© 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. 22-GEMMASCOPE-RES-131K
    4. 79686
    Prev
    Next
    INDEX
    Explanations

    programming code constructs

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

    The neuron fires on closing delimiters in code (e.g. “}” or “)”), marking the ends of code blocks or statements.

    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
     kost
    -1.07
     DÉ
    -0.96
     KING
    -0.91
     Conseil
    -0.91
     ici
    -0.91
    İR
    -0.89
     Sok
    -0.88
    vii
    -0.88
    мены
    -0.86
     stadion
    -0.85
    POSITIVE LOGITS
     if
    1.27
    tawan
    0.93
     если
    0.93
     reinen
    0.89
    cluse
    0.89
     decorazione
    0.88
     pokud
    0.88
     lunghezza
    0.88
    了一些
    0.87
     kurzen
    0.87
    Activations Density 0.001%

    No Known Activations