© 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. 34-GEMMASCOPE-RES-131K
    4. 102784
    Prev
    Next
    INDEX
    Explanations

    printmaking and related art

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

    The neuron strongly activates on the punctuation and token patterns typical of a formal bibliographic entry—author names, title segments separated by colons, lists of media (prints, books), commas, and other publication‐metadata markers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    Paint
    -1.09
    paint
    -1.07
     paint
    -1.05
     Painting
    -0.99
     Paint
    -0.93
    Painting
    -0.93
     painting
    -0.92
    塗装
    -0.91
     repaint
    -0.90
     paints
    -0.86
    POSITIVE LOGITS
     printing
    1.49
     prints
    1.34
     print
    1.33
     etching
    1.27
     lith
    1.16
     printer
    1.09
     Print
    1.05
     Prints
    1.04
     Printing
    1.03
     printers
    1.03
    Activations Density 0.031%

    No Known Activations