© 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. 70510
    Prev
    Next
    INDEX
    Explanations

    parameters, statistical concepts, or limits

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

    This neuron fires on long or repeated whitespace sequences (especially indentations) characteristic of code‐block or parameter‐listing formatting.

    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
    trucks
    -1.29
     \'
    -1.26
     alcune
    -1.23
     ‚
    -1.23
    黢
    -1.22
    ׇ
    -1.21
    cheque
    -1.20
     ʻ
    -1.17
     lige
    -1.15
     ſome
    -1.15
    POSITIVE LOGITS
     used
    1.79
     then
    1.76
     use
    1.71
     using
    1.66
    ַּ
    1.57
     Used
    1.43
     Weiterbildung
    1.42
    ֶּ
    1.39
     utilizz
    1.30
    اقرأ
    1.30
    Activations Density 0.019%

    No Known Activations