© 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. 10-GEMMASCOPE-RES-131K
    7. 59776
    Prev
    Next
    INDEX
    Explanations

    negative contracted verbs

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

    The neuron fires on negation contractions—i.e. tokens ending with the “n’t” suffix (doesn’t, didn’t, couldn’t, 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
     to
    -2.69
    endforeach
    -2.11
     your
    -2.00
     heureuse
    -1.84
    泚
    -1.83
    ⁛
    -1.81
    鳎
    -1.69
    </tfoot>
    -1.66
    绡
    -1.66
    providedIn
    -1.63
    POSITIVE LOGITS
    ↵
    2.36
    );
    2.06
    ).
    2.05
    ),
    1.98
    .
    1.92
    ?
    1.86
     -
    1.78
    )
    1.75
     
    1.69
    '
    1.69
    Activations Density 0.085%

    No Known Activations