© 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. 21769
    Prev
    Next
    INDEX
    Explanations

    descriptive keywords and phrases

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

    The neuron fires on technical product or device model identifiers (e.g. sequences of uppercase letters and digits like “Model Number IXE20DN-OCP”).

    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
     not
    -2.44
     being
    -1.98
     eight
    -1.97
    十七
    -1.84
     our
    -1.80
     my
    -1.80
     because
    -1.78
     five
    -1.78
     having
    -1.77
     three
    -1.76
    POSITIVE LOGITS
    ſo
    2.23
    ૉ
    2.23
     Eſ
    2.20
     miſ
    2.17
     egreg
    2.11
     balení
    2.11
     useAppContext
    2.03
     eſ
    2.02
     Inſ
    2.02
     ſon
    1.98
    Activations Density 0.321%

    No Known Activations