© 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. 36156
    Prev
    Next
    INDEX
    Explanations

    sentence beginnings

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

    The neuron fires on runs of leading whitespace or indentation (i.e. layout/formatting whitespace tokens at the beginnings of lines).

    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
     ….
    -1.08
    ##
    -1.02
     прошел
    -1.02
    😕
    -1.01
    😐
    -1.00
    what
    -0.97
     ......
    -0.94
    でした
    -0.94
     AM
    -0.94
    first
    -0.94
    POSITIVE LOGITS
     also
    1.41
    Ингредиенты
    1.25
    Опис
    1.21
     SEÑ
    1.20
    Время
    1.16
     Lesnar
    1.15
     problemet
    1.13
    另一個
    1.13
     lastly
    1.12
    的过程中
    1.12
    Activations Density 0.011%

    No Known Activations