© 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. 22-GEMMASCOPE-RES-131K
    7. 95823
    Prev
    Next
    INDEX
    Explanations

    ui component declarations

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

    This neuron fires on standalone numeric tokens (i.e. number literals) in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     if
    -1.51
     these
    -1.34
     However
    -1.32
    However
    -1.15
     NEVER
    -1.13
     some
    -1.12
    if
    -1.10
    はこんな感じ
    -1.06
     would
    -1.05
    śnie
    -1.04
    POSITIVE LOGITS
     and
    1.88
     cepillo
    1.49
     juguete
    1.37
     coger
    1.34
     trám
    1.32
    erializer
    1.24
     нашей
    1.23
     monedas
    1.23
     wła
    1.23
     rompecabezas
    1.23
    Activations Density 0.007%

    No Known Activations