© 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. 9384
    Prev
    Next
    INDEX
    Explanations

    programming keywords

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

    The neuron flags code “declaration” keywords and inclusion/import tokens (e.g. const, let, var, public, using, #include, namespace) that introduce variables, types, or modules.

    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
     siè
    -1.60
     Dinas
    -1.54
     forment
    -1.47
    IndentedString
    -1.40
     lakini
    -1.38
     pérd
    -1.32
    擽
    -1.32
    isches
    -1.32
     wanda
    -1.32
    phonie
    -1.30
    POSITIVE LOGITS
     with
    2.23
     new
    1.73
     more
    1.61
     people
    1.48
     horribly
    1.46
     this
    1.45
     what
    1.45
    Not
    1.42
     decreases
    1.41
    There
    1.41
    Activations Density 0.140%

    No Known Activations