© 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. 69210
    Prev
    Next
    INDEX
    Explanations

    healthcare billing and diagnoses

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

    coding, diagnostic, and technical terminology in medical and computing contexts.

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

    The neuron fires on prominent domain-specific “content” words (the key topic nouns and verbs) that typically appear in titles or headings.

    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
     chré
    -1.77
     religieuses
    -1.48
    gradu
    -1.41
    ▢
    -1.32
     chrétienne
    -1.31
    raisals
    -1.29
    larımız
    -1.29
     tasar
    -1.28
     anuncia
    -1.23
     امیر
    -1.23
    POSITIVE LOGITS
    ֶּ
    1.75
     бампер
    1.52
    īgs
    1.34
    ͎
    1.34
     correctly
    1.32
    ְּ
    1.32
     patients
    1.30
     خریدار
    1.30
    äumt
    1.29
    ּוֹ
    1.28
    Activations Density 0.020%

    No Known Activations