© 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. 70346
    Prev
    Next
    INDEX
    Explanations

    espn and sports websites

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

    The neuron activates on author and analyst names as well as media outlet and URL tokens (e.g. bylines like “Pelton,” “Youngmisuk,” and site references like “ESPN.com”), flagging those elements in sports articles.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     вокза
    -1.17
     cryptic
    -1.05
     Regularly
    -1.04
     inconspicuous
    -1.03
     Youtube
    -1.02
     THOSE
    -1.02
     metaphorical
    -1.02
    Botswana
    -1.01
    eh
    -1.00
     Adjusting
    -0.99
    POSITIVE LOGITS
    約
    1.18
     herren
    1.13
    Dữ
    1.12
     Nordrhein
    1.12
     geburtstag
    1.11
    CloseOperation
    1.10
    تهى
    1.10
     rsa
    1.09
     rígida
    1.07
    ⃙
    1.07
    Activations Density 0.006%

    No Known Activations