© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 73875
    Prev
    Next
    INDEX
    Explanations

    official titles

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

    This neuron detects uppercase titles and agency or official names (e.g., “ATTORNEY GENERAL,” “CLERK OF COURT,” “DIRECTOR OF CENTRAL INTELLIGENCE”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     '
    -1.36
    它的
    -1.31
    人たち
    -1.25
     эти
    -1.18
     αυτά
    -1.17
    锉
    -1.17
     éstas
    -1.16
     которое
    -1.16
     они
    -1.13
     aquellas
    -1.13
    POSITIVE LOGITS
     himself
    2.14
    himself
    1.44
     who
    1.33
     niya
    1.29
    girlfriend
    1.24
    ════
    1.22
    
    1.20
     niyang
    1.20
     recommande
    1.17
    gruß
    1.17
    Activations Density 0.081%

    No Known Activations