© 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. 19940
    Prev
    Next
    INDEX
    Explanations

    FC football clubs

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

    The neuron fires on proper‐noun tokens that denote football clubs or their category labels (e.g. “FC,” “AS,” club names like “Stade Malherbe Caen,” etc.).

    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
     razu
    -1.52
    робнее
    -1.48
    Â
    -1.38
     ++)
    -1.36
    ings
    -1.36
     vuestra
    -1.36
     you
    -1.33
    ۔
    -1.33
     with
    -1.33
     \&
    -1.32
    POSITIVE LOGITS
    となり
    1.44
    CIUDAD
    1.41
    ASUS
    1.40
     パスタ
    1.35
    Ingrédients
    1.35
    まっている
    1.33
    няться
    1.32
    OREM
    1.30
    Slack
    1.30
     puis
    1.29
    Activations Density 0.039%

    No Known Activations