© 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. 22-GEMMASCOPE-RES-131K
    7. 9589
    Prev
    Next
    INDEX
    Explanations

    Vienna University

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

    The neuron is keyed to named entities and technical terms—proper nouns, acronyms, and specialized multi-word 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
    atization
    -1.21
    atize
    -1.15
    neous
    -1.09
    ridescent
    -1.09
    atized
    -1.09
    thodoxy
    -1.08
    tocks
    -1.05
    preneurs
    -1.03
    ━━━━
    -1.02
    namental
    -1.02
    POSITIVE LOGITS
     zorgen
    1.17
    geç
    1.13
     uzman
    1.13
     來
    1.12
    getProducts
    1.09
    こい
    1.09
     层
    1.09
     يمكن
    1.07
     adhesivo
    1.05
    irim
    1.04
    Activations Density 0.169%

    No Known Activations