© 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. 90524
    Prev
    Next
    INDEX
    Explanations

    shrines, sacred places

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

    This neuron fires on the key topic words in list‐item entries or section headings—i.e. the prominent nouns and adjectives that label each item in enumerations.

    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
    Может
    -0.82
    anyol
    -0.80
    čila
    -0.77
     Eskimo
    -0.77
    tidas
    -0.77
    Thermal
    -0.72
    Patience
    -0.72
     Thermal
    -0.72
    Gala
    -0.71
     overheating
    -0.70
    POSITIVE LOGITS
     shrine
    1.23
     sacred
    0.97
     Shrine
    0.96
     divine
    0.95
    社
    0.92
    shr
    0.85
     purification
    0.84
     santuario
    0.84
     shrines
    0.81
     honden
    0.78
    Activations Density 0.012%

    No Known Activations