© 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. 67866
    Prev
    Next
    INDEX
    Explanations

    place names and specific entities

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

    This neuron detects named entities—words or terms that are capitalized because they’re proper names, titles, or specialized identifiers.

    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
    z
    -3.33
    .,
    -2.94
    "
    -2.91
    u
    -2.89
        
    -2.78
    反而
    -2.72
     Erlebn
    -2.63
    7
    -2.59
    真的很
    -2.59
    4
    -2.58
    POSITIVE LOGITS
    躂
    3.30
    鋨
    2.83
    硨
    2.73
    璆
    2.63
    罍
    2.63
    ཋ
    2.58
    各样的
    2.58
    ⃙
    2.56
    絝
    2.47
     Universiteit
    2.44
    Activations Density 0.011%

    No Known Activations