© 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. 98181
    Prev
    Next
    INDEX
    Explanations

    noun-forming suffixes

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

    The neuron fires on German nouns (i.e. words capitalized as nouns in German, often long compound noun forms).

    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
     to
    -3.28
     for
    -2.64
     or
    -2.52
     at
    -2.41
     from
    -2.30
     with
    -2.27
     and
    -2.19
     in
    -2.11
     of
    -2.09
     if
    -1.95
    POSITIVE LOGITS
    趿
    1.92
     apropri
    1.91
    茈
    1.87
     repuesto
    1.85
    𝒯
    1.84
     moldura
    1.82
     refroid
    1.81
     делает
    1.81
     sellado
    1.81
     rempla
    1.80
    Activations Density 0.019%

    No Known Activations