.016%

    No Known Activations