© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71073
    Prev
    Next
    INDEX
    Explanations

    agree or disagree

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

    The neuron fires on explicit first‐person statements or self‐referential opinion phrases (e.g. “I agree,” “I am,” “It’s true”).

    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
    </table>
    -0.85
    法规
    -0.84
    清洁
    -0.81
     erfolgreich
    -0.78
     arbeitet
    -0.77
     пы
    -0.77
     >
    -0.76
    [::-
    -0.75
    venta
    -0.75
     ohne
    -0.75
    POSITIVE LOGITS
    確かに
    1.79
     agree
    1.39
     indeed
    1.33
    agreed
    1.19
    Indeed
    1.14
     that
    1.13
    たしか
    1.13
     agrees
    1.10
     Agree
    1.09
     Indeed
    1.00
    Activations Density 0.090%

    No Known Activations