© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 129652
    Prev
    Next
    INDEX
    Explanations

    institute, institution, rules

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

    This neuron fires on the use of “instituted” (and closely related formal-establishment verbs) indicating that something (e.g. a law, program, or system) is being officially set up.

    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
    雱
    -2.94
    ౮
    -2.55
    不该
    -2.34
    
    -2.30
     svým
    -2.23
    又不
    -2.17
     наша
    -2.16
    艄
    -2.13
     trois
    -2.09
    ඈ
    -2.09
    POSITIVE LOGITS
    ’
    3.03
    ↵
    2.80
    ?”
    2.67
    彣
    2.58
    媖
    2.31
     Many
    2.30
    
    2.27
    ྜ
    2.27
     While
    2.25
    喼
    2.19
    Activations Density 0.003%

    No Known Activations