© 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. 130761
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on named‐entity tokens in formal technical text—things like the names of standards, organizations, and domain names (e.g. “Unicode,” “Standard,” “.gov” URLs, etc.).

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

    standards and technical specifications

    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
    bisoft
    -1.37
     hanya
    -1.23
    ׇ
    -1.11
     auch
    -1.09
     apenas
    -1.08
     efek
    -1.08
    ibilità
    -1.07
     cinco
    -1.05
    mohon
    -1.02
    llten
    -1.02
    POSITIVE LOGITS
     internazionali
    1.16
     familières
    1.09
    eterangan
    1.07
    idemiology
    1.07
     molle
    1.05
     archi
    0.99
     新款
    0.98
     particulières
    0.98
     commerciali
    0.96
    ଷ
    0.96
    Activations Density 0.002%

    No Known Activations