© 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. 100567
    Prev
    Next
    INDEX
    Explanations

    screen dimensions or scaling

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

    variable names or identifiers in programming code, particularly those related to screen dimensions, content, pages, sample rates, delivery dates, and cache durations.

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

    This neuron fires on parts of programming identifiers or field/property names (e.g. tokens in camelCase or PascalCase variable names).

    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
    definitely
    -1.75
    preferably
    -1.72
    unfortunately
    -1.72
    xiii
    -1.71
    werden
    -1.70
    菝
    -1.67
    栒
    -1.66
     of
    -1.63
    thinkable
    -1.60
     masterful
    -1.60
    POSITIVE LOGITS
     if
    2.08
     decreased
    1.88
     had
    1.87
     Putting
    1.85
    Despite
    1.81
    Even
    1.79
     strangely
    1.77
     increases
    1.77
     enormous
    1.76
     Bringing
    1.75
    Activations Density 0.033%

    No Known Activations