© 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. 10-GEMMASCOPE-RES-131K
    7. 105304
    Prev
    Next
    INDEX
    Explanations

    special characters followed by `set`, `resource`, `NAME`

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

    symbols and operators used in code and technical documentation, particularly pipes, redirects, assignment operators, and structural delimiters.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron activates on standalone punctuation symbols used for redirection or grouping in code (e.g. “>”, “&”, “{”, “}”).

    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
     although
    -1.59
     augusti
    -1.55
     when
    -1.45
     because
    -1.41
     bahwa
    -1.38
     only
    -1.38
    )
    -1.38
     still
    -1.31
     undersø
    -1.29
    ");
    -1.27
    POSITIVE LOGITS
     يبدو
    1.14
    諡
    1.13
     которыми
    1.13
    awarkan
    1.12
    Nied
    1.10
     председа
    1.09
     catalana
    1.09
     THEIR
    1.09
    Marcin
    1.09
    ánló
    1.08
    Activations Density 0.252%

    No Known Activations