© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 65350
    Prev
    Next
    INDEX
    Explanations

    anonymous objects or property initializers

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

    The neuron detects the syntax of C# anonymous‐object initializers—especially the opening brace “{” along with its commas, “new” keyword, and “=’ signs.

    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
     evenement
    -1.20
    klere
    -1.12
     jonge
    -1.10
    みなさん
    -1.09
    ㍑
    -1.06
     außerdem
    -1.06
    鹀
    -1.06
    绡
    -1.04
     behandel
    -1.03
     huevos
    -1.03
    POSITIVE LOGITS
     of
    1.64
    k
    1.23
     /
    1.10
     //
    1.08
     his
    1.08
    v
    1.05
     so
    1.04
     k
    1.02
    V
    1.00
     ......
    1.00
    Activations Density 0.005%

    No Known Activations