© 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. 34-GEMMASCOPE-RES-131K
    7. 59697
    Prev
    Next
    INDEX
    Explanations

    roles and titles

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

    This neuron fires on the framing tokens of biographical lead sentences—especially “is a … and …” constructions introducing someone’s roles or titles.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     appris
    -1.16
     kalender
    -1.09
     menyen
    -1.04
     allerlei
    -1.04
    salary
    -1.03
    ⬘
    -1.02
    ୀ
    -1.02
    Референце
    -1.00
    malige
    -1.00
     spektak
    -0.99
    POSITIVE LOGITS
     member
    1.70
     head
    1.52
     director
    1.45
     president
    1.41
     Head
    1.36
     vice
    1.36
     deputy
    1.34
     Vice
    1.26
     manager
    1.23
     Member
    1.23
    Activations Density 0.074%

    No Known Activations