© 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. 34-GEMMASCOPE-RES-131K
    7. 86849
    Prev
    Next
    INDEX
    Explanations

    built-in and standard concepts

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

    The neuron consistently fires most strongly on very common function words (articles, prepositions, auxiliary verbs) such as “a,” “the,” “in,” “is,” “being,” 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_34/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
     NSError
    -0.84
    īs
    -0.81
     ComVisible
    -0.79
    起了
    -0.77
    ailles
    -0.76
    vaig
    -0.75
    Ų
    -0.74
    unj
    -0.74
    ább
    -0.74
    Ϭ
    -0.73
    POSITIVE LOGITS
     built
    4.19
    built
    3.38
    Built
    3.31
     Built
    3.20
     builtin
    3.14
    builtin
    2.95
    BUILT
    2.69
     встроен
    2.66
    Builtin
    2.55
    内置
    2.50
    Activations Density 0.359%

    No Known Activations