© 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. 22-GEMMASCOPE-RES-131K
    7. 66072
    Prev
    Next
    INDEX
    Explanations

    defines or describes entities

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

    This neuron fires on section headings or metadata lines—i.e. titles, dates, and other structural header text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ri
    -1.22
     its
    -1.22
     POPULATION
    -1.12
    ७
    -1.08
     their
    -1.07
     It
    -1.06
     febbraio
    -1.03
    OVO
    -1.02
    PARATION
    -1.01
     cinese
    -1.00
    POSITIVE LOGITS
    Specifik
    1.63
    Étape
    1.37
    Skal
    1.35
    Neden
    1.32
     Incluso
    1.31
    Takk
    1.29
     dúvida
    1.28
     SUDOC
    1.26
     sofrer
    1.22
    viser
    1.22
    Activations Density 0.003%

    No Known Activations