© 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. 44536
    Prev
    Next
    INDEX
    Explanations

    original publications or copies

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

    This neuron activates on instances of the word “original” (and its equivalents like 原文) used to reference source texts, articles, or titles.

    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
    =
    -2.03
     you
    -1.90
    р
    -1.50
    </td>
    -1.48
          
    -1.45
    %
    -1.43
    </th>
    -1.42
    >
    -1.40
     nutzen
    -1.33
    ($
    -1.32
    POSITIVE LOGITS
    眨眼
    1.63
     originales
    1.63
    峮
    1.62
    𐐪
    1.60
     vivido
    1.59
    somewhat
    1.55
     originale
    1.54
    Furthermore
    1.51
     menyadari
    1.51
     kılı
    1.50
    Activations Density 0.027%

    No Known Activations