© 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. 72666
    Prev
    Next
    INDEX
    Explanations

    Hyde, Mercer, NAC, Notre, Cypress

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

    The neuron activates strongly on capitalized named entities (proper nouns and acronyms).

    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
    </h3>
    -3.16
    他们是
    -3.02
    我现在
    -2.73
    鹮
    -2.70
    诶
    -2.67
     treacherous
    -2.64
    </strong>
    -2.45
     naviguant
    -2.45
     turísticos
    -2.38
     respald
    -2.38
    POSITIVE LOGITS
    </b>
    3.25
    ’
    3.06
    ↵↵
    2.94
    .'
    2.44
    In
    2.34
     con
    2.28
    ”
    2.23
    杙
    2.20
    TacToe
    2.16
    ＂
    2.11
    Activations Density 0.009%

    No Known Activations