© 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. 34-GEMMASCOPE-RES-131K
    7. 33714
    Prev
    Next
    INDEX
    Explanations

    news agency abbreviations

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

    The neuron fires on the tell-tale metadata of newswire articles—things like all-caps datelines (CITY (SOURCE)) and URL or “Reuters”/“AP” bylines.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    cinct
    -0.85
     Precinct
    -0.76
    akim
    -0.73
    ėje
    -0.72
     で
    -0.71
    ECK
    -0.71
     Driscoll
    -0.70
     acquaintance
    -0.68
    िस्ट
    -0.68
     Siga
    -0.66
    POSITIVE LOGITS
     AFP
    2.95
     Reuters
    2.63
    AFP
    2.45
    Reuters
    2.08
     Agencies
    2.00
     PTI
    1.99
     AP
    1.95
     agencies
    1.94
     Reuter
    1.88
     IANS
    1.71
    Activations Density 0.071%

    No Known Activations