© 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. 102394
    Prev
    Next
    INDEX
    Explanations

    proper names and entities

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

    The neuron primarily detects proper nouns and other named entities (capitalized names of people, organizations, or specific things).

    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
    al
    -2.52
    on
    -2.47
     Empfindung
    -2.42
    ে
    -2.42
    If
    -2.41
    ?"
    -2.39
    o
    -2.38
    The
    -2.36
    out
    -2.36
    琹
    -2.33
    POSITIVE LOGITS
    '
    3.64
    㐂
    2.77
    媗
    2.73
    蚡
    2.67
    牁
    2.36
     smtplib
    2.27
    
    2.27
    itiert
    2.27
     verlie
    2.20
    Goes
    2.17
    Activations Density 0.016%

    No Known Activations