© 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. 84364
    Prev
    Next
    INDEX
    Explanations

    places names and organizations

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

    The neuron activates on capitalized proper names (e.g. “Valparaiso,” “IntelliJ,” “Achromobacter”).

    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
    骉
    -2.84
    颏
    -2.55
    genodigd
    -2.52
    鶿
    -2.50
    </h2>
    -2.50
     garantir
    -2.47
     nadru
    -2.41
    衄
    -2.31
    辚
    -2.31
     rechercher
    -2.28
    POSITIVE LOGITS
    \
    3.11
    </u>
    2.75
    <td>
    2.61
     \
    2.61
    里有
    2.45
    </sub>
    2.41
     midway
    2.39
     '
    2.38
     began
    2.36
    鏵
    2.31
    Activations Density 0.066%

    No Known Activations