© 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. 69121
    Prev
    Next
    INDEX
    Explanations

    code type declarations in programming

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

    The neuron fires on programming language type or class identifiers (e.g., names of classes, structs, templates, or other typed entities).

    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
     modelleri
    -1.11
     \&
    -1.02
    cola
    -0.98
    1
    -0.98
    </b>
    -0.98
    ⌗
    -0.96
    ях
    -0.95
     for
    -0.93
    2
    -0.93
    _));
    -0.93
    POSITIVE LOGITS
     OFFERS
    1.09
    Kde
    1.06
    אם
    1.04
    がありますが
    1.02
    Якщо
    1.02
    動機
    1.02
    hláš
    1.00
    quelize
    1.00
     fact
    1.00
     Matthias
    1.00
    Activations Density 0.005%

    No Known Activations