© 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. 33231
    Prev
    Next
    INDEX
    Explanations

    medical conditions and treatments

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

    scientific research paper text with technical terminology.

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

    The neuron reliably fires on formal academic‐style passages—especially the title and abstract sections of scientific papers.

    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
    ૉ
    -1.64
    VideoView
    -1.62
     множество
    -1.54
    androidx
    -1.52
     musíte
    -1.50
     aidé
    -1.50
     skvě
    -1.48
     mnoho
    -1.48
     خوبه
    -1.45
     estándar
    -1.44
    POSITIVE LOGITS
     at
    1.49
     particularly
    1.47
     in
    1.43
     both
    1.43
     especially
    1.34
     bombas
    1.34
     these
    1.24
    妩
    1.24
     because
    1.24
     having
    1.23
    Activations Density 0.113%

    No Known Activations