© 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. 95767
    Prev
    Next
    INDEX
    Explanations

    closing brackets and parentheses

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

    The neuron detects numeric array‐indexing expressions, i.e. digits immediately followed by a closing bracket in code (e.g. “[3]”).

    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
     maïs
    -1.65
     consé
    -1.64
    Zubereitung
    -1.63
     thoſe
    -1.60
     linge
    -1.60
     jesús
    -1.54
     remercier
    -1.52
    ་་
    -1.51
     pè
    -1.50
     faciliter
    -1.48
    POSITIVE LOGITS
     \
    1.84
     &
    1.71
     and
    1.58
     G
    1.48
     L
    1.48
     S
    1.45
     J
    1.45
     D
    1.44
     V
    1.43
     B
    1.41
    Activations Density 0.007%

    No Known Activations