© 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. 34863
    Prev
    Next
    INDEX
    Explanations

    medical and health topics

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

    specialized terminology related to medical conditions, body parts, and healthcare fields.

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

    This neuron detects isolated topic or category words—single-word nouns that serve as section headings or subject labels.

    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
    Nguồn
    -1.22
    少しでも
    -1.05
     beziehungs
    -1.03
     fordern
    -0.99
    Nhưng
    -0.99
    Бли
    -0.98
    Traduction
    -0.96
     of
    -0.96
     เน
    -0.92
     ПРИ
    -0.92
    POSITIVE LOGITS
     etc
    1.38
    やろ
    1.10
     ainsi
    1.09
    itisation
    1.09
    backers
    1.05
     منزل
    1.04
    lamps
    1.03
    thirds
    1.01
    crete
    0.98
    years
    0.96
    Activations Density 0.138%

    No Known Activations