© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 35067
    Prev
    Next
    INDEX
    Explanations

    doping, banned substances, drugs

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

    terms related to doping and anti-doping agencies in sports.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates on the substring “doping” (and its appearance in related terms and acronyms like “Anti-Doping Agency,” “ADA,” etc.), effectively detecting mentions of doping or anti-doping entities.

    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
     we
    -1.42
     (
    -1.30
    ?
    -1.20
     of
    -1.16
    ֩
    -1.14
     chré
    -1.11
     I
    -1.09
     comuns
    -1.09
     MwSt
    -1.07
     directos
    -1.06
    POSITIVE LOGITS
    ֶּ
    1.66
     Persönlichkeit
    1.48
    
    1.47
    ຣ
    1.45
     segredo
    1.44
    ַּ
    1.44
    皤
    1.41
     connaissances
    1.39
    ִּ
    1.38
    妩
    1.38
    Activations Density 0.028%

    No Known Activations