© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 25937
    Prev
    Next
    INDEX
    Explanations

    Kubernetes or blog posts

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

    The neuron spots structural or meta-text markers—things like section headings, comment keywords, hyphens in names, and other formatting/metadata tokens.

    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
     more
    -1.28
    .」
    -1.20
     they
    -1.16
    }){
    -1.06
     quattro
    -1.06
     */,
    -1.05
    ㎏
    -1.04
     not
    -1.02
     this
    -1.02
    îtra
    -1.02
    POSITIVE LOGITS
     рецепта
    1.19
     dítě
    1.16
     nabíz
    1.15
     člov
    1.12
    mphony
    1.09
     nhung
    1.09
     Newsweek
    1.09
     reflexiones
    1.06
    Hướng
    1.06
     skvě
    1.02
    Activations Density 0.000%

    No Known Activations