© 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. 10-GEMMASCOPE-RES-131K
    7. 8863
    Prev
    Next
    INDEX
    Explanations

    emphasizing a literal meaning

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

    The neuron detects markers of figurative or emphatic language—words and phrases like “literally,” “figuratively,” and simile constructions (e.g. “thin as blood”) that signal rhetorical emphasis or comparison.

    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
     how
    -1.73
     why
    -1.55
    coca
    -1.44
    rachel
    -1.32
     každý
    -1.31
     každé
    -1.31
     to
    -1.30
     nawet
    -1.27
    timberland
    -1.26
     destacados
    -1.26
    POSITIVE LOGITS
     glyphicon
    1.37
    𓐍
    1.32
     layui
    1.31
    isers
    1.30
    Ű
    1.30
    ícia
    1.26
    =";
    1.24
    formats
    1.23
     Française
    1.22
     Arquivado
    1.21
    Activations Density 0.065%

    No Known Activations