© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 15192
    Prev
    Next
    INDEX
    Explanations

    punctuation delimiters

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

    The neuron fires chiefly on punctuation separators—most strongly on commas.

    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
     this
    -2.77
     there
    -2.42
     the
    -2.20
     these
    -2.08
     We
    -2.08
     This
    -2.02
     now
    -2.02
     as
    -1.97
     each
    -1.89
    ↵
    -1.85
    POSITIVE LOGITS
     ส์
    2.11
    "",
    2.03
     stratég
    1.95
    抟
    1.91
     desinf
    1.91
     sprüche
    1.88
     profeta
    1.88
     gewi
    1.84
     différents
    1.83
     Spanyol
    1.83
    Activations Density 0.183%

    No Known Activations