© 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. 22-GEMMASCOPE-RES-131K
    4. 53939
    Prev
    Next
    INDEX
    Explanations

    patent application date

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

    This neuron responds to content-bearing nouns (especially technical or domain-specific identifiers) rather than to common function words.

    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
     dieselbe
    -1.24
     is
    -1.23
    犼
    -1.09
    より
    -1.07
     prestar
    -1.07
     проблемы
    -1.02
     penultimate
    -1.00
     pf
    -1.00
    阴影
    -1.00
    薅
    -0.99
    POSITIVE LOGITS
    secta
    1.52
    what
    1.47
     جيدة
    1.38
    /
    1.34
    るので
    1.31
    Both
    1.30
    Revenir
    1.29
     például
    1.28
    if
    1.27
     or
    1.27
    Activations Density 0.116%

    No Known Activations