© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 45282
    Prev
    Next
    INDEX
    Explanations

    acronyms or specific titles

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

    The neuron activates strongly on tokens that are part of Wikipedia’s structured metadata (infobox labels, titles, and category names).

    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
     tradis
    -0.73
    options
    -0.70
    bawa
    -0.70
     maleta
    -0.69
    askins
    -0.67
     Squire
    -0.67
     hvit
    -0.67
     Entered
    -0.66
    Страна
    -0.65
     توضی
    -0.64
    POSITIVE LOGITS
     LA
    0.76
     pyplot
    0.75
    aaa
    0.74
    zani
    0.74
     Timur
    0.73
    ဿ
    0.71
     Map
    0.71
    ANAL
    0.69
    caling
    0.69
     Ray
    0.69
    Activations Density 0.029%

    No Known Activations