© 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. 36615
    Prev
    Next
    INDEX
    Explanations

    or and either

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

    The neuron strongly activates on numeric and quantitative tokens (including digits, decimal numbers) and the coordinating term “or,” i.e. it’s detecting numerical expressions and their alternative‐linking conjunction in technical text.

    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
     acquistare
    -1.12
     fasc
    -1.01
     上海
    -0.98
    gc
    -0.96
    ListNode
    -0.96
    tc
    -0.96
     provavelmente
    -0.95
     ↓,
    -0.95
     acho
    -0.95
     Mondial
    -0.94
    POSITIVE LOGITS
     or
    3.63
     или
    2.14
    หรือ
    2.00
    或
    1.74
     یا
    1.69
    または
    1.55
     hoặc
    1.54
    /
    1.40
     alebo
    1.38
     eller
    1.33
    Activations Density 0.143%

    No Known Activations