© 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. 10-GEMMASCOPE-RES-131K
    7. 47461
    Prev
    Next
    INDEX
    Explanations

    cardiac arrest

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

    legal and medical terms related to infants, children, and patients.

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

    The neuron fires on isolated, domain-specific technical or specialized nouns (rare or jargon terms) across contexts.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     yüksek
    -1.67
     sienten
    -1.48
     hermos
    -1.43
    klere
    -1.41
     οποία
    -1.39
     amikor
    -1.39
     []:
    -1.38
    ходите
    -1.35
     jestli
    -1.34
     DÍA
    -1.34
    POSITIVE LOGITS
     that
    1.84
    they
    1.55
     /
    1.55
     include
    1.49
    through
    1.40
    Therefore
    1.39
     such
    1.37
    /"
    1.37
    However
    1.37
    some
    1.36
    Activations Density 0.083%

    No Known Activations