© 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. 130874
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on multi‐word domain‐specific terms and proper‐name phrases (e.g. medical conditions, league names, species names, game or show titles, legal citations).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    numbered lists, acronyms, or specific sequels

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     L
    -0.91
     asked
    -0.90
     Z
    -0.88
     D
    -0.85
     five
    -0.84
     takes
    -0.83
     очень
    -0.83
     religione
    -0.83
     arguably
    -0.83
     иногда
    -0.83
    POSITIVE LOGITS
    licopter
    1.11
    haviours
    1.08
    recently
    1.08
    athed
    1.05
    दार
    1.04
    gging
    1.02
    foundland
    1.02
    cidos
    1.01
    boru
    1.00
    charest
    0.98
    Activations Density 0.271%

    No Known Activations