© 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. 21001
    Prev
    Next
    INDEX
    Explanations

    names and specific terms

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

    The neuron activates on uncommon or specialized proper‐noun–style tokens—names of people, places, or technical terms.

    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
    ꜆
    -2.58
     of
    -2.31
    あぁ
    -2.30
    re
    -2.28
    씀
    -2.22
     mereka
    -2.20
     かわいい
    -2.09
    -2.09
    ner
    -2.08
    will
    -2.06
    POSITIVE LOGITS
     It
    3.36
    \
    3.00
     This
    2.55
    ↵↵
    2.33
    
    2.06
    </td>
    2.03
    ulica
    2.03
     There
    2.00
     ‘’
    1.94
     If
    1.92
    Activations Density 0.000%

    No Known Activations