© 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. 68878
    Prev
    Next
    INDEX
    Explanations

    epitaxial growth, liberal arts, revisionism

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

    The neuron fires on longer, capitalized or technical terms ending in Latin-derived suffixes (–ary, –ial, –ist/–ism, –day), for example “Black Friday,” “black market,” “Revisionist,” “Epitaxial,” and “simplicial.”

    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
     to
    -1.96
    neling
    -1.90
     during
    -1.66
     coveted
    -1.66
     Púb
    -1.62
     üç
    -1.59
    phoric
    -1.58
    نسا
    -1.57
    пой
    -1.53
    ásticas
    -1.52
    POSITIVE LOGITS
    '
    1.98
    i
    1.71
    e
    1.64
    ...'
    1.63
     置物
    1.60
    ..."
    1.59
    --"
    1.57
    /"
    1.57
    /'
    1.56
     такую
    1.56
    Activations Density 0.018%

    No Known Activations