© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 43557
    Prev
    Next
    INDEX
    Explanations

    closing bracket or quote

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

    The neuron fires on tokens that are part of bracketed or parenthetical references and metadata (e.g. square-bracket citations, asides, and numbered labels).

    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
    The
    -2.00
    ꯳
    -1.93
    логии
    -1.77
    杙
    -1.73
    嘜
    -1.69
    varlak
    -1.66
    臆
    -1.66
    Biografie
    -1.66
    larını
    -1.66
    usias
    -1.65
    POSITIVE LOGITS
     \
    2.28
    ],
    2.17
    at
    2.14
    垍
    2.11
    ,:
    2.08
    寕
    2.06
     the
    2.02
    –
    2.02
     ]
    2.00
    “
    2.00
    Activations Density 0.108%

    No Known Activations