© 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. 35755
    Prev
    Next
    INDEX
    Explanations

    names of characters and places

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

    This neuron fires strongly on proper nouns—especially character or person names.

    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
     także
    -1.06
    Jeho
    -1.01
    基本的に
    -1.01
     myös
    -1.00
    不思議な
    -1.00
     schwere
    -1.00
    гите
    -0.99
     sebagian
    -0.97
     Opin
    -0.96
     litro
    -0.94
    POSITIVE LOGITS
    ля
    1.08
    ⬈
    1.05
    อยาก
    1.02
    দ
    1.01
     deleg
    0.98
    ле
    0.97
    botas
    0.96
    LookAnd
    0.94
     st
    0.93
     chức
    0.93
    Activations Density 0.009%

    No Known Activations