© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 67498
    Prev
    Next
    INDEX
    Explanations

    closing brackets in structured data

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

    The neuron fires on the literal closing sequence “}]” (i.e. the end of a JSON array/object).

    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
    崀
    -2.69
    $
    -2.69
    羮
    -2.55
     them
    -2.38
     tratando
    -2.38
     レスト
    -2.36
    翆
    -2.34
     leichte
    -2.33
    哽
    -2.33
     blaue
    -2.31
    POSITIVE LOGITS
    )$
    2.91
    
    2.69
    ],
    2.38
    }]
    2.36
     they
    2.36
    __);
    2.30
    在我
    2.30
    ي
    2.27
    »
    2.25
    婬
    2.16
    Activations Density 0.001%

    No Known Activations