© 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. 10-GEMMASCOPE-RES-131K
    4. 47619
    Prev
    Next
    INDEX
    Explanations

    princely, atorial, ial, lish, ly suffixes and endings

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

    The neuron triggers on adjectival forms—especially words bearing morphological suffixes like English “-ial” (e.g. “senatorial,” “princely,” “editorial”) or Russian “-ский.”

    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
     '';
    -1.68
    prende
    -1.66
     через
    -1.66
    ⿻
    -1.66
     для
    -1.65
     '
    -1.63
     hailed
    -1.62
    ?
    -1.61
    📖
    -1.59
    ),
    -1.57
    POSITIVE LOGITS
     overarching
    2.17
    めちゃ
    2.16
    糴
    2.00
     endeavors
    1.99
    seems
    1.99
     aksesuar
    1.98
    〬
    1.98
     myriad
    1.95
    ほとん
    1.95
     nieuwe
    1.95
    Activations Density 0.020%

    No Known Activations