© 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. 20439
    Prev
    Next
    INDEX
    Explanations

    listing descriptors

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

    This neuron detects descriptive adjectives—especially when they occur in comma-separated series joined by “and.”

    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
     but
    -1.80
     would
    -1.74
    a
    -1.71
    We
    -1.63
     all
    -1.63
    ที่มี
    -1.61
     Schicksal
    -1.59
     which
    -1.59
     plus
    -1.59
     artikke
    -1.58
    POSITIVE LOGITS
     located
    1.93
    他才
    1.72
     bandeau
    1.70
     nè
    1.70
    chemise
    1.68
     beve
    1.61
     digitais
    1.59
     bebes
    1.58
     evangé
    1.58
     supplied
    1.56
    Activations Density 0.134%

    No Known Activations