© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 33106
    Prev
    Next
    INDEX
    Explanations

    code punctuation and symbols

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

    The neuron is picking up on punctuation delimiters—especially commas, quotes, and other symbol characters used in code and regex patterns.

    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
     with
    -1.30
     to
    -1.29
    These
    -1.23
     was
    -1.05
     is
    -1.04
    </strong>
    -1.02
     $
    -0.94
     first
    -0.94
     се
    -0.93
     by
    -0.93
    POSITIVE LOGITS
     your
    1.35
     możesz
    1.16
     the
    1.14
    選んだ
    1.09
     this
    1.09
     musisz
    1.09
     médicaments
    1.08
    狁
    1.06
     étage
    1.05
    nowu
    1.05
    Activations Density 0.008%

    No Known Activations