© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128674
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on tokens that are acronyms or other uppercase-initial identifiers (e.g. DIV, P, Query, Info, Running).

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

    states, types, and parameters

    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
    人的
    -1.04
     similarly
    -0.91
     Auch
    -0.88
     قوات
    -0.87
     universitaria
    -0.87
     their
    -0.85
    pering
    -0.85
     ivo
    -0.84
     SCHOLASTIC
    -0.83
     också
    -0.82
    POSITIVE LOGITS
     خدمات
    1.18
    撄
    1.07
    最後
    1.03
     corazón
    1.03
     AppCompatTheme
    1.02
    États
    1.01
    çek
    1.01
    plain
    1.00
     vía
    0.98
    Angebot
    0.97
    Activations Density 0.073%

    No Known Activations