© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 130079
    Prev
    Next
    INDEX
    Explanations

    The neuron detects long runs of repeated punctuation characters (like `*`, `-`, `%`, etc.) used as separators or decorative comment bars.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    separators and code comments

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     from
    -1.28
     miesięcy
    -1.14
     The
    -1.10
     rather
    -1.05
     such
    -1.05
     various
    -1.02
     each
    -1.01
     most
    -1.00
     so
    -1.00
     by
    -0.98
    POSITIVE LOGITS
    ᐯ
    1.05
     horloge
    1.03
    Максим
    1.00
     hidrat
    0.99
     caros
    0.98
     espr
    0.98
    خخ
    0.97
     DevComponents
    0.97
     école
    0.96
     dié
    0.96
    Activations Density 0.052%

    No Known Activations