© 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. 10-GEMMASCOPE-RES-131K
    7. 14887
    Prev
    Next
    INDEX
    Explanations

    Ferrari, CERN, German, Cal

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

    The neuron is a named‐entity detector—it fires strongly on capitalized proper names (especially brands, places, and organizations like “Ferrari,” “CERN,” “Maranello,” 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
    “
    -3.05
    錞
    -2.73
    踅
    -2.56
    檍
    -2.53
    栻
    -2.45
    确实是
    -2.42
    </em>
    -2.34
    不过是
    -2.27
     صفحۀ
    -2.27
    不仅仅
    -2.23
    POSITIVE LOGITS
    of
    2.84
    に
    2.27
    x
    2.25
     change
    2.06
    i
    1.98
     two
    1.98
     białe
    1.97
     speziellen
    1.93
     Füßen
    1.93
     dismiss
    1.91
    Activations Density 0.003%

    No Known Activations