© 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. 43707
    Prev
    Next
    INDEX
    Explanations

    structuring delimiters

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

    This neuron detects non-Latin‐alphabet words (i.e. tokens in scripts like Japanese or Cyrillic).

    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
    ↵↵
    -2.31
     «
    -1.93
    </b>
    -1.91
     cleverly
    -1.72
    </td>
    -1.69
     dessutom
    -1.68
     but
    -1.63
     facilitates
    -1.63
    -
    -1.61
    </h2>
    -1.59
    POSITIVE LOGITS
     jeopardi
    2.19
    ING
    2.02
     attivo
    1.95
    闥
    1.95
    GETHER
    1.91
    monials
    1.90
     esigen
    1.88
     musia
    1.86
    Ꮴ
    1.86
    っていく
    1.85
    Activations Density 0.000%

    No Known Activations