© 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. 67634
    Prev
    Next
    INDEX
    Explanations

    code comment delimiters

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

    The neuron fires on uppercase section headings, labels, and comment‐block delimiters (e.g. “KEY FEATURES:”, “DESCRIPTION”, “*/”) that mark structural metadata in technical or code documents.

    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
     director
    -1.12
    ṯ
    -0.98
     umy
    -0.98
     Gläub
    -0.96
    ḏ
    -0.95
     doda
    -0.95
    chien
    -0.95
     equilíbrio
    -0.94
     magnifiques
    -0.94
     Fisher
    -0.93
    POSITIVE LOGITS
     capitan
    1.13
    FUENTES
    1.10
     investiga
    1.05
     fama
    1.02
     efic
    1.02
    بسم
    1.01
    谪
    1.01
    海岸
    1.01
     fundamenta
    1.00
     romantis
    1.00
    Activations Density 0.001%

    No Known Activations