© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 110913
    Prev
    Next
    INDEX
    Explanations

    specific code identifiers

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

    the opening curly brace in JavaScript/TypeScript object destructuring assignments.

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

    The neuron activates on numeric literal tokens in code (e.g. integer or decimal constants).

    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
    ihr
    -2.03
    赑
    -2.03
    繻
    -2.02
    蛲
    -2.00
     estadounidenses
    -1.96
     inteligentes
    -1.96
     seguras
    -1.95
     listos
    -1.95
    denn
    -1.91
    sepatu
    -1.90
    POSITIVE LOGITS
     which
    2.59
     offer
    2.16
     from
    2.13
     --
    2.11
    v
    2.08
    e
    2.05
    m
    2.02
     provide
    1.98
    p
    1.90
     about
    1.88
    Activations Density 0.002%

    No Known Activations