© 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. 68978
    Prev
    Next
    INDEX
    Explanations

    code, comments, questions

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

    This neuron fires on long runs of whitespace (e.g. extended sequences of spaces or tabs).

    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
    苤
    -1.36
    卺
    -1.13
     langit
    -1.13
     klinik
    -1.12
    змы
    -1.11
    cahontas
    -1.08
     hunde
    -1.07
     katk
    -1.07
    WebServlet
    -1.05
     squadre
    -1.05
    POSITIVE LOGITS
    li
    1.16
     our
    1.09
    b
    1.02
     is
    1.00
    p
    0.96
    in
    0.95
     они
    0.92
    I
    0.92
    ↵
    0.91
    l
    0.91
    Activations Density 0.000%

    No Known Activations