© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 71978
    Prev
    Next
    INDEX
    Explanations

    image inclusions or references

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

    The neuron is sensitive to markup tokens that embed external content—e.g. LaTeX’s “\includegraphics”, Markdown image syntax “![](...)”, and code‐block fence markers like triple backticks.

    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
    だけど
    -1.48
     lundi
    -1.39
     doit
    -1.36
     buna
    -1.35
     aussi
    -1.34
     tapi
    -1.34
     garantie
    -1.31
     ouver
    -1.30
    чение
    -1.29
     esa
    -1.28
    POSITIVE LOGITS
     what
    1.68
     nunmehr
    1.51
     شوند
    1.50
    UCION
    1.47
    brainer
    1.45
     všetky
    1.45
     albeit
    1.45
     邊
    1.43
     会場
    1.41
    endereco
    1.40
    Activations Density 0.002%

    No Known Activations