© 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. 72361
    Prev
    Next
    INDEX
    Explanations

    closing punctuation and braces

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

    The neuron strongly responds to closing delimiter characters (brackets or braces) in code or regex.

    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
    نور
    -1.77
    杙
    -1.71
    ൎ
    -1.70
     eerste
    -1.70
    ଃ
    -1.67
    brinco
    -1.66
     這
    -1.63
    ſtian
    -1.58
    镚
    -1.55
     sağlar
    -1.53
    POSITIVE LOGITS
     we
    2.09
    end
    1.96
    }
    1.88
    hline
    1.71
    ]=>
    1.61
    ti
    1.60
     ‘
    1.60
    ’
    1.57
     setempat
    1.56
     houve
    1.56
    Activations Density 0.023%

    No Known Activations