© 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. 58659
    Prev
    Next
    INDEX
    Explanations

    propaganda and propagating

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

    The neuron strongly activates on the word “propaganda” (including capitalized or in repeated contexts), effectively detecting mentions of propaganda in the text.

    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
    ?'
    -2.59
    ?"
    -2.34
    )."
    -2.14
    我不
    -2.09
    .'
    -2.08
    🌸
    -2.02
    🥳
    -1.98
     '
    -1.98
    🎉
    -1.97
    ၀
    -1.97
    POSITIVE LOGITS
    </h4>
    3.44
    ↵↵
    3.16
    0
    2.67
    丝毫
    2.58
     ”
    2.55
    </u>
    2.45
    ”;
    2.38
     
    2.33
    不过是
    2.30
    ڧ
    2.28
    Activations Density 0.002%

    No Known Activations