© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 32680
    Prev
    Next
    INDEX
    Explanations

    (Sports Network)

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

    The neuron detects the standard newspaper dateline pattern—city names (often followed by state or country abbreviations), a parenthesis-enclosed news‐source tag, and related uppercase acronyms.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     …
    -1.14
     variety
    -1.12
     a
    -1.09
     ...
    -1.07
     vision
    -1.04
     many
    -1.03
     подоб
    -1.01
     examples
    -1.00
     descricao
    -0.99
     like
    -0.98
    POSITIVE LOGITS
     می‌گو
    1.21
     yürü
    1.11
     krä
    1.09
    uspiel
    1.05
    🔻
    1.03
    琁
    1.03
    鸰
    1.03
     händ
    1.02
    Цент
    1.02
    ête
    1.01
    Activations Density 0.058%

    No Known Activations