© 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. 3913
    Prev
    Next
    INDEX
    Explanations

    names followed by sources

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

    The neuron detects bylines and source attributions—that is, author names or publication names appended to quotes or testimonials.

    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
    ⁃
    -1.66
    ◦
    -1.55
    ➔
    -1.53
     —
    -1.49
    −/−
    -1.46
    ňte
    -1.45
     medy
    -1.45
    Ï
    -1.45
    = 
    -1.42
    stained
    -1.42
    POSITIVE LOGITS
     their
    2.11
     because
    2.05
     not
    1.98
    他们
    1.96
     if
    1.73
     of
    1.70
     from
    1.70
     melibatkan
    1.66
    เพราะ
    1.60
     first
    1.58
    Activations Density 0.038%

    No Known Activations