© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 127543
    Prev
    Next
    INDEX
    Explanations

    for loops

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

    This neuron fires on the opening curly‐brace “{” that marks the start of code blocks—particularly loops and function bodies.

    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
     most
    -1.60
     but
    -1.52
     only
    -1.49
    のは
    -1.48
    );
    -1.48
    也没有
    -1.41
    かつ
    -1.37
    但在
    -1.35
     no
    -1.35
     two
    -1.33
    POSITIVE LOGITS
     that
    2.30
    jenigen
    1.80
     distintas
    1.70
     entweder
    1.70
     its
    1.63
    請問
    1.61
     جدید
    1.59
     světě
    1.59
    Może
    1.58
    それぞれの
    1.56
    Activations Density 0.011%

    No Known Activations