© 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. 102178
    Prev
    Next
    INDEX
    Explanations

    accepted status

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

    The neuron flags prominent content nouns—topic-defining keywords that signal the main subject matter of a passage.

    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
     oiseau
    -1.59
     boute
    -1.56
     péché
    -1.53
     tempel
    -1.52
     magasiner
    -1.52
     hjer
    -1.48
     emballage
    -1.45
     diman
    -1.44
     accessoire
    -1.43
     berge
    -1.43
    POSITIVE LOGITS
     where
    1.85
     even
    1.38
     “
    1.32
    áciu
    1.28
     and
    1.27
    ”？
    1.27
     itself
    1.27
    が必要です
    1.27
    While
    1.27
    Even
    1.26
    Activations Density 0.183%

    No Known Activations