© 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. 22-GEMMASCOPE-RES-131K
    7. 93113
    Prev
    Next
    INDEX
    Explanations

    awards and voting

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

    The neuron detects four-digit numeric tokens (i.e. years or similar “YYYY” numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     потім
    -1.36
     then
    -1.16
    āciju
    -1.09
     teneur
    -1.07
    ✔️
    -1.06
     简介
    -1.05
     Painters
    -1.05
     cuillère
    -1.00
     ceramica
    -1.00
    有一些
    -1.00
    POSITIVE LOGITS
     individual
    1.16
     special
    1.14
     Best
    1.09
    ɓ
    1.07
     '/';
    1.06
     citizen
    1.05
    agan
    1.05
    vements
    1.03
    στημα
    1.02
     person
    1.01
    Activations Density 0.012%

    No Known Activations