© 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. 9829
    Prev
    Next
    INDEX
    Explanations

    describing variations and states

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

    The neuron fires on multi-syllable content words’ derivational suffixes (–ive, –ous, –ity, –ed, etc.), i.e. the endings of descriptive or nominalized English words.

    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
     unveils
    -1.91
     ALREADY
    -1.75
     majestic
    -1.73
    vemos
    -1.70
     ส์
    -1.68
     pratiquement
    -1.68
     goofy
    -1.66
     !!!!!
    -1.66
     REALLY
    -1.66
     ridiculously
    -1.65
    POSITIVE LOGITS
     because
    2.59
     they
    2.50
     for
    2.45
     when
    2.45
     that
    2.28
     although
    2.05
     if
    2.05
     but
    2.03
     what
    2.03
     in
    2.02
    Activations Density 0.000%

    No Known Activations