© 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. 10-GEMMASCOPE-RES-131K
    7. 47686
    Prev
    Next
    INDEX
    Explanations

    ocular lens

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

    medical or technical terms related to eyes, particularly "ocular" and "lens", as well as city/wheel-related terms.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron fires on tokens containing the substrings “City” or “ocular” (and similar compound terms like “All Wheel Drive”), i.e. it picks out those specific suffix‐bearing proper nouns and medical/anatomical adjectives.

    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
     as
    -1.68
     has
    -1.48
    ższych
    -1.43
     Tämä
    -1.39
     \&
    -1.39
     Since
    -1.33
     unusually
    -1.33
     was
    -1.32
     STILL
    -1.28
    哋
    -1.28
    POSITIVE LOGITS
     شركة
    1.45
     anyone
    1.39
    /−
    1.37
    rija
    1.32
     vão
    1.32
     you
    1.32
     usuário
    1.30
     your
    1.28
    れたら
    1.25
    も多く
    1.25
    Activations Density 0.099%

    No Known Activations