© 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. 73889
    Prev
    Next
    INDEX
    Explanations

    legal boilerplate and section headers

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

    The neuron fires on tokens characteristic of encyclopedic fact statements—e.g. country names, building‐related terms, and date or numeric references.

    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
     wypeł
    -0.75
     szé
    -0.69
    坏
    -0.68
    呈
    -0.67
    environment
    -0.66
    estart
    -0.66
     merit
    -0.66
     old
    -0.65
     discre
    -0.64
    сная
    -0.64
    POSITIVE LOGITS
     proie
    0.72
    흑
    0.72
     Logitech
    0.71
     Millie
    0.70
    Bid
    0.69
    gande
    0.68
    ::{
    0.68
     Pathfinder
    0.68
     Epson
    0.67
    Sharon
    0.66
    Activations Density 0.043%

    No Known Activations