© 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. 112096
    Prev
    Next
    INDEX
    Explanations

    code definitions and titles

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

    code-related tokens and technical documentation, particularly programming syntax, file paths, class names, and structured metadata.

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

    This neuron fires on fragments and full occurrences of the product name “Odoo” (e.g. split “Od” + “oo,” “Odoo Server,” “Translation of Odoo,” etc.).

    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
     which
    -1.55
    Sneaky
    -1.45
    Controversy
    -1.33
    opter
    -1.30
     linguaggio
    -1.21
     still
    -1.20
     食器
    -1.19
     échou
    -1.16
    كيل
    -1.16
     larger
    -1.14
    POSITIVE LOGITS
    
    1.21
    Si
    1.21
     своих
    1.20
    มี
    1.18
    Blur
    1.16
    在美国
    1.15
    ...”
    1.14
    hace
    1.14
    to
    1.13
    </
    1.13
    Activations Density 0.051%

    No Known Activations