© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 111204
    Prev
    Next
    INDEX
    Explanations

    people's last names

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

    This neuron is highly activated by capitalized last names (surnames) in the text.

    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
    0
    -3.84
    EN
    -2.92
    a
    -2.84
    of
    -2.72
    2
    -2.63
    AM
    -2.53
     necessárias
    -2.50
    3
    -2.50
    t
    -2.47
    <i>
    -2.45
    POSITIVE LOGITS
    
    2.56
    鵟
    2.48
    聼
    2.44
    封印
    2.42
     médias
    2.38
     ruines
    2.36
     Espanha
    2.28
     dégâts
    2.23
     március
    2.22
    消毒
    2.20
    Activations Density 0.002%

    No Known Activations