© 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. 39937
    Prev
    Next
    INDEX
    Explanations

    numbers, specific entities, and technical terms

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

    This neuron fires on the main content-bearing words—especially long, domain‐specific nouns, titles, and proper names that introduce each text segment.

    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
     in
    -1.21
     on
    -1.13
     when
    -1.11
     got
    -1.10
    These
    -1.05
     all
    -1.02
     as
    -1.00
     or
    -0.98
     взять
    -0.97
    家は
    -0.97
    POSITIVE LOGITS
     Sử
    1.20
     fta
    1.19
     foit
    1.12
     ivel
    1.10
     kto
    1.09
     vne
    1.09
     Thiết
    1.06
     handels
    1.06
     beft
    1.06
     luy
    1.05
    Activations Density 0.403%

    No Known Activations