© 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. 84146
    Prev
    Next
    INDEX
    Explanations

    data structure definitions

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

    The neuron strongly activates on programming identifiers made up of uppercase letters or CamelCase/PascalCase names (e.g. function or type names with internal capital letters).

    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
     Dinas
    -2.41
    皤
    -2.38
     siè
    -2.22
    provides
    -2.20
    ebly
    -2.20
     花纹
    -2.19
    씰
    -2.17
    aning
    -2.16
     ſol
    -2.11
    ⟣
    -2.11
    POSITIVE LOGITS
     more
    2.55
     using
    2.45
     like
    2.45
     created
    2.44
     from
    2.44
     just
    2.36
     all
    2.36
     with
    2.33
     started
    2.31
     not
    2.30
    Activations Density 0.078%

    No Known Activations