© 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. 21198
    Prev
    Next
    INDEX
    Explanations

    sources reporting news

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

    The neuron detects journalistic attribution phrases—words like “source(s),” “said,” “told,” “claimed,” “reported,” etc., that signal information coming from cited or unnamed sources.

    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
    ↵↵
    -1.45
     to
    -1.41
     juſ
    -1.30
     También
    -1.27
     också
    -1.22
    ↵↵↵
    -1.18
     minimale
    -1.18
    燠
    -1.17
     gracios
    -1.17
    fiore
    -1.16
    POSITIVE LOGITS
     concluyó
    1.45
     only
    1.42
    他們
    1.41
    interpr
    1.37
    They
    1.32
     still
    1.32
    お店の
    1.30
    德國
    1.30
     reveló
    1.30
     other
    1.29
    Activations Density 0.020%

    No Known Activations