© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 86238
    Prev
    Next
    INDEX
    Explanations

    ancient bridges

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

    The neuron strongly activates on proper‐name nouns—especially named landmarks, geological features, products, or other titled entities (e.g. “Silver Cave,” “Luoyang Bridge,” “Jade Water Basin,” “oil field,” “British,” “Snow‐Mountain With Waterfall”).

    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
    which
    -2.95
     WHICH
    -2.77
    ]
    -2.47
    ﮦ
    -2.38
     två
    -2.38
    茕
    -2.34
    ],
    -2.33
     }
    -2.31
    ৭
    -2.31
    ",
    -2.28
    POSITIVE LOGITS
    The
    2.59
    を書
    2.09
     They
    2.02
    教训
    2.02
     About
    1.99
    谡
    1.98
     Why
    1.96
    ,'"
    1.94
    狲
    1.93
    ühnen
    1.90
    Activations Density 0.004%

    No Known Activations