© 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. 72519
    Prev
    Next
    INDEX
    Explanations

    titles and designations

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

    The neuron flags named entities, i.e. proper nouns such as people’s names, organization names, acronyms, and other capitalized identifiers.

    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
     patrie
    -1.16
     still
    -1.16
    They
    -1.16
     razie
    -1.14
     have
    -1.13
    ➩
    -1.10
     offer
    -1.09
     rosto
    -1.09
    kování
    -1.09
     katholischen
    -1.07
    POSITIVE LOGITS
     reinigen
    1.30
    ᾽
    1.20
    蔟
    1.16
    ଵ
    1.14
     aujourd
    1.13
    drage
    1.10
     électriques
    1.10
     exé
    1.09
    ֿ
    1.07
     скриншот
    1.05
    Activations Density 0.053%

    No Known Activations