© 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. 72753
    Prev
    Next
    INDEX
    Explanations

    standard deviation

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

    The neuron fires whenever it sees numeric literals (particularly decimal numbers) in the text.

    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
    
    -1.42
     balk
    -1.34
    褥
    -1.28
    と呼ば
    -1.24
    igshid
    -1.20
     ghiaccio
    -1.19
    s
    -1.18
    thinkable
    -1.16
     gesprochen
    -1.15
     They
    -1.14
    POSITIVE LOGITS
     we
    1.45
     ameryka
    1.34
    ņem
    1.29
    AddHtmlAttribute
    1.27
    ^{*}$
    1.26
    ători
    1.24
     can
    1.23
    _{
    1.21
    /////
    1.20
    {
    1.16
    Activations Density 0.031%

    No Known Activations