© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 36571
    Prev
    Next
    INDEX
    Explanations

    past tense plurals in multiple languages

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

    This neuron detects sentence‐ or clause‐boundary markers—especially commas and periods (and the words that often follow them).

    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
     in
    -1.71
     because
    -1.41
     When
    -1.38
     or
    -1.36
     варіан
    -1.35
     now
    -1.30
     if
    -1.27
     should
    -1.27
    :
    -1.25
     Recently
    -1.23
    POSITIVE LOGITS
     was
    1.97
     were
    1.63
     была
    1.48
     ήταν
    1.43
     alcuni
    1.41
     było
    1.40
     molti
    1.38
     homoto
    1.36
     alcune
    1.34
     traseiro
    1.33
    Activations Density 0.039%

    No Known Activations