© 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. 22-GEMMASCOPE-RES-131K
    7. 9764
    Prev
    Next
    INDEX
    Explanations

    ANSI color codes

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

    The neuron is triggered by code tokens that refer to colors—especially hex color literals (e.g., “#f9a43e”) and identifiers like “Color” in programming contexts.

    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
    textContent
    -1.07
     évé
    -1.03
     how
    -0.98
    kaar
    -0.97
    ('/',
    -0.96
    product
    -0.95
     procedure
    -0.94
    ことができました
    -0.94
    ：“……”
    -0.93
     его
    -0.93
    POSITIVE LOGITS
     colours
    1.16
     for
    1.12
     others
    1.06
     באופן
    1.02
     colour
    1.02
     различни
    0.99
     colors
    0.98
     color
    0.98
     parku
    0.96
    axaca
    0.95
    Activations Density 0.008%

    No Known Activations