© 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. 128494
    Prev
    Next
    INDEX
    Explanations

    assembly trademark, propertychanging, dbset, scopus, ftse

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

    The neuron strongly activates on technical identifiers or proper names—particularly code‐related tokens and metadata field names (e.g. PublicKey, Token, DbSet, Scopus).

    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
    ’
    -2.30
    .
    -2.06
    ↵
    -1.98
     Again
    -1.87
     Verständ
    -1.84
     zarówno
    -1.80
    酝
    -1.78
    dersfield
    -1.75
    とはいえ
    -1.73
     pertence
    -1.73
    POSITIVE LOGITS
    t
    2.28
    ógico
    2.06
     ansik
    1.99
     grupper
    1.93
     liknande
    1.91
    s
    1.89
     fiyat
    1.88
    　　*
    1.87
     annons
    1.86
     henne
    1.85
    Activations Density 0.003%

    No Known Activations