© 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. 74540
    Prev
    Next
    INDEX
    Explanations

    grammar, word choice, and sentence structure questions

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

    The neuron flags common function words—articles, prepositions, conjunctions, auxiliaries, and similar high-frequency grammatical tokens.

    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
     récol
    -1.52
    俢
    -1.47
    ών
    -1.37
     directe
    -1.35
     réserver
    -1.34
     garantir
    -1.32
    haft
    -1.30
     trasm
    -1.30
     acheter
    -1.30
     imprimer
    -1.30
    POSITIVE LOGITS
     which
    1.63
    _
    1.61
    mathrm
    1.50
    1.50
     this
    1.48
     its
    1.45
     as
    1.43
    }$
    1.41
     some
    1.41
     these
    1.41
    Activations Density 0.063%
    No Known Activations