© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 3191
    Prev
    Next
    INDEX
    Explanations

    code structures

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

    This neuron fires on code identifiers—i.e. programmer‐defined names like class, method, and variable identifiers (especially CamelCase/PascalCase).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    –
    -1.27
    }
    -1.17
            
    -1.13
    :
    -1.12
       
    -1.08
          
    -1.07
     și
    -1.06
              
    -1.05
    ).
    -1.05
    そして
    -1.03
    POSITIVE LOGITS
     --
    1.26
    焔
    1.15
    Gründe
    1.14
    Nachteile
    1.07
    Nazwa
    1.05
    ”？
    1.03
    作品の
    1.01
    gruss
    1.00
    This
    1.00
    Beispiele
    0.98
    Activations Density 25.725%

    No Known Activations