© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 68892
    Prev
    Next
    INDEX
    Explanations

    legal and official matters

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

    The neuron strongly activates on the main content words in document titles or section headings (e.g. official report or legal case titles), essentially spotting title‐style, capitalized headings.

    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
     блюдо
    -1.20
     smör
    -1.17
     kaffe
    -1.13
     Sosten
    -1.08
    WARRANTY
    -1.06
     these
    -1.04
     valide
    -1.04
     leone
    -1.03
     piccolo
    -1.02
     vostri
    -1.02
    POSITIVE LOGITS
     Certain
    1.13
     and
    1.13
    Leírás
    1.12
    ostock
    1.09
    oncesto
    1.09
    liquez
    1.08
    êques
    1.08
     бампер
    1.05
    FormUrlEncoded
    1.05
    ruary
    1.04
    Activations Density 0.083%

    No Known Activations