© 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. 10-GEMMASCOPE-RES-131K
    7. 43430
    Prev
    Next
    INDEX
    Explanations

    negative consequences or conditions

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

    The neuron fires on occurrences of the verb “suffer” (in its various inflected forms) marking someone experiencing a harmful event.

    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
    刧
    -2.77
    憮
    -2.25
    🙂
    -2.25
    😃
    -2.14
    😀
    -2.11
    😆
    -2.08
     FORMA
    -2.03
    😉
    -2.02
    😁
    -1.99
    😄
    -1.96
    POSITIVE LOGITS
    </td>
    2.70
    .
    2.06
     a
    2.06
    </
    2.03
     suffered
    1.95
    ↵
    1.95
     autoridad
    1.73
    </h5>
    1.73
     müssen
    1.70
     policía
    1.70
    Activations Density 0.008%

    No Known Activations