© 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. 84944
    Prev
    Next
    INDEX
    Explanations

    BEGIN followed by key or license types

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

    This neuron detects document section delimiters, in particular all-caps “BEGIN” (and corresponding “END”) markers.

    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
    gucig
    -2.52
    e
    -2.50
    𝑚
    -2.19
     compagni
    -2.14
    omores
    -2.03
    یفون
    -2.00
    涠
    -2.00
     genitori
    -1.96
    髣
    -1.94
     {
    -1.93
    POSITIVE LOGITS
    '
    2.97
    1
    2.89
    7
    2.45
    3
    2.34
    5
    2.34
    ↵
    2.16
    6
    2.13
    4
    2.11
     
    2.06
    8
    1.99
    Activations Density 0.002%

    No Known Activations