© 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. 10-GEMMASCOPE-RES-131K
    7. 103242
    Prev
    Next
    INDEX
    Explanations

    `spring` `exports` `ui` `kv` `DB` `request` `glfw` `@` `hello` `test` `up` `Custom`

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

    backtick characters used in code or technical documentation formatting.

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

    The neuron fires on Markdown formatting delimiters (e.g. backticks for code spans and `**` for bold).

    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
     when
    -1.80
     particularly
    -1.72
     even
    -1.71
     for
    -1.70
     usually
    -1.66
     emphasized
    -1.65
     seldom
    -1.61
     people
    -1.58
     them
    -1.58
     improves
    -1.57
    POSITIVE LOGITS
    蘩
    1.91
    );*/
    1.86
    ▢
    1.77
    waitKey
    1.66
    )};
    1.66
    XXXXX
    1.64
    MAXN
    1.63
    QSize
    1.63
    珦
    1.59
    虓
    1.59
    Activations Density 0.014%

    No Known Activations